The Structure of a Battle

Analysis of a UN-NK Action

North of Taegu, Korea,
September 1950
Op'ertlting UndBr
Cot'Jtract with the
SSIFIE
The contents 01 ORO publica,ions, including the conclu
sians and recommendations, represent the views of ORO
and should not be considered as hoviny official Depart
ment of the Army approval, either expressed or,
PRIVILEGED COMMUNICATION
The information disclosed in this document con,tltute. proprie
tary information owned by the Operations Research Office of
The Johns Hopkins University, and is not to be used by or for
any party other than the United St.'es Government without
Ihe written consent of Ihe owner.
Any Government employee who improperly divulge, Ihi,
materi.1 is subject to Ih" criminal penalty under 18 U. S. C.,
S.ction 1905 (Septem!>er 1, 1948).

..

Ill,
r.U

C!
&1:11:
A.
.",...-..1
This document contains information affeding the noHonal
defense of the United States within the meaning of the
Espionoge laws, Tille 18, U. S. C., Sections 793 and 794.
The transmission or the revelation of its contents in any
manner to an unauthorized person is prohihited by law.
-SHIrr
1J.s
A
DUY MILITARY HISTORY INSTITI:JTE'"'l''''' rc ','," '''','"f "D
"'YO U " " j"..tIo.:.:...J. 1
- t .....
DEPART,\!ENT OF THE A.llEY
OFFICE OF THE CHIEF OF STAFF
";ashington 25, D. C.
CSPJljE (31 hn 55)
4 :hrch 1955
SUBJECT: Technical I:iemorandum ORO-T-261, uThe Structure of a i3attle. II
1. Forwarded herewith for your information and retention is a copy
of ORC-T-261, subject as above.
2. This study is a working paper of the Operations Research Office.
It was prepared by the Operations Field Office in Army Forces, Far East,
Eighth Army, and is not a part of the FY 1955 ORO Y:ork Program. It is
eonsidered that this study win be of value to the Service School System.
3. The study contains a Tery intensive analysis of a UN-M, action
north of Taegu, Eorea, September 1950. It endeavors to determine whether
oro can uncover a pattern of battle from which it may be possible to
define in measure the part played by certain factors and component
relationships, such as terrain, casualties, and fire power, and the cause
effGct mechanism by which a battle progresses. '[Cherever possible data
have been quantified with a view toward the development of computer
methods and "mr-gaming techniques of operations analysis.
FOR THE CEI3F OF !tESEARCH Ai'lD DEVELOPMENT:
/;: /. ' . "';;/:1( J /I' j' :" ,- f
./.:',.1.., " .!-.t__
1 Incl ," IiE:RBERT ''r:. i,iANSFIELD .'
ORO-T-261 GS
Chief, Operations and Personnel
Research Division
ilcgNded iJ,)CT ,:,;SIFI;,.r;
",hen sepnrated fro"Tl
Classified Inclosures
PROPERlY Of US ARMY
USAADS Library
'Fort Bliss, Texas 79916
. -. ,
I
, ' .
L
-. .,,-
. ,-
UNe
TIDS IS A WORKING PAPER
Presenting the considered results of study
by the ORO staff members responsible for
its preparation. The findings and analysis
are subject to revision as may be required
by new facts or by modification of basic
assumptions. Comments and criticism of
the contents are invited. Remarks should
be addressed to:
The Director
Operations Research Office
The Johns Hopkins University
7100 Connecticut Avenue
Chevy Chase, Md.
Washington 15, D. C.
USAADS LIBJI!AD
I. 2
blla... fa. 1.9916

('.,\ r\; .... ,
i H "Pt. ,.
.IET'
UNCLASSIFIED
IProject FECOM j.C 'j2t':U .' .. ;; ..
Technical Memorandum ORO-T-261 ,.' : .. .:\..u
t
Received: 20 January 1954 -" ._. -' ..
The Structure of a Battle
Analysis of a UN-NK Action
North of Taegu, Korea,
September 1950
by
RODert J. Best
18711.

rop-ert'! '
. Ottice tho Chiot P
Fort. J1:2.5::3, Tx. 79916
OPERATIONS RESEARCH OFFICE
The Johns Hopkins University Chevy Chase, Maryland
SHRn
UNCLASSIFIED
Published
January 1955
by
OPERATIONS RESEARCH OFFICE
7100 Connecticut Avenue
Chevy Chaso, Md.
Washington 15, D. C.
UNCLASSIFIED
SlERET
UCREt' UNCLASSIFIED
CONTENTS
SUMMARY
PROBLEM-FACTS-DIscuSSION -CoNCLUSIONS
Page
1
INTRODUCTION
5
PART I-NAIlRATIVE DIARY OF THE BATTLE
NARRATIVE DIARY FROM FRIENDLY ACCOUNTS OF THE
SECTOR NORTH OF TAEGU, SEPTEMBER 1950
DAY-BY-DAY ACCOUNTS FOR 1-21 SEPTEMBER
BATTLE IN THE
13
PART II-ENEMY AND FRIENDLY TROOPS AND SUPPORT
ENEMY UNITS
GENERAL SITUATION AND ENEMY INTENTIONS-COMPOSITION AND CoNDITION OF
UNITS AT MID-SEPTEMBER-lpENTITY AND MOVEMENT OF PARTICULAR UNITS
STRENGTRS, flEPLACEMENTS, AND CASUALTIES IN THE 13TH NORTH KOREAN
DIVISION-STRENGTR AND CAsUALTIES ON HILL MASS 314
47
FRIENDLY STRENGTHS AND CASUALTIES, AND COMPARISON WITH
. THOSE OF THE ENEMY
CAsUALTIES IN THE ENEMY'S INITIAL DRIVE-ANALYSIS OF STRENGTHS AND
CAsUALTIES 6-16 SEP 50-CoMPARISON OF FRIENDLY AND ENEMY CASUALTIES
IN THE ENEMY'S INITIAL DRIVE
78
"
\j
ENEMY SUPPORT
WEAPONS ON AND IN THE VICINITY OF HILL MASS 314-SuPPORTING ARMOR
FRIENDLY SUPPOflT
ANTILLERY IN SUPPORT-ARMOR IN SUPPORT-SUPPORT FROM THE AIR
89
liS
PART III-SYNTHESIS AND INTERPRETATION
SYNTHESIS: THE TACTICAL PATTERN IN GRAPHIC OUTLINE
DAY-BY-DAY DIAGRAMS AND HOUR-By-HoUR ACCOUNTS OF THE ACTION FOR
2-16 SEPTEMBER
143
v
SIeREl
UNCLASSIFIED
UNCLASSIFIED sUln
CONTENTS {continued)
Page
INTERPRETATION: THE BATTLE NORTH OF TAEGU IN
SEPTEMBER 1950
178
GENERAL SITUATION AND THE loCAL SETTING-CoMPOSITION OF THE OPPOSING
FORCES-CoURSE OF THE BATTLE-ESSENTIAL FACTORS DETERMINING THE
CoURSE OF THE BATTLE
pART IV-THE KEY ACTION
NARRATIVE OF THE CAPTURE OF HILL MASS 314 BY THE 3D BATTALION,
7TH CAVALRY REGIMENT 195
BACKGROUND OF THE BATTALION-PREPARATORY PRASES OF THE ATTACK-
AsSAULTS ON THE OBJECTIVE AND ITS CAPTURE-CoMMUNtCATION DURING CoMBAT
SYNTHESIS: GRAPHIC TIMETABLE OF ACTION AT THE OBJECTIVE
ON HILL MASS 314 219
ACCOUNTS OF THE ACTION FOR 1135-1530
FRIENDLY CASUALTIES AND THE EFFECTIVENESS OF ENEMY MORTAR
FIRE IN THE CAPTURE OF HILL MASS 314 247
FRIENDLY CAsUALTIES-THE EFFECTIVENESS OF ENEMY MORTAR FIRE
INTERPRETATION: THE CAPTURE OF HILL MASS 314 276
THE SITUATION AND THE SETTING-CoMPOSITION OF THE OPPOSING FORCES
CoURSE OF THE AcTION
APPENDICES
A. EXTRACTS FROM REPORTS OF INTERROGATION OF PRISONERS
FROM THE 13TH AND 1ST NORTH KOREAN DIVISIONS (II ITEMS) 289
B. ATTACK ORDER OF II SEP 50 FOR THE 3D NORTH KOREAN
DIVISION (OPPOSING THE 5TH US CAVALRY REGIMENT) 309
C. PRISONER-OF-WAR INFORMATION ON ENEMY USE OF MORTARS
AND ARTILLERY 313
D. FIllENDLY STRENGTHS AND BATTLE CASUALTIES IN THE SECTOR
NORTH OF TAEGU 319
E. SUPPLEMENTARY FRIENDLY DOCUMENTATION ON THE CAPTURE
OF HILL MASS 314 337
F. THE ACTION NORTH OF TABU-DONG IN AUGUST 1950 AS VIEWED
FROM BOTH SIDES 361
G. ENEMY AR\IOR 365
REFERENCES 287
FIGURES
l. OvER-ALL UN DEFENSE PERIMETER AS OF 9 SEP 50 9
2. FRONT LINE NORTH OF TAEGU IN THE 7TH AND 8TH CAv AREAS, 12SEP50 15
3. PANORAMIC VIEW OF HILL 314 FROM Mosr FORWARD ENEMY PoSITIONS ON
HILL 570 TO WEsr 28
4. VIEW UP TilE "IJoWLING ALLEY' FROM PoINT Jusr t\\)RTIIWEsr OF TAEGU 49
vi
UNCLASSIFIED
SHRfT
UNCLASSIFIED
S'[(REI""
CONTENTS (continued)
Page
5. ORGANIZATION OF THE l'klRTH KOREAN RIFLE REGIMENT AND. IN PARTICULAR.
OF THE Isr NK DIV. IN SEPTEMBER 1950 53
6. DAILY STRENGTH OF FRIENDLY INFANTRY AND CUMULATIVE TOTAL FRIENDLY
BATTLE CASUALTIES IN 8TH CAv ZoNE 81
7. LINEUP AND STRENGTHS OF OPPOSING FORCES ASTRIDE THE HIGHWAY.
1 AND 6 SEP 50 87
8. ENEMY WEAPONS AND FRIENDLY ARTILLERY CONCENTRATIONS 97
9. ENEMY ARMOR FOUND AFTER THE BATTLE 108
10. KNOCKED'()UT TANKS AND SP GUNS IN THE ALLEY" l'klRTH
OF T AHU-DONG III
11. TRANSVERSE COVERAGE OF VT-FUZED FIRE ON THE OBJECfIVE 120
12. SITUATION AROUND HILL MASS 314 AT 1iME OF ATTACK BY 3D BN.
7TH CAv REGT ON 12 SEP 50 197
13. CoMMUNICATION ON HILL MASS 314 DURING AssAULTS ON KNOB 2 216
14. ACCUMULATION OF CASUALTIES IN THE 3D BN. 7TH CAY REGT DURING THE
ACTION ON HILL MAss 314. 12 SEP 50 257
15. CoMPARISON OF ACCUMULATION OF CASUALTIES FROM ENEMY SUPPORTING
WEAPONS \\11TH ACCUMULATION IN CLOSE CoMBAT FROM BULLETS AND
GRENADES. 3D BN. 12 SEP 50 262
16. ACCUMULATION OF CASUALTIES FROM MORTAR FRAGMENTS AND FRAGMENTS
OF UNKNOWN TYPE IN THE 3D BN. 7TH CAY REGT IN THE CAPTURE OF HILL
MASS 314. 12 SEP 50 263
17. CUMULATIVE CAsUALTIES DIFFERENTIATED AS TO RANK. Co L AND Co I 265
18. RELATIVE DENSITY OF 120-MM MORTAR F:RAGMENTS IN VERTICAL PLANE
OF TRAJECTORY FALLING ON EAST SIDE OF KNOB 2 269
TABLES
1. PROBABLE LOCATION OF ENEMY UNITS IN VICINITY OF HILL MASS 314
ON MORNING OF 12 SEP 50 55
2. ENEMY DOCUMENT SHOWING ACTUAL STRENGTHS OF 13TH NK DIV UNITS
31 AUG 50 69
3. ENEMY DOCUMENT SHOWING ACTUAL STRENGTHS OF 13TH NK DIV UNITS
11 SEP 50 70
4. ENEMY DOCUMENT DETAILING RECEIPT AND ASSIGNMENT OF REPLACEMENTS
IN THE 13TH NK DIV 71
5. STRENGTH LEVELS OF THE 13TH NK DIV 31 AUG-ll SEP 50 74
6. STRENGTH LEVELS OF THE 13TH NK DIV FROM FIRST COMMITMENT TO
11 SEP 50 75
7. RELATION OF FRIENDLY CASUALTIES TO THE TACTICAL SITUATION 82
8. CASUALTY PERCENTAGES IN FRIENDLY COMPANIES IN RELATION TO OUT
COME OF OFFENSIVE AND DEFENSIVE HILL ACTIONS 84
9. NUMBERS OF CERTAIN WEAPONS IN MAJOR UNITS OF THE 13TH NK DIV
11 SEP 50 91
10. PARTIAL LISTING OF WEAPONS CAPTURED OR RECOVERED ON HILL MASS
314 101
11. AIR STRIKES ON HILL 570 ON 11 SEPTEMBER 133
12. AIR STRIKES ON HILL MASS 314 ON 11 AND 12 SEPTEMBER 136
13. CASUALTIES IN THE 3D BN. 7TH CAV REGT IN THE CAPTURE OF HILL
314. 12 SEP 50 250
vii
UNCLASSIFIED
.sECRET
UNCLASSIFIED
CONTENTS (continued)
Page
TABLES (continued)
14. FRIENDLY STRENGTHS, CASUALTIES, AND CASUALTY RATES IN THE CAPTURE
OF HILL MASS 314 BY THE 3D BN. 7TH CAY REGT. 12 SEP 50 250
15. BALANCE SHEET OF ROKA PERSONNEL ATTACHED TO THE 3D BN. 7TH
CAY REGT 251
16. EXAMPLES OF ASSIGNMENT OF ATTACHED ROKA PERSlNNEL IN THE 7TH
AND 8TH CAY REGTS 252
17. ROUGH OUTLINE OF THE NUMBER OF US AND ATTACHED ROKA PERSONNEL
IN THE 3D BN. 7TH CAY REGT 253
18. FREQUENCY OF AVAILABLE DATA ON TIME AND TYPE OF MISSILE FOR
CASUALTIES IN THE 3D BN. 7TH CAV REGT IN THE CAPTURE OF HILL MASS
314. 12 SEP 50 256
19. CASUALTIES IN THE 3D BN. 7TH CAY REGT IN THE CAPTURE OF HILL MASS
314. '12 SEP 50, BY TIME AND COMPANY 258
20. COMPARISON OF ACCUMULATION OF CASUALTIES FROM ENEMY SUPPORTING
WEAPONS WITH ACCUMULATION IN CLOSE COMBAT FROM BULLETS AND
GRENADES. 3D BN. 12 SEP 50 261
21. ACCUMULATION WITH TIME OF CASUALTIES FROM MORTAR FRAGMENTS AND
FRAGMENTS OF UNKNOWN TYPE IN THE COMPANIES OF THE 3D BN. 7TH
CAY REGT IN THE CAPTURE OF HILL MASS 314, 12 SEP 50 264
22. CASUALTIES BY RANK IN THE CAPTURE OF HILL MASS 314 265
23. BODILY D,STR,BUTION OF HITS BY MORTAR FRAGMENTS AND FRAGMENTS
OF UNKNOWN TYPE 270
24. RELATION OF BODILY DISTRIBUTION OF HITS BY MORTAR FRAGMENTS AND
FRAGMENTS OF UNKNOWN TYPE TO TOTAL BODY SURFACE AND TOTAL HITS 270
25. ESTIMATED ENEMY EXPENDITURE OF 120-MM MORTAR ROUNDS ON HILL
MASS 314 ON 12 SEP 50 274
UNCLASSIFIED
viii
SURET
0
UNCLASSIFIED
SUMMARY
PROBLEM
To demonstrate that battle, or at least one battle, has a pattern which re
search can uncover: to this end, to determine by exhaustive analysis of all
data obtainable the lineup of troops, the tactics of each contender, and the
course of events; finally, to deduce, and so far as possible define and measure,
the part played by certain factors and component relations, such as terrain,
the occurrence of casualties, and the effectiveness of particular fires.
FACTS
The subject of this research is a segment of the battle between United
Nations and North Korean troops in Korea. The engagement described in this
memorandum occurred north of Taegu, chiefly on and around the hill masses
designated on US Army maps as Hills 570, 314, and 660. The period of time is
from late August to late September 1950. 'The UN troops involved were mainly
from the US 1st Cavalry Division and the 1st ROK Division; the North Koreans
mainly from the 1st, 3d, 13th, and 15th NK Divisions. Particular attention is
given to what is considered to be the key action of the conflict-the capture of
Hill 'Mass 314 by the 3d Battalion of the 7th Cavalry Regiment, 1st Calvalry
Division.
DISCUSSION
The Course of battle may be determined by mechanisms of action beyond
those built into the organizations and tactical doctrine of the contending forces.
Tactical planning should embody as many of such relations as can be shown to
be predictable. The only field for relevant research is actual conflict. In this
memorandum all data obtainable on friendly and enemy movements, troops,
and support in one battle are analyzed.
Although this study is historical in method, it deals not with an entire
campaign but with a limited "vertical" slice of combat analyzed intensively.
To aid in the effort to determine whether 'or not this battle had a pattern re
course was made not only to the usual sources of information (command re
ports, war diaries, unit journals, message logs, personal testimony, etc.), but
ORO-T-261
1
UNCLASSIFIED
HuET
UNCLASSIFIED
: SUMMARY
also to tenuous information which might otherwise have been rejected. The
pattern offered is therefore not supported by a body of absolutely established
fact, but it is claimed that it is neither demonstrably inconsistent nor replace
able by a simpler alternate pattern.
CONCLUSIONS
Since wars are composed of many battles, and since battles may differ
greatly, no generalized conclusions are drawn from this study of one such
action. It is felt, however, that the consistent web of facts describing this
battle may be useful as: (a) an example to test a general model of ground war
fare; (b) a preliminary indication of key relations in the cause-effect mechan
isms by which a battle progresses; (c) a basis for defining fundamental meas
ures of effectiveness; and (d) an illustration of the conditional nature of military
action at all levels.
ORO-T-261
2
UNCLASSIFIED
SURET UNCLASSIFIED
THE STRUCTURE Of A BATTLE
$[ERET
UNCLASSIFIED
SHIET
UNCLASSIFIED
INTRODUCTION
Battle may follow its own laws. Its shape on situation maps may be merely
a silhouette of its real form. If chance is not its basis, it must conform to
mechanisms of action beyond those built into the organizations and tactical doc
trines of the jOined contending forces. Certain interactions of setting, manpower,
fire, casualties, and other factors may determine the course of battle as much
as thewills and plans of the opposing sides. Future tactical planning should ac
count explicitly for as many of the essential elements and as zrluch of the total
character of battle as possible.
Composite battle can be found only in actual warfare. in their human
machine-situational complexity, past and present conflicts are without doubt
by far the closest approximations of future wars. Old battles analyzed into
their vital components can provide the bases and much of the material for the
maneuvers and war games to work out the tactics of an evolving weapons system.
The structure of battle should be most evident when it is viewed from the
side having the lesser scope of free action-the defense. To preserve its in
tegrity the defending force must in its unit dispositions impose a spatial pattern
upon the battle, however well or poorly these dispositions may suit the actual
situation. In defense, much more than in offense, a force is obliged to maintain
the internal order of its organization and the external order of its battle array.
Dispositions necessarily follow tactical plan and are therefore almost inevitably
somewhat outdated. Since an adequate plan anticipates all eventualities, and to
some extent provides for them, the defending force must be heavily and various
ly committed. When the battle develops on enemy initiative, the defense be
comes only partially engaged; it is inevitable that only a small fraction of the
total fire power avatlable can be concentrated on the point of attack selected
by the enemy. Defensive battle exhibits the military machine committed,
stressed, and strained. It is in this condition that one should expect to find a
significant s'uccession of critical points and crucial moments.
The Subject Battle
This study is an operational research which has as a subject an entire
actual operation, a battle in which UN ground forces, though initially (later
almost desperately) on the defensive, ultimately won the victory. The study's
first purpose is to show that at least this battle had a pattern. To this end,
there is an attempt to determine by exhaustive analysis of all data obtainable
the lineup and resources of each side, the tactics employed, and the course of
events. Secondly, there is an attempt to deduce, and so far as possible define
ORO-T-261
UNCLASSIFIED
seekEr
5
SiEMT
and measure, the part played by certain factors and component relations, such
as terrain, the coordination of supporting arms, the occurrence of casualties,
and the effectiveness of particular fires. Since wars are composed of many
battles, and since battles may differ greatly, no generalized conclusions are
drawn. It Is believed, however, that an essential point is implicitly demon
strated: analytical treatment of separate aspects of combat !;ill! be related to
the over-all course of battle, provided that other major aspects and the flux
of whole chains of contingencies are taken Into account. The conditional sig
nificance of small actions is one of various typical conditions of warfare often
neglected In theoretical studies and illustrated In the subject battle.
From several points of intense conflict along the UN defense perimeter
in southeastern Korea in September 1950, that battle which included the par
ticularly threatening enemy penetration north of Taegu is selected for study.
The course of events is traced from the initial frontal attack upon the friendly
line through the following friendly counterattack and withdrawal, through a
succeeding period In which the opposing forces again came to grips, and, fi
nally, to the decisive phase of determined thrust and counterthrust at the new
line of defense which resulted In the turning of the tide. The key action, In
which friendly forces seized (and were then in position to hold) the tactical
initiative, was the capture of Hill Mass 314 on 12 September by the 3d Battal
ion, 7th Cavalry Regiment. This engagement is subjected to thorough analysis.
The particular aspects of combat studied in this key action are conSidered to
be fairly typical for the over-all battle.
The battle chosen for study is an unusually promising subject for analyt
ical dissection. The opposing forces were small and fully committed. The de
fense could not meet attacks simply by committing reserves, since they were
usually not available. It had to react with forces already deployed. As neither
side was endowed with great resources, critical action was not obscured by
waves of reinforcements and volumes of excess fire. For the observer and
analyst the confusion of several arms all brought to bear on the same target
was much less than normal. This Korean battle therefore was typical neither
of the grand engagements of the past nor of probable future large-scale con
fUcts, but if actual battle can ever be reduced to analytical terms, it should be
possible here.
Method and Sources
The study for this memorandum proceeded from the particular to the gen
eral. At the outset, all the available detailed data for both sides were analyzed
separately. The facts which emerged after further checking and correction were
then fitted together In a synthetic process of interpretation and generalization
to build up the final reconstruction. The process terminated with the elucida
tion of the essential factors that determined the course of the fighting.
There was an attempt to make the coverage of formal recorded material
especially thorough, since secondary but basic purposes of this study were to
determine the extent to which military events can be established and measured,
and to evaluate the utility of various sources of information-documents, statis
tics, and interviews-In this establishment and measurement. Recourse was
also made to a few incidental sources, such as personal diaries and affadavits
ORO-T-261 6
UNCLASSIFIED
UNCLASSI FI ED
for awards. Also, all the photographic files of the Army Signal Corps were
screened for relevant pictures, and the major press services were consulted.
The only important recorded data not studied are thought to have been destroyed.*
These include the firing charts of the artillery battalions and the tables of ammu
nition expended by the infantry units. Such data would have enriched the findings
greatly, and the lack, of course, is in principle unnecessary. On the other hand,
this study incorporates captured enemy statistics not usually available.
The value and reliability of the various sources of information are not dis
cussed as such in this study. However, the origin of all the data used is stated
at the head of each section or subsection, and completeness and reliability are
in some cases pointed out.
One aspect of the action which is now thought to be of great importance
but has been considered only in passing is the communication of combat infor
mation. It has not been ascertained whether any pertinent records that might
have contributed to this battle analysis were kept.
The intent at the outset of this work was to make every part an entirely
independent building block of the whole, so that any derived facts or conclusions
would follow strictly from the particular sources consulted. However, it was
found impossible to proceed in such a manner, one reason being that no one
source ever reported completely even a small matter, and was hardly ever
completely correct. Very often, seemingly objective facts had to be altered
somewhat when checked in a wider context. Throughout the text, such altera
tions of quoted data are indicated by comments in brackets. The process of
study, then, was one of continual feedback and readjustment.
Reliability' and Use of the Study
This study is historical in method but differs from most military history
in scope and detail. Instead of an entire campaign, the subject is a limited
"vertical" slice of combat; this slice, however, is analyzed intensively. The
battle is assumed to have a pattern which research can uncover. To define
and sketch in the pattern, use is made not only of information from the sources
usually employed-historical documents and personal testimony-but also of
both physical and numerical data not ordinarily available, and the findings of
operations research field teams, whose specialized studies included the sub
ject battle (among other areas). Where no better data exist to bring events into
intelligible relation, recourse is made to tenuous information which might other
.. wise be rejected; in some cases purely arbitrary assumptions of cause and ef
rect are introduced. The resulting pattern therefore is not entirely supported
by objective eVidence, but it should neither be demonstrably inconsistent nor
replaceable by a simpler alternate pattern.
This memorandum is not concerned with personalities, although it is rec
ognized that the mind and character of individuals are intimately involved in
every phase of battle. Personal factors in the higher command echelons are
a familiar subject of historical study. The writer feels that they are important
.At the writer's request,the Far I<..:ast Command made an exhaustive st:arch for captured maps and docu
ments of the 1st and 13th NK Divs, reports of interrogations of prisoners of war cruTied out within the 1st
Cay Div, firing charts and records of the 9th, 99th, and 77th FA Bus, and records of ammunition consumption
and' expenditure by particular infantry units for the period covered by this study-all, as of summer 1953,
with negative results.
ORO-T-261
sultn
UNCLASSIFIED
7

at all levels but that no Individual has more than very limited freedom to im
pose a new departure on the course of events within his sphere of Influence,
especially in such an action as the subject defensive battle. Usually the ma
terial situation itself-as appreciated-determines at once the nature of the
responses deemed appropriate. It is in part in the choice of response, but far
more in the adequacy of response, that Individual make-up and Interpersonal
relations are of influence. The intention of this study is primarily to relate the
events that actually occurred and deduce their self-determinlng continuity, and
only implicitly to suggest the adequacy of a scheme of operations, the coordina
tion of an attack, or the defense of a position. As the men involved are consid
ered in terms of function rather than indiViduality, they are as a rule designated
by military title rather than by name.*
The consistent web of facts describing the subject battle may be useful as
a basis to test a general model of ground warfare. The key factors and relations
appearing In this battle may well serve as a basis for further research and study,
since they may also be important In other battles. The placing of small actions
and particular concentrations of fire in the context of over-all battle may serve
as a suggestive basis for defining some fundamental measures usually applied
and summed over an entire battle. Certain parts of the memorandum, such as
the relation of battle casualties to the outcome of an action, or the effectiveness
of enemy mortars in a specified tactical situation, may, withln their historical
and military limitations, have utility as data, since detailed information on such
points is otherwise almost totally lacking.
,..It is planned to append to the final version of this study a list of names of the men prominently involved
in the battie, those consulter! for Lasic data, and those who criticized the draft.
8 ORO-T-261
UNCLASSIFIED
SICAH
I
Q3BISSV1:lNn
SEl'T
Fig. I_Over_oll UN defense perimeter
os of 9 Sep 50.
The UN perimeter is indieoted by the heavy
broken line. Arrows denote enemy thrusts;
lhe eird.. enclose. area of ...emy penetra
tion in the 8th COY oreo, Doto from FECOM
DOily Intelligenee SummClry 2923.
ORO-T-261
UNCLASSIFIED
9
UNCLASSIFIED
rart I
NARHATIVE DIAHY OF THE BATTLE
UNCLASSIFIED
SlERE1
UNCLASSIFIED
NARRATIVE DIARY FROM FRIENDLY ACCOUNTS OF THE
BATTLE IN THE SECTOR NORTH OF TAEGU, SEPTEMBER 1950
This part is a day-by-day account of the action of friendly units in the
regimental sector due north of Taegu during the period 1 - 21 Sep 50. The
fighting so delimited in space and time had a clean-cut beginning with the
launching of the enemy offensive on Tabu-dong; it ended cleanly with decisive
enemy defeat in the same area. The intervening enemy advance was stopped
short of Taegu at Hills 570,314, and 660, and then the tide slowly receded. The
narrative below is concerned chiefly with the subject sector and in particular
with engagements on these hills, but it also gives a running summary of the
evolving tactical situation in the adjoining areas.
The enemy attacks north of Tabu-dong prior to the period studied are of
present interest chiefly because of the enemy's use and loss of virtually all his
armor. These actions are described in Part II under "Supporting Armor" and
are summarized in App F.
In this part events are reported as seen at the time through many friendly
eyes and as recorded in the command reports, war diaries, journals, and mes
sage logs of the units involved.'-24* Such records are the usual documentary
sources of battle data. The accounts which follow necessarily reflect the in
completeness and emphases of the sources. It would be impractical to give the
authority for every statement, but important basic reports are often quoted,
especially in cases of disagreement, and the source is,always indicated.
Up to 2400 on 9 September K-zone time was used. Clocks were then turned
back 1 hr to I-zone time, which was kept from then on. Sunrise on 12 September
was at 0547 I-time; sunset at 1821. Positions are given in terms of the lOOO-yd
coordinates of the World Polyconic Grid system in use at the time, and may be
located from the marginal scales in the basic map, Fig. 2, even though the cor
responding grid is not ruled. The position at any time of any friendly unit (com
pany) may readily be found in the diagrams of Part ill, whether or not the unit
is mentioned in the text which follows. The diagrams may also serve to keep
the over-all action in perspective.
*Heferences 1 to 24 represent all known formal reports pertinent to this study; several units apparently
did not submit command reports.
ORO-T-261
UNCLASSIFIED
13
UNCLASSIFIED
1 SEPTEMBER
The renewed enemy offensive against the UN perimeter began on the night
of 31 Aug -1 Sep 50 with massive attacks along the southern Naktong front in
the zones of the 2d and 25th US Inf Divs. With part of the 2d Div's 9th Regt
overrun and its 23d Regt penetrated as deeply as 8000 yd, the outlook was
grave. In addition to drastic measures to avert a breakthrough toward Pusan,'
a diversionary counterattack on the northern front was deemed necessary. The
Eighth Army commander issued an emergency order (dated 1045, 1 September)
to the commanding general of the 1st Cav Div and the chief of staff of the ROK
Army directing that their forces attack without delay to seize and secure the
following line:
1st Cay Div: 1138.7-1457.3 to 1139.3-1458.9 to 1138.9-1460.5 to 1137.9-1463.0 to
1138.7-1463.5 to 1143.1-1462.6 to 1144.3-1463.3 to 1146.8-1465.5 to
1149.8-1466.2 to 1150.4-1467.2 to 1151.2-1469.0 and
ROK forces: 1151.7-1470.2 to 1154.8-1471.5 to 1158.4-1475.7 to 1159.0-1477.0 to
1162.8-1481.4 to 1155.0-1482.0 to 1168.0-1480.5 to 1169.0-1480.6 to
1172.4-1482.5 to 1177.4-1483.5 to 1179.0-1483.3.
The line held by the 5th Cav Regt at that time, along the east bank of the
Naktong River, would require little or no adjustment. East from the Naktong
and hinged on it, the northern front would make a great swinging advance to
ward the northwest. The line of the 7th Cav Regt would move forward a vari
able distance, up to 4000 yd, and its continuation in the sector of the 8th Cav
Regt would advance up to 6000 yd. At the time the order was transmitted, the
11th Regt, 1st ROK Div had just completed relief of ROK units occupying posi
tions running northeast from the 8th Cav Regt right flank along the general line
1157-1462.5 to 1161-1467 (excluding Hill 527), and the 12th Regt, 1st ROK Div
was continuing to defend against light enemy pressure along the adjacent seg
ment of the front from 1161-1467 to 1165-1469.5. Further to the east, the 1st
Bn of the 7th ROK Regt and the 15th ROK Regt were continuing to counterattack
northward against enemy penetrations of the line from 1177 -1465 to 1180-1466.5.
It is then evident that the newly arrived 11th ROK Regt and the 12th ROK Regt
were called upon to advance approximately 9000 yd, the already counterattack
ing 15th ROK Regt even further.
The attack plans formulated by the 1st Cav Div provided initially only for
advance to that portion of the newly specifiec! line west of the point 1144.3-1463.3.
At 0900 on 2 September the 7th Cav Regt, using at least two battalions, was to
seize Hill 518 (1142.0-1461.5) and Hill 346 (1139.7-1461.8). The 8'th CavRegt
was'to support the attack with six of its 4.2-in. mortars and cover the 7th Cav
Regt right flank by advancing to Hill 290. (A hill of this designation does not
appear on the map and coordinates are not given, but the numeral is repeated
in many reports. Perhaps the 280-meters-plus knob 1800 yd southeast of Hill
518 at 1144.0-1460.7 was meant.) Subsequently, the 8th Cav Regt was to advance
to the line 1150-1465, 1152.5-1465, further east and not as far north as specified
in the Army order.
*The 1st Prov Mar Urig ..... as committed in the zone of the 2d Div. The two units comprising the anny
reserve, the 24th Inf Uiv and the 27th llrit Brig, were alerted but not committed here.
ORO-T-261 14
UNCLASSIFIED
SlCRH'
1167 1149 1150 1156 1157 1158 1159 1161 1162 1163 1164 1165
c1-:
1465 :":"-'; J r
"
$6 65 1- 1465
1-464 ,..
1464
1463 :--'
'- 1463
.--:.,
-,
1462

1462
" .:;i '<;><""",0",'<
1461 "
89
- 1461
"/".'0,'.

1460

1460
\,
1459
1458
, J
"
I
/
- 1458
\: 86, ,;',{
,'MCUN-S,""
1457 -/
C
1457
..... 1456
,01.
1455
1454
- 1454
1453 +)
'/
- 1453
/
..:. 1452
>,
- ' .. --,.. .. \
- 1449
-- 1448
1446
:.:-1444
Fig. 2-Front I ine north of T oegu in the 7th
and 8th Ca. Areas, 12 Sep SO. ,:-1439
Plot shows locations of: divisional and regimental
headquarters; the 99th FA Bn (10s.mm howitzers
..:.1438
supporting 8th Cay); the two batteries of the nth
FA Bn (105-mm howitzers supporting the two bat
talions of the 7th Cay); md one battery of the 9th

FA Bn (105mm howitzers attached to Diy Arty).
"'""
The 1000-yd grids are numbered; light lines are
1()()().m gri ds.
, . .-" ......1436
l' -'----'-- C
( ..
( , s''. 0...-..: '-,'
?q
,
r ....
_ t I}..i I r-,F
'1.1435
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168
UEitET
UNCLASSIFIED
The 8th Cay Regt, upon which attention centers in this part, moved up to
the Tabu-dong area on or about 29 Aug 50'to relieve the 1st ROK Div and 3d
Bn, 27th Ini Regt, which had remained in the area. The positions held by the
various companies of the 8th Cay Regt on 1 September were essentially those
shown in the diagram for 2 September in Part m. The high ground north of
and west of the highway was occupied by Co A and Co C; Co B was
spread along the ridge of the west to contact the 7th Cay Regt. Company G
blocked the highway by occupying the hills on either side. The remainder of
the 2d Bn was disposed to the east in an arc to tie in with the 11th ROK Regt.
The 3d Bn, 8th Cay Regt, the regimental reserve, was in position south of
Tabu-dong astride the highway. According to the regimental operations orders,
the Bth Cay Regt was at 70 percent 'of authorized strength, its morale was good,
and its combat effectiveness was 75 percent.
There was little action on 1 September, as both the 1st Cay Div and the
1st ROK Div prepared to attack. Far to the east, the 15th ROK Regt seized
Hwasan-dong (1178.5-1467). The 11th ROK Regt planned to take Hill 527 (1159.3
1464.2), but apparently did not. Most of the 8th Cay Regt contacts with the
enemy were on the regiment's left front, that is, in the area of the 1st Bn. As
early as 0445 on 31 August, the 1st Plat, Co C had used machine-gun fire and
grenades (both rifle and hand) to repulse an attack by an enemy force of unknown
size. During the morning of 1 September, enemy troops were seen digging in on
the high ground on the west side of the highway across from Co G positions on
Hill 448. Somewhat to the west, a disaffected enemy officer surrendered to
Co C at 2040. Four hours later, the ,preliminary information he supplied was
reported in the following telephone message:
020100. From 8-3. 8th Cay. Report on prisoner picked up at 1147.9-1464.9. A
South Korean who could speak English interrogated the prisoner picked up by the 1st
Platoon, C Company, 8th Cavalry, with the results as follows: the prisoner is an officer
of the I&R Platoon of the enemy forces, and he left the North Koreans to surrender at
the time he was captured this evening, 1 September. The prisoner stated that during
the day (1 September) groups of four or five North Koreans moved into the vicinity of
C Company's front, at which point apprOximately 400 North Koreans were to assemble.
This force when assembled had the mission of making an all-out attack supported by
one (1) tank. He further 'stated that he (the POW) thought th'e attack would come off
early in the morning (2 September) ....1*
The 8th Cay Hv Mort Co (see 2 diagram in Part III for positions)
fired continually from 0800 to 1930. In the evening, Co A and Co C exchanged
small-arms, fire with small groups of enemy. and Co G received 50 rd of enemy
mortar fire.
North of the physical boundary between US and ROK forces, but in the ROK
area of responsibility, a continuing enemy build-up was observed in a valley a
long the Kunwi - Tabu-dong road in the vicinity 1158.8-1470.5. Enemy support
ing weapons were observed to total 31 artillery pieces plus SP guns. This tar
get was bombed by friendly aircraft with 500-lb' fragmentation bombs. During
night of 1 - 2 September fire was again noted coming from enemy guns in
this area.
The subsequent (onnal interrogation report on this prisoner is reprol]uccIl in full us hem r ill Apr A.
ORO-T-261
UNCLASSIFIED
17
UNCLASSifiED
Along a secondary valley leading south from the principal enemy build-up
area, a patrol from Co E, 8th Cav drew fire in the afternoon. ROK forces re
ported later that there were two hundred enemy troops and four 120-mm mortars
still further south along this valley, across from their lines at 1158.0-1463.7.
The 11th ROK Regt.made plans to block further enemy movement down this
valley by taking Hill 527.
2 SEPTEMBER
Though the situation in the zone of the 2d Div remained serious, the enemy
penetrations had been contained and some of the lost ground retaken. The ab
sent 3d Bn (reinforced) of the 23d Regt, 2d Div was urgently needed by its still
intensely engaged parent unit. The battalion remained attached to the 1st Cav
Div and held positions on the division's left flank, along the Naktong River. In
this sector, continued crOSSing of the river and infiltration toward Taegu by
small groups of enemy were of persisting concern. To relieve the battalion, it
was planned to commit the current army reserve, the 27th Brit Brig and Co C
of the 73d Med Tk Bn. Though these units remained temporarily in position
some forty or fifty miles south-southwest of Taegu, they were attached to the
1st Cav Div effective 2400, 2 September in an order dated 1345, 2 September.
At the same time, the 19th Regt of the 24th Div, which had been reorganizing
after severe losses, was ordered to move up with attachments to 1169.9-1379.0
as the new army reserve, to prepare for possible commitment in the zone of the
2d or 25th Inf Div. The other two regiments of the 24th Div were already in the
line elsewhere.
, The only regular infantry reserve of the 1st Cav Div was the 16th Recon
Co, which had abandoned and burned its organic tanks in the August fighting.
This company was not fully available, Since several hundred enemy troops had
infiltrated on the division's left flank, and the company, with attached tanks of
the 70th Tk Bn, had been sent to positions south and west of Taegu to block any
attempt by these troops to reach the city. The 8th Engr Comb Bn, the division's
organic engineer unit, was the immediately available reserve, plus, of course,
those tanks of its armor attachment (the 70th Tk Bn) that had not yet been as
signed missions forward.
The 7th Cav Regt jumped off in the attack at 0900, as scheduled. Company
B, 8th Cav Regt had been ordered to cover the attack by occupying Hill 290, appar
ently 1800 yd southeast of Hill 518. Company B reported that it had occupied its
objective in the morning, without opposition. At 1230, the divisional command
ordered Co L, 8th Cav to make a similar move to Hill 400, which is at 1142.5
1464.0, 2200 yd north of Hill 518 and 5 miles west-northwest of Tabu-dong.
Company L, the 1st and 2d Plats of the 8th Cav Hv Mort Co (the 1st Plat for
direct support of the 7th Cav), and a section from Co M, 8th Cav moved out by
truck at 1410. At 1726 the task force reported that it was ready to jump off for
its objective on completion of the artillery preparation. The attack was sus
pended shortly after it began, however, when the enemy launched an offensive
against the 8th Cav itself.
On the 8th Cav front, the remaining 3d Plat of the Hv Mort Co fired con
tinually from 0600 until about 2300, when it was ordered to displace.
ORO-T-261 18
UNCLASSifiED
UNCLASSIFIED
The 52 ground-support plane strikes listed in the 1st Cav Div Tactical Air
Control Party {TACP} log on 2 September were all directed at Hills 346 and 518
in support of the 7th Cav Regt attack. No air strikes hit the enemy build-up op
posite the 8th Cav Regt.
Formal interrogation of the prisoner picked up by Co C on the evening of
1 September yielded fuller and somewhat different information than had first
obtained. He was the S-3 of the 19th Regt, 13th NK Div. According to him, the
23d and 21st NK Regts were located to the west of the highway, the remainder
of the 13th NK Div to the east; the 13th Div was to make a full-scale attack at
dusk {not dawn} on 2 September. At 1700 all friendly units were alerted about
the impending attack.
The enemy attack materialized as predicted, beginning at about 1815 with
mortar and artillery preparation. A first thrust straight down the highway was
stopped by Co' G. A second thrust appeared soon after down the ridge to the
east of the highway. This apparently was the attack later reported as follows:
031330. Spirit 2 [S-2, 99th FA BnJ. Heavy mortar and high velOCity fire coming
from small town at apprOximately 1156.1-1466.9. Mortar from 1155.5-1465.8. Enemy
closely followed barrage in attack. Enemy heavily loaded with grenades and small arms,
no packs. Enemy did not follow trails but attacked across country in waves. Enemy fol
lowed closely behind withdrawing troops, hurling grenades. No enemy small-arms fire
at that time. While attacking, enemy screamed loudly.3*
At 0320 on 3 September a special flight of two F4U's strafed and rocketed
enemy artillery at 1155.6-1469.8, and another strike was requested for 0700.
Enemy artillery was evidently well dispersed, as guns at 1155.8-1467.7 had been
observed firing at 2100 on 2 September.' The attacks mentioned above were first
thought to be by two enemy regiments. The estimate was later reduced to one
regiment.
One platoon of Co F was overrun. After an unsuccessful attempt to retake
the lost ground, the company was forced to withdraw to the right to the vicinity
of Hill 742, and then further to the 'rear. Company H also was hit. The platoon
of Co G that was on Hill 346, on the right side of the road, was driven off. Thus
the enemy pushed a wedge into the center of the 8th Cav line and threatened to
overrun the 3d Plat of the Hv Mort Co, the only heavy-mortar support available;
at 2335 the platoon was ordered to withdraw to the vicinity 1153.3-1455.6.
A different version of the same report was also recorded and is of interest because it gives the time of
the observation and, apparently incorrectly, the origin of the troops that made the attack, an important point
in later analysis.
"031340. From Spirit-A sergeant from the 99th FA Battalion while OP in Scrappy sector at 021830
observed the enemy attack and discloses the following facts of the attack. The enemy attacked immediately
after the enemy b8lTage of mortars ceased. The enemy came from a village at 1156.1-1466.9. They left their
packs and attacked cross country. Upon reaching Qur linea they used mostly hand grenades when engaging
in close combat with our men and seemed to be conserving A/W ammo. The North Koreans came hollering
and screaming when in the attack." 1
Unless the attack described was that against units of the 11th HOK Hegt on Ilill-SS8late on 2 September
or during the morning of 3 September, the report would seem to be in error in ascribing the origin of the
troops, rather than of the supporting fires, as at the stated village, for the following reasons: (a) the enemy
troops would have had to come 3 miles cross-country and climb 200 to 500 m to attack the 8th Cav; (b) this
report was logged out of chronological order more than an hour after it was received (i.e., after items dated
1600), whereas the log previously cited
3
is strictly chronological; and (c) this report has errors of typing,
etc. (which have been corrected), whereas that above does not.
ORO-T-261
UNCLASSIFIED
19
UNCLASSifiED
The ROK forces on the 8th Cav Regt right flank had not made the preven
tive attack on Hill 527; instead, their positions on nearby Hill 558 were the ob
jective of a third enemy thrust. This appeared at about 2300 but was repulsed.
Company E, 8th Cav, in position on theadjacent ridge, was not subjected to great
pressure, but was in a difficult situation with enemy penetrations on each side.
Apparently no attack was made on the left flank against the 1st Bn, 8th Cav, al
though Co A and Co C engaged in fire fights with probing enemy elements.
3 SEPTEMBER
On the night of 2 - 3 September, the enemy began a massive counterattack
against the attacking 7th Cav Regt. This was launched on the left; of the enemy
drive against the 8th Cav toward Tabu-dong. To the east, the enemy offensive
against the ROK Army also got under way. FollOwing the repulse of the attack
late on 2 September on units of the 11th ROK Regt occupying Hill 558, a new
push 3000 yd to the east, at Hill 655, began. This was against the 3d Bn, 11th
ROK Regt, at 0200 on 3 September. This attack was still under way at 0400,
but ROK units had at that time lost no ground. Some time later, the 11th ROK
Regt left flank was again attacked, and then Hill 558 fell to the enemy. The
earliest report states: 031205. Tel Message. From Shamrock [1st Cav Div
Arty]. The left battalion of the 11th ROK Regiment has been pushed off Hill
558, and have fallen back to the next high ground. 'E' Company of Scrappy fell
back to 'walled in' area, 1155-1459 grid square.
1
Somewhat later the G-3 Section, EUSAK, in Taegu, received the following
message: "1330. Message from KMAG. 2d [3d?] Battalion, 11th Regt ROK
attacked by estimated two enemy battalions and withdrew to pOSitions in vicinity
1159.2 -1461.5, which is better defensive ground. The attack is being contained
.21
Another report at 1335 stated that ROK forces had withdrawn somewhat
further, to 1159.5-1460.5.
Taken together, the following reports indicate that the enemy penetrated
straight ahead between Hills 558 and 665 from his observed build-up area at
1158.0-1463.7, bypassing ROK units still on the southern slopes of these two
hills:
031500. Tel Message. From Capt Kiernan. From Hill 538 (1160.8-1460.8) there
are friendly troops. They can see 400-500 enemy troops at 1158-1460 headed toward
"Walled City.'
1845. Talked to Capt Cunningham [KMAG, G-3 Section] at 1845, relative the gap
between tbe 11th ROK Reg! and 8th Cav reported to be existing at that time by 1st Cay
(Col Carlson). ROK reports that they believe E Company of 8th Cav to be at 1156.3
1460.2. If that is the case. the situation ~ h o u l d be under control, because the ROKs have
a company at 1157.2-1462.1 and a company at 1159.5-1461.2 and are attempting to make
contact with elements of Cav. Further, the ROK 11th Regt have requested an overlay of
Cay positions in order to determine their pOSitions in order to better effect lateral contact.
2t
Whether or not Co E, 8th Cav had ever moved as far east as the line 1156.3,
it had already withdrawn to the friendly rear at the time of the later report (see
below). The 11th ROK Regt may in fact have moved as far west as the line 1153.3.
ORO-T-261 20
UNCLASSifiED
.SEun
$ERn
UNCLASSIFIED
The EUSAK Periodic Operations Report No. 159, covering the period 1200-2400,
3 September, locates the 11th ROK Regt at 1153.3-1462.3 to 1156.2 -1462.8, with'
other elements from 1159.2-1459.6 to to 1159.2-1463 to 1160-1463
to 1160-1464 to 1160.9-1464.8 to 1161.3-1465.4 to 1161.6-1466.4. The central
element apparently moved back still further, for at 1950 ROK forces were re
ported on Hill 756 at 1158.2-1458.0, along the ridge east of the "Walled City."*
If these reports are correct, the. ROK regiment had been split, leaving a gap of
approximately' five thousand yards open all the way to the Walled City. How
ever, the left ROK element, which early in the day moved west, covering the
least precipitous approach, had been left isolated by the withdrawal of Co E,
8th Cav Regt, with which it would otherwise have been in contact; thus the gap
had been widened.
By daybreak, the situation in the 8th Cav sector was as follows: the enemy
thrust had split the 2d Bn and penetrated some 3000 yd to the friendly, second
line of defense; Co F had been scattered by the spearhead and had withdrawn
to the rear to reorganize; Co G had moved across the road toward the west and
joined Co A and Co C, which, while they held their main positions, had pulled
back somewhat. The enemy was then left in possession of the high ground on
both sides of the Taegu road almost as far south as Tabu-dong. Company E,
which was flanked both by this penetration and by the enemy advance against
the ROK forces, started to withdraw at 1000, and rejoined the 2d Bn at 1600.
Meanwhile, Co I and Co K of the 8th Cav Regt, Co A of the 8th Engr, and
the tanks of the 2d Plat, Co B, 70th Tk Bn, had moved up somewhat to man a
new line through Tabu-dong. Friendly infantry blocking the main enemy advance
then comprised Companies I, K, and H; also the Sv Co and Co A of the 8th Engr
Comb Bn. In addition, there were the two battalion headquarters companies;
most of Co M; the 3d and Hq Plats of the Hv Mort Co; and the Hq of Co B; 70th
Tk Bn, with the 1st and 3d Tk Plats. Batteries A and C, and presumably B, of
the 99th FA Bn, were not far to the rear along the Taegu road in the vicinity of
Hill 570. (The 9th FA Bn was to the rear of the 7th Cav Regt in the vicinity
1147.8-1450.8.) During the day, Co C, 8th Comb Engr was also attached as in
fantry, with the mission of occupying Hill 516, just west of Tabu-dong; they did
so, against some enemy resistance. The regimental CP was moved back during
the day from its location on the highway just south of Tabu-dong to 1153.2-1456.1,
2 1/2 miles to the rear. Companies B and L and attachments were ordered to
button up during the night of 3 - 4 September at their positions to the west. .
The enemy was quick to consolidate the penetration almost to Tabu-dong,
with consequent splitting off of the 1st Bn from the regiment. At 0900 enemy
troops were observed digging in on the hiU just east of Tabu-dong at ii50.3
1460.5, and at 0955 somewhat further to the east along the same ridge on Hill
466 (1152.0-1460.4). At 1130 enemy mortars were seen on Hill 673, which over
looks Tabu-dong from the northwest. Armor and artillery also were brought
forward. At 1030 a tank was spotted on the road within 2 miles of the villalte.
(the one tank reported by the just after his capture?), and later in-the
afternoon 9 "tanks" were observed about 1 1/4 miles from it. (Most or all of
The "Walled City of Kasan is not really a city but the crown of a mountain (Hill 902) enclosed by re
taining walls. It is of religious significance to the Koreans.
ORO-T-261
UNCLASSIFIED
21
UNCLASSIFIED
these either were SP guns or the hulks 'of previously destroyed armor.) Two
T34 tanks towing field pieces were reported knocked out by Co I as they ap
proached its position astride the road north of Tabu-dong, and an accompanying
column of troops was dispersed.
T.he enemy moved into the gap between the US and ROK forces. During
the day, several hundred enemy troops were seen moving toward the south and
east at 1157.5-1461.5 and 1158.5-1462.3. Some .enemy forces had evidently al
ready slipped through past the western rim of the Walled City well to the south,
for at 1445 Co K, 8th Cav was reported driven off Hill 570, which was to its
right rear. (The four coordinates reported located the hill mass without ques
tion, but perhaps the designation of the company is incorrect.) The next day
the Scty Plat reported that what was estimated to be an enemy regiment had
moved eastward through the gap toward ROK forces.
The controlling position of the 902 -m mountain crowned by the Walled
City was much appreciated. From the south, the I&R Platoon of the 8th Cav
Regt, with an attachment of 150 South Korean police, was sent up to maintain
contact with the enemy. It arrived there at about 1610, but by 1955 had with
drawn to the village of Namchang at the southern base of the mountain, after
enemy opposition had been encountered. Company D of the 8th Engr Bn was
attached to the 8th Cav Regt and at 0500 given the mission of occupying the
Walled City. Plans were made to reinforce it with Co E, 8th Cav Regt. Engi
neer Co D made a preliminary move up from Taegu but returned in the even
ing when its mission was postponed.
Since the only permanent infantry reserve of the 1st Cav Div was the 16th
Recon Co, two prOvisional rifle battalions were organized from headquarters
and service personnel of various divisional units and designated as Task Force
Allen (Brig Gen Allen, Asst Comdr of the 1st Cav Div). Though they were never
committed, they had the mission of being ready for movement to any part of the
division area within 30 min after being alerted. The composition of one of the
two battalions was:
Personnel Strength
Co A, 15th QM Bn 107
Co B, 70th Tk Bn 110
Co C, Band and Repl Co 101
Co D, 8th Engr Bn 167
485
In its move up to relieve the 3d Bn of the 23d Inf Regt, the 27th Brit Brig
completed its move into the 1st Cav reserve area at 1153-1442 on the north
side at Taegu at 2125. Though the brigade was not available to the cavalry
commander for general commitment, it was in position to block a drive from
the north.
4 SEPTEMBER
In the sector north of Taegu, the intention of the command of the 1st Cav
Div was to reestablish a continuous line of defense across the road at Tabu
dong and eastward along the ridge to include the Walled City, where contact
ORO-T-261 22
UNCLASSIFIED
UNCLASSIFIED
with the ROK Army would be made. Though the 8th Cav remained out of phys
ical contact with ROK forces, the regiment, according to one observer's report
at 0620, was still confident of closing the gap. From the ROKA side, closure
was to be effected by redeployment of the 15th ROK Regt on the left flank of the
11th ROK Regt, but the lead unit, the 3d Bn, 15th Regt was not expected to move
into position until 0430 on 5 September.
Commitment o(the engineer companies completed the line at Tabu-dong.
Here the enemy attack continued. At 0415, Co A, 8th Engr Bn repulsed an attack
against its positions on the ridge east of the village at the cost of seven casual
ties. Engineer Co C, on Hill 516 just west of Tabu-dong, was now subjected to
intense pressure and cut off. Covered from the west by the fire of the tanks
of the 2d Plat, Co B, 70th Tk Bn, it fought its way north onto Hill 673 to join
the 1st Bn, 8th Cav. Meanwhile, to block any further enemy advance south of
Tabu-dong, the 2d Bn, 8th Cay (H and Hq Companies?) had been ordered to
occupy the hill at 1150.2 -1458.2, and by dark on 3 September had done so with
out opposition.
The enemy continued to attack just east of Tabu-dong. Company I, 8th
Cav was pUshed off the high ground at 0800, and its four attempts to retake it
ended without success. Company A, 8th Engr Bn maintained its positions to
the right on the ridge at 1151.3-1460.4, and Co K and Co F, 8th Cav, remained
deployed further east through the grid square 1152-1460. During the morning,
Co B rejoined the 1st Bn, and Co L and its attachments returned to bolster the
line by replacing Co F. However, the positions of the 3d Bn became untenable
when the enemy gained an observation point on Hill 466 and was able to call
down observed artillery and mortar fire. The Tabu-dong line was then t e m p o ~
rarily abandoned. The battalion withdrew about a mile to the vicinity 1152.0
1459.0 and set up a perimeter defense. The units of the 2d Bn remaining in
the area were ordered to withdraw to tie in with the 3d Bn, but they stayed
through the night in the locations they then held.
The friendly operation to reoccupy the Walled City seems to have been
two-pronged. Around 1345, Co D, 8th Engr reported that it was engaged in a
fire fight about a half mile southwest of the Walled City while on the way up.
At 1550 the company reported that it had advanced to within 200 yd of the ob
jective, and at 1627 that forward elements were within the Walled City. It is
thought that Co E, 8th Cay, met its augmentation of 125 South Korean troobs at
the former regimental CP at 1100 and was moved by truck south around Hill
Mass 314 and then north to the village of Namchang, southeast of the Walled
City, at 1157.0-1457.2. Here, at 1530, Co E passed through the 1&R Plat and
moved up to the Walled City through a mortar and artillery barrage. It was
reported that this fire caused some of the inexperienced attached Korean
,troops to flee. At 1730 Co E was pinned down by small-arms and automatic
weapons fire and remained on the southern slope of the mountain during the
night.
At 1845 a message from the ROK II Corps reported that a prisoner had
stated that 800 enemy troops were in the Walled City and that 3 additional bat
talions were massed just to the north. The head of the enemy penetration was
said to be" at the village of Namchang. Friendiy troops further forward within
the walled area soon discovered that they were cut off.
ORO-T-261
UNCLASSIFIED
23
UNCLASSIFIED
SURE'1
To block a possible enemy attempt to penetrate far to the south from the
Walled City, the ROK 5th Spl Tng Bn, a unit composed of South Korean police,
was moved up from Taegu at 0220, 4 September to blocking positions at Cbilgok
(1153-1448). (This battalion apparently was maintained at nearly the strength
of US battalions, for its strength on 15 September was 630.)
5 SEPTEMBER
During the night of 4 - 5 September, the 27th Brit Brig moved from the 1st
Cav Div reserve area on the north side of Taegu at 1153-1442 to the division's
left flank, where it took up positions along the Naktong River, relieving the 3d
BCT of the 23d Inf Regt. Nevertheless, a considerable gap still remained be
tween the 2d and the 27th Brig.
In the 8th Cav Regt sector, the situation was reported as follows:
050530. Saber 3 - Situation report as of 050530. Blue in perimeter defense 500
yard radius in vicinity 1152.0-1459.0. White ordered to move out and establish line to
west of and in contact with Blue. (Orders to White via officer over. telephone.) R ~ d . no
change except B Company ordered to move to vicinity 1147.0-1461.0 and establish road
block. Tanks to begin patrol at dawn along MSR from Blue to junction of MSR and Burma
Road [Junction at 1153-1445]. All 8th Cavalry forces have been committed. Out of com
munication with White and Blue. Radio contact with Red. Battery A, 99th FA receiving
small arms fire from vicinity 1154.0-1455.5. Enemy infiltration as far south as 1154.0
1455.5. Report on D and E: E Company advanced through village and up hill toward
Walled City, receiving excellent artillery support. No small arms fire, some mortar
fire. Continues to move to contact D.3
Company B was to block the Waegwan-Tabu-dong road to protect the bat
talion's left flank. The Intention of the cavalry command apparently was still
to regain Hills 449 and 466, but any operations to this end necessarily presup
posed success In holding the Walled City. In the event that the situation should
deteriorate here, the two provisional battalions of Task Force Allan were alerted
to prepare to occupy defensive positions on the 8th Cav right flank.
At 0540 Co D, Engr was reported deployed along the wall at 1156.2 -1458.6,
with the enemy directly to the north. Company E, 8th Cav moved up from the
position it had held on the slope during the night and joined Co D at 0700. The,
two companies (and remaining 60 attached ROKA troops) were Intensely engaged,
and shortages of ammunition and other supplies became critical. Attempts were
made to resupply by airdrop, but, perhaps because of the bad visibility, all sup
plies landed In enemy-held territory. Plans to resupply the companies by an
armored convoy were made, but at 1330 the divisional command ordered the
companies to withdraw at once, and the operation became a rescue. The convoy,
made up of elements of the Sv Co and the Hv Mort Co, with tanks of Co B of the
70th Tk Bn, cleared the regimental CP area at 1430 and reached the base of Hill
902 at about 1500. By 1540 Co E was already coming down from the Walled City.
Because of severed communications, the message to withdraw was delayed In
reaching Co D until 1930, when the company had to fight out of the Walled City.
Company E had three officers and 99 men. Company D reverted to the control
of its parent engineer battalion.
24
ORO-T-261
UNCLASSIFIED
jlQHT
UNCLASSIFIED
Further enemy reinforcement from the rear was reported at 0945, when
an estimated 400 enemy troops were seen moving south along the Kunwi - Tabu
dong road, at grid square 1155-1469, in columns of 25 with a wide interval be
tween men. At the close of the day it was estimated that at least four enemy
battalions were in the Walled City.
In the Tabu-dong area, action slackened markedly as the enemy brought
up reinforcements. Small enemy elements slipped through gaps in the friendly
line and appeared to be trying to set up road blocks, but a force composed of
the 16th Recon Co and tanks of the 70th Tk Bn' was able to keep the main supply
route (MSR) clear. After attempts to bring up tank-towed artillery had failed,
on 4 September, the enemy resorted to manhandling. Throughout the day of 5
September, troops carried ammunition and hauled field pieces through the vil
lage to the south. As described in the subsection on friendly armor support, in
Part II, this movement provided perfect targets for the tanks of the 2d Plat,
Co B, 70th Tk Bn, which remained in position on the road west of Tabu-dong.
6 SEPTEMBER
As the friendly position north of Taegu continued to deteriorate, the enemy
penetration toward Yongchon in the eastern portion of the northern front became
increasingly menacing. On 5 September the remaining army reserve was com-'
mitted to meet this new threat. The infantry reserve consisted of the US 19th
Regt and certain other units of the 24th lnf Div, and had been located approxi
mately forty miles south-southeast of Taegu at 1169.5-1379.0. With all forces
now in action, the top command echelon began to move from Taegu back to
Pusan. As of the morning of 6 September, Korean Army Hq and part of the
EUSAK staff had been reestablished in Pusan.
North of Taegu, Hills 518 and 464 were falling to the enemy in the 7th Cay
sector. The 8th Cay had been infiltrated, and the gap on its right flank had only
partly been closed by moving up companies of the ROK police battalion to sepa
rated positions on Hill 660 and Hill Mass 314. The 8th Cay Regt and the 15th
ROK Regt had not yet joined forces. On the morning of 6 September, the Eighth
Army commander authorized a general withdrawal by the 1st Cay Div to con
. solidate the front. The confirming formal order, dated 1545, 7 September, was
addressed to the commanding general of the 1st CavDiv and the chief of staff of
the ROK Army. It stated that "contact will be established and maintained, re
peat, maintained" and specified as the new line of the 1st Cay Div: 1139-1439
to 1142-1441 to 1144-1443 to 1146-1448 to 1146-1451 to 1147-1454 to 1150-1456
to 1152-1457 to 1154-1457 to 1156-1455 to 1160-1455. On the right Side, this
line is based on Hill Mass 570, Hill Mass 314, and Hill 660. The course of the
crucial fighting on these hill masses* is described in the succeeding pages of
this narrative.
*The tenn ".Hill Mass denotes the entire ridge defined by the surrounding valleys. On Hill L\lass 570
there are other peaks lower than the 57Q-m peak. notably the one at the southern end, called Hill 373. The
314--m point from which Hill Mass 314 derives its designation is not a peak at all, and the northern portion
is over 500 m in elevation.
ORO-T-261
,SEEKn
UNCLASSIFIED
25
UNCLASSIFIED
In accordance with a divisional order, the 1st Bn, 8th Cav at 0350 on
6 September was preparing to break contact and move back to the new line of
defense by way of the 7th Cav MSR. The 3d Bn, 8th Cav was also planning to
displace, and the 2d Bn, 8th Cav was already gradually withdrawing to tie in
with it. To the rear, the two provisional battalions of Task Force Allen were
again alerted for possible quick commitment, and they remained in this emer
gency condition throughout the day.
At 0520 an enemy attack drove back outposts of the "3d Bn, but the battalion
hold its main pOSitions. At the same time, the enemy attacked the regiment's
rear from flanking positions on Hill 570, and for a time succeeded in cutting the
MSR with a fire block across the Taegu road (see below under Hill 570).
The move of the 1st Bn to the west during the morning left the tanks of the
2d Plat, Co B, 70th Tk Bn cut off. Around 1430 they: were gl!ided eastward from
the air along the Waegwan road to Tabu-dong, while covered by friendly artil
lery fire brought down on the enemy positions north of Tabu-dong and on the
road block further south. From the village, the column of tanks turned south
onto the Taegu road and ran a gantlet of enemy antitank weapons without losses,
to rejoin Co B Hq and the 3d Bn, 8th Cav northwest of Hill 570. This action is
described in the subsection on friendly armor support in Part II. Company C,
8th Engr, which had been with the 1st Bn, 8th Cav, moved down the ridge bet
ween the 7th and 8th Cav sectors to Hill 714. Here it was attacked from the
south in an engagement lasting 2 hr, and then subjected to continuing small-arms
fire.
By 1925 most of the 1st Bn had withdrawn to the vicinity of 1143.1-1458.8.
As of 2015, the 2d Bn had closedwith the 3dBn 3/4 to 1 mile north of Tongmyongwon,
and the 3d Bn maintained its previous positions.
Hill Mass 570 (Including Hill 373)
At about 0630, Btry C of the 99th FA Bn, in position along the Taegu road
in the vicinity of the 480-m peak on Hill Mass 570, was subjected to mortar
fire for 20 min, and then attacked by enemy troops who moved down from Hill
570. The enemy were driven off by machine -gun fire. A prisoner taken later
said that the battery was the express objective of the two attacking enemy
companies. The enemy troops withdrew somewhat and set up a fire block across
the Taegu road. A friendly task force composed of Co E (which had been re
trieved from the Walled City only the preceding afternoon), the I&R and Scty Plats
of the regimental Hq Co, a platoon of the 16th Recon Co, and five tanks of the
70th Tk Bn (two from the 3d Plat Co C, two from the 3d Plat, Co B, and one
from the 4th Plat, Co B) moved up the road from the south during the morning
and after a rather prolonged engagement succeeded in clearing the road. The
action is reported by the 16th Recon Co as follows:
0730 hours. Our force jumped off with one-half of the troops and two of the tanks.
going along a dry creek bed about six hundred yards to the left of the road block as the
flanking forward element. The second element, with the remaining three tanks, moved
along the road to the block, coordinating their fire with that of the flanking element.
Within six hours the road block had been broken, and the road was open to Tabu-dong,
which permitted contact to be made once again with tbe main group of the 8th Cavalry
Regiment, which had been cut off. OUr casualties were one KIA and five WIA. The
enemy's losses were estimated at more than one hundred KIA.9
ORO-T-261
26
UNCLASSIFIED
SUREr
..sURH
UNCLASSIFIED
The tank battalion reported that small-arms fire was received in this
action and that small-caliber antitank fire scored about twenty hits on the tanks
without producing more than minor damage. The platoon claimed that it de
stroyed 2 antitank rifles and inflected 200 casualties!4 Company E took the
prisoner mentioned above. Among other things, he stated that two companies
of the 3d Bn, 1st (14th) Regt, 1st NK Div had made the attack (see the section
on enemy units in Part II) and that prior to the attack his company had a strength
of 51 though the entire 3d Bn had a strength of only 153. It is possible that the
1st Bn of the enemy 14th Regt also participated.
Clearing the fire block did not eliminate enemy forces in the 8th Cav rear.
During the evening, large numbers of enemy troops were observed by a liaison
pilot on the ridge on the southeast side of the 570-m peak, and his plane was
fired on. At 2000 a friendly air strike hit enemy troops on the top and east
side of the hill mass with bombs, rockets, and strafing.
Hill Mass 314
The following message from an air liaison flight indicates a very quiet
situation north and east of Hill Mass 314:
1145. From Lt Greer. Two American tanks in pass at 1156.4-1452.8. The road
running south to north just east of these tanks [that is, between Hills 314 and 660] shows
no sign of defense or of defensive troops. Road is passable for small vehicles. One
group of men (approximately 8 or 10) was seen coming southeast Qut of Wall City at
1157.5-1456.7. South Korean artillery battery in position at 1160.9-1451.7. Two small
groups of men seen moving along path running around hill in a generally southeast di
rection at 1160.3-1455.5 [the ridge northeast of Hill 660]. No enemy vehicles or .con
centrations of troops seen.
1
At 1415, ROK observers on Hill 660 reported that enemy troops were on
Hill Mass 314. Friendly artillery then fired on its west side. To block any
further enemy advance, US officers and the I&R Plat, 8th Cay Regt led two
companies of the ROK 5th Tng Bn onto the southern pa,rt of the hill mass. At
1750, the Korean Military Advisory group (KMAG) reported that patrols of the
1st Cav Div and 15th ROK Regt had made contact at 1156-1453 (southern tip of
the hill). Friendly forces did not engage the enemy on the northern part of the
hill mass. The location of the enemy troops was reported at 2030 by a liaison
pilot as the northwest side of the hill at 1156.0-1455.5.
Hill 660
As reported under "Hill Mass :114
n
above, but little enemy activity was
observed north and east of Hill 660. This hill was occupied by one company
of the 15th ROK Regt, and except for patrols, no activity was reported.
7 SEPTEMBER
Plans were made to eliminate the salient from which the enemy was har
assing the 8th Cav. Attacking toward the rear, the 3d Bn was to take Hill 570
while the 2d Bn covered the attack by blocking the Tabu-dong road. The 1st
ORO-T-261
"'EERU UNCLASSIFIED
27
00
c:
z
n
r
::J>
en
en
...,..,
""
,...,
------,
t:I
Hill 660
.\ : ~ ~ ~ w " Knob I /
Fig. 3-Panoramic view of Hill 314 from most forward enemy poistions on Hill 570 to west.
Most infantry action took place on both sides of the ridge above the dotted line. The distance
from the dotted line to the upper limit of the ridge appears highly foreshortened.
Scale: Hill 660 is 3500 m from viewing position; Hill 923, 6300 m. Points on ridge vertically
in I ine with these peaks are 920 m apart.
UNCLASSIFIED
SUREr
Bn was also ordered to fight rearward to establish a road block at 1148.6-1455.6,
thus denying the enemy the possibility of penetrating readily from the west and
linking up with the forces on Hill 570. At 1524 the 1st Bn was further order to
occupy a new line running from 1145.0-1455.5 to 1152.8-1457.0. The battalion
reported at 2200 that it had fought back almost to the new positions.
Hill Mass 570
It was clear that the enemy was on the hill in force. At 0730 troops were
observed at 1154.05-1455.99, and at 0820 friendly artillery fired on enemy
mortars observed just east of the 570-m peak in the vicinity 1155.0-1456.0.
The situation as seen from the ,air at 1140 was reported as follows:
Spirit [99th FA Bn] air observer, saw approximately 150 North Koreans on Hill
570 and approximately 100 North Koreans at 1154.3-1456.8 [north end of the ridge].
Both groups were taken under "time'" fire with good results. The observer also saw
50 troops at 1154.5-1455.9 [southwest side of the 570-m peak] and 50-75 troops at
1154.9-1455.7 going through pass to 1154.9-1454.9 [moving south down ridge to the
480-m peak]. On Hill 373 at 1155.1-1454.1 observed 50-75 troops. Targets in Par
[sic] 2 not fired on due to weather conditions and low gas supply in plane, forcing it
to return to base.
1
If these estimates were accurate, at least an enemy battalion fairly up to
TO strength was on the hill mass (see the section on enemy units in Part II).
The value to the enemy of an observation point on Hill 570 was demon
strated at 1430, when Btry A, 99th FA Bn was for an unspecified period sub
jected to 1 rd of artillery fire every 2 sec and forced to displace.
At 1535, after leaflets had been dropped, a friendly air strike hit Hill
570. Seven planes strafed, dropped napalm, and then strafed again. The entire
hill was reported covered, and the mission reports described the results as
"good" or "excellent." At 1900 fires were seen burning on the east side of the
hill.
Hill Mass 314
United states observers reported from their vantage point on Hill 660 that
they estimated two platoons of enemy troops on the northeast slope of the hill
mass. These were fired on by friendly artillery. Many enemy troops were also
observed just to the northeast in the village of Kisong-dong and on the ridge
further to the northeast. Enemy movement from the Walled City down the trail
to the south was noted.
Other observers noted activity at the northwest end of the hill mass,
adjacent to Hill 570. At 0620, enemy mortars were seen at 1155.8-1455.3.
Nearby, at 1155.8-1455.4, at 1007, enemy troops were observed, and at 1045
the number of troops was estimated at 50, located at 1155.9-1455.8.
Hill 660
The 1st ROK Div reported its line stabilized as of 1150. The hill was still
occupied only by Co F, 15th ROK Regt.
ORO-T-261
UNCLASSIFIED
29
UNCLASSIFIED
8 SEPTEMBER
The 8th Cav sector remained very quiet through the early morning hours.
During the preceding night, however, an enemy force had infiltrated from the
east, blown up the bridge on the Taegu road at 1153.5-1452.5 near the south
western tip of Hill 373, and mined the bypass. The block caused some difficulty,
and the mines damaged two tanks, which were moving up to protect the 99th
FA Bn, and a jeep. The 3d Bn planned attack on Hill 570 was not affected,
however, as its supporting tanks were already to the north.
The bulk of the 2d Bn covered the 3d Bn in its attack to the rear (see
below) and then withdrew, arriving at the assembly area at the southern tip of
Hill 314 at 1155.8-1452.6 by 1250. Later it took up positions south and west
of Hill 373, which remained unoccupied.
Pending the completion of the 1st Bn move to the east to cover the regi
ments left flank, Co F was placed in position across the valley west of Hill 263
to block any enemy move from Tabu-dong along this corridor parallel to the
Taegu road. By 1300, the 1st Bn right flank was already moving into position,
apd at 1407 Co C was engaged in a fire fight at 1151.6-1456.3, in the vicinity
of Hill 263. Company F moved back to take up a new blOCking position across
the Taegu road at Hills 225 and 373.
On the 8th Cav Regt right flank, the 15th ROK Regt reported at 1445 that
the enemy was attacking at Hill 923, 2 1/2 miles northeast of Hill 660. It was
feared that massive enemy forces in this vicinity would swing west against the
8th Cav Regt unprotected right flank. Later reports stated, however, that ROK
forces were containing the enemy, though with difficulty.
At this time, when the 8th Cav situation had become more difficult, ammu
nition shortages became more critical. Stocks on hand in the five artillery
battalions in the zone of the 1st Cav Div totaled 12,400 for 105-mm rd and 3803
for 155-mm rd, and reserve supplies at the Taegu supply point were very low.
Except in case of emergency, a daily ration of 50 rd of 105-mm ammunition
per piece was imposed. Mortar ammunition was also short, and daily expend
itures were limited to 15 rd per 81-mm mortar and 10 rd per 4.2-in. mortar.
Hill Mass 570 (!ncluding Hill 373)
With the support of tanks of the 70th Tk Bn and the mortar platoon of
Co H, the 3d Bn moved south at dawn to attack Hill 570. Because of bad weather,
there was no air support or air observation. By 1010, the first objective, the
knob at the northwestern end of the hill mass, had been taken with little oppo
sition. A second peak was also taken, but apparently against resistance. in
formation on the succeeding action is meager, perhaps in part because commu
nication with the battalion was poor. It appears that when two companies
attacked the main ridge in parallel an unexpectedly strong enemy force counter
attacked and drove between them. (Casualties were considerable in all three
rifle companies. See Table 02.) The battalion withdrew, first to its first ob
jective and then from the hill mass. It was ordered to proceed to the rear,
where it became the regimental reserve.
ORO-T-261
30
UNCLASSIFIED
SHR!r
SiElEr
UNCLASSIFIED
Undetailed and unconfirmed reports state that a third thrust on Hill Mass
570 was made by one platoon of provisional infantry of Co A, 7lst Tk Bn. The
company had lost its tanks in previous fighting and had been formed into two
rifle platoons, which on 21 August were committed as a security force for the
8th Cav forward CP. After 8 September, at least one platoon held positions on
the front line in the 1st Bn area.
In support of the attack, the 99th FA Bn fired "numerous" concentrations
(the number of rounds expended was not recorded). This fire was reinforced
by that of the 9th FA Bn, which fired on and silenced enemy mortars and artil
leryat 1154.56-1456.37 (the ridge on Hill 570), a mortar at 1154.74-1456.30
(just north of the 570-m point), and a mortar at 1155.13-1456.19 (down the slope
east of the 570-m point), among other targets.
Hill Mass 314
The 16th Recon Co was assigned the mission of occupying and holding the
northern part of the hill mass. It relieved the two ROK police companies on the
southern part and moved without opposition onto the central portion, one platoon
apparently establishing positions on Knob 2, the others in" the vicinity of Knob l.
Late in the morning, the I&R Plat of the 8th Cav Regt moved from the north
western slopes of Hill 660 to join the Recon Co on Hill 314. The day passed
here without incident.
Hill 660
By 1645 two companies of the ROK 5th Tng Bn had moved east from Hill
314 onto the northwestern slopes of Hill 660. A company of the 15th ROK Regt
apparently remained on the crest. There were some reports of massive enemy
attacks on the eastern side of the hill, but confirming evidence is lacking.
9 SEPTEMBER
The commanding general of the 1st Cav Div assigned the tlth Cav Regt the
missions of taking Hill 314 (the current mission of the 16th Recon Co) and of
preparing to capture and hold Hill 570. The 3d Bn, 7th Cav Regt, in division
reserve at Taegu, was attached to the 8th Cav Regt to first capture Hill 373, if
not already occupied by the 8th Cav Regt, and then to aid in the capture of Hill
570. The plan of attack formulated by the 7th Cav Regt on the evening of 8 Sep
tember set the time of the attack as 1100 on 9 September. The 3d Bn, 7th Cav
moved out of the Taegu rest area at 0700, but the attack was postponed because
of bad weather, and the battalion took up blocking positions across the highway
south of Hill 373.
The front of the 1st Bn, 8th Cav, was inactive. In the late afternoon a
tank-infantry patrol by the 2d Plat, Co B, 70th Tk Bn and troops of Co F pro
ceeded up the highway almost as far as Samsan-dong without observing enemy
activity.
Hill Mass 570
At 1215 the divisional command ordered that Hill 373 be occupied. Company
G did so and maintained positions there throughout the day. There was little
aggressive activity on either side. Around 0930 a flight of four F51's dropped
napalm on Hill 570 with the "area well covered," according to air observel's.
ORO-T-261
HU!f
UNCLASSIFIED
31
sum
UNCLASSIFIED
Hill Mass 314
During the night of B - 9 September, the 16th Recon Co received small-arms
fire from enemy units which had also moved onto the hill. The first enemy
attack materialized at 0700. It consisted of a frontal assault and attempt to
envelop friendly units by way of their left flank. Forward units on Knob 2 with
drew BOO yd to Knob 0, and the bulk of the friendly force at Knob 1 moved
back 300 yd to Knob O. The Recon reports: "When we reorganized, our
casualties came to three KIA and fifteen WIA. We were unable to estimate how
many casualties we had inflicted on the enemy."
A patrol was sent forward; by 1050 it had almost reached Knob 2. Re
ceiving no opposition, it moved into the former positions on Knob 2 and found
that they had not been occupied by the enemy. Near Knob 3, 250 yd further
forward, a group of 20 enemy soldiers was observed. The patrol was attacked
by this group and also drew fire from its left flank. The Recon Co had by this
time followed the patrol to occupy Knob 2. It sent a message, relayed at 1626,
reporting opposition by one strong enemy platoon armed with automatic weap
ons. Its later report stated further:
1630 hours: counterattacking in an attempt to retake the hill, we had a recoilless
rifle team [from Co E, 8th Cay] and artillery support and got to within two hundred yards
of our objective [Knob 3], only to be driven 'off by intense enemy fire. Due to the fact
that night was rapicily approaching, we withdrew to the line of departure [to reorganize
for the night]. KIA, 5; WIA, 6.'
Sandbags were sent up, and the Recon Co dug in around Knob 1. Companies
A and C of the ROK 5th Sp Tng Bn (police) were sent again onto Hill 314 from
Hill 660 to reinforce the Recon Co on the southern part of the hill mass. Com
pany A was reported attacked at 1B55,.apparently in the same action as the
Recon Co.
Hill 660
Late in the afternoon, Hill 660, which was occupied only by Co F 15th ROK
Regt and a scouting party of the 1st Cav Div, received mortar and artillery fire
from the north. During the evening, the ROK force repulsed an attack by an
estimated enemy battalion. Company B, of the 5th ROK Bn, was placed in a
backup p o s ~ t i o n just southwest of Hill 660 and due south of Hill 314 at coordi
nates 1156.5-1452.0. During the night,it closed in on an estimated 300 enemy
troops who had infiltrated to the southwestern slopes and were now in the
vicinity of Todok Temple at 1157.9-1453.0. At 0200 on 11 September, Co F,
15th ROK Regt reported that it had repulsed an attack by what was estimated
to be an enemy company.
10 SEPTEMBER
Still another gap persisted in the friendly line of defense, this one between
the left flank of the Bth and the 5th Cav Regts. In an effort to close it, the
1st Bn, 7th Cav, which was attached to the 5th Cav, attacked Hills 345 (1147.6
1454.4) and 392 (as reported-presumably Hill 312 at 1148.9-1452.6). The gap
ORO-T-261
UNCLASSIFIED
SUREr
32
UNCLASSIfIED
remained, however, when the enemy held and went on to take Hill 232 (1148.8
1451.4). The separation between the two regiments was further.incr!,!ased when
CoA and Co C of the 8th Cav were pulled in from the area 1150.6-1456.5 to the
vicinity 1151.2-1455.8 in order to tighten the regimental front.
For the first time, high-velocity flat-trajectory artillery fire was received
along the Taegu road from the north. The CP of the 8th Cav Regt was moved
back one mile from 1153.9-1449.7 to 1153.4-1448.1, the village of Chilgok.
With Hill 660 only lightly held by ROK forces, the right flank of the 1st
Cav Div was still vulnerable. At 0700 the 2d Bn, 7th Cav, which was in division
reserve at Taegu, was alerted to be prepared to occupy the hill. As the 3d Bn,
8th Cav was ordered to take over the attack on Hill Mass 314, the attack on
Hill Mass 570 planned for the 3d Bn, 7th Cav was again postponed, presumably
to keep a regimental reserve.
Hill Mass 570
The day passed quietly. At 1025 an air observer reported that nothing was
visible on the ground. Air strikes scheduled for the morning had to be post
poned until the weather cleared. Finally, at 1835, a flight of four F80's appeared
and remained in the area until 1905, strafing and rocketing the highwaywest of
Hill 570 and strafing the other hills along it. No results were noted from the
air, but on Hill 373 Co G, 8th Cav reported being strafed even though panels
were displayed. Around 1900 an uhknown number of F51's attacked the ridge.
From 1902 to 1930 an F82 struck at the pass traversed by the highway north
west of Hill 570. This plane expended 2400.rd of 50-cal ammunition and dropped
two 500-lb bombs on troops in a rice paddy, with unobserved results.
Events on Hill Mass 570 late in the evening of 10 September are discussed
under 11 September.
Hill Mass 314
During the night of 9 - 10 September, the enemy attempted to seize the
southern part of Hill Mass 314. The 16th Recon Co reports:
0400 hours: the enemy carried out two banzai attacks in company strength, but
they were repulsed with no friendly casualties. When daylight c ~ e . we counted 14 KIA
around our positions and 38 at the bottom of the hill [presumably the bottom of the saddle
between Knobs 1 and 21. all enemy.'
Companies A and C of the 5th ROK Tng Bn were pushed back somewhat
in this action.
Around 0700 the 3d Bn, 8th Cav was assigned the mission of capturing
Hill Mass 314. The 16th Recon Co was relieved and attached to the 2d Bn, 8th
Cav. Companies A and C of the ROK battalion were attached to the 3d Bn, 8th
Cav and remained on the hill in backup positions.
By 1155, units of the 3d Bn, 8th Cav were moving into pOSition on Knob 1.
By 1235 they had undergone mortar fire. There was a short artillery prepara
tion, and at 1330 Co K jumped off. At 1435, the 99th FA Bn expended 50 rd on
"enemy troops," probably on Hill Mass 314, with "casualties high" reported.
By 1515, after having again been subjected to mortar fire, Co K occupied Knob 2.
Here it was confronted by an enemy platoon backed up by an estimated two .
ORO-T-261
UNCLASSIFIED
33
UNCLASSIFIED
UERn
companies 'at Knob 3. Artillery was called in on Knob 3 (none is reported by
the 99th FA Bn), and friendly troops moved back below Knob 2. Here they
remained 'at least until 1840, for a message relayed at this time by a pilot stated
that red panels were observed on the ridge at coordinates ll56.5-1455.3, Le.,
just forward of Knob 2. It was further observed that personnel dressed in white
were on the north slope and a few personnel in white on the south slope. These
figures may well have been civilian labor parties working for the respective
opposing sides (see Fig. E 1).
The situ.ation at the end of the day's fighting was reported in a relayed
message:
102000. White 5 reports that the 3d Battalion will hold on position of first ridge
line south of objective and attack at 110650 Sep. SK Company will attack and secure
bridge south of objective. I Company on third ridgeline behind K and L Companies
blocldng road on trail to east. 22 North Koreans killed. Two enemy companies on
objective and one platoon forward of K Company position.
1
The 22 enemy dead were apparently killed on the hill in the day's fighting.
Hill 660
An unelaborated report relayed at 0300, II September mentioned heavy
fighting between the peak of Hill 660 and the village of Kisong-dong. Reports
are not available from Co F, 15th ROK Regt which was on the hill. At 1235,
small-arms and mortar fire on the hill were noted by observers elsewhere,
and at 1355 an observer reported groups of enemy moving from Hill 660 along
the l'idge to Hill 313, 1 mile to the south. Just to the east of Hill 660, the left
flank of the 15th ROK Regt was under heavy attack. At 2259 the I&R Plat,
7th Cav Regt reported a fire -fight on the west side of Hill 660.
The 300 enemy troops reported at Todok Temple the night before were
still there at 2300, shouting and singing. They were hemmed in by Co B of
the 5th ROJ( Tng Bn and reserve units of the 3d Bn, 8th Cav.
11 SEPTEMBER
Enemy efforts were concentrated north of Taegu. To the west the sector
of the 27th Brit Brig remained quiet, but in the sector of the 5th Cav Regt and
attached 1st Bn, 7th Cav Regt, seesaw fighting of attack and counterattack con
tinued around Hill 203. Both sides also attacked in the subject sector of the
8th Cav Regt. To the east, however, enemy attacks against the ROK 1st Div
slackened, and ROK forces began to patrol to the north.
Hill Mass 570
At 2024, 10 September two enemy patrols of about twenty-five men were
driven off Hill 373 by Co G, 8th Cav. Two enemy NCOs taken prisoner stated
that their unit of 115 men was to attack at 2100. They further stated that they
were members of the 14th Regt (1st NK Div) and unfamiliar with the regiment
to which they were assigned (presumably the 19th Regt, 13th NK Div).
ORO-T-261
34
UNCLASSIFIED
SUREr
UNCLASSIFIED
The attack, by enemy first estimated at 115, then 200 to 300, came at 2345,
the enemy using grenades and automatic weapons. Company G's left flank gave
way, but the attack was repulsed with the support of fire by the 99th FA Bn. By
0210, 11 September, Co G had reestablished its line. At 0320 the enemy was
observed regrouping on the western finger of the 480-m peak behind Hill 373.
It was planned to send up Co E in the morning to bolster the defense of
Hill 373. The second attack came at 0710, before the reinforcement arrived.
The assault was again turned back, but this time pOSitions on the forward slope
of Hill 373 were overrun and Co G suffered many casualties. A message re
layed at 0801 reported that there were 35 WIA. Official casualty statistics are
presented in Part II.
While making these frontal assaults, the enemy wab also perSistent in
attempts to infiltrate, as the following report of enemy intentions had anticipated:
101700. From G-2. Just received from reliable confidential sources the follow
ing additional information: The 1st and 13th NK Divisions are having difficulty in co
ordinating their advance. The 1st NK Division commander had been ordered to barri
cade a major road in the 1st Cay Div prior to sunset 10
The road in question must be the Taegu -Tabu-dong road and the barricade in
question must be contemplated in rear of our front troops, perhaps in the vicinity of
the blown culvert (1153.6-1452.2) or further to the south so that our movements and
resupply will be hampered.
1
The attempt materiaU:<ed. Around 0100 a group of an estimated fifty
enemy was engaged by a friendly highway patrol at the bridge at coordinates
1153.4-1452.8, a quarter of a milenorth of the expected spot but still well to
the rear of Hill 373. The friendly patrol was driven back to a second bridge
and then pushed off that one,. When it returned to the latter bridge, the enemy
had apparently withdrawn. A prisoner's account that seems to fit this incident
appears in Part II.
The situation at Hill 373 was precarious. Already the enemy had pene
trated to the rear in company strength, if only briefly. As shown in the 8th
Cav operations overlay dated 1000, 11 September (see Part III), Co G had been
pushed back to the fingers on the southern slope, Co F on its left was spread
out across the valley and had already been infiltrated, and Co B was on the far
side of the valley on Hill 225. From his position on Hill 314 the CO of the 3d
Bn, 8th Cav observed Hill 570 "teeming" with enemy troops. At 0725 a mes
sage came from the commanding general of the 1st Cav Div announcing that
the 3d Bn, 7th Cav would again postpone its planned attack on Hill 570 (and
373) pending development of the situation at Hill 373 and the outcome of the
final assault on the northern part of Hill 314 by the 3d Bn, 8th Cav.
The timely intervention of the mo&t intense air attack yet directed against
Hill 570 appears responsible for the respite necessary for the reinforcement
of Co G. Hill 570 was the target of at least twenty fighter-bombers in a period
of an hour and a half. A flight of four F51's appeared at 0830, dropped
napalm, and strafed, with results claimed to be "good." At approximately
0850 another flight of four F51's rocketed and strafed the ridge, with undeter
mined results. Four more F51's attacked shortly after 0855, dropping napalm
with "excellent" re;:;ults, ctrafing with "fair" and a 20-mm gun
ORO-T-261
UNCLASSIFIED
SlEREr
35
UNCLASSIFIED
with their rockets. At around 0905 still another flight of four F51's dropped
napalm and rocketed, with "good hits" claimed. Finally, at about ()950, one
more flight of four F51's bombed and strafed the ridge, with unknown results.
It is uncertain fi-om the confliCting reports whether a flight of four FBO's that
were over their target from 1115 to 1140 expended their 16 rockets on Hill 570
or on some other hill.
The effectiveness of these strikes is indicated by a ground observer's
report at 0940. Having moved down from the 570-m peak toward the southwest,
150 enemy troops believed flushed by the air attack were observed milling
around in the saddle (1154.7-1455.4) between the 570-m and the 4BO-m peaks.
They were taken under artillery fire and fled to the road on the west side of
the hill, where they were again taken under fire and dispersed by the tanks of
Co B of the 70th Tk Bn.
At 0801, Co E was already on the way up to Hill 373. After it arrived,
Co G moved out at 1200 in a limited advance to stabili>:e the situation on the
hill by occupying the positions lost previously on the northern slope. During
this action it sustained additional casualties from intense enemy mortar fire.
Throughout the day, the 99th FA Bn fired a total of 472 rd on targets of
opportunity, in support of 8th Cav troops engaged on Hills 373 and 314. Of
22'misslons fired, only 2 were to repel enemy attacks.
As further reinforcement on Hill 373, the 16th Recon Co, which had been
relieved on Hill 314 by the 3d Bn, 8th Cav, joined Co E and Co G, and was sta
tioned in the forward positions. That the enemy offensive had only been delayed
was demonstrated on the night of 11 - 12 September by the following attack
reported by the Recon Co:
During the night our company was placed in front of the battalion's position. Dur
ing the early morning hours an estimated reinforced enemy c o m p ~ y attacked our pasi
tion at several points, but was driven off by artillery, mortar, automatic weapons, and
small-arms fire. Over one hundred enemy dead were counted the follOWing morning,
and there were no friendly casualties .... There were no further attacks on our posi
tions during the remainder of the night.
9
The enemy dead counted by the Recon Co had probably accumulated from all
the actions to date.
Hill Mass 314
Units of the 3d Bn, 8th Cav moved out toward Knob 3 at 0730. Details on
the subsequent action are lacking. The operations overlay dated 1000, 11 Sep
tember snows Co K on Knob 1, Co I on Knob 0, and ROK forces on the
southernmost slope of the hill (see 'Part III). These may not have been the
actual locations of the units at that time. Reports of no specified time state
that friendly force s were driven back from their objective by intense mortar
fire from Hill 373. (Hill 373 was partly in enemy hands during the morning,
but the fire probably came from Hill 570 or beyond.) At 1340 direct fire was
also being received from a gun in the vicinity of the cemetery, 1 J;2 miles to the
northwest. It would appear that the 3d Bn did withdraw as shown by the over
lay,.and then, later, made an attempt to capture the northern objective. A
message at 1415 reported that friendly units had reached the 1454.5 line, the
middle of the saddle between Knobs 1 and 2. At 1515 the advancing units were
ORO-T-261 36
UNCLASSIFIED
SECRET
UNCLASSIFIED
counterattacked by an estimated 200 enemy troops and forced to withdraw to
their original positions," evidently those shown on the overlay.
Air support was sent and by 1540 was working over the northern part of
the hill mass. The TACP log of the 1st Cav Div lists an air strike on Hill
Mass 314 at this time by four F51's of the 39th Ftr Sq. Unfortunately, nO
corresponding mission report is available. The log states that the planes were
armed with rockets, napalm, and 50-cal ammunition.
The TACP log also lists a second strike on Hill 314 at approximately 1015
by four FBO's of the Bth Ftr Sq. The corresponding mission report is available;
it specifies as the point of attack the proper coordinate of longitude for Knob 2
but gives a coordinate of latitude more northerly than any target scheduled
after 4 September. The report states:
Contacted at 1600 and was directed to Wildwest [1st Cav Div] control, Taegu area.
Contacted Mosquito Wildwest and had a ridge to be strafed at '3605-12835 [Coordinates
of Knob 2: 36 00 _128 35']. Fired 15 x 5 HVA rockets in troop positions and strafed
area. Results not observed but controller informed flight after first pass NK troops
started to give up positions and run. In area from 1600 to 1325.
18
In agreement with this description, the TACP log states that the planes
were armed with rockets and 50-cal ammunition. Referring either to this or
to the preceding strike, it further remarks: "bodies seen blown in air, perfect
results ."
During the air strikes, Co I and Co K withdrew to Knob O. Company L,
Bth Cav, which had been in reserve, then passed through these units to occupy
this pOSition. Company I and all but a platoon of Co K that was retained for
right-flank security'were relieved.
The night passed without incident on Hill 314. But, as shown in Part II,
the enemy was bringing up new forces.
Hill 660
At 0015 a message was received from unspecified ROK units stating that
enemy troops on Hill 660 were moving. These apparently were the enemy
troops which had been at Todok Temple. At 0410 the 15th ROK Regt reported
that there had been no activity of any Significance during the night of 10 - 11
September. There were no later reports.
At 0905 the 15th ROK Regt, looking down on any movement of the 1st NK
Div troops from the ridges northeast of Hill 660, was shelled by guns of the
13th NK Div at Tongmyongwon, 4 miles to the west.
12 SEPTEMBER
It had been planned to relieve the entire Bth Cav Regt with the 7th Cav
Regt, the key operation being the capture of Hill Mass 570 by the 3d Bn, 7th
Cav. The related moves by the several battalions of the two regiments were
specified in the following order received on the afternoon of 11 September:
111655. Message from Commanding General, 1st Cavalry Division -lao 5th Cav
alry will effect relief of 1st Battalion, 7th Cavalry by 111800 Sep. 'lb. 7th Cavalry will
(1) Upon relief of 1st Battalion by elements of 5th Cavalry, move 1st Battalion to vicin
ORO-T-261
UNCLASSIFIED
s!nn
37
UNCLASSIFIED &ECRU
ity of bridge (1153-1442) under regimental control. (2) Relieve ROK forces now on
Hill 660 (1158-1453) with 2d Battalion beginning 120600 Sep 50. (3) 3d Battalion will
attack, seize and hold high ground from 1153.9-1456.9 to 1154.6-1456.9 to 1154.6-1456.0
to 1154.8-1455.0. Upon completion of mission, 7th Cavalry zone of responsibility ex
tended from Division's rigbt boundary to line 1152.0-1457.5, 1153.0-1446.0, road to 7th
Cavalry. 1e. 8th Cavalry will: (1) Hold present position until 3d Battalion, 7th Cav- .
alry has reached and secured objective. (2) When objective of 3d Battalion, 7th Caval.ry
is secured, 8th Cavalry (minus 1st Battalion) reverts to Division Reserve in vicinity
1153-1442. (3) 1st Battalion will remain in present position under control and lOgistical
support of 7th Cavalry when objective area is secured.
3
This order apparently assumed the capture of the northern part of Hill
Mass 314 by the 3d Bn, 8th Cav. Since this did not occur, and since the battalion
at the close of 11 September was in rather bad shape on the southern part of
the hill mass, the major attack by the 3d Bn, 7th Cav was rescheduled for Hill
Mass 314, as recorded in the following message: .
Saber 5 [Executive, 1st Cav Div] called - Skirmish Blue will not go on scheduled
mission but will take ground now held by Scrappy Blue. All operations under Skirmish
6. SW boundary approximately along vertical grid line 1155 south along east side of
Hill 570 and to valley north of Skirmish position, then east to boundary. Scrappy Blue
to remain in position until passed through, when they will assist Skirmish Blue to maxi
mum extent by automatic weapons and mortar fire. Skirmish Blue plan begins one hour
later than scheduled. Scrappy continue to hold Hill 373 and improve positions. No at
tack. Move forward only to improve positions.
3
The 1st Bn, 7th Cav, which had retaken and was holding Hills 203 (1145.8
1452.6) and 174 (1146.9-1451.7) was relieved at 1830 on 11 September by the
3d Bn, 5th Cav which was redeployed from the west. The release of the 1st
Bn to division reserve at Taegu merely compensated for the commitment of
the 2d Bn, 7th Cav. On the evening of 11 September the 5th lnf Regt was or
dered to send one battalion to Taegu at once. The 3d Bn, 5th lnf closed in its
Taegu location at 0330 on 12 September.
With the 13th NK Div continuing build-up there was a general increase in
the tempo of enemy activity, and 13th Div artillery and mortar fire reached a
new intensity. Starting at 0420, enemy artillery walked up the Taegu highway
from the 8th Cav Regt CP at Chilgok* to the front lines. Company G on Hill 373
and Co F across the valley and highway just to the west reported at this time
that they were subjected to enemy artillery and mortar fire. At 0515, enemy
fire was again received down the valley, when the 9th FA Bn (see Fig. 2 for
current location)* reported that 100 rd of artillery fire had come in. At 0607
enemy mortar fire was again reported falling on Hill 373, and at 0715 Co B
on Hill 225 on the western side of the valley reported that enemy mortar fire
had killed 4 men and wounded 1 ~ . t At 0730 some artillery fire was still falling
down the valley on the 9th FA Bn. Around 0855, 120-mm mortar fire landed at
the foot of Hill Mass 314. Then there was a lull.
Chilgok was 10,000 m and the 9th FA Un 11,000 m south of the 13th NK Div Arty positions (POW infor
mation). Whether from 122-mm howitzers or 76-mm pieces, then, fire was at almost maximum range.
tCompany C on 10 September reported that it received a mortar concentration of 30 to 50 rd around 1630.
but suffered no casualties because it was well dug in.
ORO-T-261
38
UNCLASSIFIED ,SEERET
UNCLASSIFIED
SEERET
With haze lying in the valleys until late in the morning little of the enemy
fire could have been observed by the enemy. Enemy artillery (Qut not mortars)
apparently remained quiet until 1640, when the 99th FA Bn was taken under fire.
Hill Mass 570
Aside from the fires noted above neither any enemy activity nor friendly
moves were reported on the hill mass.
Hill Mass 314
Until the arrival of the 3d Bn, 7th Cav around 0800, there was no reported
activity on the hill mass. The battalion's move-up and the subsequent battle
on the hill are the subject of the detailed narrative of Part N.
Hill 660
At 0815 it was reported that the night of 11 - 12 September had been ex
ceedingly quiet on Hill 660; not a shot had been fired.
To eliminate the weak point in the line of defense, the 2d Bn was committed
to relieve ROK police units on Hill 660. The battalion cleared the reserve area
at Taegu at 0650 and moved by motor toward the objective by a route well to
the east of the "Bowling Alley."' By 0830 the battalion had reached and cleared
its assembly area. Moving up the ridge that runs north to Hill 660, Co F and
Co G had reached the vicinity of Hill 339 by 1220. Company E at this time was
moving to occupy Hill 516, somewhat to the northeast, and the battalion CP had
been set up at the southeast side of this hill. Hills 660 and 516 were occupied
without incident by 1400.
13 SEPTEMBER
In view of the intense enemy pressure maintained against the 5th Cav Regt
at Hills 203 and 174 and the possible breakthrough here to the corridor to Taegu
traversed by the railroad, the continuing gap between the 5th and 8th Cav Regts
remained menacing. The concern of the 8th Cav regimental command was re
corded as follows:
122045. Message to Saber 5 - Attached overlay shows my front line units as of
1900 this evening and the location of 5th Cavalry units as I last heard of them late this
afternoon. The gap between the units is a matter of great concern to me and is in my
estimation a very definite threat to securing the MLR. My estimate is based on terrain
consideration and on known enemy activity on my left flank. Troops presently backing
up the center, which I consider the most likely avenue of approach, could not be diverted
in time to prevent an envelopment or the passage of enemy troops down my left flank.
I deemed it advisable to call this to your attention for such action as you consider neces
sary. Scrappy 6.'
*The tenn "Bowling Alley' was applied first to the north-south valley north of Tabu-dong and later, as
here, to the valley running from Tabu-dong to Taegu.
ORO-T-261 39
UNCLASSIFIED
SECKEl
UNCLASSIFIED
SEEm
To patrol this gap, a task force of normally noncombat personnel from the
5th Cay Regt (Task Force Wolfe) which had moved up to protect the OP on Hill
260, was to sweep northward along the ridge to contact the 8th Cay in the vicin
ity of 1149.0-1456.0. During the day, forward observers and other patrols
noted the presence of several enemy groups in the area of the gap, and these'
were fired on by friendly artillery.
At 1200 on 13 September, I Corps became operational. The 1st Cay Div,
heretofore directly under Eighth Army control, was among the units assigned
to it. Attachment of the 5th RCT to the corps gave a net increase in friendly
strength in the Taegu area, as well as improved organizational control.
Hill Mass 570
Reinforced platoons of Co E felt out enemy positions with the following
, reported results:
131715. White sent out two patrols on H!ll 373. Knocked off an enemy two-man
outpost just below crest of next high ground between 373 and 570 [the 480-m peak].
Contacted a strong point of two machine guns and 20 rifles. Withdrew and brought
mortar and arty fire on enemy position with unknown results. Leading elements have
secured 300 yards in front of previous position.
3
Contact with the main enemy force was further reported:
From 8-3 Division Artillery: 8-3 of 99th FA Battalion reported that observer
with patrol from 2d Battalion, 8th Cavalry at 2220 hours observed two enemy companies
massing at 1154.8-1455.3 [the saddle between the 480- and 570-m peaks]. The 99th FA
Battalion massed one battalion volley on the enemy troops, repeated the concentration
ten minutes later. Excellent results ....1
During the night of 13 -14 September, a patrol from Co I, 7th Cay reached
the pOint 1154.8-1456.4 on the east side of the hill mass, near the 570-m peak,
and found only small, disorganized enemy groups. A prisoner was taken who
had been on Hill Mass 314 during the battle on 12 September and who was from
the 1st NK Div.
Hill Mass 314
Except for sporadic enemy mortar fire, the day passed quietly.
Hill 660
No action was reported. During the day the CP of the 7th Cay Regt moved
up from its previous location just north of Taegu at 1153.5-1442.2 to 1161.3
1449.0, to the right and rear of the new positions of the 2d Bn.
14 SEPTEMBER
In the morning, Task Force Wolfe reported that it had been ambushed at
Hill 401 and had withdrawn to Hill 260. At 0740 the 5th Cay Regt reported
that its units on the 8th Cay left flank were disposed as follows: Co K, 5th
Cay, east of Hill 203 (1145.8-1452.6); Co I, 5th CaY, Hills 312 (1148.8-1452.6)
ORO-T-261
40
UNCLASSIFIED
SEEREY
UNCLASSIFIED
'EeRn
and 232 (1148.7-1151.4); and Task Force Wolfe, north, on, and south of Hill
260 (1149.2-1448.9).
A patrol fro!" Co I, 8th Cav moved to the top of Hill 401. A prisoner
taken from the 23d NK Regt stated that both the 23d and 21st NK Regts were
in the vicinity of Hill 401. Several days before, another prisoner had stated
that these two regiments had been joined to make an effective force. Patrols
from the 1st Bn of the 8th Cav .Regt maintained contact with the enemy to the
north, along the regiment's left flank.
To the northeast, the ROK Army opened its drive westward along the
ridge leading to the Walled City. The newly arrived C Bty, 17th FA: Bn (8-in.
howitzers) fired its first missions this day on targets in the Tabu-dong area.
Hill Mass 570
Patrols of 3 to 10 men were sent out toward Hill 570 from Hill 373 at 1045
by the 16th Recon Co and by Co E and Co G of the 2d Bn, Bth Cav. From Hill
Mass 314 small enemy groups could be seen withdrawing before the slow advance
of the friendly troops.
The three companies jumped off in the attack at 1430, after an artillery
preparation and air strikes at 1345 and 1415. Support of the infantry consisted
of the two batteries of the 9th FA Bn, one battery of the 99th FA Bn, and the
tanks of the 2d Plat, Co B, 70th Tk Bn. Progress of the attack was reported
as follows:
1615. From Sth Cavalry. The 2d Battalion has ~ o v e d past its first objective,
that is, the first big bald bill on the nose [4S0-m peak?] and they are moving on the
second objective....
1730. '" Scrappy units are now as far north as 1154.6-1456.3 (beyond the 570-m
point]. There were 70 enemy dead counted and a total of 5 POWs captured on Hill 570
thus far.
2125. From S-3, 7th Cavalry: ... In the Sth Cavalry attack on Hill 570, the 3d
Battalion, 7th Cavalry claims they killed at least 200-300 North Koreans as they came
off the, bill with 75mm recoilless rifles and macbine guns1
The action ended with the 2d Bn's front line just north of the 570-m peak.
Whether or not the 7th Cav claim is accepted as accurate, the fact that its 3d
Bn had targets indicates that many of the enemy troops on Hill 570 fled to the
north or northeast as well as to the northwest, where they were seen on the
Taegu road. That the retreat was precipitous is suggested by the discovery
(at 1400, i5 September) of the bodies of four Americans, obviously murdered,
at 1154.4-1456.3, the point of farthest friendly advance on 14 September. They
had been bayoneted or shot from short range. They had been tagged, prior
wounds had been dressed, and their hands were tied behind their backs.
Casualties in the attacking companies were light. In Part II of this memo
randum the attack and enemy withdrawal are described briefly from the enemy
viewpoint by prisoners captured shortly afterward.
Hill Mass 314
No significant friendly activity was reported except for fire in support of
the 8th Cav capture of Hill 570. Between 0100 and 0300, harassing fire fell
on Hill Mass 314 every 15 min.
ORO-T-261
UNCLASSIFIED
SEtRE{
41
UNCLASSIFIED
S[EltET
Hill 660
No activity was reported.
15 SEPTEMBER
The 10th AAA GP (strength 99) and the 68th (strength 776) and 78th (strength
368) AAA Bns were attached to the 1st Cav Div. In the absence of enemy air
craft, the 90-mm guns of the 68th AAA Bn, which had only landed :n Korea on
6 September, fired its first missions on enemy ground targets on 15 or 16
September. The other AAA units may have been used actively in the same role.
Hill Mass 570
At 0420 the 2d Bn, 8th Cav was attacked by an estimated ten to twenty
enemy troops; they were driven off by 0440.
The enemy was not abandoning the hill. During the morning an observer
reported enemy activity in the two villages nearby in the valley to the east.
At 1400 tank and mortar fire were received. A later report stated: "1720.
From 8th Cavalry: My troops on Hill 570 have had 4 KIA and 40 WIA from
mortars so far today. Artillery can't seem to stop it.,l By evening the fire
slackened off to a few occasional rounds. A prisoner from the 13th NK Div
captured early the next morning near Hill 660 'pinpointed eight 120-mm mortars
at Tabu-dong.
Hill Mass 314
At 0900, the 3d Bn, 7th Cav Regt sent out a large patrol to reconnoiter
and to contact the 15th ROK Regt to the east. The battalion contacted the
enemy and suffered several casualties.
At 2210 some mortar fire was received on the OP on the hill.
Hill 660
Except for 12 to 15 rd of 14.7-mm fire received at 2300, no enemy activ
ity was reported.
16 SEPTEMBER
With the arrival of its 1st Bn the entire 5th Inf Regt was present. Its
total strength of about 2580 significantly increased friendly capabilities in the
1st Cav Div area.
Hill Mass 570
Enemy attacks were reported during the day, as follows:
1050. From Scrappy 3: 8th Cavalry states they are receiving heavy small arms
fire from enemy force north and east of Hill 570. One platoon of the 8th Cavalry had
advanced beyond Hill 570 but was called back. The enemy force had an estimated
strengtb of 150 and attacked toward Hill 570 from the northeast.
42 ORO-T-261
UNCLASSIFIED
S&(RET
UNCLASSIFIED
'EERH
1200. From Scrappy 3. Enemy attack 'stlll in progress on Hill 570. We still
hold the hill. Appears to he a lessening of pressure hy the enemy. Some movement
to the west of the hill. Cannot hit with artillery.
1245. Scrappy 2 reports that at 1100 hours the 2d Battalion was attacked hy 225
enemy from 1154.4-1457.0. Attack dispersed by mortars. Used 100 rounds.
1350. From Spirit 3. At 1153.4-1459.6 there are reported to be 800 enemy troops.
At 1150.0-1460.6 are four 120mm mortars. The 99th FA Battalion will fire a TOT on
these two areas.
1825. From G-2. Spirit 2 reports FO of Company E, 8th Cavalry states North
Koreans are pounding Hill 570 continuously with artillery and mortars (predominately
estimated to be 82mm mortars). Mortars are located at 1153.9-1456.9 and 1154.1
1458.1. Artillery is located at 1152.0-1459.1 and 1152.7-1457.9. The observer esti
mates he saw 300 enemy KIA by artillery fire and 100 KIA by small arms. Along the
west slope there appears to be more dead, but he was unable to count them. There is
enemy traffic on north and south road west of Hill 570. Our forces have withdrawn to
south part of Hill 570, the point on the map where the 570-mark shows.'
Hill Mass 314
At 0200 a ROK company attached to the 3d Bn, 7th Cav broke up an attack
by an enemy platoon, described by a prisoner as a suicide mission to capture
Hill 660.
Hill 660
The 2d Bn, 7th Cav was relieved on Hill 660 for offensive action further
to the west in the 5th Cav Regt sector. One company of ROK police moved back
onto Hill 660.
17-21 SEPTEMBER
As a consequence of the successful US landing at Inchon on 15 September,
the plans for the break-out from the UN perimeter were put into effect on 16
September, with attacks all around the perimeter. Within the zone of the 1st
Cav Div the main effort got under way on 17 September. The division's initial
thrust northwest to the Naktong River was made by the 5th Cav Regt, the attached
5th InfRegt, and the 2d Bn of the 7th Cav Regt. The 3d Bn, 8th Cav Regt partici
pated in the offensive by attacking objectives on the regiment's left flank.
Against stubborn enemy resistance, this general effort gradually produced gains.
In the 8th Cav zone, the aggressive patrolling of the 1st Bn produced sharp
reactions. Up to 18 September the 2d Bn was still being subjected to intense
enemy fire and to attacks against its pOSitions on Hill Mass 570. On 19 Sep
tember a coordinated friendly operation was ordered, in which the 2d Bn was
to take all of the hill mass, while the 1st Bn, with tank support, was to patrol
north to Tabu-dong and then seize the village. The remainder of Hill Mass 570
was taken with little opposition, but the 1st Bn was stopped short of Tongmyon
gwon by enemy resistance which caused at least 31 casualties in Co B, and by
the fact that the tanks could not advance beyond a blown bridge and mined bypass
at 1153.2-1455.0.
ORO-T-261
UNCLASSIFIED
SEERff
43
SECKn
UNCLASSIFIED
The 8th Cav renewed its attack to the north on 20 September. Company B,
8th Cav Regt, and two platoons of tanks of Co B, 70th Tk Bn again moved up
the highway, while Co F, 8th Cav Regt apparently advanced from the newly won
forward positions on Hill Mass 570. After an advance of about a thousand yards,
the friendly attack was stalled by another blown bridge and mined bypass in the
vicinity 1153.0-1456.7 and by intense crossfires of artillery and automatic weap
ons from the approaches to the Walled City and the hills west of Tabu-dong.
Company B suffered at least 23 casualties, and Co F at least 22; three tanks
were put out of action by AT mines (all-wooden construction and difficult to
detect). Friendly units withdrew to their original positions.
To the northeast, ROK forces continued to gain toward the Walled City in
their frontal attack along the ridge, but the fighting was seesaw and they were
stopped by the tenacious defense of the Walled City itself. They did succeed,
however, in slipping past it and penetrating well to the north. Turning this
advance to the south and west, they now threatened to envelop the Walled City.
On the morning of 19 September, they took the road junction 3 miles north of
Tabu-dong-the area from which the 13th NK Div had launched its offensive on
2 September -and at 1200 the 1st Bn of the 15th ROK Regt moved on south to
take up positions on the west side of the highway only 1 3/4 miles north of Tabu
dong, at 1141.3-1462.8. By this time, a ROK company had taken Hill 902 within
the Walled City. As of 1240 on 20 September, 2000 ROK troops had moved into
the area between the road juriction 3 miles north of Tabu-dong and a pOint within
1 3/4 miles of the village. Despite its continuing resistance, the 13th NK Div was
definitely cut off. During the night, its chief of staff slipped away to the south
and at 0700 on 21 September surrendered to the 8th Cav Regt at 1153.3-1454.3.
The 1st and 3d Bns of the 7th Cav Regt were withdrawn from their positions
adjacent to the 8th Cav Regt on 19 September and recommitted to the west to
relieve the 5th Cav Regt. After attacking westward in the Waegwan area, the
1st Bn was given the new mission of making a long, encircling movement behind
enemy lines along the Waegwan-Tabu-dong road, with the 3d Bn following and,
if necessary, relieving it for attack on Tabu-dong. On 21 September, the 1st
Bn and tanks of Co C, 70th Tk Bn covered the more than 5 miles to Tabu-dong
and took the village, at 1400, at the cost of few casualties. To the south, the
enemy continued to oppose the 8th Cav (Co B had at least 15 more casualties
on 21 September) and tanks of Co B, 70th Tk Bn. Finally, at 1700, the 2 tank
infantry forces made contact on the Taegu road 1 1;2 miles south of Tabu-dong.
Many prisoners were taken; for example, Co C, 7th Cav reported 35. They
were chiefly from the 19th NK Regt. However, the enemy did not surrender en
masse, but slipped away into the hills.
All around the perimeter, enemy lines were crumbling.
ORO-T-261 44
UNCLASSIFIED
UNCLASSIFIED
5ECRU
Part II
ENEMY AND FRIENDLY THOGPS AND SUPPOHT
UNCLASSIFIED
SECRET
UNCLASSIFIED
ENEMY UNITS
This section is based chiefly on reports of interrogation of prisoners of
war and on captured enemy documents. Those journal entries which give in
formation on prisoners just captured or observations of enemy movements are
also used. Because the enemy situation has been worked out in from
these data, it is natural that the present identification of some units differs
from that carried in intelligence estimates current at the time of the action
and necessarily based on less complete information. Very little recourse is
made to general intelligence analyses of the tilne, partly for this reason and
also so that the findings will follow directly from the raw data. Though the
present analysis has not been examined by 1st Cav Div personnel, the enemy
situation in outline has been discussed with the former S-2 of the division, and
he proffered no alternate view.
Much of the possible utility of the interrogation reports for the purposes
of this study was not realized, because of the incompleteness, inconsistency,
and inaccuracy of the reporting of data furnished by the prisoners. There are
sometimes discrepancies in the reporting of the same event or unit designation
or date even within one document. The prisoner-of-war (POW) tags called for
the time and place of capture and the friendly unit concerned, but often this
valuable information does not appear in the report, and one must accept such
data on these points as may have been given during interrogation" by an ignorant
and confused enemy recruit.
Many of the interrogation reports are reproduced in full or part in App A.
In the text below, these are designated appropriately; in any case, the interrogat
ing unit is specified. Most of them are referred to by their Allied Translator
and Interpreter Service (ATIS) number. Those of these which were viewed in
earlier identical form are referred to by Advanced ATIS (ADVATIS) numbers
or by 164th Military Intelligence Service Detachment Interrogation (l64th MISDI)
designations. Several reports of interrogations carried out within the 1st Cav
Div are also cited. These reports are the richest in tactical information and
the most consistent in the facts reported.
GENERAL SITUATION AND ENEMY INTENTIONS
At the beginning of September 1950, three enemy divisions launched an
offensive to capture Taegu from the north. Disposed along a line running east
from the Naktong River in the vicinity of Hill 518, the 3d NK Div faced the 7th
ORO-T-261
UNCLASSIFIED
SiEREf
47
UNCLASSIFIED
Cav Regt. To the east there was a gap of rough terrain almost unoccupied by
either side. Due north of Taegu the 13th NK Div was concentrated astride the
highway which leads south to the city through the village of Tabu-dong. Among
the rough ridges on its eastern flank, the 1st NK Div opposed units of the 1st
ROK Div.
The 13th NK Div had followed the 1st NK Div through Sangju and as far
as Naktong-ni. From here, it fought south and east until it was stopped short
of Tabu-dong by the 27th RCT. The 13th Div's mission in the new offensive
was recorded in a notebook by its chief-of-staff'in a meeting 30 Aug 50, as
follows:
(a) To operate in our Div main offensive direction date 2 Sep. Three days for
preparation for attack. Begin attack at 1800 hours 2 Sept. (Arty firing for 15 minutes).
Mter capture of Taegu City, advance 15-20 km south from Taegu and give Taegu strong
defense.
(b) Supreme CG directive rush on 6 kIn at the first time.
(c) Advance before daybreak.
(d) 1st Div at the left side, 3d Div at the right side.
(e) (n) .
Ammunition: every private 2 ~ 6 , Regt 0.5, Div 0.5, Arty 1.5, Arty 1, Div 0.5.
[Basic load of regimental artillery was 60 rd per piece, and 120 rd per piece were
issued at this time (Item 1, App A).]
Provision cooked for 2 days, raw food for 2 days.
UISONG 3000 men.
ANDONG 1000 men.
Commanding Post 170 men.
MU JONG
CG, 2 Corps
The entries for Uisong and Andong refer to replacement pools, and this
subject is analyzed later in this section. As reported in Part I, the attack by
the 19th Regt, 13th NK Div was disclosed by the 19th Regt S-3, who surrendered
on 1 September. Nevertheless, the attack was carried out on the evening of
2 September just as ordered. From the food allowance it would seem that a
period of four days was allowed for the first phase of the offensive. As noted
in Part I, by 6 September the enemy had advanced little more than 4 of the 6 km
prescribed for the first rush.
After the penetration by the 13th Div and the wide advance of the 1st Div
which it facilitated, both divisions were given the same mission: reach Taegu
via Chilgok in the Bowling Alley. (Figure. 4 is a photograph of this corridor
and the highway just north of Taegu.) Intelligence relayed at 1745 on 10 Sep
tember" reported that the commander of an enemy corps would attempt to take
Chilgok by envelopment. The now rather spread-out units of the 1st Div would
.. SutTendered on 21 September. See 164th MISDl 0930 or its apparent repror!llction ,\'1'15 1468. This
lengthy interrogation report is unusually rich in tactical detail and in LackbfounJ infonllation on the 13th
NK Div and North Korean forces in general. An earlier, briefer interrogation of the same prisioner, ATI5
1293, is concerned mainly with strategic matters.
ORO-T-261
48
UNCLASSIFIED
HUEr
SURE1' UNCLASSIFIED
evidently have to converge sharply upon this avenue of advance. Prisoners of
war stated that the operations of the two divisions were coordinated but offered
no details. On 10 September informants behind enemy lines also reported that
the enemy was having trouble in coordinating the offensive, and that a party of
high-ranking officers was to visit the front. This visit seemingly took place
on 10 September, a quiet day; at 1700 an official-looking party was observed
in the valley between Hill 660 and the Walled City. Two or three enemy tanks
were reported at 1747 at 1153-1456, further south than they had ever been. Dur
ing the night and following morning intense enel'lY pressure was applied.
Wi,,,, World
Fig. 4-View up ihe "Bowling Alley from pOiili just norihwesi of Taegu.
View to the north taken 18 Sep 50 at coordinates 59.4-72.7 (m). Shows the road
which was the axis of enemy advance from Tabu-dong.
Appendix B is a battle order of the 3d NK Div drafted on 11 September.
On 12 September the entire division, which was now opposed by the 5th Cav
Regt, was to attack in coordination with the "motorized artillery regiment"
on its right flank* and the 13th NK Div on its ieft. Since the 13th Div had been
*According to a prisoner captured at 1137-1446 on 16 Septemher, the 65th :\lecz Hegt, l05th NK Armd iliv,
minus tanks, only crossed the Naktong over the damaged hridge at Waegwan at 0600 on 14 September, and
then moved south. 1
ORO-T-261
UNCLASSIFIED
5EfRU
49
,S.(HT
UNCLASSIFIED
rather inactive after its offensive stalled south of Tabu-dong, it would appear
that it was now in condition to take over from the 1st NK Div the general offen
sive in the 8th Cav zone.
Several pOints in the 3d Div order are of note as probably typical of enemy
operations at this time: (a) The friendly defense was considered to be a dls
continuous line of independent outposts. (b) The enemy planned to advance by
taking hill mass after hill mass rather than by slipping through the friendly
defenses. (c) All three regiments were to attack abreast without a reserve
regiment. (d) Artillery was to "protect infantry with continuous fire until end
of an attack.'
According to the 13th Div CofS North Korean units were not limited by
geographical boundaries, and Simply regarded territory in line with their
assigned objectives as their area of operations.. Overlapping of the 1st and
13th Divs as they converged upon Taegu was therefore to be expected. There
is evidence that at least for a brief time Hill Masses 570 and 314 were both
occupied by units of both divisions. This could have been as much the result
of poor control as the execution of a plan, for, according to the CofS, when
the 13th Division made a limited advance toward the end of August it lost all
contact with the regiments, and they were located only after the division staff
officer scoured the hills for them. On the other hand, control of small units
on occasion may have been formally transferred. For example, prisoners from
the 14th Regt, 1st NK Div said they were unfamiliar with the regiment to which
they were assigned on Hill 570 (this was certainly the 19th Regt, 13th NK Div).
There was little if any consideration of strategic withdrawal at any time
in the enemy maneuvering north of Taegu. The CofS stated that news of the
UN landing at Inchon on 15 September had not reached him at the time of his
desertion on the night of 20 - 21 September, and that the only plan for his division
was to defend in place. The course of the fighting confirms this statement.
According to a prisoner from the 1st NK Div, almost daily conferences of its
top command were held during the third week of September when the division
was being enveloped by ROK forces, but suitable defensive tactics could not
be arrived at because of the rapidly changing situation.
COMPOSITION AND CONDITION OF UNITS AT MID-SEPTEMBER
Analysis of information obtained from POWs indicates that by mid-Sep
tember enemy companies were often committed singly or in pairs rather than
as components of entire battalions. Enemy divisions appear to have become
as much pools for the receipt, organization, and expenditure of low-grade man
power as integral tactical units. Such operating procedure would have been
consistent with the enemy's replacement situation at this time. The introduc
tion of rather large groups of recently conscripted and almost untrained North
Koreans appears to have firstoccurredtowardthe end of August, and the large
poo Is of impressed untrained South Koreans were first drawn on at the end of
that month. By mid-September regiments were reported composed of as much
as 70 percent replacements, a considerable proportion conscripted in South
Korea.
*But see the 3d Div battle order, App B, in which regimental sectors are defined.
50 ORO-T-261
UNCLASSIFIED
SHREl
UNCLASSIFIED
Sinn
Morale of th-e surviving trained North Koreans, who formed the hard cores
of units, was probably still high because of the prospect of imminent success,
but the morale of the recently conscripted North Koreans was poor, that of the
South Koreans certainly very low. In the interrogation report ATIS 1149 (Item
7, App A) a North Korean recently drafted into the 14th Regt, an experienced
unit which had trained in Manchuria, stated:
Morale was very low. The men were aware that the majority of tp.em were un
trained and that the units were understrength. Consequently. the men were very reluc
tant to become involved in aggressive warfare. Some of the men refused to advance
while the unit was under fire, and subject saw three men of his platoon being killed by
his platoon leader for this reason. The unit commander had much difficulty in control
ling the troops under fire because the inexperienced men were panicky.
It is not surprising that almost all soldiers who became POWs had been
replacements.
With insufficient weapons and such a high proportion of poorly trained
troops, the enemy appears to have devised appropriate tactics. One inter
rogation report (ATIS 1103; Item 8, App A) describes how recruits, some of
them unarmed, were driven into a suicidal assault before the guns of a second
wave of trained troops. An intelligence study made at that time states that it
was usual for the enemy to attempt to wear down friendly defenses by the sheer
pressure of masses of untrained troops before the decisive attack by more
combatworthy units. The following rather curious statement by a prisoner
(Item 9, App A) suggests how necessary such piecemeal commitment must have
been: POW stated the regiment stops at one place and sends out just one
company to the front line. If the company is wiped out another company is sent
out. Whenever a company is about to be annihilated a runner informs the
battalion commander." Further illustration of this technique may be found in
ATIS 1149 (Item 7, App A).
Replacements were sent up to forward units but apparently not to any
appreciable extent to those in active combat. Thus, as shown below, 14th Regt
units on the 480-m peak on Hill Mass 570 were not appreciably strengthened
during their attacks on Hill 373, even though 19th Regt units just to the rear on
Hill 570 itself were receiving massive consignments of replacements.
In the detailed analysis of unit movements which follows, the enemy's
personnel situation must be kept in mind. Companies were apparently kept in
combat until a very high percentage of casualties had been sustained, but very
often this occurred in a single engagement. Remaining personnel were then
withdrawn as a nucleus to which green replacements could be attached to bring
the unit back to somewhere near normal strength. In the 1st NK Div in general
there was evidently a high degree of depletion around 10 September. One pri
saner, a senior lieutenant and commander of the division's antitank battalion,
stated (ADVATIS 0875) that he heard that the 2d and 3d Regts were ordered to
withdraw from the front line and assemble at Kunwi on 16 September.* As
appears in the detailed analysis, the 14th Regt was in much the same need,
The same document also gives the date as 10 Other POW information I states that the deci
sion to withdraw two regiments to Kunwi was made known in a meetinR of the 1st NK Kiv held on 14
September.
ORO-T-261
51
UNCLASSIFIED
UNCLASSIFIED
SUREl
but replenished its depleted companies near the front at the Walled City while
some of its other companies remained in combat.
An indication of the rapidity of attrition of enemy units is furnished by
the interrogation report (ADVATIS 0936) of a medical officer of the 13th Div,
a major captured on 23 September. He restated the fact 'widely reported by
prisoners that the enemy dead were left where they fell. Front-line soldiers
were not permitted to carry the wounded to the rear, even though there were
not enough medical personnel to collect, evacuate and treat them. Eighty per
cent of the wounded who might otherwise have been saved died because of the
primitive medical facilities and complete lack of vehicular transport at the
regimental level. The major apparently was referring to cases requiring
evacuation, for he stated that soldiers with flesh wounds were treated immedi
ately and sent back to the front. His estimate that the number of wounded was
about the same as the number killed would agree with the following assumed
situation: on the battlefield, one killed in action to every three wounded (as
with the US Army); of every three wounded about one-third wounded seriously
(most of these also die). Thus, of every four enemy casualties, one is KIA,
one is DOW, and two are WIA. In view of the massive flow of replacements,
it is immaterial when the WIAs return to duty.
With such enormous transfusions of replacements flowing into unengaged
units, the total strength of a diVision or regiment on a particular day is not a
good indication of the effective strength of the force opposing friendly units.
The proper quantity would appear to be the number and actual strengths of the
companies in active contact. With such an improvisatory enemy situation it
is difficult to account for individual companies. It appears logical to assume
that newly committed companies had been padded to somewhere near TO strength,
if not to TE equipment. For units that had been in sustained combat a much
lower average strength must be assumed.
Figure 5 is given as a basis for strength calculations. It is based on
testimony of a 1st Div officer captured on 16 September. The many subunits
'in the organizational chart that he omitted, intentionally or otherwise, have
been filled in from statements made by a 15th Div officer captured on 14 Sep
tember. The latter prisoner also estimated the standard strength figures
shown with the different units. These amount to a regimental strength of 2423,
a figure which agrees with the approximate total of 2500 given by other prisoners.
This probably was the assigned regimental strength at the time; the figure of
2810 carried by US intelligence would seem to be authorized full strength. The
number given by the CofS of the 13th Div was 2700. Numerous prisoners
mentioned actual regimental strength figures in the range 1000 to 1200 and
gave the strengths of their own companies as around 60, proportionately low
in relation to the TO company strength of 150 given in Fig. 5.
The omission of subunits from the battalions of the 1st Div shown in Fig. 5
may be significant, and may represent deterioration in the organization due to
lack of weapons, equipment, and trained men. Therefore, where more definite
information is not available,a battalion is assumed to be made up, on the
average, of: 2 combat-worn rifle companies, perhaps incorporating speCialized
remnants, with a strength of 60 each; a newly committed rifle company with
a strength of 120; and a newly committed 82 -mm mortar or machine-gun
ORO-T-261 52
UNCLASSIFIED
UNCLASSIFIED
Slnn
xx
1st NK Div
L.."'T'"--,Regt
Other units attached to division:
45-mm AT -Gun Bn, SU-76 Bn, Engr Bn,
r:::-""'::>I
2d

Arty Comm Bn, Med Bn, Reeon Co, Trans Co,
"-.,...:'" Regt "----'''' Regt
Sety Ca, Gd Sqd
I

IBn

IBn
,
I I L..-....J
1
76-mrn How 76-mm How
Ca [bca I 11Ca

1
600ca
76-mm How* 12Q-mm Mort
20 Ca 53 Ca 50 Ca 70 Ca 100
Trans 5ig SMG
Plat Ca 20 150 Ca 150 Ca 150 Ca 60
82-mm Mort 45-mm AT Gun
Plat 60 Ca
12 Plat 8 Plat 18 Plat 20
Supply HMG
r---I r- -I r- -,
SOURCES
I I Plat I 1 Plat
I I Sqd
Heavy boxes: I Corps Interrogation Report 528-LDI
I___J L __ -I L __ J
0031. POW a Sr Lt, CO 14.\-mm ATRifle Co,
Reeon
45mm ATGun Bn, 1st NK Oiv, captured 161100
Sep 50 at Maegoktong (11581465)
Light boxes and strength figures (at left of boxes):
ADVATIS 0853. POW a Sr Lt, CO 7th Co, 3d Bn,
45th Regt, 15th NK Oiv, captured 14 Sep 50 at
Imgo-myon (vicinity 1190-1450)
*Original says "Gun"
Broken-line boxes: Other
Fig. 5-0rgcnization of the North Korean rifle regiment and, in particular,
of the 1st NK Div, in September 1950.
53 ORO-T-261
UNCLASSIFIED
SEun
$Inn
UNCLASSIFIED
company with a strength of 40. Total battalion strength is then 280; regimental
strength would be 1120.
Information supporting the assumptions above can be found in the testimony
or prisoners captured in the area under study. For example: a prisoner from
the 3d Bn of the 14th Regt(ATIS 1155), which had fought on Hill 570, stated that
as of 11 September, even after replacements were received, the strength of
his 7th Co was only 75 and that the equipment of the company, other than rifles,
then comprised 2 heavy machine guns and 3 light machine guns; the prisoner
from the 5th Co, 2d Bn, 2d Regt captured on Hill 314 on 12 September (ATIS
1142; Item 10, App A) stated that the battalion had a strength of 100 and con
sisted of 3 rifle companies, a heavy machine gun company, and a mortar com
pany, but that the mortars were issued to the rifle companies during engage
ments (since presumably the mortar company itself was so depleted it was
unable fully to use them). The other prisoner who had been on Hill 314 during
the battle on 12 September was from the 2d Bn of the 14th Regt. This battalion
had been badly beaten at Hill 373 and evidently was withdrawn to Kasan. There
on 11 September the prisoner was one of 80 replacements joined to its 4th
Co, bringing the company strength to 120, or almost the TO level. The HMG
Co was brought to a strength of 40 and equipped with 5 machine guns, and both
companies were recommitted to combat, this time on Hill 314.
IDENTITY AND MOVEMENT OF PARTICULAR UNITS
In this subsection an attempt is made to reconstruct from all available
data the identity and movement of the battalions and companies of the 13th and
1st NK Divs after they moved into the Tabu-dong area. Especial attention is
given to dispOSitions on Hills 373, 570, and 314 as of 12 September. Since much
of the information is incomplete, ambiguous, even self -contradictory, and was
often supplied by uneducated farmers only just impressed into enemy service,
it is inevitable that the connections made between the reported facts are often
deVious and tenuous. It is believed, however, that the pattern of events which
has been reconstructed is not in conflict with any confirmed friendly observa
tions of enemy disposition and movement.
The first of the subdivisions which follow is devoted to the units of the
13th Div. It divides naturally into two periods-that before and that after the
lull in the division's activity around 8 September. The second subdivision
follows the respective regiments and battalions of the 1st Div throughout the.
over-all period under study. Table 1 summarizes the location of units of both
divisions on 12 September.
The 13th North Korean Division
Unit Locations through 8 September. Information on the movement of the
13th Div to the Tabu-dong area appears in the report of interrogation of the
S-3 of the 19th Regt, sections of which are reproduced as Item 1, App A. By
19 August, the 23d Regt was deployed 1 kIn southwest of that road junction 3
miles north of Tabu-dong on the northern slope of the hill mass called Yuhak
san, the 21st Regt was deployed 1 kIn to the southeast of the junction, and the
19th Regt was in reserve in the Sangjang-dong area, about 3 1/4 miles to the
ORO-T-261
54
UNCLASSIFIED
UNCLASSIFIED
northwest on the Sangju road. On 20 August, the 21st Regt moved across the
Taegu road to the southeast slopes of Yuhak-santo escape intense shelling, and
the 19th Regt moved up to take its place. The 2d Bn in turn became disorgan
ized by friendly artillery fire, and this contributed to the failure of the division's
attack on 21 August. After the attack, the units fell back to their originalposi
tions and maintained them through the time of the S-3's desertion on 1 September.
TABLE 1
PROBABLE LOCATION OF ENEMY UNITS IN VICINITY OF
HILL MASS 314 ON MORNING OF 12 SEP 50
Location
Unit
Hill Mass 314 480-meter peak 570-meter peak Elsewhere
19th Regt, 13th Div
Ist Bn
Una
2d Bn
Bn, Not 570
b
3d Bn Bnb
14th Regt, 1st Div
lst Bn 2 Co 2 Co unaccounted
2d Bn
Bnb,e
3d B ~ 2 Cac 2 Cac 2 Co to NEa
2d Regt, lst Div
lst Bn Bn
c
2d Bn 3 Co
3d Bn
Unlocated
Total present,
AM, 12 September 1 Bn, 5 Co Remnants 1 Bn (+?b)
8Located on positive evidence.
bLack of direct evidence for location.
cUnits probably destroyed, depleted, or withdrawn in previous fighting
at or near specified locations.
The following logged message gives explicit information evidently furnished
by the S-3 shortly after his capture:
021920K. From G-2. Interrogation of POW of 19th Regiment, 13th Division
reveals:
At 1153.5-1465.7 on trail Is the headquarters of the 13th NK Inf Div [or rather,
the 19th Regt, 12th Div?J.
At 1150.9-1464.4 is the headquarters of the 23d NK Regiment.
At 1149.8-1462.9 [?J is the headquarters of the 21st NK Regiment.
At 1150.2-1466.95 are 5 x 120mm mortars.
At 1150.3-1466.90 are 4 x 76 mm guns.
At 1151.3-1466.0 are 3 x 45mm guns.
At 1151.5-1466.0 are 3 x 45mm guns.l
The rather forward location specified as the site of the 13th Div Hq is
doubtJess that of the 19th Regt Hq, which of course was well known to the pri
soner and was not otherwise recorded. The stated location of the headquarters
ORO-T-261
UNCLASSIFIED
SEU"
55
UNCLASSIFIED
.sHIEf
of the 21st Regt falls just within friendly lines and is probably the site as of 21
August, when the regiment attacked and then fell back somewhat to the rear of
its original positions. All the weapons positions are in the vicinity of the road
junction. It therefore seems very likely that the 13th Div was still astride the
road, with the 21st and 23d Regts on the west side and the 19th Regt on the east.
As may be seen from the strength figures of Table 2 (shown later in this section),
such deployment would have produced the best balance of 13th Div forces across
its axis of advance. Furthermore, a prisoner from the 1st Bn, 19th Regt captured
on 12 September (Items 2 and 3, App A) stated that he joined his unit on 3 Sep
tember northeast of Tabu-dong in the vicinity of Hill 742.
Some information on succeeding movements appears in the interrogation
report (ATIS 1518) on a pOlitical officer, a captain attached to the 19th Regt
45-mm AT-gun Co, who was captured just south of Tabu-dong on 24 September.
The report states that the 19th Regt moved to Tabu-dong around 3 September
and that:
o/a 6 September. All the artillery pieces (122mm howitzers, 76mm mountain
guns, 76mm AT guns and 45mm AT guns) of the 13th Division were In position along
the hills locaied approximately two kilometers south of Tahu-dong. All the infantry
troops had advanced further south at this time. POW's company (45mm AT Company)
was attached to the 45mm AT-gun Battalion, which was directly under the 13th Divis
ion. All the artillery units Waited two kilometers south of Tahu-dong until 22 Septem
ber 1950.
Few prisoners from the 13th NK Div were taken eluring the first phase of
the enemy offensive. The meager evidence which follows is taken from the
several pertinent interrogation reports, indlcates that all three regiments were
close to the front, and tends to confirm the conclusion that the 21st and 23d
Regts initially were to the west of the highway. All three regiments had ap
parently been committed, at least in part, by 8 September.
If the following report can be taken as apprOximately true, it would indicate
that the 19th Regt had ceased offensive operations by 5 September to receive
replacements, and that the now possibly strengthened 21st Regt had moved over
to the east to take over the 19th Regt pOSitions, while the 23d Regt remained
on the west side:
051915. From G-2. Scrappy 2 reports... have captured a prisoner from tbe Medi
cal Detachment, 19th Regtment, 13th Nortb Korean Division. He claims that tbe 19th
Regtment is 3,000 strong and is located at Tabu-dong (1150.1-1460.4). He states tbat
the 21st Regtment is located on Hill 466.3 at 1152.9-1460.4: the 23d Regtment, 13tb Divis
ion is on Hill 449 at 1150.7-1459.5. He states that tbe 19tb Regtment's Headquarters is .
at the village of Tabu-dong: the lleadquarters of tbe 21st and 23d Regtments are located
on top of tbeir respective hills. Five artillery guns were located last night (2400 bours)
at Soya (1150.8-1460.1): five mortars were located on top of Hill 449 at 1150.7-1459.5.
1
This prisoner was evidently the subject of ADVATIS 0720, which states
that a member of the 19th Regt First Aid Plat was captured on 5 September at
1152.0-1459.0, a point along the Taegu road 1 1/4 miles southwest of Tabu-dong.
On 8 September a patrol from Co A, 8th Cav Regt captured a private of
the 1st Bn,21st Regt at 1149.5-1459.5, just south of Tabu-dong and west of
ORO-T-261
56
UNCLASSIFIED
UNCLASSIFIED
the highway between Hills 516 and 449 (164th MISDI 0789). He stated that his
company had previously been attacked at Tabu-dong by US forces.
At 0700 on 8 September Co A, 8th Cav also took a private, (who had sep
arated from his unit) of the 2d Bn, 23d Regt at a point reported as 4 km north
of Tabu-dong but certainly elsewhere-from the Co A location, probably some
where to the west of the highway and 4 km south of Tabu-dong (l64th MISDI
0796)
Unit Locations after 8 Sep.tember. The following report indicates that the
regiments of the 13th Div were still in the vicinity of Tabu-dong on 8 September:
090320. From 8-2, 8th Cavalry. POW picked up at 090220 Sep is a wire man of
the 13th Div Signsl Battslion. The North Korean 19th Regiment is located on and in the
vicinity of Hill 466 (1152.0-1460.4). The 21st North Korean Regiment is located on the
east slope of Hill 516 (1148.9-1459.9) as of 082400. The 13th Division's OP [sic, CP?1
is located by a bridge (1150.8-1461.0) on the northwest side of the road approximately
1,000 yards north from the road Junction at Tabu-<iong (1150.1-1460.3) as of 082400 .... '
The interrogation report on this prisoner contains no tactical information.
The locations of the' 13th Div major units on 20 September were given by
its CofS (l64th MISDI 0930) as follows: 2M Regt south of Tabu-dong at approx
imately 1150-1457, 21st Regt east of Tabu-dong, and 19th Regt along the south
ern wall of the Walled City. The 13 Div CP was at 1150.3-1461.0. The principal
over-all changes in position from 8 to 20 September were, then advance of the
23d Regt to the south and displacement of the 19th Regt to the east. More
detailed information follows.
At 1830 on 10 September a rifleman of an enemy patrol was captured just
north of Hill 225 by Co B, 8th Cay. The prisoner, who was wounded, said that
he belonged to the 19th Regt and that the regiment was on both sides of the
'Taegu-Tabu-dong highway in the vicinity 1153.0-1456.6, and that the 21st and
23d Regts of the 13th Div were to the west. This information would place the
19th Regt on Hill 263 and perhaps Hill 351, as well as on Hill 570.
A prisoner from the 3d Bn, 23d Regt (see below), who was captured on 12
September in the vicinity, 2 1;2 miles southwest of Hill 373, stated that the 21st
and 23d Regts were located to the north of Hill 307, and that the 23d Regt had
a strength of 1400. The only possible location for such a large unit would be
Maebong-San, Hill 624. The presence there of the 23d Regt would mean that
the 19th Regt was deployed no further to the east than Hills 351 and 263.
On 10 September a private deserted the 3d Co, 1st Bn, 19th Regt after
it had undergone artillery and air attack on Hill 570. He surrendel'ed at 0800,
12 September to Co F, 8th Cav at the southwest edge of Hill 373 near the high
way. Reports of two interrogations of this soldier are available, and both are
reproduced in full in App A. Item 2, App A gives the interrogation by the 1st
Cav Div, Item 1I that by ATIS (1158). The prisoner, a replacement from North
Korea, said that he was assigned to the 1st Bn on 3 September at a hill thought
by the interrogator to be Hill 742, somewhat north of Tabu-dong and 2 miles east
east of the highway. The 1st Bn departed from this location for the Walled City,
to leave for Hill 570 on 9 September with an estimated strength of 200. The
1st Bn Hq was located at the north end of the Hill 570 ridge. According to the
prisoner, when last he had been with his own platoon, 10 of the original 17
men were left.
ORO-T-261
S{(RET UNCLASSIFIED
57
UNCLASSIFIED
This private was one of a group of 1000 replacements from North Korea
who passed through Wonju on 28 August. Further information was obtained
from 3 South Koreans who had been impressed at Wonju into another group of
1000 and included in a detachment of 300 also dispatched from there around
29 August as replacements for the 1st Bn, 19th Regt. The 1st Cav Div report
on the interrogation of these men is reproduced in full in App A as Item 4. The
3 men, with 27 others, arrived at Hill 570 on 10 September and were assigned
to the Mort Co, 1st Bn, 19th Regt, bringing its strength up to 60 men. They
stated further that the 1st, 2d, and 3d Bns of the 19th Regt were all in the vicinity
of Hill 570, and that the strength of the entire regiment, including replacements,
was about 1000 men. This figure is confirmed by enemy statistics compiled on
11 September (see Table 3 later in this section) listing for the 19th Regt a
strength of 1024 (846 privates, 148 NCOs, and 30 officers).
The fact that on 12 September Hill 570 was occupied by one enemy battal
ion with a strength of 200 or 250 is confirmed by the following report:
121330. Dlvarty S-2 reports: ... At 1154.9-1456.1 (saddle on Hill 570) [on pres
. ent map these are coordlnates of a point on the east side of the 570-m peak] 100 people
dug In (west side of saddle). Also, on east side of saddle ammo boxes, 50 soldiers and
mortars dug in. At 1154.9-1455.0 (top of Hill 570) [on present map, the coordinates of
the 480-m peak] 100 people dug in....'
The estimated total of 250 enemy personnel observed would be just right
for one battalion of the 19th Regt at this time. The observation of "50 soldiers
and mortars agrees remarkably well with the prisoners' statement that the
60-man mortar company, with certainly 1 mortar (and probably at least 3),
was deployed on the ridge within a 200-m circle.
As a result of the attack on Hill 570 on 14 September by the 2d Bn, 8th
Cava number of additional prisoners from the 1st Bn, 19th NK Regt were taken.
The subject of ATIS 1165 was captured on 14 September, and may be one of the
three prisoners reported above; the subjects of ATIS 1193, 1203, 1215, 1218,
and 1237 were all taken on 15 September. Several of their statements on the
8th Cav Regt attack are of interest. ATIS 1203 states:
ola 10 Sep. 19th Regiment Headquarters located apprOximately three kilometers
south of Kasan. Tbe 70 men were separated into squads and took positions along the hills
located apprOximately four kilometers south of Kasan. The troops dug foxholes and re
mained there for about four days.
14 Sep. US troops started to attack witb artillery, mortars, machine guns. The
1st Company [of the 1st Bn] retreated to the 19th Regiment Headquarters area (three
kilometers south of Kasan). After the firing ceased in the evening, the troops advanced
toward the US positiOns. About 50 men, Including the POW, came as close as 50 meters
to the US positions.
14 Sep. [night of 14-15 Sep]. The 50 men were discovered by US troops. The US
troops fired machine guns, and many North Koreans were killed. Some of the North
Koreans escaped, hut the POW and two others were captured hy US troops at the point
1154.6-1456.1 [570-m peak].
ORO-T-261
58
UNCLASSIFIED
S.EER!T
UNCLASSIFIED
On the night (14 Sep) [14-15 Sep] that POW and two others advanced to the. US posI
tions, POW heard from his platoon leader that most of the men in the 1st Battalion, 19th
Regimeni were killed.
The enemy withdrawal in the face of the attack is confirmed by the subject
of ATIS 1193, a soldier of the 2d Co, 1st Bn, 19th Regt: "14 Sep. At 1500 hours
the regiment was attacked by alrcraft, and at the same time artillery shells
began to fall in the area. The shelling and air attack lasted until 1900 hours.
Losses sustained unknown. The 19th Regiment withdrew immediately after
the attack lifted."
What remains is to establish the location of the 2d and 3d Bns of the 19th
Regt. One interrogation report, 164th MISDI, is available on the 2d Bn, 19th
Regt. The subject, a second lieutenant of its 5th Co, captured on 20 September
when he deserted his unit, stated that from 28 August to 20 September he was
in the mountains 2 km southeast of Tabu-dong in the vicinity of the Walled City.
It is then quite possible that from early September-not 28 August-the 2d Bn
remained on the abutment of the Walled City ridge lying just north of Hill Mass
570, and was the reserve battalion of the 19th Regt.
The 3d Bn, 19th Regt may have occupied the fingerlike hill mass (Hills
449, 351, 263) which runs southeast from Tabu-dong. In a recent communica
tion, the former commanding officer of Co B, 8th Cav, which occupied Hill
225 and the unnumbered hill just to the north, stated that at the time one rein
forced enemy company was estimated to be on Hill 263. The enemy force stayed
there until a platoon of Co C, 8th Cav occupied Hill 263 without casualties about
19 September.
Further information on the 19th Regt was supplied by a prisoner from the
3d Bn, 23d Regt, whose capture on 12 September was recorded as follows:
121030. POW captured in vicinity 1152.8-1451.6 [west side of the Taegu road,
south of Hill 373] was member of 3d Battalion, 23d Regiment. Reported his battalion at
200 strength moved to new location 1600 meters northeast of 1151.0-1452.4.'
121300. From Scrappy 2 [8th Cav Regt S-2]: POW states that at 121030 the 3d
Battalion of the 23d Regiment, estimated strength of 200, was located In grid square
1150-1452. Report Is unconfirmed. POW also states that as of 121100 the 19th Regiment
of the 13th Division, with estimated strengtb of 1000 was on Hill 570 (1154.5-1456). POW
stated further that the 3d Battalion of the 1st Infantry Division [sic] was moving today to
join forces at 1156-1455 [northern part of Hill Mass 314]. . . Time and place where
3d Battalion departed from unknown. I
Unfortunately, no interrogation report is available for this prisoner. It
is thought for the following reasons that the reported move to Hill 314 was to
be made by the nearby 3d Bn, 19th Regt of the prisoner's own 13th Div: (a)
Within the 1st Div, the 3d Bn would have to be that of the 2d or 3d Regt, since
the 3d Bn, 14th Regt has been accounted for (see below). The prisoner would
probably have no knowledge of these units, as they were disposed to the east.
(b) The prisoner meant the 1st Regiment was following the common enemy
usage of referring to the lowest numbered regiment of any division as the
"1st Regiment." .
ORO-T-261
UNCLASSI FlED
SEREr
59
UNCLASSIFIED
SEhEr
The following observations confirm that some battalion, presumably the
3d of the 19th Regt, 13th Div, did move from the 13th Div area toward Hill Mass
314 on 11 September:
111415. From 1st ROK: We have an OP on Hill 660 that observed troops and sup
plies estimated at one battalion moving on trail 1152.7 -1457 .5. [These are tbe coordinates
of Tongmyongwon, the village at the nortbwest edge of Hill 570 where the highway meets
the trails that lead to Hill 314.]
111500. From Artillery Liaison Officer with ROK: ... Estimated one battalion
of North Korean troops moving east along road at 1152.70-1457.55 to 1155.0-1456.8 [from
Tongmyongwon along the trail that skirts the northern rim of Hill 570 to a point 3/4 mile
from the northwest edge of Hill Mass 314].1
Further observation of continued enemy movement in this direction would
have been masked to observers on Hill 660 by Hill 314 itself. Since the enemy
force was definitely coming from the 13th Div area, and since the 3d Bn of the
19th Regt remains unlocated, it is assumed that they were one and the same.
It is further assumed that on the evening of 11 September, the 3d Bn, 19th Regt
moved onto Hill 314.
The assumption that units of the 13th Div moved to Hill Mass 314 on 11
September is strongly supported by another move in this direction observed
on 12 September. A report states:
121330. Divarty 8-2 reports: ... Town at 1152.6-1457.5 [Tongmyongwon] was
full of soldiers; soldiers moved from town on trail to 1153.6-1457.8 [village of Tumo at
the northernmost point of Hill 570], then along trail southeast to 1154.6-1457.3 [vicinity
of the cemetery, 1 mile northwest of Hill 314].1
Since this move was made during the battle on Hill 314, it could well have been
a preparation to send up reinforcements.
The seeming reinforcement of units of the 1st Divan Hill 314 (see below)
by units of the 13th Div would only repeat the situation on Hill Mass 570. Further
more, ADVATIS 0875 gives the statement of the commander of one of the com
panies of the 1st Div AT Bn that he had heard that the 19th Regt, 13th Div had
"reinforced" the 1st Div. According to the CofS of the 13th Div, the 19th Regt
did not replace the' 1st Div, 2d and 14th Regts at Kansan until 19 September.
Therefore, any reinforcing prior to 17 September, the date of the prisoner's
statement, must have been south of the Walled City, -Le., on Hill Mass 314,
Hill 570, or both. The presence 'of troops of both divisions on Hill Mass 314
is also suggested by the fact that toward the end of the battle they fled or with
drew to Hill 570 as well as to the north and east, i.e .,back to their parent units
and in the directions from which they had come.
The plan of attack on Hill Mass 314 by the 3d Bn, 7th Cav Regt (App E)
stated that the hill was held by units of the 13th NK Div, but the basis for this
estimate is not known.
There is little further information on the 23d Regt to add to that for its
3d Bn on 12 September cited above. On 23 September a private of its ~
(ATIS 1465) was captured. He stated that the last known strength of the 23d
Regiment was approximately 300.
No information is available on the 21st Regt.
ORO-T-261 60
UNCLASSIFIED
,SHREl
Sinn UNCLASSIFIED
The 1st North Korean Division
The 1st Div arrived at Kunwi on' 17 August (164th MISDI 0824) and was
intensely engaged with the ROK Army. The objective of its 14th Regt, and
probably of the entire division, was the mountain Palgong-san (Hill 1192, 10
miles east-southeast of Tabu-dong and 10 m ~ l e s northeast of Taegu). At
1157-1472 on the road south of Kunwi one valley leads southward toward the
Walled City and another runs southeast toward Palgong-san. Both valleys were
used, but the main weight of the division was sent into the one leading south
ward. The division's CP, Arty, and'SP guns were all located in the vicinity of
Kogok-tong (1153.3-1468.2). According to a prisoner from the 1st Div, SP-gun
Bn who was captured on 20 September (ATIS 0871), the battalion advanced to
Tabu-dong on or about 1 September, and withdrew on or about 8 September.
Most of the 14th Regt moved on through the first Su-dong, 3 miles north of the
Walled City, to confront ROK units at the second SU7dong, 2 1/2 miles tothe south
east.* The 2d Regt and the 2d Bn of the 14th Regt appear to have moved along
the southeasterly valley. The 3d Regt seems to have followed and moved be
tween the other regiments. Thus, by 6 September, when all three were reported
attacking abreast, the line from west to east was: 14th Regt, 3d Regt, 2d Regt.
One rather puzzling bit of evidence suggests that some troops of the 3d
Regt, perhaps only a specialized unit, were sent to the Tabu-dong front to back
up the 19th Regt, 13th Div. At 1545 on 4 September a tank dozer of Co C, 70th
Tk Bn was sent up to the 8th Cav Regt forward positions. Its loss was reported
as follows: "One EM that was with the dozer from Company C reported to the
CP with the information that the dozer accomplished its mission and threw a
track while trying to turn around. Enemy antitank, mortar and artillery fires
were falling. The crew abandoned the tank.-
1S
On 22 September, after Tabu-dong had been retaken, the disabled dozer
was found north of the village. The following statements, apparently in English,
were reported by telephone as painted on it: " ...on side: 'Sep 4th Destroyed
by 3d Regiment, 1st Division'; on dozer blade: 'Victory is with the People
Army. Taegu and Pusan Will Fall with the Speed of Winds.'15
As noted above, the 1st Div SP-gun Bn was said to have moved to Tabu
dong early in September and to have remained there till around 8 September.
Around 31 August, after the division received 2000 replacements, it was
stated by prisoners that all three regiments were attacking abreast, and also
that the division's mission had in some way been changed; The change appears
to have been the reassignment of objectives directly to the south of the 14th
Regt and than rather later, to the 2d Regt, since a thrust of this direction was
obviously necessary after the 13th Div offensive had stalled just south of Tabu
dong. Palgong-san, however, was still being attacked by the 2d Regt as late
as 10 September. All prisoners from the 3dRegt were taken east of the Walled
City. It appears almost certain that the 3d Regt was never engaged with US
forces. According to the 13th Div Cofs, the 19th Regt moved onto the southern
slope of the Walled City on 19 September, replacing the 14th and 2d Regts, which
had occupied this area, and which now moved back to a point 1 km southwest
of Kogok-tong .
The two villages have the same name and are less than two miles apart.
ORO-T-261
UNCLASSIFIED'
HCRE.
61
UNCLASSIFIED
S5EtrET
Information furnished by POWs on the movements of the various battalions
of the 14th and 2d Regts is given below. The 14th Regt was also designated the
1st Regt. At the same time, one prisoner might report his regiment as the
14th, and another, from the same company, as the 1st. Wherever the designa
tion "1st" was given, it has been changed to "14th."
A private of the 1st Co of the 1st Bn, 14th Regtwho surrendered to ROK
forces near the Walled City on 31 August, said he had joined his unit the same
day at a point 4 miles north of the Walled City (ATIS 914). On 3 September a
member of the 2d Co of the same battalion deserted, and surrendered at 0800
on 4 September to the ROK Army at Yongchu (1158.5-1459.9). It is evident
that the battalion did not move far, if at all, between 31 August and 3 September.
Its continued presence in the 8th Cav area is indicated by the statements of the
subject of ADVATIS 0747, a private of the 1st Bn HMG Co. This unit, formed
entirely of North Korean replacements, joined the battalion on 3 September.
After an engagement with US forces on 6 September (in which the company lost
15 of its 80 men), the prisoner sought out US forces, to surrender on 7 September
at 1153.3-1453.2, a point on the road to Taegu near Hill 373. The status of the
battalion on 2 September is suggested by the following data on its 1st and 2d
companies (furnished by the prisoners cited above):
1st Bn, 14th NK Regt
Strength data
1st Co 2d Co
Original strength
Strength 24 August 50
Replacements 26 August 40
Losses from ROKA, 27 August 20
Strength 28 August 70
Strength 2 September
120
98
These two companies, if not the entire 1st Bn, appear to have been with
drawn to the north to receive replacements, and then, on 12 September recom
mitted on Hill 314. The subject of ATIS 1418 (Item 5, App A), a private from
the 1st Co, 1st Bn, 14th Regt, was captured by ROK forces on 19 September at
a place specified as Namsan-dong, 3 1/2 miles northeast of Hill 314. Probably
*The following explanation of regimental designations within the 1st Div appeared in OB Annex to G-2
~ U S A K P/R No. 61, 11 September 1950: "1st Rifle Division: Originally, the infantry regiments of the
divi!;Jion were numbered 1st; 2d, and 3d. Upon recall of Korean personnel serving with the Chinese Commu
nist Forces and subsequent activation of the 6th Rifle Division, a regiment-the 14th Infantry-of the 6th
Division was transfetted into the 1st Rifle Division. This was done for the obvious purpose of boosting
combat effectiveness ot the 1st Rifle Division through integration of battle.wise and politically indoctrinated
personnel of the 6th Hifle Division. In turn, a regiment-the 1st Infantry-of the 1st Division was assigned
to the 6th Division. Both regiments retained their original fonnation and personnel. Further, recent reports
indicate that the original regimental designations were also retained. This fact coupled with the North
Korean system of arbitrarily referring to the lowest numbered regiments as the 1st Infantry, etc., has caused
endless confusion in the identification of infantry regiments of the 1st and 6th Rifle Divisions. Although
the picture still remains unclear, for the purpose of clarification, infantry regiments of the 1st Rifle Division
will hereafter be referred to as the 14th, 2d, and 3d Infantry."
ORO-T-261
62
UNCLASSIfiED
.SHI'ET
UNCLASSIFIED
Namwon-dong is meant, as this village, just northeast of Rill 314, was reached
by the advancing 1st ROK Div on 19 September, whereas the division was already
1 3/4 miles beyond Namsan-dong on 17 September. Furthermore, Namwon-dong
agrees with the prisoner's statement about an attack that must have been on Rill
314, since only here did US forces attack on 12 September:
12 September. 1st and 2d Companies advanced to a point 16 kilometers north of
Taegu and took up positions on a high hill. The hill was attacked by US troops with"
machine gun and mortar fire, and the two (2) companies fled from the hill towards the
north. POW deserted and hid in a village at the foot of the hillfor a period of a week.
On 24 August a soldier oi the 5th Co of ihe 2d Bn, 14th Reg!; was taken in
the vicinity of Palgong-san (ADVATIS 0714). On 26 August a private from the
same company was captured by the 1st ROK Div at Taesa-dong (1161-1467.7),
a village along the valley leading to Palgong-san (l64th MISDI 0781). A trained
soldier, he had on 23 August rejoined his unit at this point, where, he said it was
being reorganized. At 0900, 31 August a member of the 6th Co, 2d Bn was cap
tured at 1127.0 [an impossible coordinate, perhaps a juxtaposition of 1172.0]
1462.4 after he had separated from his unit (ATIS 884). If the first coordinate
actually was 1172.0, it would again appear that the battalion was well to the
east of Tabu-dong on 3i August. The prisoner stated further that on 3i August
the 6th Co had a strength of 50; the 2d Bn, 175, and the 14th Regt, 1000.
A soldier of the 8th Co of the 3d Bn, 14th Regi was taken prisoner by Co E,
8th Cav Regt at 1030, 6 September at 1153.8-1454.3, a point on the Taegu road
at Rill 570 (ADVATIS 0725). It was evidently this prisoner who supplied the
following preliminary information:
061715. From G-2. Elements of the 1st North Korean Division .are hitting our
right flank. The road block is now an attack. 300 enemy are located at 1154.2-1455.0,
since 1130. We have a POW from the 3d Battalion, 1st [14thl Regiment, 1st Division,
who states: (a) the regimental headquarters of the 1st Regiment, 1st Division is on top
of Hill 570 (1154.6-1456.1), (h) the 3d Battalion, 1st Regiment has 6 HMG, 3 mortars,
which are at 1154.2-1455.0, (c) the mission of the enemy 3d Battslion was to destroy
US artillery located at former "B" Battery, Spirit [99th FA Bn], (d) three regiments of
the 1st North Korean Division are attacking abreast to capture Taegu, (e) the CP of the
1st NK Division was in Wall City on 4 Septemher, (f) the ammo supply is good and soldiers
are getting two meals a day, (g) the 7th and 8th Companies of the 3d Battalion, 1st Regi
ment, 1st Division came down hill 1154.2-1455.0 at 0530 today and went to road with no
resistance; companies were dispersed by our MG fire; the two companies were, at 061130K
Sep, at bottom of hill on west slopes, (h) he has seen no Chinese in his battalion.
1
After the engagement on 6 September, Co 7 and Co 8 of the 3d Bn appear
to have been withdrawn to receive replacements. One of these replacements is
the subject of ATiS 1147, which reports that he was a member of the 7th Co,
3d Bn, 14th Regt and was captured by ROK forces on 19 September in the vicinity
of Palgong-san,an 1192-mmountain 5 miles east of Rill 314. The report states:
8 September. 800 men arrived at Taedun (1150-1460) [coordinates of general area;
more exact location is 1156-1461] and were assigned to various units of the 1st Division.
Details unknown. Fifteen men, subject included, were assigned to the 7th Company, 3d
Battalion, 14th Regiment. Location of other units, 14th Regiment unknown. POW issued
weapons and uniforms. They came from unit
ORO-T-261
SHRn
UNCLASSIFIED
63
UNCLASSIFIED SURlT
12 September. The 7th Company was deployed on the ridges approximately 4 kilo
meters south of the castle (KASAN). Disposition of other units unknown. The 7th Company
did not make any contact with the enemy hetween 8 September and 12 Septemher. At dawn
12 September, the 7th Company was surprised by ROK forces and suffered heavy casualties.
Losses unknown. POW taken prisoner.
The 7th Co probably was located on 12 September at Hill 783, 4 kIn east
southeast of the Kasan enclosure, for by 12 September ROK forces had reached
this point in their offensive. The 8th Co, 3d Bn, 14th Regt is established in
proximity with the 7th Co on 9 September by the testimony of three prisoners
who participated in the same raid. Their statements also confirm that the
same regiment was known to enemy soldiers as both the 1st and the 14th. On
9 September in the vicinity of the Walled City a group of 15 officers and enlisted
men was organized to raid Taegu airport. This mission fits in with the assumed
location of the companies. During the advance of the group, four men were
wounded in a fire-fight, and on 11 September the raid was broken up by ROK
forces with four enemy killed and seven taken prisoner. Interrogation reports
are available for three of the prisoners. The subject of ATIS 1273 was a junior
sergeant of the 7th Co, 3d Bn, 1st (sic) Regt; the subject of ATIS 1282 was a
private from the 8th Co, 3d Bn, 14th (sic) Regt; the subject of ADVATIS 0869
was a private of the 8th Co, 3d Bn, 1st (sic) Regt.
The iirst iniormation on Co 1 and Co 2, 3d on, 14th Regt is as oi 10 Sep
tember, the date a replacement from North Korea save for his assignment to
the 2d Co north of the Walled City. He was captured in the vicinity of Hill 401
on 12 September. The interrogation report, ATIS 1149, which is somewhat
muddled and requires considerable interpretation, is reproduced in full as
Item 7, App A. From the point of capture it appears that the actions described
by the prisoner were the skirmishes south and west of Hill 373 reported by
friendly forces on the evening of 10 September and again on the evening of 11
September:
o/a 11 Sep 50: 1st and 2d Companies, 3d Battalion were ordered to march south,
objective unknown. Remaining units of the 1st Regiment remained at the above location
[the Walled City. As noted above, the 7th Co and 8th Co, 3d Bn remained here.J The
1st and 2d Companies marched south through the castle enclosure to a ridge approxi
mately two kilometers south of the castle [presumably Hill Mass 570J. At this point at
about 2000 hours, the 2d Company was ordered to proceed further south, while the 1st
Company remained at the position. The 2d.Company proceeded south to a point approxi
mately five kilometers south of the castle and was ambushed by UN forces. In the ensu
ing battle the strength of the 2d Company was reduced from 60 to 14 men. The remain
ing 14 men retreated to the position of the 1st Company. At about 2200 hours the 1st
Company with the 14 men of the 2d Company proceeded southeast from its position to a
point about five kilometers south-southeast of the castle. where it was also ambushed.
The engagement lasted until about 2400 hours. At the end of the engagement the combined
strength of the 1st and 2d Companies was apprOximately 70 men. When the engagement
ended. subject remained concealed. He then became separated from his unit.
Further information on the 1st NK Div was obtained from a deserter from
the 19th Regt who was later captured by US units. While hiding in the village
of Kwiam-ni on the northwest side of Hill 373 on the evening of 11 September,
this soldier observed a group of 30 soldiers of the 1st Div withdrawing to the
ORO-T-261 64
UNCLASSIFIED
S ~ E R O
SRR'ET
UNCLASSIFIED
north, and was told by one of them that the area belonged to the 1st Div (1st Cav
Div Interrogation Report 0085; Item 2, App A).
On 7 September friendly observers estimated that two enemy platoons were
on the northern part of Hill Mass 314. The first contact here was on the night
of 8 - 9 September, when the 16th Recon Co was attacked at its positions around.
Knob 2 by a group of unspecified size. In this intense action the Hecon Co with
drew. When it reoccupied the positions during the day, it was confronted by what
was estimated to be a strong enemy platoon. Evidently one of the original two pla
toons was still intact. On the night of 9 - 10 September the enemy attacked in
company strength. After this action, 52 enemy dead were counted in the vicinity
of Knob 1. According to the assumptions discussed earlier, under "General
Situation and Enemy Intentions," total enemy casualties were probably around
200, the equivalent of almost 2 companies.
Taking over the mission of capturing the entire hill mass, the 3d Bn, 8th
Cav Regt reported in its attack on 10 September, that it was faced by one enemy
platoon apparently backed up by two companies. It would seem that two fresh
companies had COme up to reinforce the remaining equivalent platoon, and that
an entire battalion of four companies had now been committed.
The only mention of the identity of this battalion is this message: "111515.
Unit f i g h t i ~ Blue [3d Bn, 8th Cav Regt] is 1st Battalion, 2d Regiment, 1st Divi
sion (NK)." The basis for this report is not known. No interrogation reports
are available on members of this battalion.
In the fighting on Hill 314 on 11 September, casualties were fairly heavy
in the 3d Bn, 8th Cav Regt. It is assumed that casualties to the enemy were
also high, that the effectiveness of enemy units was destroyed, and that surviv
ing troops of the 1st Bn, 2d Regt were relieved by morning of the next day.
According to ADVATIS 0711 the 2d Bn, 2d Regt attacked the Walled City
on 2 September, moved on, and deployed approximately five hundred meters
south of it on 5 September. The prisoner, a So lIt/! Korean private of the battal
ion's 82-mm Mort Co, had been assigned to the company on 1 September, and
was captured by ROK forces south of the Walled City at 1400, 7 September.
All or part of the battalion then apparently moved on to the east. The G-2
situation map for 9 September shows the 2d Regt, 1st Div making a two-pronged
attack on ROK forces northwest of the 1192-m mountain called Palgong-san,
6 1;2 miles east of Hill 314. On 10 September ROK forces took a prisoner at
approximately the site of these attacks. He was from the 6th Co, 2d Bn, 2d
Regt. The report of his interrogation by US forces (ATIS 1103) is given as
Item 8 in App A.
The latter prisoner, also a replacement from South Korea, joined the 2d
Regt on 8 September 8 km north-northwest of Palgong-san. He said that the
entire 2d Regt, with a strength of 1200 (including 200 replacements) marched
south at 1000 9 September and occupied the summit of Palgong-san (probably
really the high ridge running north from the summit). Here, at 0530 on 10
September, the prisoner's company, consisting of 100 replacements was driven
into the assault before the guns of 50 tralned soldiers and dispersed. It is not
known whether the nucleus of the 6th Co survived.
On the afternoon of 11 September and night of 11 - 12 September new enemy
forces came up on Hill Mass 314 from the north, northeast, and northwest. The
preparation for reinforcement from the northeast is suggested by the following
ORO-T-261
UNCLASSIFIED
SHIET
65
UNCLASSIFIED
report: "111500. From Artillery Liaison Officer with ROK: 200 North Korean
troops unloading supplies at 1158.8-1456.0 [1 1/4 miles northeast of Hill 314] ..,
From this direction, the 1st Div is thought to have sent up three companies
from the 2d Bn, 2d Regt (and two companies from the 1st Bn, 14th Regt-see
above).
The prisoner captured on Hill Mass 314 by the 3d Bn, 7th Cav Regt on 12
September was a private from the 5th Co of the same 2d Bn, 2d NK Regt. Two
reports of interrogation of this soldier are available, one by the 1st Cav Div
and one by ATIS (1142). Both are reproduced in full as Items 9 and 10 in App
A. They are discordant and internally inconsistent, especially with regard to
locations, directions, and distances. The inconsistency is not surprising, as
the prisoner was from extremely distant North Korea and had just arrived in
the battle area, traveling always by night. His reported capture by Co I, 7th
Cav Regt agrees with firsthand information supplied by the commanding officer
of Co I. The two reports are in complete agreement as to the prisoner's organ
izational affiliations. Though the movements of the prisoner's unit are hard to
trace on a map, the following general data seem factual: marching south over a
mountain road, the prisoner's group of replacements reached the 2d Regt at
0400 on 11 September at a mountain northeast of Taegu. After assignment, the
prisoner, with the 85 men of the 5th Co, left the hill at 1800, 11 September.
At 0500 on 12 September, the company arrived at Hill 314, deployed in a wide
area, and dug foxholes.
A second enemy soldier who had been on Hill 314 during the battle on 12
September was taken prisoner two days later when a patrol from CO I, 7th Cav
Regt moved from Hill 314 across the Taegu road onto the eastern slope of Hill
Mass 570. A first report states:
140825. From 8-2, 7th Cavalry: A POW captured last night on Hill 314 (1456.4
1453.8) was from the 2d Battalion, 2d Regiment, 1st North Korean Division. He states
the enemy had 700 men on Hill 314 prior to the attack of the 7th Cavalry and the enemy
suffered 200 KIA with many of the remainder WIA. The unit is now located in the temple
at 1157.2-1457.9....'
A later message appears to refer to the same prisoner but differs some
what:
140915. From 8-3, 7th Cavalry: OUr patrols were about 1000 yards behind Hill
570 (1154.8-1456.4). We took a POW who had been on the hill attacked by our 3d Battal
iOD. He is from the 3d [sic] Battalion, 2d Regiment, 1st North Korean Division. He
states that at least 200 in his battalion were KIA and most of the rest wounded, his bat
talion commander is a Chinese Communist, and that elements of the Division are at
1157.2-1457.9 trying to get organized.'
Extracts of the one interrogation report that might represent this prisoner
are reproduced as Item 11, App A. He was from the 4th Co, 2d Bn, 2d Regt of
what he said was the 14th NK Div. Up to this time the 14th Div had never been
reported in contact with UN troops, and according to the assistant commander
of the 45th Regt, 15th NK Div (ADVATIS 0958) was used for security around
Seoul. Since the prisoner had only joined his unit on 11 September, it seems
certain that he had mistaken the designation of the neighboring 14th Regt as
ORO-T-261 66
UNCLASSIFIED
SHRn UNCLASSIFIED
that of his own parent division. It is therefore assumed that he was from the
2d Regt, 1st Div. The 5th Co and HMG Co of the prisoner's 2d Bn, as well as
his own 4th Co are then placed by the prisoner's testimony on Hill 314 on 12
September. Since he gave strength and equipment figures for 14 September,
as well as for 11 September, he must still have been with the battalion when
it reoganized after the battle. (See "Strength and Casualties on Hill Mass
314," later in this section.)
The subject of 164th MISDI 0784, a North Korean conscriptee, was one of
2000 replacements received by the 1st Div. On 28 August he was assigned to
the 7th Co of the 3d Bn, 2d Regt at a point 7 kIn southeast of Kunwi; he was
captured on 30 August 9 kIn southeast of Kunwi. The company was said to have
a strength of 120 and the battalion a strength of 500 (including 300 replacements).
Except for what seems to be a typographical error in one of the reports cited
above, there is no later recorded mention of the 3d Bn, 2d Regt. It apparently
remained to the east of the 1st Cav Div zone.
The 3d and 15th North Korean Divisions
The location of all three rifle regiments of the 3d NK Div in the
1145-1452 on 11 September is definitely fixed by the captured attack order of
the division (App B). At least two of the regiments had been further to the
north. The evidence from the few relevant interrogation reports follows.
Date of Reference or
Prisoner's unit capture Data on capture interrogation report
1st Bn, 7th Regt, 3d Div 041530 Sep At 1143.7-1454.7 Killed
l
2d Bn, 8th Regt, 3d Div 051200 Sep 4 km NE of Waegwan ATIS-971
3d Bn, 8th Regt, 3d Div 7 Sep By Co G, 7th Cay
HMG Bn" 7th Regt, 3d Dlv 7 Sep At 1143.8-1457.9 ADV ATIS-0746
1st Bn, 9th Regt, 3d Div 8 Sep At 1143.8-1457.9
1st Bn, unkn Regt, 3d Div 8 Sep At 1143.8-1457.9 ADV ATIS-1743
As the division had fought at and to the south of Waegwan at the end of
August, it evidently had displaced to the north and east by 4 September. One of
the 3d Div prisoners captured on 7 September stated that a regimentai CP was
located at 1147.7-1458.6.
The prisoner captured on 5 September (ATIS-971) stated that elements of
the 15th NK Div were on the 3d Div left flank. The only additional evidence of
the presence of units of the 15th Div is the following:
041700. From G-2. Skirmish-2 reports that two POWs were captured in their area.
One is wounded and will probably not live to get to Division. He had quite a. bit of infor
mation. He is a FO captured on Hill 464 (1143.7-1457.9). He estimates [that there are]
1200 enemy on Hill 518 and back of the ridge witb120mm mortars at 1141.8-1461.2,
also states there are six 82 mm mortars, three 120mm mortars on north side oi Riii
518, one 120mm mortar 1141.8-1461.2. POW states there is plenty of ammo for the
mortars, states that one regiment or 2d Battalion [sic], a total of 1800 troops, is located
west of river In town by railroad track (coordinates unknown at present). POW is from
1st Battalion, 28th Regtment, 15th North Korean Division.
ORO-T-261
UNCLASSIFIED
67
SHin
UNCLASSIFIED
Sapphire [77th FA Bn) captured the other POW in the vicinity 1141.3-1458.2. He is
from the 1st Battalion, 28th Regiment. POW came to Hill 464 last night hut didn't want
to fight. His captain threatened to kill him if he didn't fight. POW escaped when his
captain went to sleep. He has substantially the same information as the above POW. He
also states there were 1800 enemy in the vicinity 1141.8-1461.2. The 77th FA Battalion
is firing into these areas.
1
Supporting interrogation reports are lacking. The bulk of the 15th NK Div
unquestionably was located to the east in contact with the ROK Army in the
vicinity of Yongchon. Many prisoners were captUred from its 45th, 48th, and
50th Regts, from the Arty Regt the SP-gun Bn, and other divisional units. The
parent division of a 28th Regiment is not known to the writer and may perhaps
be an earlier or mistaken designation of another unit. Whatever the designation,
an entire regiment may have been attached to the 3d NK Div, just as the 15th
Div's 45th Regt was attached to the 13th Div for a period in the latter part of
August. (ADVATIS 0670 directly confirms this attachment, as reported in
Item 1, App A.)
STRENGTHS, REPLACEMENTS, AND CASUALTIES IN THE 13TH NORTH
KOREAN DMSION
The personnel situation of the 13th NK Div can be outlined from the begin
ning of the war to 31 August from information supplied by the S-3 of the divi
sion's 19th Regt, who surrendered on 1 September (see Item 1, App A). The
division was formed from conscripted troops in March 1950. It crossed the
38th parallel on 26 Jun 50 in the wake of the general enemy attack on South
Korea, but did not experience resistance until subjected to friendly air attacks
and artillery fire in its approach to the Naktong River in mid-August. From
the account given by the S-3, and from his estimates of strengths, casualties,
and replacements, one can follow the attrition of the division in its crossing
of the Naktong and through its battle with the ROK Army. Table 6, at the end
of this subsection, gives the appropriate data.
Statistics captured with the chief of staff of the 13th Div on 21 September
afford a basis for putting together a more detailed picture of the situation in
the period 31 August-ll September, the time during which the 13th Div attacked
the 8th Cav Regt. The figures also throw some light on the manner in which the
enemy units were committed. Though the enemy statistics in their orginial
form have already been presented and analyzed in a previous ORO report,25
they are given again in this study, since more complete information now avail
able leads to a rather different lineup of friendly and enemy forces.
In the enemy's personnel system, forward units did not requisition needed
replacements, nor, so far as is known, was there any formal, routine system for
reporting daily strengths. Rather, the responsibility rested with higher head
quarters, which, it is inferred from the captured statistics analyzed below, per
iodically directed that censuses of all subunits be made, usually in a period of
quiet before a new phase of the offensive, and then allocated available replace
ments accordingly. By September a time lag as long as a week might occur
before a particular group of replacements actually completed the march for
ORO-T-261
68
UNCLASSIFIED
SURn
UNCLASSIFIED
ward and reported to the unit to which it was assigned. It is quite likely that
sometimes a group did not arrive intact, if at all, or reported to the wrong unit.
From the enemy figures on the strength of the 13th Div and its subunits
on 31 August and 11 September, and from the enemy data on the receipt and
assignment of replacements during this interval, it is possible to make a per
sonnel material balance yielding the over-all number of losses sustained.
Since the quantitative validity of this computation rests on the interpretation
of the data on replacements, particular attention is given below to their meaning.
TABLE 2
ENEMY DOCUMENT SHOWING ACTUAL STRENGTHS
OF 13TH NK DlV UNITS 31 AUG 50
a
. Unit
Pvts NCO. Officers Total
DivHq 267 63 93 423
19th Regt 806 (694)b 161 90 (66)b 1057 (878)b
21st Regt 382(230) 51 53(25) 486(305) ,
23d Regt 535 (521) 163 7)(71) 769(751)
13th Arty Regt 492 204 89 785
, Tog Bn 135 23 12 170
Sig Bn ISO 46 24 220
Engr Bn 132 35 21 188
AT Bn no 74 22 206
SP-gun Bn 18 29 14 61
Medics 46 47 20 n3
Recon Co 63 16 2 81
Trans Co 30 16 7 53
Unassigned 314 314
Total 3480 928 518 4926
Total
(Cn action) (1445) (327) (162) (1934)
aOocument one loose sheet, handwritten, carried on person of Sr Col Lee Hak
Ku, CofS 13th Div, NKPA, who suttendered to US forces at 1153.3-1454.3 on
21 Sep 50.
hAll parentheses quoted from original document. These figures for troops in
action presumably also represent the date 31 August. Note that columns are
totaled as indicated, with totals of entries in parentheses given in parentheses.
Table 2 is the accounting of 13th Div strength as of 31 August. The par
enthetical "in action" figures are of incidental interest. In the 23d Regt, 98
percent of the personnel are listed as in action; in the 19th Regt, 83 percent;
and in the 21st Regt, 63 percent. This variation suggests different degrees of
commitment of the subunits of each regiment rather than, as with the US Army,
absence of assigned personnel for medical or other reasons, since in general
the enemy took account only of groups, not individuals. Figure 5 (given earlier
in this section) is an organization chart of a North Korean division constructed
from information supplied by POWs. Nearly exact TO strength figures are
given for each of the regimental subunits. If three infantry battalions and, say,
the 120-mm mortar company are considered as being "in action" while the
ORO-T-261
UNCLASSIFIED
UERff
69
UNCLASSIFIED
remaining seven specialized companies are not, then, just as with the 19th Regt,
83 percent of regimental personnel are in action. Similarly, if two of the
infantry battalions and, say, the 76-0000 howitzer, 120-0000 mortar, and 45-0000
AT -gun companies are considered to be in action, then 61 percent of regimental
strength, almost as with the 21st Regt, is found to be in action. Though the
determination of commitment of the specialized companies on this basis is
questionable, the situation of the infantry battalions seems established: as of
31 August, only one infantry battalion of the entire division was not "in action."
Table 3 is a similar listing of the strengths of units of the 13th NK Division on
11 September, but it does not include "in action" notations.
TABLE 3
ENEMY DOCUMENT SHOWING ACTUAL STRENGTHS
OF 13TH NK DIY UNITS 11 SEP 50'
Unit
pvts NCOs OHicel's Total
Hqs 267 63 93 423
19th Regt 846 148 30 1,024
21st Regt 439 131 53 623
23d Regt 521 159 71 751
13th Arty Regt 739 205 71 1,015
Tog Bo ]13 21
]I 145
Sig Bn 146 48 23 217
Engr So 266 33 19 318
AT Bo 174 74 20 268
SP-gun Bn 23 27 16 66
Medics 51 42 20 113
ReeoR Co 59 II 2 72
Trans Co 29 22 7 58
Unassigned 1,021 8
1,029
TO 7,777 2,954 1,061 11,792
Present
h
4,694 992 436 6,122
ShOft 3,083 1,962 625 5.670
8i)ocument (dtd 11 Sep 50) one loose sheet, handwritten, carried on \Jer
son of Sr. Col Lee lIak Ku, CofS 13th Uiv, NKPA, who surrendered to S
forces at 1153.3-1454.3 on 21 Sep 50. Note the differences between this
table and the similar listing for 31 August (Table 2).
LThe horizontal line marked "Present" rather than the line marked "TO"
gives the totals of the columns above.
Table 4 is an accounting of the disposition of batches of replacements
received at the 13th Div several replacement depots on the dates given. The
number of men and the place of origin (whether the group came originally from
North or South Korea) are stated for each consignment. The date of the docu
ment is unknown, but would seem to be 11 September or shortly before, as this
was the date of a number of related documents with which it was captured.
The table also gives the distribution of each batch QY assignment to the division's
* This replacement table shows a total of 1065 men unassigned, whereas the divisional strength table of
11 September shows 1029 unassigned.
ORO-T-261
70
UNCLASSIFIED
SiEln
UNCLASSIFIED
various units as of the date the f i ~ r e s were compiled. Thus it is seen that with
the exception of one-third of a very large group of South Korean replacements
received on 31 August, virtually all those received on or before 2 September had
been disposed of by 8 - 11 September, while three-quarters of those received on
6 September and all of those received on 8 September had yet to be assigned.
When did replacements received at a certain depot and subsequently assigned
to a unit actually join that unit? First, the locations of these centers must be
considered. All are situated on main roads in the 13th Div rear area; Kunwi
and Uisong also may have served the 1st Div, and Sangju the 3d Div. A town
named Angae cannot be located on any map or in the gazetteer; it is thought
TABLE 4
ENEMY DOCUMENT DETAILING RECEIPT AND ASSIGNMENT
OF REPLACEMENTS IN THE 13TH NK DIY
(Document undated)
DATE, 23 Aug 31 Aug 2 Sep 6 Sep 8 Sep
PLACE, ANGAE SANGJU U1SONG KUNWI KUNWI SANGJU
FROM, NK SK SK NK NK NK
TOTAL,b 961 319 2271 314 219 163
Hq 20 100 'Xl
19th Regt 309 459 200
21st Regt 245 97
, 23d,Regt 245 208 114
13th Arty 301
Tng On 12B 250
Sig nn 50
Engr Bn 140 50
AT Bn 31 190
Medics 10 8
Total
c
Assigned
Unassigned
d
959
2
319
0
1540
731
314
0
50
169
0
163
aUocument one loose sheet, handwritten, carried on person of Sr Col Lee Hak KU
t
CoES 13th
Div, NKPA, who surrendered to US forces at 1153.3-1454.3 on 21 Sep 50.
bTotal replacements received: 4247.
cThese entries have been added by present author by subtraction of unit totals from number
of replacements received by the division.
dThe total number of unassigne(l replacements derived from the enemy's figures is 1065. This
value does not quite agree with the total of 1029 given in Table 3.
likely that it was the road junction in the valley at the coordinates 36 23' 13"N,
12816'18"E, variously known as Angye-myon, Ankei-men and Angyejang. The
apprOximate distances in road miles from these depots to Tabu-dong are, read
ing from north to south: Sangju, forty; Angye, thirty-five; Uisong, thirty; and
Kunwi, eighteen.
Interrogation reports are available for prisoners who probably had been
included in two of the consignments listed- in Table 4. The three subjects of
the 1st Cav Div Interrogation Report No. 0088 (Item 4, App A) correspond to
the group received at Uisong,on 31 August. According to their statements, they
ORO-T-261 71
UNCLASSIFIED
SHIrt
UNCLASSIFIED
were impressed into the NKA on or about 22 August at Wonju. On or about 29
August they were part of a group of about 1000 replacements which set out for
Andong, where they were told they would haul ammunition. However, they were
sent further forward and on 7 September arrived at a place called Son gyo-dong,
on the highway 5 miles north of Tabu-dong; the prisoners heard that this was
the site of the 19th Regt Hq. Details are not given on this move, but the route
followed would have taken them through Uisong, Kunwi, and Hajang-dong. Of
the 1000 men arriving at 19th Regt Hq, 300 were further assigned to the regi
ment's 1st Bn, which they jOined on Hill 570 at sunset ,on 10 September.
The prisoners' story fits in well with the interpretation of the table of re
placements, i.e., that on 31 August 2271 South Koreans were received by the
13th Div at Uisong, of whom 459 were subsequently aSSigned to the 19th Regt.
From the two sources it is then deduced that traveling from Uisong to regi
mental headquarters, a distance of 25 miles, took 7 days in all, The prisoners'
statements alone show that the move from regimental headquarters to the even
tual combat unit, a distance of 9 miles, took 3 days.
Some of the South Koreans in the group of 2271 received at Uisong on 31
August were seemingly sent forward quickly but not committed to battle. The
subject of ADVATIS 0721 stated that he was one of 3000 South Koreans conscrip
ted in the vicinity of Wonju. His group left there on 27 August, and traveling via
Andong and Uisong reached the viCinity of Tabu -dong on 3 September, where he
was aSSigned to the 3d Bn of an unknown regiment and division. The strength
figures given by the prisoner seem to be much too large for the units represented,
but the statements that of 370 men in the new unit only 70 of its original troops
remained and that 100 men were unarmed may be significant. Up to the time
when the prisoner deserted at 2400 on 5 September, the unit had not been engaged
in battle, He surrendered to US troops at 0600, 6 September at 1151.7-1459.4,
a point near the road, 1 1/4 miles southwest of Tabu-dong.
The 1st Cav Div Interrogation Report No. 0085 and ATIS Report No. 1158
(see Items 2 and 3, App A) both represent the same prisoner, a North Korean
who was'apparently in the group of replacements received at Kunwi on 2 Sep
tember, He was inducted on 19 August and after three days of training sent with
a group of 1000 recruits to Seoul and then further south. On or about 28 August
the group arrived at Wonju. They continued to Kunwi, where the prisoner was
issued a rifle. The date of arrival there is not reported but could very well have
been 2 September, when, according to the 13th Div replacement table, 314 North
Korean replacements were received, Moving on, the prisoner on 3 September
arrived at a point several miles north of Kasan, where he was assigned to a
company of the 1st Bn, 19th Regt. The same day, his group of 200, which he
thought an independent battalion, moved on to Kasan. It stayed there until 9
September, when it finally displaced to Hill 570. These statements again check
the replacement table, which shows that 200 of the 314 replacements received
at Kunwi were assigned to the 19th Regt. Contrary to the prisoner's impreSSion,
it would seem that the group of 200 was not a battalion itself but consisted
simply of replacements designated for the several companies of the 1st Bn; the
group waited at Kasan until the 1st Bn had moved from a position to the west
onto Hill 570. .
The two examples just presented indicate that the tactical situation of the
combat units rather than the distance replacements had to march or the time
required to process them along the way was primarily responsible for delaying
ORO-T-261 72
UNCLASSIFIED
SURfY
~ R E I -
their arrival at the front. Green replacements, especially unreliable South
Koreans, could hardly be incorporated into units currently in active combat.
In particular, the 2271 South Koreans received at Uisong on 31 August could
hardly be used in the offensive on 2 September. That their presence was taken
into account, however, may be Seen in the notes made by the CofS on 30 August at
a conference for planning the offensive: Uisong 3000 men, Andong 1000 men."
The replacements received on 23 August certainly were introduced into forward
units during the lull before the offensive. The 319 South Koreans from Sangju
received at an unknown time between 23 and 31 August, however, were apparently
not yet assimilated. The strength table for 31 August shows 314 replacements
unaSSigned. This must have been the Sangju group, not yet used because of too
recent arrival, lack of training or arms, or unreliability. By the time the re
placement table was compiled between 8 and 11 September all had been assigned.
From the foregoing considerations it is concluded that only the replacements
received by the division on 23 August are included in the various unit strengths
listed on 31 August.
There is little doubt that almost all those replacements received by the
division through 2 September who were subsequently assigned to units did actu
ally arrive at those units to be counted in the strength list of 11 September, as
by 8 or 9 September the 13th Div had again stopped attacking. It is also likely
that the 50 replacements received on 6 September and subsequently assigned to
the Sig Bn joined by 11 September, as this unit was probably somewhat to the
rear.
Losses in the period 31 August-ll September can now be computed by add
ing to the strength of each unit on 31 August the number of replacements con
sidered to have arrived after this date and subtracting the strength on 11 Sep
tember. This procedure is straightforward. and produces credible figures, with
the following few exceptions: the 21st Regt, which was indeed in reserve, has a
balance of minus 40, the engineer battalion minus 80, the SP-gun company minus
5, and the transportation company minus 5, while the training company, with a
strength of about 160, has the incredible positive balance of 403 losses. It is
most probable that the training company served as a replacement depot within
the division and that the apparent 403 losses actually were men passed on to
units showing negative losses. Entirely arbitrarily, the bloc of 128 South Kor
eans received by the division between 23 and 31 August, and subsequently assigned
to the training battalion, has been given to the 21st Regt. To avoid distributing
the remaining excess personnel of the training company, all divisional units
except the four regiments and the AT Bn have been lumped together. The results
are given in Table 5.
In the table it is seen that in the 11-day period there was a turnover of about
one -third of the division's strength, which remained roughly constant. The actual
rate at which losses occurred probably varied greatly and most likely was high
est in the early days of the offensive. Casualties were probably very low during
the week before the offenSive began on 2 September, because there were almost no
friendly air attacks. For computing average casualty rates, all losses are con
sidered to have occurred in the nine days of fighting from 2 through 10 September.
Since the strength of most units was about the same at the end of the period as
at the beginning, the simple average of the two has been used. The resulting loss
rates are shown in Table 5 as percent of average strength per day. Their inter
pretation in the light of the tactical situation is given elsewhere.
ORO-T-261
-SEERU
73
UNCLASSIFIED
SRltT
The data on the replacements also permit calculation of the number of
South Korean draftees in each unit as of 11 September. For this purpose it is
assumed, as all eVidence indicates, that prior to the period considered the divi
sion was entirely North Korean. As suggested by the data presented above,
the South Korean replacements shown in Table 5 arrived at their units late in
the period, after the fighting bad died down; therefore, it is also assumed that
all losses shown were North Korean personnel. The ratio of :gouth Korean
replacements to unit strength on 11 September thus gives a conservative figure
for the fraction of South Korean personnel. As seen from the table, for the
division as a whole the figure is 3'1 percent, and this is typical of most of the
TABLE 5
STRENGTH LEVELS OF THE 13TH NK DIV 31 AUG -11 SEP 50
(A synthesis of enemy statistics)
Strength levels
In! Regts
Arty
Regt
Bn
All other
units
Entire
13th Div 19th I 21st I 23d
Strength 31 Au!!"", 1057 486
7@ 785 206
1009 4612
Replacement. 31 Aug-Il Sep:
Dates received, piace
of origin
23-31 Aug, SK 1213" 31 160 319
31 Aug, SK 459 97 208 301 190 'l5S 1540
2 Scp, NK
200 114 314
6 Sep, NK 50 50
Total 1716 711 1091 1086 427 1777 6835
Strength 11 September 1021 623 751 lOIS 268 1412 5093
Losses 31 Aug-ll Sep 692 ssh 340 71 159 365 1742
Average strength 1040 555 760 900 240 1360 4850
Loss tate, percent per dare 7.4 1.Sb 5.0 0.9 7.4 3.0 4.0
Percent South KQrenns, 11 Sep
45 36
h
213 30 83 30 37
~ b e s e repiacements arbitrarily reassigned from training battalion.
h Arhitrary figure. See F ootnot:e 8.
eBased on nine days of comhal, 2-10 Septemher.
division's component units. From the same assumptions it also follows that
only 47 percent of the personnel composing the division's units on 31 August
were still present on 11 September.
When the data of Table 5 are combined with the information furnished by
the S-3 of the 19th NK Regt, a balance of losses and gains in personnel can be
made for intervals throughout the period from the first actio!! of the 13th Div
in mid-August through 11 September 1950. Table 6 gives the data.
According to the 5-3, the 13th Div went on the defenSive on 24 August. It
is of particular interest to compute its attrition as of this date. If it is assumed
that half the losses in the period 21- 31 August occurred by 24 August (the only
active fighting was on 21 August) then the strength of the 19th Regt on this date
had fallen to about 1100, while the combined strength of the 21st and 23d Regis
'14
ORO-T-261
UNCLASSIFIED
SlEREr
------
UNCLASSIFIED
SUitT
had fallen to about 2250, or again to around 1100 per regiment. The 13th Div
offensive really stalled for the first time-and for ten days-when its strength
had fallen to about 40 percent of the TO figure.
TABLE 6
STRENGTH LEVELS OF THE 13TH NK DIV FROM
FIRST COMMITMENT TO 11 SEP 50'
Strength 1evel s
Rifle Regts, 13th Div
19th
I
21st
I
23<1
Entire
13th Div
Original TO strength
Losses from Air & Arty on
move to NBktong
Losses crossing the Naktong
Strength 16 August
Losses 16, 17 August
Strength 18 August
Losses 18 August
Strength 19 August
Losses 19 August
Strength 20 August
Losses 21 August
Losses from Air & Arty
21-31 August
Replacementt-l 24 August
Strength 31 Auguste
Losses from 8th Cay
31 August_ll September
d
Replacements 31 August
11 September
e
Strength 11 September
e
2,500+
b
2,500
700
1,800
300
1,500
1,500
750
d
182 (30g
e
)
1,057
692
659
1,024
2,50O+h 2,500.,1>
700
1,800 2,500
800
1,000 2,500
400 400
600. 2,100
100
2,600
700
d
618 (590
e
)
486 769
88 340
225 322
623 751
12,OOOC
2,000
10,000
800 (96I
e
)
4,926
1,742
2,223
6,122
8Figures as given by the 5-3, 19th NK Regt, who sUITendered 1 Sep 50 (Item 1, App A}, except as
otherwise noted.
bEstirnate, consistent with balance; TO strength 2,815.
cTO strength 11,792; see Table 3.
dEstimate, consistent with balance.
eFigures from enemy documents.
One rough check can be made of the data in Table 6. A medical officer of
the 13th Div stated that from 24 August to 20 September 800 wounded of the 21st
Regt were treated at his field hospital (ADVATIS 0964). If, as estimated above,
the regiment's strength was about 1100 on 24 August, by 11 September it had lost
from all causes about 600 men, for its strength by then was 486.
STRENGTH AND CASUALTIES ON HILL MASS 314
The summary listing of enemy units on Hill 314 which follows is derived
from the analysis given earlier in this section under Composition and Condi
ORO-T-261
Hun
UNCLASSIFIED
75
SEERrf
UNCLASSIFIED
tion of Units at Mid-September."- Strength figures are added if given by POWs;
otherwise, they are estimates in accord with the assumptions discussed at the
beginning of this section.
Time and unit Strength Source
Before 12 September:
1st Bn, 2d Regt, 1st Div
On 12 September:
4th Co, 2d Bn, 2d Regt, 1st Div
5th Co, 2d Bn, 2d Regt, 1st Div
HMG Co, 2d Bn, 2d Regt, 1st Dlv
1st Co, 1st Bn, 14th Regt, 1st Div
2d Co, 1st Bn, 14th Regt, 1st Div
3d Bn, 19th Regt, 13th Div
Total present, 12 September
280 . Assumed
120 POW
85 POW
40 POW
60 Assumed
60 Assumed
280 Assumed
645
It is interesting to compare this total with the statement by the POW in
the report quoted earlier, in "Composition and Condition of Units at Mid-Sep
tember" that 700 enemy troops were on the hill at the time of the attack by the
3d Bn, 7th Cay Regt.
Total enemy strength may have been lower but hardly higher. According
to those members of the 3d Bn, 7th Cav who were on the ridge between Knobs
2 and 3 in the perimeter defense maintained in the days after it was captured,
the ridge was very crowded; they could not imagine it holding many more than
700 men. On 15 September the number of US personnel present for duty in
Companies I, K, and L totaled 477. ROK troops integrated into the 3d Bn who
were present for duty totaled 115. In addition, a platoon of Co M and the equiv
alent of a platoon of personnel of the Bn Hq Co were probably on the ridge.
Total friendly strength then numbered about 670, and this population was close
to the capacity of the ridge.
A rough count of enemy bodies gave a figure of about 250. Most of these
were found in the wooded crown around the ridge rather than on the ridge itself.
It is likely that the enemy disposed of his dead just as US troops disposed of
enemy bodies on occupying the ridge, that is, they were pulled out of the fox
holes and rolled down the slope. The number counted would then be product of
all the fighting on the hill. The prisoner mentioned above said that 200 men
were killed on 12 September, but it appears that at least 74* had been killed
previously. In the final action, many of the enemy troops immobilized by non
fatal wounds were undoubtedly killed by friendly supporting arms or by. the in
fantry when the objective was overrun; on the previous days many of the enemy
wounded may well have been evacuated. On the average, perhaps three out of
eight men hit died on the hill. Total enemy casualties would then have been
* Of these, 52 r e p o r t ~ d by 16th Recon Co on 8, 9 September; 22 by 3d Un, 8th Cay Hegt on 10 September.
76 ORO-T-261
UNCLASSIFIED
UNCLASSIFIED
about 665, out of about 925 enemy troops committed to battle on the hill. The
breakdown of all enemy troops committed would then be, of 925 men committed:
250 bodies left on hill (167 KIA; 83 WlA, then KIA); 415 WlA, escaped from hill;
and 260 not hit, escaped from hill.
If it is assumed that none of the wounded remained with their units after
they regrouped, then enemy units would be at 28 percent of original strength
after the battle. The prisoner from the 4th Co, 2d Bn, 2d Regt gave the resid
ual strength of his company and the HMG Co as 40 percent of original strength.
Since the medical officer from the 13th Div stated that soldiers with flesh wounds
were immediately sent back to the front after field treatment, it is more rea
sonable to assume that perhaps 100 out of the 415 wounded estimated to have
left the hill suffered only trivial wounds and remained with their units.
ORO-T-261
UNCLASSIFIED
SUtEr
77
Si.fRtT
UNCLASSIFIED
FRIENDLY STRENGTHS AND CASUALTIES, AND COMPARISON WITH THOSE
OF THE ENEMY
The first subsection below presents statistics on friendly casualties sus
tained in the first phase of the enemy offensive. Though they are not complete
and have not been corrected by special study, they serve for comparison with
the data derived earlier on casualties in the 13th NK Div. This comparison is
made later in this section.
A more rigorous accounting of friendly strengths and casualties in the
8th Cav Regt area for the period 6-16 Sep 50 also follows; the sources of the
data are discussed there.
CASUALTIES IN THE ENEMY'S INITIAL DRIVE
Table D 1 is an accounting of friendly strengths and casualties in the 8th
Cav Regt in the period 1-9 Sep 50. In addition to all the regiment's companies,
those engineer companies attached as combat troops appear. Company A of the
71st Tk Bn was also attached as a provisional rifle unit during this period but
is not included because it did not become actively involved in the fighting (five
battle casualties, of which three KIA, in period 1-9 September). Also, the 5th
ROK Spl Tng Bn (South Korean police), which moved into backup positions on or
about 5 September, does not appear. The ROK troops integrated into US com
pimies, however, are represented in both the strength and casualty figures,
except in the case of the 8th Engr Bn, whose morning reports have not been
consulted. Though the artillery batteries were close to the front line and
did become somewhat involved with infiltrating groups of the enemy, they are
not included, since their casualties never reached a high number on anyone
day.
Strength figures are given as of the time of the enemy attack, i.e., the
close of2 September, and are presumed to be accurate, since there was very
little combat on the preceding days. The figures given in the morning reports
for the succeeding days are not considered to be accurate and so have not been
used. For example, the morning report of Co E for 5 September gives the
number of US personnel present for duty as 169, but the regiment's operations
report for this date states that on withdrawal from the Walled City Co E had
a strength of 102 (3 officers and 99 men), which may indicate that integrated
ROK troops were included in the morning report. It is unlikely that replace
ments were received in the first days of the offensive. Company strengths may
then be taken as the figures for 2 September minus casualties to date.
ORO-T-261
78
UNCLASSIFIED
SURET
UNCLASSIFIED
The casualty statistics of Table Dl are provisional. They were prepared
by a machine run made by the Adjutant General's Office, FECOM in January
1951 and should be replaced by more complete data from currently available
corrected machine records, as has been done for the period 6-16 Sep 50 for
the data given in Table D2. As pointed out below, the 1951 listing accounts for
only about 75 percent of the names now carried in machine records. The dis
crepancies are thought to be due chiefly to incomplete reporting for certain
companies on those days when many casualties 'were sustained. From the over
lapping coverages of the two tables in App D for the period 6-9 Sep 50 it may
be seen, for example, that whereas 52 casualties were listed for the 3d Bn,
8th Cav in 1951, the number is now carried as 95. On the other hand, for the
2d Bn in the period 6-16 Sep 50, the earlier listing carries 91 percent of the
casualties of the later tabulation.
ANALYSIS OF STRENGTHS AND CASUALTIES 6 - 16 SEP 50
Table D2 is a day-by-day listing of strengths and casualties of infantry
units in the 8th Cav Regt zone in the period 6-16 September. For complete
coverage of each unit, the data on troops attached from the South Korean army
are tabulated in parallel and also combined with the figures on US personnel
to give the totals. In addition to the 8th Cav Regt, which is completely repre
sented, the 16th Recon Co, which was attached to the regiment and fought as
infantry throughout this period, and the 3d Bn, 7th Cav Regt, which was attached
and moved up to the area on 10 September, are included. The companies of the
8th Eng Bn, which had fought as infantry with the 8th Cav in the Walled
City and south of Tabu-dong, are now shown because they were no longer attached
as combat troops. Company A of the 71st Tk Bn was still attached as a provi
sional infantry unit but did not become involved in active combat and so is not in
cluded (the only battle casualty was one KIA on 18 September). No data are
available on the 5th ROK Tng Bn, whose Co B did briefly become engaged with
infiltrating enemy on Hill 660 and whose Co A and Co C, though never intensely
engaged, did back up friendly forces on Hills 314 and 373. The 2d Bn, 7th Cav,
which occupied Hill 660.on 12 September, is not included because this hill was
never actively fought on and has therefore not been taken into the area under
study here.
For US personnel, the strength figures have been taken from the morning
reports of the individual companies and correspond to the entry "present for
duty" as of the close of the preceding day. As the reports are replete with
directions for small retroactive corrections and adjustments, these numbers
are not considered precise. Since they are used for the computation of casualty
rates, and in sum, for gross comparisons of friendly and ene(llY strengths, this
correction has not been deemed necessary.
The statistics on US casualties have been furnished by the Adjutant Gen
eral's Office, FECOM, and are the product of a recent machine run. The totals
are rather larger than those shown in a previous report,25 which was based on
a machine run made a few months after the action. The earlier listing included
only about 75 percent of the casualties presently carried in the records. One
reason for the incompleteness of the previous figures was the inadequate ac
ORO-T-261
SUtt1T
UNCLASSIFIED
79
UNCLASSIFIED SH'RtT
counting of casualties hurriedly evacuated from the theater. The statistics of
Table D2 are still considered to be somewhat inaccurate as to the precise date
a casualty occurred, especially in the KIA cases. Misdating is shown, for
example, in the listings of the 16th Recon Co for 9 and 10 September. Though
correction is attempted in using the data in interpreting tactical developments,
the figures have not been changed in Table D2.
Attached ROK troops were all enlisted men. In the 7th and 8th Cav Regts
they were integrated into US squads rather than attached as small homogeneous
Korean units. The number attached to each company was available from occa
sional entries in the morning reports, which did not account for them from day
to day. Nevertheless, from entries giving the number assigned, in blocs or by
"K" number (see Table 16), and from subsequent notations as to casualties and
other gains and losses, it was possible to work out the number present on each
day. Sometimes this balance sheet for a certain company could be checked by
. an entry that mentioned the number present for duty. Furthermore, medical
data on casualties occurring in the period 11-14 September among all Korean
personnel attached to the 1st Cav Div were obtained from the A&D sheets of the
division's clearing company. Through the K numbers and knowledge of the
assignments by the numbers in blocs, the mention of individual casualties in the
morning reports could be checked. It is concluded that the data in Table D2
are substantially complete but that the dates are not necessarily accurate. It
has not been discovered why the casualty rate among the attached Koreans in
8th Cav Regt units was lower than the US rate in the same units. .
The opposing forces were in contact at or On Hills 624, 225, 570, 373, and
314. The area of active defense against the enemy penetration is therefore
defined to include these hills and the area more or less directly to the friendly
rear. It is impossible to define just what units constituted the essential infantry
defense in this area, even though the situation was clear -cut at the line of contact
and, by 12 September, all regular reserves had been committed to the vicinity
of the front line. The ROK Tng Bn, though inexperienced, was on hand and
available for use in an emergency. To the rear of the area as defined above
several provisional infantry battalions could readily be assembled from miscel
laneous speCialized US Army personnel of quartermaster, headquarters, armored,
and other units. For the present purpose, the defensive infantry force is taken
to be those units appearing in Table D2. There is a lack of rigor in including the
specialized companies of the 8th Cav Regt with its infantry battalions, but this
is felt to be of little weight in comparison with other uncertainties.
Daily total infantry strength in the defense area is plotted in Fig. 6.
Straight lines are drawn because the individual points are somewhat question
able. In particular, the strength figures for the several days just after the
intense fighting on 10-11 September probably did not fully account for the cas
ualties sustained.
To show the accumulation of losses by the defensive infantry force as a
result of enemy action, the daily grand totals of battle casualties as given in
Table D2 have been cumulated and plotted in Fig. 6. The symmetry of the plot
is striking. In the period 6-10 September casualties were sustained at the
roughly constant rate of about 58 per day. This period can be characterized as
that of enemy initiative with increasing pressure, stemmed by limited friendly
counterattacks. The apparent enemy bid to force a decision began on the night
ORO-T-261
80
UNCLASSIFIED
1200
UNCLASSIFIED
of 10-11 September, was checked, and then, before it was fully developed,
decisively turned back with the friendly capture of Hill Mass 314 on 12 September.
On these two days friendly casualties increased at a rate of about 175 casualties
per day, three times the previous rate. The period 13-16 September was then
much like the 6-10 September phase, except that the initiative was now on the
Enemy Initiative
Li mited AttacksCounterattacks
1000
W
z
..J
Q
..J
0
800
e
w'"
I
..J
r-6--
I
:l
"'"
I
'"
600
U
I
w
"'" --e--.Q..- ..... ..Q.i
..J
SlIengtb
l-
l
"'"
'"
w
400
~
I
..J
"'"
:l
:l
'"
U
200
Casualties
o
5 6 7 8 9 10
'2
10 '"
~
'"
1;;
=
'"
f
~
'"
ILl "
"0
'"
. ~ "
0;
'"
0
6000
Friendly Initiative
Limited Attacks
> - ~
Counterattacks
=0:;
5000
.-
"
'"
-
'" '"
- '"
u.."
W
a l ~
Z
, , ~
..J
E 1;;
- ~ z
'"
0;=
4000
"'
of
0
'"
e'"
:<
I
--"1)---.11---
C>
z
3000
w
I<
I
'"
>
I<
I
Z
..
2000 "'"
:!:
..J
I
"'"
0
I
1000 .
11 12 13 14 15 16
DATE, SEPTEMBER 1950
Fig. 6-Daily strength of friendly infantry and cumulotive total
friendly battle casualties in 8th Cay zone.
friendly side. Friendly casualties again increased at a roughly constant rate,
almost the same rate as before. Comparison of the strength and casualty plots
in Fig. 6 shows that the fluctuation of the absolute casualty rate does not reflect
gross variation in the total number of men present and therefore must reflect
the development of the tactical situation.
The relationship of friendly casualties to the tactical situation is outlined
by the data of Table 7. On each day of the period 6-16 September, those gen
ORO-T-261
SEE1m'
UNCLASSIFIED
81
SiEMT
UNCLASSIFIED
TABLE 7
RELATION OF FIlIENDLY CASUALTIES TO THE TACTICAL SITUATION"
Date, Sep 50 Place Company
Casualties
b
Action Outcome
No. I Rate, %C
5
6
City
West of
E/8
E/8
21
16
)
12
9
)
Fight into Walled City
and hack
Task force attacks
loin Co. 0, 8th Engr Bn,
both companies
withdraw
Road cleared
Hill 570 enemy fire block
across Taegu road
6 Unknown H3/8 10 8 Unknown, rear guard? Unknown
7 ----------------
------------------------
8 Hill 570
{L/8
K/8
[/8
27
15
14
15
}
9 Friendly attack
7
Friendly withdrawal
8-9 Hill 314 16 Recon 32 17 Enemy attack Friendly withdrawal
9 Hill 314 16 Recon 20 12 Friendly attack Friendly withdrawal
9-10
Hill 314 16 Reeon 0 0 Enemy attack Friendly stand
10
ID-!l
Hill 314
lIill 373
K/8
G/8}
18
36
!l
r
O
)
Friendly attack
Enemy attack
Friendly withdrawal
Slight friendly with
drawal, friendly stand
11 Hill 373
{ 22
(12) }
(12)
Limited advance,
enemy mortar fire
Slight advance, then
friendly stand
11 Hill 314
{ [/8
K/8
27
16
14 }
12
Friendly attack Friendly withdrawal
11-12 Hill 373 16 Recan 0 0
E.nemy attack Friendly stand
12 Hill 225 B/8 18 8
Enemy mortar fire Friendly stand
12 Hill 314
{L/7
[ /7
62
58
34 }
34 Friendly attack Friendly capture of hill
K/7 23 13
13 Hill 225 B/8 11 5 Unknown Unknown
14 Hill 570 E/8 9 4 Friendly attack Friendly captureofhill
15 Hill 570
{ E/8
F/8
16 Recon
16
9
5
7
6
4
}
Limited advance, small
counterattack, enemy
mortar fire
Short friendly advance
15
16
Hill 401
Hill 307
L/8
H 1/8
18
10
14 }
10
Friendly move to
occupy ridge
Friendly capture of
Hill 401, then 307
16 Hill 570 F/8 6 4 Enemy attack Short friendly withdrawal
aAll cases of (a) ten or more casualties per cO:llpany per day, or (b) fewer than ten casualties per company
per day but company directly attacked or was attacked itself.
bCasualties and casualty rates include HOKA augmentation and the IIA.
CHate expressed.as percentn( company strength per day. Parentheses show estimated breakdown of combined
casualty figure for two attacks.
ORO-T-261
82
UNCLfl.SSIFID
UERn
UNCLASSIFIED
eral circumstances have been determined in which (a) any company sustained
ten or more casualties on one day or (b) any company sustained less than ten
casualties and yet is known to have made an attack or to have been attacked
itself. The number of casualties in each company on each day is given in the
table. From this numlier and the strength of the company at the beginning of
that day, the casualty rate has been computed and listed. Except in those cases
where many casualties resulted from a concentration of enemy fire in the absence
of an attack, it is assumed that most of the casualties occurred in some phase
of an attack, whatever the caus'e might be. Where the daily casualty figure was
the result of two separate actions, the fraction attributable to each has been
estimated.
To avoid a misleading implication, data from an earlier statistical tabu
lation are included for Co E for 5 September when it was engaged with the enemy
at the Walled City. It is possible that all the casualties listed for 6 September
occurred when the company participated in the action of an armored task force
which cleared an enemy fire block on this date, but it seems more likely that
many of these casualties actually occurred at the Walled City.
Since all attacks, friendly and enemy, and all effective enemy concentra
tions of fire are represented in Table 7, it is evidently a partial schedule of
Significant action in the entire defense area. To be complete, of course, it
would have to show those presumably numerous cases where, in the absence
of friendly infantry action, friendly air attacks and concentrations of artillery
and other supporting fires inflicted a significant number of casualties on the
enemy.
There is reason to suppose that the development of the general situation
depended on how many friendly casualties were sustained at each point of contact
of the opposing forces. Since friendly battalions were not as a rule backed up
closely, if at all, by a reserve battalion, it was important to them to avoid being
crippled by enemy action for they could not easily be relieved. Perhaps most
often only two rifle companies were involved in a hill action, with the third held
in reserve. To be sure, if one company were rendered quite ineffective, as by
being overrun, the elasticity of the entire battalion would be gone, but lesser
losses, whether concentrated in one platoon or distributed over the entire com
pany, would not destroy its functional unity and there would be opportunity to
repair the unit with repblcements. One would therefore expect that only a cer
tain maximum percentage of casualties would be acceptable in any rifle com
pany. Individual platoons might, however, be expendable.
As the company is the smallest administrative unit, no data on the distri
bution of casualties among platoons exists. (Some suggestive information on this
point is considered, however, later in this memorandum in the section on friendly
casualties in the capture of Hill Mass 314.) In Table 8 the data on individual
companies presented in Table 7 are rearranged into several categories accord
ing to which side took the aggressive action and which at the end of the engage
ment held the contested point, a hill or a knob on a ridge. No friendly or enemy
defending force was ever overrun.
The following generalizations express the contents of Table 8 and may
not be at all validfor prediction: (a) Friendly companies occupying a hill facing
enemy forces and subjected only to enemy fire held the pOSition even though 12
ORO-T-261
SiEREf
83
UNCLASSIFIED
percent casualties had been sustained. (b) Friendly companies occupying such
a hill and assaulted by the enemy held even though 4 percent casualties had been
sustained, withdrew from forward positions, then held when 10 percent casualties
had been taken, withdrew from the contested point when 17 percent casualties
had occurred. (c) Friendly attackers did not capture a hill if more than about
14 percent of the strength of one company became casualties. Exception: the
two assualt companies that finally captured Hill 314 each had a casualty rate
of 34 percent.
TABLE 8
CASUALTY PERCENTAGES IN FRIENDLY COMPANIES IN RELATION
TO OUTCOME OF OFFENSIVE AND DEFENSIVE HILL ACTIONS
Company casualty rate
b
(in percentages)
when result was:
Action
Friendly Friendly Friendly
capture of hill withdrawal stand
Friendly attack 3 cases: 4 cases:
34 (34, 13) 15 (9, 7)
14 (10) 14 (12)
4 12
11
Enemy attack 3 cases: 2 cases:
17 o
10 o
4
Enemy fire only No cases 2 cases:
12 (12)
8
BAll are actions in which defending friendly or enemy force made stand or with.
drew, was never overrun.
bThe highest company casualty rate in anyone action is given, with casualty
rates of other companies actively involved stated in parentheses.
As to casualties, the capture of Hill 314 by the 3d Bn, 7th Cav was not
so exceptional as it may seem. As discussed in a later section, about half the
casualties in Co L and Co I occurred during and after the long inactive period
when they were pinned down by mortar fire after the second unsuccessful assault.
The company casualty rate had reached about 15 percent at this point, and the
attack might have been called off if the general situation had not been so critical.
Also, it seems certain that the hill would not have been captured without the
unusually effective air strike and outstanding leadership. In regard to taking
further offensive action, the battalion was crippled for some days. This was
not true of the other units that had withdrawn from unsuccessful attacks; for
example, the 3d Bn, 8th Cav went on to take various pOints of high ground in
the vicinity of Hill 401.
ORO-T-261
84
SiERfl

...
UNCLASSIFIED
SEUEI
COMPARISON OF FRIENDLY AND ENEMY CASUALTIES
IN THE ENEMY'S INITIAL DRIVE
Data given earlier and in this section permit a fairly clean-cut comparison
of friendly and enemy casualties in the first phase of the enemy offensive when
the deployment of the units of both sides, as represented in the diagrams of
Part III, is taken into account. For the comparison to be valid, the 1st NK Div
must not have engaged the 8th Cav Regt. The evidence given earlier indicates
strongly that this was so but does not prove the point. It is therefore possible
that the ratios of enemy-friendly casualties calculated below are somewhat too
"low. .
As shown earlier", in Table 5, from 31 August through 10 September the 13th
NK Div's 19t1lRegt with an initial strength of 1057 is calculated to have sustained
692 casualties; its 23d Regt, with a strength of 769, 340 casualties; its 21st Regt,
with a strength of 486, 88 casualties. It was estimated that one enemy regiment
opened the offensive by attacking the 8th Cav Regt's 2d Bn; this was almost
certainly the 19th Regt. In all likelihood, the weaker 23d Regt, which initially
faced the 8th Cav Regt's 1st Bn, was moved up to reinforce the 19th Regt. The
very weak 21st Regt doubtless remained in reserve or in a quiet sector. The
strength and casualty figures suggest at once why the 1st Bn was subjected to
only very light pressure and why the offensive stalled south of Tabu-dong.
Enemy casualties of course accumulated from friendly air attack and
artillery fire as well as from infantry action. By 8 September contact had
dwindled to clashes of outposts, and after the friendly withdrawal on this date
there was no contact at all in this sector on 9 September, and perhaps 10 Sep
tember as well. There was one air strike on 1 September and none on 2, 3, 4,
or 5 September. Since the offensive had halted by 6 September, it is likely that
the enemy did not overtly offer targets for the air attacks which were delivered
on 6, 7, and 10 September. The casualties in the two enemy regiments were
therefore probably sustained chiefly in the first days. For further analysis,
80 percent are arbitrarily assumed to have occurred before 6 September. If
it is assumed that no replacements were received during the first phase, the
following balance of personnel can be made: initial strength 31 August, 19th
Regt-1057, 23d Regt-769; casualties 1-6 September, 19th Regt-557, 23d
Regt-272; strength 6 September, 19th Regt-500, 23d Regt-497.
These figures include casualties caused by friendly supporting fires among
regimental units to the rear of the front line, such as the regimental artillery
and mortar companies. A proper comparison therefore includes casualties in
friendly units just to the rear. In Table D1 casualties considered inflicted by
the 13th Div's 19th or 23d Regts in any units, front or rear, are italicized.
Company B is excluded because on 2 September it moved to the west and pre
sumably into contact with the 3d NK Div; Co E is excluded from 5 September
on because of its move into the Walled City and contact with the 1st NK Div.
With such exclusions a total of 298 friendly casualties is attributed to the 13th
Division in the period 1-6 September. If this total is assumed to be only 75
percent complete, the estimated actual figure is 397. The comparison for op
posing regiments, may then be made: enemy casualties, 829; friendly casual
ties, 397; ratio, enemy/friendly, 2.1.
ORO-T-261
$oHREr
UNCLASSIFIED
85
UEREr"
UNCLASSI FI ED
A comparison of the number of casualties in the opposing units at the front
line is of particular interest. On the enemy side, estimates of strengths of sub
units and their casualties may be made for the 19th Regt. In accord with the
data and assumptions given earlier in this part, the regiment is estimated to
have had the following make -up on 1 September: total strength was 1057; of
which, 1 battalion has 160; 2 battalions (280 each), 560; CP units, 1705 artil
lery, mortar companies etc., 167.
From the data of Table 5, it is calculated that in the period 1-6 September
the 13th Div miscellaneous divisional units (with the omission of the antitank
battalion, which was in a special situation) lost 17 percent of their strength
(applying here again the 80 percent correction for casualties through 10 Sep
tember). If the U'gimental rear units had a similar but somewhat greater
hazard, an average depletion of 25 percent may be assumed for them. Of the
337 men not in the rifle battalions of the 19th Regt, then, 84 are calculated to
have been casualties. If this figure is subtracted from the 557 casualties for
the entire regiment, the result is 473 rifle-battalion casualties. Of their com
bined initial strength of 720, they then had only 247 men left, or 34 percent of
the orginial number. One or two battalions very likely were depleted even
more.
If itis assumed that two battalions (of 280) of the 23d Regt attacked, and that
the remainder of the regiment remained in the rear and suffered 25 percent
depletion, then 200 casualties are attributed to the two battalions-. This figure,
when added to that of the 19th Regt, gives a total of 673 casualties in enemy
rifle battalions.
Friendly front-line troops may be taken to be the rifle and heavy weapons
companies and the combat engineers. From Table D1 it is calculated that the
13th Div inflicted 240 casualties on these units. If the 75 percent figure of
completeness is again taken into account, the number is raised to 320. The
comparison for opposing front-line units may then be made: enemy casualties,
713; friendly casualties, 320; ratiO, enemy/friendly, 2.2.
It is appropriate to compare the depletion of certain friendly companies
with that of the enemy rifle battalions. Companies F and G were penetrated
. by the enemy's frontal attack. According to Table D1, both had an initial strength
of 160. Through 5 September Co F suffered 63 casualties and Co G, 38. If the
casualty-reporting for these hard-hit units is taken to be 65 percent complete,
as suggested earlier, and if it is assumed that no replacements actually joined
them at the front, then for Co F the remaining strength would have been 39 per
cent of the original figure, and for Co G, 63 percent.
The deployment of the opposing forces at the beginning and end of the
enemy's first push is shown with attached strength figures in Fig. 7. Only units
astride the highway, i.e., in the area of penetration, appear. On the west flank,
the bulkofthe 8th Cav Regt's IstBn, with Co L and elements of Co M (and with a
strength of 617), was, in effect, opposed by the very weak 21st NK Regt (strength
.486), though there was little actual contact. By the time of the withdrawal here
on 6 September friendly strength had increased to about 800, so there was
little danger of envelopment by the enemy.
ORO-T-261
86
UNCLASSIFIED
SlUff
---
--
0 2-3 September
::0
FRIENDLY "N",MY
0 MLR
I
>-l
I 2d Line 1st Line
Forward 'Reor
.....
'" Support Co line Co Support Co Une Co
'"
3 Bns/l9th Reg' 720 19th Regt (-) 337
C/Eng' 175 K 172 H 147
=G160
A/Eng' 175 I 166 A 167
2 Bns/23d Regt 480 23d Regt H 289
(2/.l) M 88 H2 95 F 160
H3 104 E 160
=
1200 626 542 338 242 647
~
~
__880
---------- 889
~
~
1826 1769
5000 ,/0
6 September
FRIENDLY
Support Co Line Co
M 121 A/Eng' 163
C
H3 101 I 157
2:
('") SV 131 K 167
r-
H 124 L 168

H2 80 F 63
C/)
C/)
~ 5 7
718
~
~
-
1275
", .
0
MLR
ENEMY
<
Forward Rea,
19th and 23d 19th and 23d
Regts Regts
C
2:
('")
r
:l;.,.
0-'
C/)
r
487 500
2J
r.,
~
Fig. 7-Lineup and strengths of opposing forces astride the highway,' and 6 Sep 50.
Strength figures for 6 September assume no replacements received by either side.
0
UNCLASSIFIED
The 13th NK Div evidently achieved its limited advance to Tabu-dong and
beyond at the cost of temporary loss of combat effectiveness. Though its
casualties were little more than double those on the friendly side, the low initial
strengths did not afford leeway for much depletion with retention of organiza
tional structure. The advance was evidently achieved by the concentration of
force in a very small area, i.e., against Co G and Co F. Because of the limited
scope of the offensive, it would seem that friendly units on the second line of
defense could easily have moved up to meet the enemy attack and might well
have'held the original line. There were, of course, reasons why the reserves
were not so committed.
ORO-T-261
88
UNCLASSIFIED
SIEHl
UNCLASSIFIED
ENEMY SUPPORT
Enemy supporting arms originally comprised armor, artillery, and mor
tars. Most of the tanks and SP guns were lost north of Tabu-dong just before
the period under study here. As the circumstances of this unsuccessful employ
ment of armor are of considerable interest, this prior action is analyzed in
a succeeding subsection. Only fragmentary data are available on the number
and dispOSition of enemy mortars and artillery pieces at the beginning of Sep
tember 1950, but detailed and almost comprehensive data are available for the
time when the battle was joined on Hill Mass 314. This section is therefore
chiefly concerned with dispositions in the vicinity of Hill Mass 314 on 12 Sep
tember, and coverage is extended to include a subsection on small arms and
automatic weapons in direct support of enemy infantry on the hill mass at this
time.
WEAPONS ON AND IN THE VICINITY OF HILL MASS 314
Supporting Weapons in General
One conclusion reached in the earlier section on enemy units was that by
12 September units of both the 1st and 13th NK Divs were or had been on Hill
Masses 314 and 570. The artillery and mortars used by the enemy in reSisting
the attack on Hill Mass 314 by the 3d Bn, 7th Cav Regt were almost entirely
those of the 13th NK Div. The reasons why the 1st NK Div did not actively
support its own forward units are considered to be: (a) it did not have many
supporting weapons, (b) its weapons were so located that only a very poor shot
could be had at friendly units on the southern part of the hill mass, and (c) its
rear units could not adequately observe friendly movements there. The evidence
for these statements follows.
Information on the heavy weapons of the 1st Div is to be had from reports
of interrogation of several prisoners of war. In the report ATIS 1430 a senior
lieutenant from the 1st Div's AT Bn located the division CP and its artillery
regiment 5
1
/
2
miles north of Hill 314 in the vicinity of the coordinates 1156.7
1464.4 on 15 September. He stated that the 2d Bn of the regiment had two 76
mm guns, and the 3d Bn two 122-mm howitzers. No mention was made of a
1st Bn. Further details from this prisoner are available in a field interroga
tion report by the 1st Cav Div. In this report the artillery regiment is located
at 1157.7-1468.4 and stated to have, perhaps in all three battalions, three 76
mm guns and three 122-mm howitzers. The division was evidently very deficient
ORO-T-261
SlEREr
UNCLASSIFIED
89
UNCLASSIFIED
in these weapons, since the normal number with a full complement of personnel
is 18 of each. The prisoner added that all tanks in the 1st Div had been sent to
the 13th Div. From another POW, there is some evidence that the 14th NKDiv's
SP-gun Bn moved to Tabu-dong early in September and then withdrew on or
about 8 September. Since the 1st Div was advancing on Taegu directly from the
north through rough terrain, it would appear that available heavy equipment,
including artillery, had been assigned to the 13th Div because of the good roads
leading from its area directly to Taegu. This would be a concrete example of
the kind of intradivisional coordination discussed in the previous section.
No detailed information is available on the heavy mortars or artillery of
the 14th and 2d Regts of the 1st Div. The fact that they were committed piece
meal and moved onto Hills 570 and 314 at night from the ridges to the north and
east suggests that heavy equipment was not brought along. Also, the broad
open valley south ofthe Walled City in the 1st Div area (see Fig. E9) did not
offer suitable intermediate positions in defilade for artillery or heavy mortars
even if the division had any.
Furthermore, the configurations of Hill Mass 314 and Hill 660 simply did
not offer good targets. The gentler reverse slopes of Hill 660 could be reached
by howitzers and mortars, but such fire would be pointless without observation
and without frontal assault, impossible up the steep forward slopes (see Figs.
3 and E8). The US positions on the southern part of Hill 314 could also be
reached by high-angle fire, but again, such fire would not be useful without
observation of friendly units and the movements of the enemy's own troops.
By 12 September the ROKA advance had reached the vicinity of the 783-m point
on the northeast ridgeline, leaving the enemy even distant observation only from
here to the 900-m elevation of Kasan. Very little could be seen of Hill 314
south of Knob 2.
Circumstantial evidence also suggests that 1st Div mortars of 82-mm size
or greater were not involved in the fighting on Hill 314. Previous to 12 Sep
tember, and especially on the 11th, 8th Cav units had fought back and forth for
many hours each day on Hill 314; however, theywere not punished with anything
like the intense mortar barrage directed at the 3d Bn, 7th Cav. As discussed
in Part IV, it is surmised that on the night of 11-12 September the 3d Bn of the
19th Regt, 13th Div moved onto Hill 314, establishing a new and potent position
of four 82-mm mortars at the north end of the hill. Also, Hill 314 was now
included in the field of fire of the 13th Div 120-mm mortars (if it had not been
before). From Hill 570 friendly movements on the ridge and the left side of
Hill Mass 314 were nakedly displayed, as shown by the photograph from this point
(Fig. 3). From one central observation post on Hill 570, the fire of both groups
of mortars and of regimental artillery could be controlled.
Number and Type of Mortars and Artillery Pieces
Authoritative enemy data are available from the enemy on the number of
weapons in the 13th NK Div. A comprehensive count of personnel and weapons
in all 13th Div units was made on 11 September. These statistics happened to
be in the pocket of the CofS of the 13th Div when he surrendered later in the
month. Table 9 presents the figures on weapons in the 19th Regt, 21st Regt,
23d Regt, Arty Regt, and certain other units. For comparison, TE figures for
full-strength infantry and artillery regiments are included. As shown in a
ORO-T-261
90
UNCLASSIFIED
SInn
StERn
UNCLASSIFIED
previous study, 25 the division was hardly underequipped, in proportion to its
reduced strength. For its manpower at that time, the 19th Regt, as an outstand
ing example, had 275 percent of the normal number of 120-mm mortars and
275 percent of the normal number of 76-mm guns. The 1st NK Div certainly
was in no such favorable position. The enemy document used in constructing
Table 9 does not specify the model of each weapon listed, but, among 76-mm
pieces, does distinguish those of the type for regimental support (howitzers)
TABLE 9
NUMBERS OF CERTAIN WEAPONS 11\ MAJOR UNITS
OF THE 13TH NK DIY 11 SEP 50
Weapon
Actual numhersc
Inf I
Regt
Arty
Regt
19th
Regt
I 21st
Regt
I 23d I Acty I SMG
Regt Regt Sn
I AT-gun
Sn
I SP-gun
Qn
LMG 118 13 7 15 8
HMG 54 12 ,12 5 6 2
AAMG 6 2 3 5
14.5-mm AT Rifle 18 12 2 12 34
45-mm AT Gun 12 4 4 5 II
82-mm Mortar 27 4 2 6
12(}.mm Mortar 6 6 2
76-mm How 4 4
76-mm Gun m 8
122-mm How 18 9
SP Gun 2
8Data from captured enemy document, which states "These statistics were made 11 September 1950."
bSource: Document, one loose sheet, handwritten, carried on person of Sr Col Lee Hak Ku, CofS of the
13th Div, NKPA, when he surrendered to US forces at 1153.3-1454.3 on 21 Sep 50. For the complete table
see Hef 25.
cUS intelligence estimates as of September 1950.
and those for divisional artillery (gunS). The distinction between guns and
howitzers and among the different models of howitzers is important. Artillery
fire can usually be distinguished from mortar fire by the scream of the pro
jectile at supersonic speeds and, in the case of guns, perhaps by the flat tra
jectory. High-angle fire from one weapon, the 76-mm howitzer M1943 (which
has an extremely low, subsonic muzzle velocity about ihe same as mortars
at long range), probably could not be distinguished from mortar fire. As ade
quate data on the models of its weapons are lacking from the 19th Regt itself,
recourse must be made to information from other sources, including the inter
rogation of prisoners from other units. It is not known whether the prisoners
volunteered the model designations or merely chose among pictures of various
weapons shown them. .
The Asst Regt Cmdr of the 45th Regt, 15th Div (ADVATIS 0958; Item 2,
App C) gives the assignment of weapons within his division as: r>iv Arty Regt
122-mm How M1938, 76-mm AT Gun M1942 in all-purpose role; Regt Arty
ORO-T-261
UNCLASSIFIED
SECRET
91
UNCLASSIFIED
76-mm How M1927; and Div AT Bn, Regt AT Co-45-mm AT Gun M1942.
A number of additional interrogation reports indicate that the only 122-mm
howitzer supplied the North Korean Army was the M1938 model, and the only
76-mm field gun the M1942 model.
The following weapons can be identified in photographs of the Tabu-dong
area taken in October 1950: 45-mm AT Gun M1942; 122-mm How M1938; 76-mm
How M1927.
Besides the M1927 howitzer, 76-mm regimental artillery could have in
cluded the M1943 howitzer or the M1938 mountain gun. A private who had been
with the 19th Regt Hq from March 1950 to 7 September (ADVATIS 0879) stated
that Regt Arty had orginally been an M1938 Gun Bn, though he had no idea of the
situation around the time of his capture on 12 September. A junior lieutenant
from the Arty Bn of the 6th Regt, 2d Div (ADVATIS 0837) stated that his unit
was equipped with M 1943 howitzers, and a sergeant from the AT Co of the 9th
Regt, 3d Div (ADVATIS 0927) also said that his regiment had M1943 howitzers.
However, a lieutenant from the 18th Bn, 21st Regt, 13th Div (ADVATIS 0926)
stated that as of 18 September he saw ten 76-mm pieces, M1927 and M1942,
emplaced along the hills 5 km southeast of Tabu-dong; some of the M1927
pieces may have belonged to the 19th Regt.
The former S-3 of the 3d Bn, 7th Cav Regt said that from Hill 314 he
positively identified an enemy 76-mm mountain gun up on Hill 570. This weapon
is exceedingly similar to the M1943 howitzer; neither is very different from
the M1927 howitzer.
The Technical Intelligence Branch, G-2, Dept of the Army, reports that
ordnance intelligence teams have scoured the battlefields and in the entire
Korean war picked up only one model of 76-mm howitzer, the M1927, with the
exception of one M1910 model that had been with Chinese forces.
Though it is possible that the enemy may have destroyed other models, it
is concluded that the regimental artillery of the 19th Regt, 13th Div consisted
of 76-mm howitzers M1927.
Though most prisoners reported the M1942 long-barrel model of the 45
mm antitank gun, mention was also made of the M1937 short-tube version. The
leader of the 2d Plat, Co B, 70th Tk Bn, which ran the gantlet of these weapons
north of Tongmyongwon on 6 September, noted that both types were present.
Former personnel of the 99th FA Bn stated that air obserVers had noted the
distinctively long barrels of the M1942 model, but that these weapons in general
were hard to locate from the air. Probably both types were present on 12
September.
There is no problem in the identification of mortars. The CofS of the
13th Div (ADVATIS 0900) stated that at the time of his capture on 21 September
the only 120-mm mortars the division possessed were the five model M1938's
buried at the north edge of Hill 570. These almost certainly had belonged to
the 19th Regt. At any rate, the M1943 modification which had been issued to
the 15th Div (ADVATIS 0755) differed only in minor respects. As for 82-mm
mortars, the CofS identified the nine remaining in his division on 21 September
as M1937 model's. Very likely all 14 in the division on 11 September were of
this model.
ORO-T-261 92
UNCLASSIFIED
SUiU.
UNCLASSIFIED
Location of Mortars and Artillery Pieces
Evaluated data on the exact locations of enemy weapons are not available.
Such information might be expected from the artillery observation battalion,
but it was in no position to cover the front. Recourse must be made to obser
vations from any source whatever.
The 1st FA Obsn Bn arrived in Korea on 25 Aug 50, under strength (61%'
of TO), and without full equipment." Batteries B and C were attached to the
1st Cavalry Division when it took over the defense of the Tabu -dong area at the
end of the month. Though full equipment had not arrived on 4 September, it
was considered that sound, flash, and radar operations could be undertaken.
Battery B went into positions in the sector of the 7th Cay Regt and Battery C
with the 8th Cav Regt in the vicinity 1153.5-1450.3.
Battery C's radar went into operation at 1115 on 4 September but closed
down in the evening when the battery was withdrawn due to the uncertainty at
the frpnt line. Sound and flash equipments were installed on 9 September. The
battery's three OPs were located in a cluster: No. I at 1156.7-1445.7, No.2
at 1155.3-1445.5, and No.3 at 1154.1-1444.8. The radar was located nearby
at 1156.1-1446.2. The first radar target, an unidentified movement in the vi
Cinity of the Walled City, was picked up on the evening of 9 September.
The character of the terrain placed the observation battalion at a severe
disadvantage. Because the ridges and peaks decreased in elevation from the
enemy to the friendly positions, much territory was screened from visual and
radar observation, and the deep valleys probably hindered accurate sound rang
ing. From BtryC'sthreeOPs only a very short segment of the enemy front
line could be seen, but all of the Taegu-Tabu-dong road from south of Hill 373
to within a few thousand yards of Tabu -dong was included in this narrow angle.
From enemy sources it is known that the 13th NK Div Arty was along the road
within the visible stretch north of Tongmyongwon (12,000 to 13,000 yd from the
OPs), and the 19th NK Regt Arty was very likely at Tongmyongwon. Battery
C's flash observers were then suitably placed to observe enemy artillery if it
fired, and since virtually all of the valley north of Hill Masses 570 and 314 was
hidden from them, there was no need to divert attention from the Taegu-Tabu
dong road.
On several occasions from 9 through 21 September, Battery C's observers
noted enemy movements, and several times they looked for enemy guns already
observed by other units. The following list gives all enemy gun positions actu
ally picked up by the flash observers throughout the period and logged by the
1st FA Obs Bn:13 '
10 Sep. 1500. Gun firing at 1153.15-1456.20 [on road at Hill 263]. 1st Cav Divarty re
ports gun too close to friendly line to fire on.
1520. Tank at 1153.0-1456.7 [on road, just north of previous position]. Battery
C flash OP adjusts fire of 9th FA Bn, several direct bits reported.
12 Sep, 0440. Location [unspecified] of enemy gun reported by flash OP, wliich at 0515
adjusts fire of 9th FA Bn.
1445. Enemy gun observed at 1152.50-1458.00 [on road, 500 yd north of
Tongmyongwon] .
14 Sep, 0810. What appear to be flashes observed at 1153.87-1451.75 [on road in friendly
held territory].
ORO-T-261
SEERU
UNCL,A,SSIFIED
93
S5EMT
UNCLASSIFIED
There is evidently rather strong negative evidence to confirm the POW
report that 13th NK Div Arty did not fire.
Doctrine for an artillery observation battery calls for a sound base within
3000\to 5000 yd of the front line. The enemy never took positions closer than
within approximately 6500 yd of Battery C's OPs, yet its sound base was with
drawn on 10 September to a new location southwest of Taegu. Battery B of the
1st FA Obs Bn then extended its sound coverage to include the sector of the
Taegu-Tabu -dong road. The only significant target of Battery B' s sound rang
ing logged through 20 September was at 1255 on 13 September, when an enemy
gun of unknown class was located at 1152.37-1456.95, just south of Tongmyongwon
and on the west side of the road.
To determine the source of the fires received by friendly units on Hill
314, one must first put together the reports of observers from various units on
the ground and in the air and then combine them with the observations of both
the men subjected to these fires and POWs.
Enemy weapons were not readily observed from the air. As the reports
from the artillery air-observers are presumably included in the unit logs, the
writer has not consulted the former air observers of the 1st Cav Div's artillery
battalions. An Army observer who flew regularly in one of the 1st Cav Div
mosquito planes said that enemy mortars were very hard to distinguish from the
usual flying altitudes, and that airmen had great difficulty in identifying a partic
ular ridge in relation to the map. Also, security required that they be told little
and report little via radio. The writer has not communicated with the air ob
servers of the 1st Cav Div's artillery battalions.
Figure 8 is the plot of enemy weapons on and around Hill 314 arrived at
after consideration of all available evidence. In first form, it was made by
the former S-3 of the 3d Bn, 7th Cav Regt over two years after the action, from
memory and without the aid of any references whatsoever. Reports of enemy
weapons logged at the time of the action were then keyed to this plot as a basis
for comments and correction by several officers formerly with the 77th FA Bn
and, independently, by the artillery liaison officer who had been on Hill 314
calling in the battalion's fire during the attack on 12 September. From the
remarks of these officers a few changes were made. Further adjustments were
made to fit the testimony of POWs, statements by various infantry personnel
interviewed and, as a last, rigorous condition, the 13th Div's
Concentrations of friendly artillery fire shown in Fig. 8 are discussed in
the next section of this memorandum. It may be noted here that most of the
positions of enemy supporting weapons were not subjected to counterbattery
fire during the battle on 12 September. In its anti mortar role, the 9th FA Bn
fired on various positions rather to the rear, as it was generally surmised that
the troublesome 120-mm mortars could be anywhere, even back with the artil
lery near Tabu-dong. It now appears, however, that all enemy mortars were
well forward. Of 13th Div units in the area, only the 19th Regt had mortars,
and it and its weapons were in the vicinity of Hill 570.
The sites of reported enemy supporting weapons are discu5sed individually
below. Where no weapon is shown, the reason is given. Unless otherwise
noted, a given position was plotted by the former S-3 of the 3d Bn, 7th Cav, and
was accepted by the artillery officers without comment.
ORO-T-261
94
SiEttET
UNCLASSIFIED
UNCLASSIFIED
Slnn
Site 1. A map drawn by the former CofS of the 13th NK Div (ATIS 1468)
shows five 120-mm mortars buried at this location. According to a former
officer of the 99th FA Bn, there was a rumor at the time that the enemy had
actually carried his heavy mortars up on the ridgeline of Hill 570 in order to
secure greater range in the direction of Taegu. Since there is reason to believe
that all the 19th Regt 120-mm mortars were located at Site 4 on 12 September,
such a move would have had t.o be between then and 14 September. An active
mortar position at Site 1 was not reported in any of the logs or by any of the
8th or 7th Cav personnel interviewed.
The presence of a gun in the vicinity of Site 1 is confirmed by the follow
ing report (made during the final battle for Hill 314): "121310. From Arty
Liaison Officer with 7th Cavalry:' Receiving possible tank fire from 1155.0
1456.9 or north of these coordinates." The day before, a gun had fired from
this vicinity at the 3d Bn, 8th Cav Regt in 'its attack on Hill 314. "111340.
From Arty Air Observer: In draw in grid square 1155-1457 there is a gun firing
direct fire at hill at 1156.5-1455.5 [Hill 314J."
This weapon is presumed to be one of the two SP guns of the 13th Div. It
could not have belonged to the 1st Division, for a prisoner from the SP Bn of
this division (ADVATIS 0871) reported that from 8-16 September this unit, with
two SP guns, remained at Hyoryong, well to the north. As shown in Table 9,
the 13th Div also had two SP guns, and one of these must be the subject of the
following report (sometime on 12 September): " ...The 1st FA Observation
Battalion reported a self-propelled enemy gun and adjusted the 9th FA Battal
ion, scoring two direct hits on the target.,'0
These hits apparently had little effect, for a prisoner from the 13th Div's
SP Bn (ADVATIS 0925) stated that the two SP guns were only destroyed by air
attack on 21 September. The enemy lost the guns as of this date but they were
not damaged. According to the former leader of the 2d Plat, Co B, 70th Tk Bn
in the breakout around 21 September two SP guns were captured intact at
Tongmyongwon (one was d r i v ~ . l some miles), and a third was reported to be
nearby in the .hills.
Site 2. These four guns were added to the plot by former officers of the
77th FA Bn. The following message sent during the battle appears to refer to
these pieces but locates them somewhat inaccurately at the location shown as
Site 3: "121455. From Shamrock [Div ArtyJ: From photographs the location
of four guns at 1152.8-1457.2 has been confirmed."
Two days later further reports were made on this position. The coordinates
in the following message are plotted as Site 4 but probably were meant to re
present Site 2: "141315. From Divarty Air [throughJ Shamrock 3: Found three
artillery pieces and one tank at 1152.6-1457.6"
Another report at about the same time speCifies coordinates close to Site
2: "1415.55. From Divarty: ...Nathan Hale [1st FA Obsn BnJ reports suspected
two pieces at 1153.58-1457.88. ,I
The four pieces plotted at Site 2 are considered to have been 76-mm how
itzers M1927, the infantry support artillery of the 19th Regt. One of the four
could conceivably have been dragged up on Hill 570 the next day and been iden
tified there as a 76-mm mountain gun, leaving three to.be observed at the ori
ginallocation on 14 September. The confirmatory evidence for Site 2 is nega
tive but strong. As discussed later, the weapons thought to be at Sites. 5 and 6
ORO-T-261
SECRn
UNCLASSIFIED
95
UNCLASSIFIED
by personnel of the 7th Cav Regt simply weren't there, or they would have been
observed by units of the 8th Cav Regt occupying adjacent sites. The only pos
sible site for regimental artillery, the only logical one in view of communication
with the observation point on Hill 570 and defilade, is Site 2, as shown.
As given in Item 2, App C, a regimental commander of the 15th NK Div
who was captured on 28 September, stated that the 76-mm regimental howitzers
would be emplaced at least 2 km to the rear of a hill on whose top the observa
tion point was located. Site 2 is 1'j. km behind the OP on Hill 570; it would there
fore be a forward pOSition. Since it was enemy doctrine to move artillery into
new emplacements at night, and since the first friendly observation of the posi
tion at Site 2 was apparently on 12 September (though 11 September was clear)
it is likely that the 19th Regt howitzers only were emplaced on the night of 11
12 September at Site 2.
Site 3. See under Site 2.
iiU. The enemy's 120-mm mortars are definitely located here on 12
September by the report: "121515. Scrappy Red 5 [Executive, 1st Bn, 8th Cav]
reports that all Forward Observers report 120mm mortars are located at
1152,6-1457.5. Further reports that all movement draws fire from these
mortars."l
According to the former CO of Co B, 8th Cav, the mortar fire that caused
apprOximately twenty-five casualties in Co B on Hill 225 around 0730 on 12
September was the first such intense fire received. The position at Site 4 was
hidden from direct observation from Hill 225 and the outpost hill just to the
north, and undoubtedly was responsible for these casualties. It apparently only
became active on 12 September.
All six 120-mm mortars of the 19th Regt have been plotted at Site 4.
Enemy doctrine (see ADVATIS 0755, Item 1, App C) called for six mortars in
line. They are shown in two groups of three because fire on Hill 314 was noted
as coming in two adjacent concentrations and in 3-rd salvos. Though this
location would seem to leave the mortars vulnerable to counterbattery fire and
south-to-north passes by friendly aircraft flying up the valley, it would permit
the mortars to fire from a reverse slope on friendly positions on Hill 225 and
the southern slopes of Hill 624, as well as on Hill 373, Hill 314, and elsewhere.
The extent of the cover and camouflage offered by the town of Tongmyongwon
is not known.
The four 45-mm antitank guns of the 19th Regt have been placed by the
writer in the vicinity of Site 4 on the basis of the following reports:
131200. From D1varty: Field piece at 1152.48-1457.42 fired on and silenced.
131420. From Scrappy 2 [8th Cay S-2]: Reports two AT field pieces and ammo
dump at 1152.9-1457.8, reported by FO of Spirit [9th FA Bn]. Arty was put on them but
could not reach the guns. OUr tanks ran into the fire of these pieces at 1305 but with
drew....1
Personnel of the 8th and 7th Cav Regts reported that for several days
before 12 September friendly movements across the mouth of the valley drew
fire from small-caliber high-velocity weapons. Air observers' reports, as
related by the former S-3 of the 99th FA Bn, stated that 45-mm AT guns were
responsible.
ORO-T-261
96
S.EERrI'
UNCLflSSIFIED
UNCLASSIFIED
5eRET
1153 1154 1155 1156 1157 1158 1159
1458
1458
1457
1457
1456
1455
1454
1453
1452
1153 1154 1155 1156
Enemy Weapons
-I
t
t
US Bar y Mort
~
LMG Gun or How
HMG (;
SP Gun
1157 1158 1159
Fire by:
0
77th FA Bn, Btrys B
& C (105-mm How)
@ 9th FA Bn, elements
(155-mm How)
1160
Fig. a-Enemy weapons and friendly artillery concentrations.
Thirteen probable sites of enemy weapons ore given. lOOOyd grids
are numbered; light lines indicate lOOO-m grids. Caliber of enemy
weapons is given in some instances.
ORO-T-261
UNCLASSIFIED
HER!T
97
UNCLASSIFIED
Fire at the southern mouth of the valley was also received from a self
propelled gun according to the former leader of the 2d Plat, Co B, 70th Tk Bn.
On one patrol up the highway in the vicinity of the village of Chungsan, the tanks
received high-velocity flat-trajectory fire, judged to be 76-mm. Night after
night, fire of this same type landed against the knoll down the valley at the
tanks base position and never.landed anywhere else, as if there were some
limitation on the traverse of the weapon. Clearly this limitation consisted of
the walls of the corridor formed by the southwe stern slope of Hill Mass 351
263 and the northwestern slopes of Hill 570. Situated in the assumed position,
very close to the knoll north of Tongmyongwon at the coordinates 1152.7-1458.4
the presumed SP gun would have an exceedingly narrow fan of fire, but one that
would just include the knoll at the tanks' base position and, 6500 yd up the valley,
the village of Chungsan.
Site 5. The former S-3 of the 3d Bn, 7th Cav, who had observed the mor
tar fire received during the battle on Hill 314, later placed four 120-mm mor
tars at this site. No logged reports can be found to confirm this improbable
position. Instead, for the reasons given above, these mortars have been located
at Site 4.
The situation overlay of the 77th FA Bn for the period 101700-111800 Sep
tember shows a tank atthis site, probably the SP gun thought to have been based
in the vicinityof Site 4.
Site 6. Enemy artillery was placed at this improbable location by several
officers. It is considered that it was not artillery but one of the 13th Div's two
SP guns which had been temporarily moved down on 11 September from its
normal position north of Tongmyongwon, and was again moved to an advanced
location on the morning of 12 September under the cover of mortar fire. The
evidence for this conclusion follows.
(a) The former CO of Co B, 8th Cav, which was on Hill 225 with outposts
on the next hill to the north"adjacent to Site 6, stated in a recent communica
tion that no enemy gun positions were ever observed in the valley. Neither
were any ever observed by the personnel of the tanks that patrolled the valley.
(b) Around 0730, Co B, 8th Cav was subjected for the first time to intense
120-mm mortar fire, and sustained many casualties.
(c) At 0730 a gun was firing nearby, according to the following report:
"120735. From S-3, 7th Cavalry. I&R Platoon reports flash from artillery
pieces in vicinity 1152.9-1454.5.
1
(d) These coordinates are of a point somewhat further south along the
road and are plotted as Site 7. Friendly tanks at this time were not patrolling
the valley.
(e) Later in the day, the tanks of the 2d Plat, Co B, 70th Tk Bn found a
neat, untended pile of boxed ammunition along the road in the vicinity of the
village of Chung san.
(f) Several days after 12 September, the former S-3 of the 3d Bn, 7th Cav
while on reconnaissance up the road found a large number of 76-mm shell cases
in the vicinity of the village of Samsan-dong. These could either have come
from an unusual expenditure of ammunition by a single SP gun or could have
been the litter of the destroyed ammunition pile otherwise located slightly to
the north.
Site 7. See under Site 6.
O R O - T ~ 2 6 1
98
UNCLASSIFIED
SHUT
.sHRtT
UNCLASSIFIED
Site 8. Mortars of unspecified caliber are located on the ridge of Hill 570
by the report: "132015: From G-2: S-3, 8th Cavalry reports Heavy Mortar
Company [positions in the valley to the rear of Hills 225 and 373] destroyed
two enemy mortars at 1154.4-1455.7. Suspect more in this position." The
mortars are considered to have been at the somewhat different coordinates of
Site 9, and are so plotted.
The presence of an automatic weapon at Site 8 is confirmed by the message:
"131420. From Scrappy 2 [8th Cav S-2]: Enemy MG at 1154.7-1456.0.,1
Site 9. The coordinates in the following report made during the battle for
Hill 314 seem interchanged with their parenthetical explanations: "121330.
Divarty S-2 reports: At 1154.9-1456.1 (saddle on Hill 570) [.ak]100 people dug
in (VV side of saddle). Also on east side of saddle ammo boxes, 50 soldiers dug
in. At 1154.9-1455.0 (top of Hill 570) 100 people dug in.,1
As noted in the earlier section on enemy units, this observation of per
sonnel in the saddle on Hill 570 corresponds with the location given by POWs
for the Mort Co, 1st Bn, 19th Regiment (see Item 4, App A). The three platoons
of the company were in line one or two hundred yards apart. The site in the
saddle is the only one possible for such a disposition. One more mortar is
shown than originally plotted by the former S-3, 3d Bn, 7th Cav, in order to
fit the relevant report quoted under Site 8 and attributed to Site 9. At least
one of the pieces was a captured US 60-mm mortar. Men on Hill 314 were
quite definite about having received fire with the distinctive burst of this caliber
of shell.* More conclusively, this caliber was speCified by One of the prisoners.
A hearsay report passed on by a prisoner of a rear unit of the 1st NK Div had
it that this division collected and issued US material. Since the 13th NK Div's
statistics do not mention US items, it is suspected that if 60-mm mortars were
on Hill 570 they had been the property of the 14th Regt, 1st NK Div, turned over
to the 19th Regt, 13th NK Div when the former regiment's units on the hill had
been decimated.
Again according to the 19th Regt prisoners, on 12 September 200 rd of
ammunition were on hand for the mortar of one platoon. By 14 September one
third of this amount had been expended. The prisoners also stated that the
mortar company had telephones, presumably linking it with the observation
post on the 570-m peak.
Further confirmation of the presence of mortars at Site 9 is afforded by
this report: "121835. White 2 [S-2, 2d Bn, 8th Cav] reports...four mortars
and some machine guns on reverse slope of hill opposite G Company. White
mortars taking this pOSition under fire now.... "
The fermer S-3 of the 3d Bn, 7th Cav was rather positive about the presence
of a 76-mm mountain gun at Site 9. The date of the observation is not known.
This weapon is assumed to have been on Hill 570, possibly brought up by 1st
NK Div units.
As Table 9 shows, the 19th Regt possessed easily enough automatic weap
ons to account for those plotted near Site 9.
Site 10. The former platoon sergeant of the P&A Plat of the 3d Sn, 7th
Cav stated in a recent personal interview that during the later stages of the
battle on 12 September, while he was bringing up ammunition to the troops then
Soviet 50-mm company mortars, with a maximum range of 875 yd, could only have reached halfway to
Hill 314.
ORO-T-261
SUI" UNCLASSIFIED
99
UNCLASSIFIED
advancing north from Knob 2, he heard the regular firing of enemy mortars
from defilade at the northern end of the hill mass. When he returned on the
next trip, the firing had stopped. Later, while on reconnaissance around the
hill, he discovered the position from which the fire had come. Of four 82-mm
mortars that had been emplaced, only one base-plate remained. Nearby, 600
rd of 82-mm ammunition were stacked in small piles. There were no empty
boxes to indicate the number of rounds that had been fired. Single -strand tele
phone wire led from the position.
Mortars at Site 10 were not definitely reported by anyone else, though
several officers, when shown only the mortars plotted at Site 11, said they
thought enemy mortars had been at the north end of the hill.
It is assumed that the four 82-mm mortars at Site 10 belonged to the 19th
Regt and were all it had of this caliber. They almost certainly were controlled
from Hill 570, since no enemy observation of the concentrations falling in the
saddle just south of Knob 2 was possible from Hill 314 itself.
Site 11. The location of three mortars here was confirmed only taCitly by
the artillery personnel who were shown the plot. The mortars represented were
probably those already definitely located at Site 10. Since that position accounts
for all 82-mm mortars of the 19th Regt, mortars at Site 11 would have to be
120 mm in caliber. Their location here appears unlikely, because (a) they probably
could not have been quickly and completely evacuated at the end of the battle,
(b) In any case, the position would have been discovered by the P&A Plat recon
naissance, (c) the position would have been in full view of US troops on Hill 373,
and (d) for 120-mm mortars, such a position would have been needlessly far
forward.
Site 12. The only evidence for a position here is the following report.
"140825. From S-2, 7th Cavalry: POW captured last night on Hill 314 states ...
that three large guns are 300-400 yards. east of temple 1157.2-1457.9."' Since
no fire was received from this area, no piece has been plotted.
Site 13. No weapon is plotted at this location because the sole evidence is
the report: "141555... Nathan Hale reports a field piece at 1156.64-1457.85."1
Enemy Weapons on Hill Mass 314
Information from the enemy is available on the weapons possessed before
and after the battle by two of the companies on Hill Mass 314. The source is
the interrogation of a prisoner thought to be from the 2d Bn, 2d Regt, 1st NK
Div (see Item 11, App A). Personnel and weapons of the prisoner's 4th Co
and of the HMG Co are listed as of 14 September. Parallel "original" figures
are also given. These are taken to represent the situation as of 11 September,
because (a) the prisoner only joined his unit as a replacement on this date, (b)
he was one of 80 replacements assigned to the 4th Co, and this massive rein
forcement would explain the almost TO strength figure, and (c) he was issued
60 rd of ammunition, as listed in the "original" column.
As reported by the prisoner, before the battle 108 of the 120 enlisted men
in the 4th Co had M1891/ 30 rifles with 60 rd of ammunition, and 12 men had
submachine guns. Each man had two grenades. Personal weapons are not
listed for the 40 enlisted men of the HMG Co, but it was well equipped, with
five Maxim heavy machine guns and a box of ammunition for each.
ORO-T-261
100
UNCLASSlFlED
.sRf(y UNCLASSIFIED
By 14 September both the 4th Co and the HMG Co had been reduced to about
40 percent of their "original" strengths. Personal weapons were lost in pro
portion, Le., each survivor kept his weapon. All five machine guns were retained.
Presumably all losses occurred in the battle on Hill Mass 314 on 12 September.
TABLE 10
PARTIAL LISTING OF WEAPONS CAPTURED OR
RECOVErlED ON HILL MASS 314
Materiel
Amount & type of materiel listed by:
7th Cay Periodic I' 5-3 Jnl,
LogisticsReptNo.36
8
3d Bn, 7th Cav
b
Additions & comments
by 5-3, 3d Bo
Enemy materiel captured
LMG
HMG
Submachine guns
Rifles
Other
US materiel recovered
MG
Auto rifles
Carbines
Rifles
Other
1, Soviet 7.62-mm DP
90, Soviet 7.62-mm,
M1891(30
2. cal 30, M1919A4
2, M1918A2
6, cal 30, Ml
50, cal 39, Ml
2 mounts, tripod,
cal 50, ~ 1 3
1, Japanese
1
23
72, Korean
e
1, cal 30-32
I, cal 50
2, BAR
51. Ml
1 8I-mm Mort
1 SeR-300 radio
Many more captured
i\'lany more than 72 captured
d
3 long AT guns
Pistols
1\lany grenades
Total of 6 US 1\:lG were
recovered
At least a dozen were
recovered
8Covering period 161800 to 171800 September.
bEntry dated 161800 September.
cAll North Korean weapons were of Soviet' origin.
dMany rifles ,:"ere not cOllnted becallse they had been Ilsed as improvised splints and litter poles.
Table 10 is a partial listing of enemy weapons captured on Hill Mass 314
and of US weapons recovered. The basic listing is an entry in the S-3 journal
of the 3d Bn, 7th Cav. Over two years after the action, comments and additions
to this list were made by the former S-3 himself, and these are summarized in
the adjoining column. Another list appeared in one, and only one, of the 7th Cav
periodic logistics reports, this one on the day after the battalion's listing. Since,
except for the capture of Hill Mass 314, the 7th Cav Regt did not advance any
where until 17 September, and then made only limited gains, it is likely that all
of the heavier weapons listed were taken on Hill 314, Some of the rifles and
carbines may have been picked up after skirmishes with enemy patrols that
penetrated US rear areas. The type of rifle picked up, M1891/ 30, was the same
as that reported by the prisoner from the 8th NK Co.
Entries that appear to refer to the same weapons are tabulated in parallel.
In aggregate, the list roughly checks the layout of weapons plotted in Fig. 8 by
101
ORO-T-261
UNCLASSIFIED
SHirr
UNCLASSIFIED
the former S-3 of the 3d Bn, 7th Cav. In all, nine light machine guns and nine
automatic rifles appear. Table 10 includes two light machine guns of enemy
origin and at least two US machine guns; according to the prisoner's list, five
Soviet machine guns had been carried off. The fact that roughly twelve BARs
were recovered perhaps allows for nine of these weapons to have been used by
the enemy. The antitank rifles may also have been used against friendly troops
or automatic weapons. No enemy mortars appear in the table or in the plot of
the hill. The former sergeant of the P&A Plat was certain that none had been
picked up. It is of course possible that the light portable 50-mm Soviet com
pany mortars may have been used and removed.
Not all recovered US equipment was listed officially, because such men
tion would have required shipment of the weapons to the rear for accounting,
etc. when they were needed at once to replace previous losses. In total, large
amounts of US equipment had been abandoned prior to 12 September in various
hill actions, including the fighting on Hill Mass 314. There is no information
on the amounts of US ammunition in enemy hands.
Individual weapons (burp guns and Soviet and US rifles and carbines) total
170. The unknown number of pistols captured would increase this figure some
what. Nevertheless, if, as the previous analysis has indicated, 645 enemy troops
were present, this figure seems rather small. Estimates by US personnel and
prisoners of war place the number of enemy killed at several hundred. Ifthese sev
eral figures are correct and the majority of enemy troops were armed, then surviv
ing enemy soldiers must have withdrawn from the hill in good order, carrying
their weapons with them and even, as was their doctrine, retrieving the weap
0ns of the dead. This conclusion agrees with the testimony given by the pri
soner cited above, to the effect that survivors of the action were all still armed
and, in particular, that all five machine guns of the HMG Co were retained.
Figures E6 and E7, photographs taken after the battle, show the litter of
enemy equipment around the enemy foxholes at the very crest of Knob 2. Of
weapons, only rifles are to be seen.
SUPPORTING ARMOR
A broad study of enemy armored operations in Korea is available."'* In
formation obtained from captured enemy tank personnel and given in that study
indicates that one depleted regiment-either the 107th or 109th, and probably
the latter -of the 105th NK Tk Div advanced toward the UN perimeter along the
Sangju-Tabu-dong road. Each of these regiments had had about forty tanks at
the start of the war. After friendly forces swept back from the perimeter,
fifteen T34/ 85 tanks were found along this route, all of them south of the Nak
tong River. The total number that reached the sector could not have been much
greater. Almost all of them were knocked out before the 8th Cav Regt took over
the defense of Tabu-dong at the end of August 1950. Most of the SU -76 guns of
the organic battalion of self-propelled guns of the 13th NK Div, the enemy infantry
that advanced along the Sangju-Tabu-dong.axis, had by this time also been
destroyed. .
A detailed study of tank-vs-tank engagements in Korea is soon to be published by the Operations
Research Office.
102
ORO-T-261
UNCLASSIfiED
SURn

In renewing the offensive in September, the enemy was forced to adopt
tactics no longer based on thrusts spearheaded by tanks; with further changes
in tactics, a breakthrough to Taegu was almost achieved. Perhaps the freeing
of infantry maneuver from the limitations of tank movement was in part respon
sible. It is also pertinent to inquire whether enemy tanks, if they had still been
available and if they had been suitably employed, might not have tipped the scale
at the critical stage of the battle-in September. The engagements in August
throw light on these questions.
Armored Thrusts toward Tabu-dong
The tank-led drives on Tabu-dong were stopped by the 27th Regimental
Combat Team. Detached from the 25th US Inf Divas the EUSAK reserve, the
27th Inf Regt was committed to the defense of Taegu in the latter part of August
1950 after the enemy penetrated the line of the 1st ROK Div north of Tabu-dong. 23
On the afternoon of 18 August, after air strikes covered the sector of advance,
the 27th Regt moved out from blocking positions it had occupied at Chilgok and
Hill 373. It was supported by Co C, 73d Tk Sn (M26's), with the 1st Plat of
tanks attached to 1st Sn of infantry and the 3d Plat to the 2d Sn. '" Supporting
artillery comprised two batteries of the 8th FA Sn (105-mm howitzers) and
two batteries of the 37th FA Sn (105-mm howitzers); for the 37th FA Sn, this
was the first combat mission in Korea. Targets along the axis of advance
as far south as 1150.3-1462.6 were fired on. On the RCT's left and right flanks,
ROKA units advanced. Though enemy opposition to the RCTwas light (11 cas
ualties tabulated by the AGO for 18 August), the advance was halted 3000 to
4000 yd north of .Tabu-dong in the vicinity of the twin villages of Singi and Soi
ri because the ROK units on the flanks had been slowed by rough terrain and
enemy small-arms and mortar fire. The 1st Sn, 27th Inf deployed on the hills
along the west side of the road, the 2d Sn on the right.* The tanks of the 1st
Plat, Co C,_ 73d Tk Sn formed a line across the road in a dry stream bed, and
the 3d Plat remained behind. Sy then, it was dusk.
The descriptions of engagements which follow represent reconciliations
of somewhat discordant information from diverse sources. These include the
27th RCT's rather undetailed historical report,23 from which a tabular summary
of action in the period 18-26 August is derived and presented as App F; the
afteraction report of Co C, 73d Tk Sn;" pertinent extracts from the tabular
listing of enemy armor claimed destroyed by Co C, 73d Tk Sn (Table GI);
and lists and descriptions from surveys (discussed more fully below) of enemy
armor later found along the road (Table G 2).
In addition, the tank company's own afteraction report" and data of especial
value from recent interviews with its former personnel were available. t The
unit reports of the 8th and 37thFASns, although they contained only fragmentary
data of use for the present purpose, served to confirm the dates and general
character of the actions. The designations in parentheses identify the tanks,
so far as can be determined, in Fig. 9 and the lists of App G, and the notes
relate the data to Fig. 2.
*It might seem that this defensive line was the same as the later deployment of the 8th Cav's 1st and
2d Bos astride the foad in approximately the same place, hut the 27th Regt's forces were probably much
less widely spread.
t Ur Dr. A. Coox, Operations Research Office.
ORO-T-261
103
UNCLASSIFIED
,SURn
UNCLASSIFIED
S5ERn
The 27th RCT's advance was promptly countered. As darkness fell on
18 August, friendly infantry observed enemy tanks flanked by troops 800 yd
to the north (village south of Chonpyong-dong). The report radioed to the
infantry battalion's CP was overheard by Co C; 73d Tk Bn. Two M26's of its
1st Plat moved out in column onto a stretch of the road just north of a blown
bridge (at 1151.5-1461.8) and flanked by rows of trees. The platoon's remain
ing tanks in the stream bed were in a line roughly perpendicular to the road,
two on the west and one on the east. The five M26's of the 3d Plat were on the
road somewhat to the rear and in column about 75 yd apart. It was now 2200 and
quite dark. As the 1st Plat lead tank moved slowly forward, the men inside
could hear the clanking of several enemy tanks, about 550 yd away, as they
approached around a bend. Then they stopped.
The lead T 34 opened fire. Its fir st round was 25 yd short. The second hit
a small vehicle on the road between the opposing forces and set it on fire. The
blaze revealed a column of five T34's, with the lead tank about three hundred
yards from the forward M26. Some troops were in the ditches behind them.
(The 27th Inf report states that enemy troops were following in vehicles, per
haps further behind.) With their guns and machine guns, two more of the T34's
opened fire on the M26's to the rear and on the friendly MSR. Perhaps the two
advanced M26' s were hidden by the trees. The leading M26 opened fire on the
foremost T34 (1/7). According to the M26's gunner (acting as platoon leader),
the first round (HE) hit the front plate of the enemy tank. Expenditure of five
rounds ofHVAP then killed it, one round passing all the way through. As the
M26's in the stream bed opened up, the four surviving T34's turned and re
treated, and the enemy foot soldiers dispersed. The two advanced M26's
moved back to the stream bed, and the tanks of the 3d Plat withdrew to a se
cond position.
During the night, the enemy continued to probe with infantry and tanks
but did not reach the positions of the friendly tank company. The M26's of the
1st Plat fired blindly, and both enemy and friendly artillery were active. At
one point, a T34 supposedly abandoned among some houses off the road approxi
mately five hundred yards away (Singi-Soi-ri)opened fire on the M26's of the
1st Plat with half a dozen rounds. They missed but were coming closer when
the enemy tank was silenced by a hit on its left side by a 3.5-in. bazooka round
fired from its left front.
According to the infantry's report"' three enemy tanks were knocked out
by close-in artillery and bazooka fire on the night of 18-19 August.
The 3d Plat, Co C, 73d Tk Bn was ordered to advance at daybreak on 19
August to protect the flank of the 2d Bn, 27th Inf against new attacks. The
platoon leader moved his three effective tanks up to forward positions, station
ing his own in the lead on a rise. The T34 knocked out the night before on the
road (1/7) was now smoldering. Approximately seven hundred yards from his
own position, there was an abandoned SU -76 (which had been damaged by a US
AT-mine N/ 11) facing north. It was now destroyed with 1 rd of HE. Amid the
houses 800 yd away (south of Chonpyong-dong), 2 or 3 stationary T34's were
visible (0/12, P /13). The infantry reported that they had been knocked out
from the air and added that direct fire was now coming from a house 3500 yd
away (beyond the road junction?). Renewed fire confirmed the report. The
104 ORO-T-261
UNCLASSIFIED
SURO UNCLASSIFIED
M26 fired two or three rounds at the house and silenced the T34 presumed to
be inside. Later the enemy tank began firing again and had to be resilenced in
the same way.
The next enemy attack came late on 20 August. It was pitch dark, and the
leader of the 3d Plat stated that his gun was aimed by estimating the location
of the road. Again the approach of enemy armor was reported by the infantry
before the clanking was heard. This time, the M26 fired first and missed. The
enemy tank replied and also missed, and then turned around and retreated as
the M26 fired several more times. The infantry report states that tanks and
SP guns had led an unsuccessful enemy attack in which the enemy troops had
been deployed behind. No enemy armor was claimed destroyed by friendly
infantry, armor, or artillery.
Fire still came from the supposedly de'stroyed enemy tanks in the build
ings further up the road (presumably 0/ 12, P /13). It was evidently an enemy
practice to send up small crews to man still-functioning tank guns. On the
morning of 21 August an armored task force consisting of the 1st and 3d Plats,
Co C, 73d Tk Bn and one company of the 27th Inf was sent approximately one
thousand five hundred yards beyond the MLR to clean out the villages, recon
noiter and mislead the enemy as to friendly intentions. The US tanks destroyed
at short range an SU -76 (M/ 10) in the pass to the right of the road and two
T34's (0/12, P/13) parked in the yard of the school south of Chonpyong-dong
(at 1151.9-1464.3). To date, therefore, friendly armor had destroyed one T34
in battle and further damaged two already immobilized T34's and two already
immobilized SU -76's. Their crews had further observed three or four more
T34's amid houses. The infantry report on the task-force operation states
that six T34's and two SP guns were observed and now destroyed with thermite
grenades if they had not been destroyed already. Three additional tanks were
therefore apparently further damaged by the infantry, perhaps one or more of
them off the road amid the houses at Singi-Soi-ri.
The enemy still had five intact T34's, but probably no more. The S-3 of
the 19th NK Regt, who surrendered on 1 September (Item 1, App A), stated that
the only tanks he knew of were the five he saw on 21 August. He added that they
all took part in the attack on this date.
The defense against this attack was carefully planned. On the afternoon
of 21 August, the crews of the 2d and 4th Plats of Co C, 73d Tk Bn took over the
tanks of the 1st and 3d Plats, respectively. The five tanks now of the 2d Plat
remained deployed in the creek bed to the left of the road; the five tanks now
of the 4th Plat were moved up in staggered column on both sides of the road,
the leader stopping just beyond and to the right of the T34 knocked out on the
night of 18-19 August (1/ 7). Company G, 27th Inf occupied the point of high
ground just to the east. The tanks took up pOSitions carefully planned fC!r cam
0uflage and fields of fire. Tank guns and bazookas were zeroed in on the road.
It was another moonless night. Enemy preparatory' fires began around
2200. Much or most of the fire at the MLR seemed to be direct, high-velocity
fire, from only several thousand yards away. In the next hours perhaps two
hundred and fifty rounds came in. The US artillery concentrated on the road junction
ORO-T-261
105
UNCLASSIFIED
$.EER"
UNCLASSIFIED UERn
and may have beaten off the enemy temporarily. The fullest account of these
events is from the 73d Tk Bn report:
At about 2300 hours 21 August 1950, an estimated eight T34 tanks, four 76-mm
self-propelled guns, an unlm.own number of wheeled vehicles and one regiment of infan
try were observed advancing toward our positions along the axis of the road leading into
Kumwha-dong [the twin villages of Singi and Soi-ri are shown as Kumwha-dong on the
1:250,000 msp], which was about 2,000 yards to our direct front. Our artillery was di
rected upon the enemy, causing him to halt his advance and disorganizing his attack.
[The S-3 of the 19th NK Reg! gives a different reason for the halt.] However, at spproxi
mately 0130 hours on 22 August, our artillery 11fted, and the enemy maneuvered his tanks
and SP -guns into position in the town, and began firing AP and HE into our pOSitions, ex
pending about 75-80 rounds total. We did not return his fire at this time, bec""se of our
Inability to bring effective direct fire upon him. At 0300 hours, 22 August, tanks were
heard maneuvering to our front, but because of darkness we held our fire. At 0325 hours,
tanks were definitely heard advancing towarp. our positions amid screams and apparent
commands . ...24
The enemy tanks could be tracked by their noise. Enemy infantry had
removed a first line of US AT mines strewn across the road. Now the leading
T34 stopped at the second line, and some fifty yards from the leading M26.
Three vehicles piled up behind it. The order came to fire. The first round
from the leading M26 hit the leading T34 (JI 6?) abreast of the driver. At the
same time, the. bazooka scored a hit. The tank did not burn. Following the
plan, the M26 shifted fire to the second T34 as rapidly as possible. The first
round of HVAP missed, but the second hit this T34 (HI 5) on the right front
slope and went through the Side, bogie, and track into the ground. The third
enemy vehicle, an SP gun (K/ 8), had been firing up the draw toward Kumwha
dong. The gun now traversed and fired the only shot at the friendly tanks. It
hit the ground, and the leading M26 set the SP gun on fire with its fourth shot,
a WP rd. Shifting fire back to the first T34, it expended the remainder of the
original 9 rd in its ready rack. Three additional M26's opened fire, and the
leading T34 started to burn. The fourth enemy vehicle, a T34, turned and
escaped. Of the estiinated platoon of enemy infantry accompanying the tanks
(ten of these riding on the first T34), ten or fifteen were killed and the rest
were dispersed. The bulk of the enemy infantry and the remaining vehicles
never approached close enough to be engaged by friendly tanks.
The prisoner from the 19th NK Regt stated that of the five T34's on hand
for the attack of 21 August only one survived, to be stationed subsequently at
the road junction. The decimation of enemy armor would explain why the
engagement on the night of 21-22 August was the last direct challenge.
In the early hours of 23 August, an SP gun took up a position 5000 yd
from the MLR and scattered apprOximately twenty-five rounds over friendly
pOSitions. The 27th RCT reported another attack during the night of 24-25
August in which friendly artillery destroyed two more T34's. Confirming evi
dence is lacking. It would seem that either additional SP guns were knocked
out or that knocked -out hulks were further damaged.
Enemy resources in armor on 2 September appear to have consisted of
one T34 tank and a small number of SP guns. As reported elsewhere in this
study, enemy plans for the attack on 2 September included support by one tank.
106 ORO-T-261
UNCLASSIFIED
SURn
UNCLASSIFIED
This tank was in fact observed moving behind the attacking enemy infantry on
3 September. Presumably, it is also the one !mocked out with a bazooka round
by Co I, 8th Cav north of Tabu-dong on 4 September (unidentified). By 2 Sep
tember, the 13th NK Div very likely was already reduced to the two SP guns it
is !mown to have had on.11 September. This lack would explain the move of the
SP -gun battalion of the 1st NK Div to Tabu -dong on or about 4 September, as
reported by one of its members taken prisoner, though there is no ready and
convincing reason to offer for its reported withdrawal on or about 8 September.
An additional bit of evidence is presented below to support the very likely con
clusion that the two SP guns aggressively active south of Tabu-dong in the vicin
ity of Tongmyongwon on 11 and 12 September belonged to the 13th NK Div. One
of these two, out of all the enemy armor committed in the attack on Taegu from
the north, was the only vehicle to survive and fall intact into friendiy hands.
Post-Mortem on Enemy Armor
The number, type, and location of !mocked-out enemy armor found after
the battle in September serve to check the several sources of data on the
engagements in which it was involved. It is !mown that the North Koreans
repaired damaged tanks at Sangju. Slightly damaged vehicles may well have
been evacuated from the Tabu -dong area during the three to four weeks of enemy
occupation. Furthermore, it was an enemy practice to cannibalize damaged
armor, and several of the vehicles found north of Tabu-dong show definite
evidence of such usage (0/12, R/15). Whatever the validity of these arguments,
thoroughly destroyed enemy_ armor should have remained on the spot.
The remains were studied over a month after the most important engage
ments. On 27 Sep 50 an operations-research team, Survey No.1, examined
friendly and enemy armor which had been knocked out along the roads in the
Tabu-dong area. On 11 Oct 50 a second team, Survey No.5, covered the same
ground. The two teams were made up of different individuals and the surveys
were independent. Survey No. 1 was made by an infantry officer and an oper
ations analyst; Survey No. 5 by an ordnance officer and an armor officer. The
results of these and other surveys have been analyzed in a previous ORO
report
2S
but were not related to the battles that produced the armor casualties,
. nor was this considered possible. It is therefore of considerable interest to
reconcile the two surveys with the battle reports.
The enemy tanks and SP guns found by the two teams are listed in Table
02 with the original notations and evaluations of the probable causes of the
observed damage. Positive correlation of the duplicate sets of data was made
possible by the series of photographs t a k ~ n in each survey. The."Remarks"
have been added by the present writer.
In each survey the location of each specimen was plotted only roughly on
a 1:250,000 map. In an effort to locate the remains more definitely on the
1:50,000 map used throughout this study, the sequence and background detail
of the photographs, the notations made by the survey teams, and the additional
data furnished (without benefit of a map) by the former crews of Co C, 73d
Tk Bn have been considered. The resulting most likely pOSitions are plotted
in Fig. 9. The apprOximate location of the front line of the 27th RCT and the
southernmost line of the 8th Cav Regt are also indicated.
ORO-T-261
UNCLASSIFIED
HEKn
107
18 Aug
L-_____
UNCLASSIFIED
SRRtT
T/_
Probably koocked out by artillery
Knocked out by various means,
destroyed by US tankson dates
21 Aug'
21 Aug?
All
Fig. 9-Enemy armor found ofter the battle.
T34185 tanks are indicated in parentheses; the
designations without parentheses i ndieate SLJ..76
guns. Symbol to the left of slosh isdesigna
tion of Survey No.1; to the right, of Survey No.
--
-
-
....
5. Vehicles were found on the side of the road
- __ _ Front line of 8th Cay,
indicated. -
12 Sep 50
indicated 21 Aug (0/12)
Koocked out by Air attack
,..--"
, , (L/9)
"-
....
21-21 Aug K/8
Knocked out by US tanks and
bamokas in enemy attacks
on dates indicated
(P/13) 21 Aug
L-----N/11 19 Aug
-J-------M/10 21 Aug
--....
- 'Approximate front line of 27th ReT, 18-26 Aug 50
(Several T34's amid houses, knocked out before
21 Aug and not found by survey tearns)
(T34 knocked out 21-22 Aug, evacuated by US?)
(J/6) 4 Sep
108
ORO-T-261
UNCLASSIFIED
SUHT
UNCLASSIFIED
All the enemy tanks and SP guns which Co C, 73d Tk Bn reported as
destroyed by one means or another can be correlated with remains found
later, as indicated parenthetically in the preceding narrative. The reports
of enemy armor hit in battle by friendly tanks or further damaged by them are
in most cases confirmed by attributions of damage to 90-mm tank guns in
Survey No.5. Remains were not found to confirm the 27th RCT's claim of a
total of 13 enemy tanks destroyed in the period 17-25 August. The surveys
may have missed any T34's which may have been in houses off the road or
which may have been evacuated by friendly forces. As for the rest, it may be
that enemy armor that did not approach closely enough to be observed by Co
C, 73d Tk Bn was knocked out but only slightly damaged and at once evacuated.
It seems more likely, however, that the enemy practice of mannlIjg the guns
of immobilized tanks caused certain tanks to be hit and reported destroyed
more than once. The RCT's claim of six SP guns destroyed agrees with the
number later found north of Tabu-dong. '
The probable date and cause of destruction of each enemy tank or SP
gun is indicated in Fig. 9. Since almost all of them were found along the road
and facing south, most enemy armor was apparently knocked out while
advancing.
The enemy division to which the SP guns belonged can be inferred from
the numbers stenciled on their turrets and visible in some of the photographs
taken by the various surveys. The first figure of a 3-digit deSignation and
the first two digits of a 4-digit designation appear to represent the parent
diviSion of the SP-gun battalion. Without exception, such numbers visible on
SU-76's along the Taegu-Tabu-dong axis are 4-digit, and begin with "13."
They therefore appear to have belonged to the 13th NK Div. All numbers visible
on the many SP guns knocked out in the Waegwan area are in the range 300 to
399, suggesting as the parent division the 3d NK Div, which was in fact the one
in the area. Elsewhere, an SP with a number in the "600" series was found
along the Naktong, and one with a number in the "800" range along the north
ern part of the perimeter. All visible numbers on enemy tanks found along
the Tabu-dong-Sangju road are in the "300" series. The significance of the
tank designations is unknown.
Role of Armor in the Enemy's Tactics
To evaluate the enemy's use of armor it is necessary to deduce his tactics
from the over-all character of the battle. Appendix F presents a general de
scription of the battle north of Tabu-dong in August 1950 as reported by friendly
and enemy sources. These accounts have not been confirmed or corrected by
further data, but they agree fairly well.
While the 27th RCT was advancing to the Tabu-dong area on 18 August
against the negligible opposition of scattered units probably of the 1st NK Div,
the 13th Nk Div was fighting toward it from the northwest. The 13th Div renewed
its attack on ROK forces at 0500, from pOSitions on the Sangju-Tabu-dong road
approximately eight miles from the road junction north of Tabu-dong. As shown
in Table 6, three rifle regiments,of the 13th Div were at about 50 percent TO
strength, and in the fighting on 18 August each of them sustained several hun
dred casualties. The fighting ended at 2100, evidently with a clean-cut pene
ORO-T-261
109
UNCLASSIFIED
SHRfT
UNCLASSIFIED
tration to the road junction almost at the time the 27th ReT was going into
position just to the south.
The tank-led enemy thrust on the evening of 18 August was probably not
a counterattack in reply to the 27th ReT's advance but simply an attempt by
the 13th Div to exploit its breakthrough and advance as far and as fast as
possible. Up to that time it had never made contact with US forces and there
fore had no reason to depart from the doctrinal practice which enemy armor
followed at the start of the war: overt display of tanks ahead of infantry columns
in order to intimidate opposing forces" The 27th ReT's report of enemy
infantry following the tanks in trucks agrees with this interpretation. Further
more, no deliberate attack was made against the 27th Inf, if one may judge from
its negligible casualties on 18 August.
The friendly casualty list suggests that at no time did the enemy infantry
make determined assaults in an effort to capture the positions held by the 27th
Inf. The daily number of battle casualties in the entire regiment during its
stay in the Tabu-dong area are:*
Day. Aug Casualties
18 11
19 19
20 25
21 6
22 7
23 5
24 0
25 14
26 0
Figures are not available for the flanking ROK units, which may also have
been engaged by the 13th,NK Div. The data in Table 6 indicate that in the period
21-31 August the 13thDiv's three rifle regiments lost about fifteen hundred men.
The huge disparity of the losses on each side seemingly follows from the
enemy's tactics. Until 18 August the 13th Div had probably not been opposed
by more than meager artillery fires and not by any armor at all. It evidently
attempted to break through the 27th ReT with massive drives down the valley,
very likely the kind of attack which had succeeded against the ROK Army. As
described in App F, the massive formations were now beaten off by US firepower
before they ever reached the friendly MLR.
Armor continued to spearhead the enemy attacks. It was finally destroyed
as a tactical entity on the night of 21-22 August in what appears to have been
the most determined attempt to penetrate to Tabu-dong. On this occasion, the
leading vehicles got well beyond the friendly front line, only to be stopped
deCisively by the carefully planned ambuslf of the friendly tanks. There is good
evidence that in this attack the enemy infantry was disorganized by friendly
artillery fires concentrated on the vicinity of the road junction, the obvious
.The figures are from a machine run on 11 Dec 50 by the Strength Accounting Branch. CHO. Fb:C. and
may perhaps be less complete than later, corrected tabulations.
ORO-T-261
110

SE(R!T
SiERET UNCLASSIFIED
o.
Survey No. J, 27 Sep 50
ORO-T-26.1
UNCLASSIFIED
HCREl
111
UNCLASSIFIED
b. Survey No.1, 27 Sep SCi
112 ORO-T-261
UNCLASSIFIED
SUR!T
UNCLASSIFIED
c. Survey No. J, 27 Sep 50
Fig. lO-Knockedout tanks and SP guns in the 'Bowling
Alley' north of Tabu-dong.
o. The enemy view of 18 August and 2 September, looking south
toward Tabu-dong. The destroyed SU-76 is Q/14. b. View the
lead M26 of Co C, 73d Tk Bn would have had on the night of
21-22 August. The T34/85 (J/6) in the foreground and SU.76
(K/8) in the left rear were probably knocked out in the ambush
on this night. The vehicle in the center rear is unidentified and
may be a truck. c. The bottleneck just north of Tabu-dong,
view looking north. This M4 tank dozer was abandoned by Co
C, 70th Tk Bn on 4 September after throwing a track while
attempting to back away from enemy direct fire.
ORO-T-261
SlEln
UNCLASSIFIED
113
UNCLASSIFIED
assembly area for such obvious tactics as the enemy had been using. The
timing was also seriously upset by the curious failure of the enemy command
reported in App F. Except for the small number of troops riding on the tanks
in this action, the enenlY infantry. did not penetrate to support the leading armor,
so it was expended for nothing.
Figure 10 gives several views of the narrow valley north of Tabu-dong and
of knocked-out enemy armor. In this sharply canalized terrain, US tanks simply
canceled out the shock action of enemy armor. The defense had the advantages
of knowing the route the enemy tanks must take, of cover and concealment, and
of dispersion for greater total firepower. This damming of enemy armor made
the enemy's infantry tactics unsuitable, as the enemy demonstrated in September
when he took to the hills to attack.
The divisional battalion of self-propelled guns is intended for direct
support of the infantry. Operating singly and boldly in September, but now to the
rear of the infantry, the remaining SU-76's had a physical and nuisance value
out of proportion to their small numbers. If the enemy had had more of them
for use from forward positions at night or on days of nonflying weather, his
climactic drive on Taegu would have been much stronger. Tanks could have
been used in the same way.
A recent book on Russian doctrine contains some interesting statements
derived from Soviet publications in 1940:
.. . Mechanized and tank formations (divisions) are not considered useful for de
velopment of a breakthrough in mountainous terrain. Medium tank battalions are assigned
to rifle formations for close support in the m o u n t a i n ~ .... The use of artillery [In the
mountains], especially in direct-fire roles, is considered very important.... *
R.L. Gartboff, Soviet Military Doctrine, p. 378. The RAND Corporation, Santa Monica. 1 May 53.
ORO-T-261
114
UNCLASSIFIED
UNCLASSIFIED
FRIENDLY SUPPORT
The subsections which follow are devoted to artillery, armor, and air sup
port of friendly forces, with particular attention to their part in the friendly
capture of Hill Mass 314. The material on fire during this action should be
read in connection with the narrative description in Part IV. Because of the
lack of detailed data, coverage of the other periods of the battle is less thorough.
ARTILLERY IN SUPPORT'
In the absence of firing charts and detailed records, the following proce
dure was used to reconstruct the role of supporting artillery in the attack on
Hill Mass 314:
(a) The general situation of artillery in support of the 7th Cav Regt was
discussed with the former commander of the 77th FA Bn.
(b)' The former S-3 of the 77th FA Bn was given a map of enemy weapon
positions prepared by the former S-3 of the 3d Bn of the 7th Cav Regt, all per
tinent entries on enemy weapons and fire in the journal of the 1st Cav Div, and
a reconstructed timetable of the attack on Hill Mass 314 on 12 September. With
several colleagues (the former Asst S-3 of the 77th FA Bn and the former SurvO
of 1st Cav Div Arty), the former S-3, 77th FA Bn then plotted the concentrations
fired by several artillery battalions.
(c) The former ArtyLnO of the 77th FA Bn, who was on Hill Mass 314 call
ing in supporting fires during the battle, was furnished the same material. He
gave his observations and comments.
(d) The revised firing plot shown in Fig. 8 (given in the earlier section on
enemy support) was then constructed.
(e) The revised plot and the general artillery situation in the 8th Cav Regt
area were discussed with the former S-3 of the 99th FA Bn.
As a result of this research, the pattern of artillery fire on the objective
is known with fair certainty. Data on counterbattery fire are inherently diffi
cult to obtain and are also known less adequately for the reasons given below.
Artillery: Support in General in the Sector North of Tae!m
With the commitment of the 3d Bn, 7th Cav to the attack on Hill Mass 314,
the right-flank boundary of the 8th Cav Regt was defined by a north-south line
running between Hill Mass 570 and Hill Mass 314. The 99th FA Bn (105-mm
howitzers), which had been continuously in support of the 8th Cav, at this time
ORO-T-261
UNCLASSIFIED
SHRn
115
UNCLASSIFIED
was responsible only for targets west of this line, while the 77th FA Bn (105
mm howitzers) assumed support operations of the 7th Cav east of this line on
Hills 314 and 660. Figure 2 (given earlier) shows the positions of the 99th FA
Bn and of Btrys B and C of the 77th FA Bn. Battery A of the latter battalion
had been detached for duty with the British Brigade.
Elements of the 8th FA Bn (155-mm howitzers) attached to Div Arty were
also in position near the other battalions at the coordinates 1153.5-1447.5. The
battalion apparently had only two batteries. On 12 September the attached ele
ments were made available to the 77th FA Bn for countermortar fire. Shortly
after Hill Mass 314 was captured Btry C, 17th FA Bn (8-in. hOwitzers) arrived
in the area for general counterbattery fire. Its first mission was fired on 14
September. On 15 September supporting fires were further increased by use
of the 68th AAA Bn (90-mm guns) against ground targets.
Ammunition, in short supply, was not critically low in the artillery battal
ions.- Stocks on hand were still close to basic loads, though VT fuzes were
scarce and were hoarded. So that the imminent emergency would not find stocks
depleted, daily expenditures were restricted. On 8 September Div Arty restricted
expenditures of 105-mm ammunition to 50 rd per gun per day, except in emer
gencies. This restriction was to be lifted for the attack on Hill 570. Then, on
11 September, Div Arty announced that the 77th FA Bn was limited to 25 rd per
piece. The next day, the artillery preparation scheduled for the attack on Hill
314 was entirely canceled. When the outcome of the attack became uncertain,
the restriction was lifted, and perhaps as many as a hundred VT -fuzed rounds
were fired in direct support.
Daily expenditures are reported by the 99th FA Bn as: 9 September, 774
rd; 10 September, 206 rd plus harassing fire at night; 11 September, 472 rd;
12 September, 490 rd plus harassing fire at night.
The average daily expenditure was then about five hundred roundS, or ap
prOXimately twenty eight rounds per piece per day. Thus, the limitation of 25
rd per piece per day imposed on the 77th FA Bn on 11 September was in line
with expenditures by the 99th FA Bn.
For all the artillery battalions, the only table of expenditures by type of
target is that reported by the 99th FA Bn for 10 September.
As coordinates of targets are not specified, one can only guess which fire
was on Hill Mass 314; however, since the only action during the hours covered
was the attack of the 3d Bn, 8th Cav on Hill 314, it is likely that much of the fire
centered about this hill. The attack jumped off at 1330.
According to the former S-3 of the 99th FA Bn, 75 percent of the missions
fired in the 8th Cav area were unobserved and probably not very effective. There
was great difficulty in locating enemy gun pOSitions; firing also suffered from
Of artillery ammunition, only lOS-mm shells appear on the critical list for Ammunition Supply Point
NumLer 2 at Taegu. The following numLers of founds of critically low items were on hand on 11 Septemher
1950: lOS-mm howitzer, 1,971 (44- percent of busle load of one FA Bn); 60-rnm rnOftar, 4,92B; 81-mm mortar,
11,247; 4.2-in. mortar, non!'; 76-rnm tank gun, 4,068; 90-mm tank gun, 548; 3.S-in. rocket, none; carbine,
896.000.
The b:USAK Artillery Status Hepart covering the period 1500, 8 Sep-1500, 9 Sep states that 1st Cav
Div Arty expended 1,575 rd of unspecified caliber, leaving 12,400 rd of 10S-mm and 1,814 rd of ISS-mm
ammunition on nand. T h e s ~ , figures are apparently the totals for the 61st, 8U, 77th, and 99th FA fins. A
separate entry for the 9th FA Bn states that 580 rd were expended and that 1,989 rd of ISS-mm ammunition
were on hand. A complete set of status reports is not availaLle.
116
UNCLASSIFIED
SHIET
UNCLASSIFIED
the lack of meteorological data and subordinate personnel capable of computing
corrections properly. The observed-fire missions, on the other hand, were
considered effective. At some time the enemy observation post on Hill 570 was
taken under observed fire with a single gun until a dir ect hit was obtained and
enemy personnel were seen fleeing from the point. Unfortunately, it is not known
EXPENDITURES BY 99TH FA BN, 10 SEPTEMBER
Expenditure,
Time Target rd Results
1010 Enemy field piece 5 Silenced
1015 Large concentration of troops 14 Dispersed
1025 Enemy troops 18 Dispersed
1107 Enemy personnel on ridge 16 Dispersed
1137 Large troop concentration 48 Unobserved
1205 Field piece 7 Unobserved
1220 Troop coneentr ation 26 Unobserved
1240 Registration 2
1400 Bridge 10 Not completed
1430 Automatic weapons 10 Not completed
1435 Enemy personnel 50 Casualties high
Total 206
whether or not this incident occurred on 12 September. The 99th FA Bn fired
continually, day and night. R e q ~ e s t s for missions were of the greatest variety
and far more than could be met with the official ammunition allotments. Unlike
personnel of the 3d Bn, 7th Cav, men of the 8th Cav do not have in their memo
ries a single, intense action.
Reports of Artillery Units on 12 September
Targets are not listed in detail in the artillery journals. Documentary in
formation on rounds fired at particular coordinates and times is probably un
obtainable, as the original firing charts are thought to have been destroyed.
The journals do, however, list total expenditures on each day. These statements
are used as limits in the estimates given in the next section, imd are quoted here.
Operations of the 77th FA Bn on 12 September are summarized as follows
in the division artillery journal:'o "The 77th FA Battalion fired two missions
on enemy troop concentrations and several in support of friendly troops. 'C'
Battery displaced to an alternate position." In its own unit history12 the battal
ion reports on operations in the period 111800 September to 121800 September:
The battalion remained in position continuing its mission of supporting the 7th
Cavalry Regiment. Nine missions were fired during the period, expending 128 rounds
of l05mm ammunition, principally on enemy troop concentrations, in support of the ....
attack made by tbe 2d [sic] Battalion during the hours of darkness on Hill 314.
The last statement above is considered to be very much in error. It is
unlikely that the 77th FA Bn fired at all on 12 September before the 3d and 2d
Bns, 7th Cav moved out toward their objectives after daybreak. Since the 2d
Bn was merely relieving ROK units on Hills 660 and 516 and encountered no
ORO-T-261
117
UNCLASSIFIED
SHIrl'
UNCLASSIFIED
S.E(RrT
opposition, it probably received no artillery support. All the 77th FA Bn ex
penditure would then have been in support of the 3d Bn attack on Hill 314, which
is summarized by the same source:
At 1115 hours, the 3d Battalion, 7th Cavalry Regiment, also jumped off with the
aid of extremely effective air strikes and artillerybombardment [there was no artillery
preparation] was successful in securing hill at 1515 hours, at which time the battalion
dug in and reorganized....
Operations of the 99th'FA Bn on 12 September are reported by Div Arty:
"The 99th FA Battalion fired 10 missions, of which 7 were harassing, two on
enemy troops and one on an enemy gun. The troops were dispersed and the
enemy gun silenced." In its own war diary" the 99th FA Bn reports on 12
September: "Sporadic high-velocity artillery fire from the enemy hit in the A
and B Battery area. The battalion fired 14 missions with a total of 490 rounds
expended. The battalion also fired barassing fire during the hours of darkness."
Operations of the 9th FA Battalion on 12 September are reported by Div
Arty:
The 9th FA Battalion fired one observed and one unobserved mission on enemy
field pieces and several missions on enemy troop concentrations. An estimated 150
casualties were reported. The 1st FA Observation Battalion reported a self-propelled
enemy gun and adjusted the 9th FA Battalion, scoring two direct hits on the target.
No further information is available on the 9th FA Bn, as no unit journal
was submitted. It would seem that only two batteries were in Korea, and that
it was without its own headquarters battery.
Direct-Support Fire on Hill Mass 314 on 12 Sep.tember
Figure 8 gives the revised plot of concentrations fired on 12 September in
the vicinity of Hill Mass 314. Only for the 77th FA Bn is it reasonably complete
and accurate. The targets at Sites 2 and 3 originally 'shown for Btry C, 17th FA
Bn (which had not yet reached the area) have been assigned to the 9th FA Bn.
Some of the concentrations just north of Hill Mass 314 may have been fired by
the 99th FA Bn at the request of the 77th FA Bn, and not the 77th itself, as men
tioned below. .
Only the 77th FA Bn fired on the objective. The writer has deleted two
concentrations plotted on Knob 1 because this point had already been overrun
when the artillery began firing. One of the concentrations plotted on Hill 570
has also been deleted to accord with the distribution of the total expenditure
estimated below. Replotting the enemy 82-mm mortars at Site 10 has also
eliminated the concentration at Site 11, the originally plotted position. The
three concentrations between Knobs 2 and 3, fired between the first and second
assaults, were later refired when the attacking units were again pinned down
between the second and third assaults. The concentrations beyond Hill Mass
314 were fired to prevent the enemy from reinforcing his units on the objectives;
'r
those on Hill 570 were apparently intended to inhibit the fire of enemy support
ing weapons. About 90 percent of all rounds fired were VT-fuzed.
118 ORO-T-261
UNCLASSIFIED
HCREr
SUUT UNCLASSIFIED
It is assumed that one battery-volley of 6 rd was fired On each of the 77th
FA Bn concentrations with the exception of the second mission at the objective,
when, it is assumed, both batteries fired each of the three concentrations. The
total expenditure of 128 rd reported in the battalion's journal would then be dis
tributed as fo llows:
Mission Estimated expenditure, rd
Registration 6
Three I-battery volleys fired on
objective after first assault 18
Three 2-battery volleys fired on
objective after second assault 36
Seven concentrations beyond
Hill 314 '
42
Four concentrations on Hill 570 24
Marking target for final air strike 2
Total 128
This distribution of fire from two batteries is in line with the tabulated
expenditures of the three batteries of the 99th FA Bn given earlier.
The fire of the 77th FA Bn on the northern part of Hill Mass 314 was
called in by the forward observers and by the artillery liaison officer, who
initially occupied an observation post adjacent to the 3d Bn CP on Knob 2 and
later moved up with the assault troops to a position just short of Knob 2. In a
communication in 1953, among his observations on various aspects of the battle,
he gave some information on friendly artillery fire.
Registration fire was laid down on the objective by 0930, when the fog had
cleared sufficiently. The first supporting fire was supplied after the initial
assault on Knob 2 had been repulsed and most friendly troops had withdrawn
back into the saddle. The time would then have been about 1210. Up on the
ridge, any enemy were located in foxholes, which were well concealed with
brush. The second occasion for supporting fire by the artillery was when the
assault troops were again pinned down short of Knob 2 in the linear depression
along the backbone of the saddle known as "the shallow." The ridge forward
of Knob 2 was thoroughly worked over by artillery fire, which was then called
off for the final air strike. Immediately, the final, successful assault was made.
An observation by the former commander of Co I, who was with the artillery
liaison officer and others in the most forward position in the shallow, is per
tinent. He said the torsos of enemy soldiers could be seen projecting above
several foxholes. When the ridge was overrun soon after, no masses of enemy
troops were encountered, only wounded enemy soldiers in the foxholes.
Though the objective was reported "throughly worked over" by artillery,
it is evident from the meager expenditure of ammunition that not very much of
the area of the objective could have been covered at anyone time. It is of in
terest to calculate how much effect a reasonable distribution of fire along the
objective ridge would have had on enemy troops on the flanks of the hill mass.
After the second assault, the artillery liaison officer and one of the two forward
ORO-T-261
SER!1
UNCLASSIFIED
119
UNCLASSIFIED SRRtr
observers moved up to the shallow just short of the crest of Knob 2. From here
they could see straight down the ridgeline about halfway to Knob 3 but neither
into the declivity at this point or behind the little knoll just beyond, nor much
to either side.
A reasonable distribution of fire in the absence of observed enemy troop
concentrations would be battery volleys uniformly spaced between Phase Line
2 and Phase Line 3 with the guns firing in parallel sheaf. It is assumed that
the transverse spacing of the bursts was 25 yd, and that each shell exploded
15 yd aboveground and had a radius of effective fragmentation of 25 yd. Along
the ridgeline, about 240 of the 380 yd from Phase Line 2 to Phase Line 3 would
then be covered by effective fire. Across the ridgeline, about 165 yd would be
covered. Whether this fire reached the side slopes as well as the top of the
ridge depended on the variable width of the plateaulike top. Figure 11 shows
the transverse spacing of a 6-rd volley at the cross section of the objective
along Phase Line 2. The second assault is thought to have been stopped by
enemy troops located along this line to the left. According to the present as
sumptions the artillery fire would hardly have reached them. At the waist of
the objective the fire would have covered some of the side slopes. On the
other hand the long finger running northwest from Knob 3, and the assumed
command post in the bunker on it, could hardly have been well covered.
5SO
VT Bursts, Parallel Sheaf, 25-yd I"terval
~
'"
w
> soo
~
z
o
>
~ 4SO
w
.J
W
Crosssectio" of Objective
alo"g Phase Li"e 2
400
200 ISO 100 50 o 50 100 ISO 200
DISTANCE FROM CENTER OF RIDGE, METERS
Fig. 11-Transverse coverage of VT -fuzed fire on the objective.
The former officers of the 77th FA Bn stated that forward observers re
ported the fire "effective." In the absence of observed enemy troops, the basis
for this judgment is not known, unless it was the otherwise unreported inhibition
of small-arms fire. When the objective had been taken, many of the enemy dead
found down the wooded slopes were deemed artillery-fire casualties. Air attack
and the fire of the organic weapons of the 3d Bn, notably that of the 81-mm mor
tars, could just as well have been responsible. All the same, denial of the top
of the ridge must also be considered an important effect.
ORO-T-261
UNCLASSIFIED
SUR"
120
UNCLASSIFIED
SURfT
Counterbattery' Fire on Enemy' Supporting Weapons
The normal primary targets for friendly counterbattery fire did not exist,
since enemy heavy artillery was lacking and medium artillery was inactive.
The divisional 76-mm guns and 122-mm howitzers fired very little, if at all.
At the front line, the few SP guns and the 45-mm AT guns did fire effectively,
but they did not present good targets, the former because of their mobility, the
latter-especially the short-barrel 45-mm AT guns-because of their inconspic
uousness. In any event, these flat-trajectory weapons were more of a threat to
fixed installations more or less toward the friendly rear than to friendly troops
in the attack or defense.
The really troublesome enemy supporting weapons were mortars and per
haps 76-mm howitzers. As noted in the section "Enemy Support," US flash ob
servers and radar were in no position to spot these defiladed weapons, and sound
ranging equipment had very little success. Such counterbattery fire as US 105
mm, 155-mm, and, later, 8-in. howitzers, could bring down on them depended,
first, on a spotting by a ground or ai'r observer who had only a map for reference
and second, on indirect fire. This was usually unobserved, and, according to an
artillery officer, probably inaccurate. When fire was observed, the invariable
procedure was to adjust successive individual rounds. According to a report of
a survey of enemy pOSitions in the Tabu-dong area just after the battle," the
enemy would simply stop firing after the first round came in, and the gun crews
presumably would take to the well-constructed bunkers nearby. The weapons
themselves were rather well protected.
There was also a significant time lag due to friendly organization. The 7th
Cav troops attacking on Hill Mass 314 on 12 September were being punished by
enemy supportingweapons located in the 8th Cav area of responsibility. Clearance
from the 99th FA Bn was required for the 7th Cav's supporting 77th FA Bn to
fire on targets in the former unit's area. In the ordinary case of a special mis
sion against sporadically active enemy artillery, such clearance was quickly
forthcoming. During an intense action, when communication links were broken
or clogged, it was a difficult and slow process to obtain clearance.
Counterbattery fire does not appear to have inhibited enemy fire on Hill
Mass 314 Significantly. Figure 8 shows that the 77th FA Bn did indeed fire on
targets on the ridgeline of Hill Mass 570, and the 99th FA Bn may also have
laid down concentrations there. While this fire may have been effective in stopping
the fire of automatic weapons, 60-mm mortars, and the 76-mm mountain gun
thought to have been on the ridge, it appears to have left untouched the more
potent weapons in defilade to the rear. The sensitive target on the ridge, the
OP, was not affected, if one can judge from the fires of the weapons it presum
ably controlled (see Part IV).
It is possible that the 99th FA Bn on 12 September did fire on the weapons
that were punishing the 3d Bn, 7th Cav on Hill Mass 314, but most of the 490 rd
this battalion fired on 12 September were apparently haraSSing missions. The
9th FA Bn was used in general in a counterbattery role, but it fired only two or
perhaps three such missions on 12 September. Two have been plotted in Fig. 8.
Unconfirmed reports from the infantry state that enemy mortar positions active
during the battle were located by the artillery air observer and neutralized by
friendly fire. The casualty analysis given in Part IV, however, indicates no
diminution in the intensity of enemy mortar fire until almost the end of the action.
121
ORO-T-261
SiER!f
UNCLASSIFIED
UNCLASSIFIED
ARMOR IN SUPPORT
A study"" has been made of the operations of friendly and enemy armor
in the Korean campaign up to 21 Oct 50. Annex 24, App K of this reference is
a discussion of armor in the infantry division by an officer of the 1st Cay Div.
The operations considered were those of the division's organic armor unit,
the 70th Hv Tk Bn. Annex 23, App K of the same report is a critique of the
same operations by an officer of the tank battalion. Though neither annex
argues from specific data, pertinent information is given on the background,
organization, and methods of the tank units.
The tank-vs-tank actions of the 73d Tk Bn in support of the 27th lnf Regt
north of Tabu-dong in August 1950 are described under "Supporting Armor" in.
the section "Enemy Support" of Part II.
General Situation of Armor in the 1st Cavalry DiviSion
The 70th Tk Bn had been stationed at the Armored School at Fort Knox
for three years and used for demonstrations, school problems, etc. Though .
the individual tank platoons had completed training, the battalion as a whole
had had only 30 days of training in the field. After being brought up to combat
strength with about 250 new men, it was shipped to Korea and committed to
action with the 1st Cay Div on 14 August 1950. Prior to combat, the tanks and
infantry had had no opportunity to train together as smal1 units.
In the absence of medium tanks, the battalion of heavy tanks was moved
up to support the infantry regiments and essentially ceased to be an independent
divisional force. One tank company, with approximately twenty-two tanka, was
attached to each infantry regiment. Each infantry battalion in the line was usu
ally supported by a tank platoon. The need.,to retain some flexibility of organ
ization made it impossible to station uncommitted tank platoons with uncommit
ted infantry battalions. Armor and infantry perso,!nel, therefore, did not work
together between engagements, and did not get to kIiow each other. In combat
there was little more than the tenuous linkage of the SCR-300 radio.
Companies A and C were equipped with M4A3E8 medium tanks, which
carry a 75-mm gun; Company B with M26 medium tanks, which ha:ve a 90-mm
gun. Company B natural1y was used along the MSR, the Taegu-Tabu"dong road.
There were no tank-vs-tank encounters in the sector of the Taegu-Tabu
dong road in the period 1-22 Sep 50. Eight tanks operating along the road were
damaged by enemy mines: one on 6 September, two on 15 September, one on 19
September, three on 20 September, and one on 22 September. No tanks in the
Tabu-dong area were put out of action by enemy fire.
The 16th Recon Co, also organic to the 1st Cay Div, lost all its M24 light
tanks prior to September 1950.
Qperations of Friendly Armor in the First Phase of the Enemy Offensive
On 1 September the tanks of the 70th Hv Tk Bn were distributed as fol1ows:
Co A in the general zone of the 5th Cay Regt, Co B in position on the MSR some
what north of Taegu, and the 1st and 2d Plats of Co C in direct support of the
7th Cay Regt. For the attack on 2 September, Co C's 4th Plat was also attached
to the 7th Cay and ordered to move up on its right flank to positions at 1148.3
122 ORO-T-261
UNCLASSIFIED
UNCLASSIFIED
SiEKn
1460.7, about 1700 yd west of Tabu-dong on the Waegwan-Tabu-dong road.
Company C's 3d Plat was attached to the 16th Recon Co and ordered to move
south of Taegu to the vicinity 1148-1425 to constitute a security force against
possible enemy attempts to reach Taegu by infiltration from the southwest.
The 4th Plat, Co C moved up to the specified position west of Tabu-dong
but apparently was not required by the 7th Cav. As a result, it left this area
on 2 September, just at the time the enemy was about to launch his drive on
Tabu-dong against the 8th Cav. The following reports are relevant:
1423 [2 Sep] S-3 received COncurrence from CO of 7th Cavalry Regiment to move
4th Platoon, Company C from its location on the Shelf Road to the Battalion CP area.
1700. 4th Platoon, Company C crossedIP (MSR and Shelf Road) to return to the
CP area.
1705. Received from G-2 Interrogation of POW Report No. 0076 dated 02 Septem
ber 1950.
15
The interrogation report mentioned is most likely a report of interrogation
within the 1st Cav Div (not now available, but No. 0075 is dated 01 Sep, and No.
0088, Item 2, App A, is dated 12Sep 50) with the subject the S-3 of the 19th NK
Regt, who surrendered to the 8th Cav at 2040 on 1 September. If so, the above
logged item would confirm the report' that, all friendly units were alerted around
1700 regarding the impending enemy attack disclo'sed by the prisoner. In the
case of the tank battalion, this warning evidently came via the .formal channel
of an interrogation report, and insofar as it might have concerned the 4th Plat,
Co C, came just too late. '
Once the enemy attack materialized, plans to counter it with armor were
promptly made. At around 2330 on 2 September the 3d Plat, Co B was alerted
to make a possible counterattack with infantry.-' The 2d Plat, Co B remained in
pOSition, while at 0830 on 3 September Co B's 1st Plat was moved up from the
bivouac area (1162.1-1437.9) to 1152.8-1457.3, the village of Tongmyongwon on
the Taegu-Tabu-dong road. 'The 3d Plat, Co C and the 16th Recon Co were
moved back north to Taegu. '
It was the 2d Plat of Co B, not the 3d, that was sent to the area of enemy
penetration. Early on 3 September, its six tanks moved up to Tabu-dong and
west along the Waegwan road to take up positions at 1149.0-1460.5, close to
those vacated by the 4th Plat, Co C the day before. In the move 'up the tanks
destroyed one 76-mm field piece of an unknown type.
The 4th Plat, Co C was sent back from the area of the tank battalion's CP
to the 7th Cav early on 3 September and joined its company's 1st and 2d Plats.
The movements of Co B's platoons during the days were reported as follows:
2150 [3 Sep] . CO of Company B reported to the CP with information on the fol
lowing locations for the company as of 032100K: 2d Platoon in position at curve in road
at 1150.4-1460.0, 4th Platoon In defense position at 1152.6-1457.5 with Battery A, 99th
Field Artillery, 3d Platoon in defensive position at 1152.8-1456.4 with Battery A, 99th
Field Artillery [sic], 1st Platoon In defensive position at 1152.7-1455.4 with the 9th
Field Artillery Battalion. The 1st Platoon at 031610K Sep 50 attacked from positions
from coordinates 1150.4-1460.0 along the road west (MSR) to Company C positions at
apprOximate grid square 1142-1459 with Company C, 8th Engineers providing infantry
ORO-T-261
UNCLASSIFIED
SEElEY
123
UNCLASSIFIED
.SEun
support on bridges south of road. The tank. platoon returned on the same route at ap
proximately 031740K Sep 50. No enemy were fired approximately 1,000 rounds
of caliber .30 per tank and 2 rounds of HE per tank. Planned for 04 September: 4th
Platoon on same route with same support, RecaD [Plat1 on main road.
iS
While the Waegwan-Tabu-dong and Taegu-Tabu-dong roads were being
patrolled by other tanks, the 2d Plat, Co B remained in position on the Waegwan
road. At daybreak on 4 September, three enemy antitank guns which during the
night had been wheeled into positions straddling the road suddenly opened up
on the tanks. Two were 45-mm AT guns and one was a 76-mm M1942 all-pur
pose gun (recognized by its muzzle brake). A fourth 76-mm M1942 piece was
not manned. The three enemy guns had fired only one round apiece when all
four were destroyed by the immediate return fire of the tanks. A total of only
8 rd was reqUired, and the engagement lasted only a minute or so.
Around midnight on the night of 4-5 September, reconnaissance intelli
gence was received from the 8th Cav that the road junction at Tabu-dong was
in enemy hands. It was planned to take counteraction with 8th Cav troops and
to sweep to Tabu_dong from the west with tanks attached to the 7th Cav. How
ever, as shown earlier in this memorandum, the situation continued to deterio
rate. In the general, limited withdrawal of the 8th Cav Regt down the Taegu
road, the 3d Plat, Co B, 70th Tk Bn pulled back too.
On the flank of an enemy salient, the men of the 2d Plat, Co B were in good
pOSition to observe the enemy move up. Throughout 5 September, enemy troops
were spotted carrying ammunition and dragging field pieces south through Tabu
dong. On the crossroad only 1200 yd to the side of this traffic, the tanks had a
field of fire like a shooting gallery. A total was claimed of nine enemy weapons
destroyed, of which seven were 45-mm AT guns and two were 76-mm howitzers
(no muzzle brake, among other identifying features).
On the night of 5 September, the men of the 2d Plat thought that the
infantry had withdrawn, leaving them stranded, and they were surprised when a
supply truck with fuel and ammunition got through to them from the west. The
1st Bn, 8th Cav was still in position but had no communication with its support
ing tanks. By roundabout routing, information on the situation was relayed from
the battalion to the rear. The battalion was then ordered to withdraw to the west,
and the tanks were ordered to fight to the east and back. The fo llowing order
was relayed to the 2d Plat: "060550. Stateside Baker 6 [Hq, Co B, 70th Tk Bn]
Message for Baker 26: Return to my pOSition through your old sector at 0700.
Shoot village on way. Fire several shots at low elevation to north.'
While the 1st Bn was withdrawing and the tanks were preparing to move,
enemy positions north of Tabu-dong and the enemy road block south of the vil
lage were fired on by friendly artillery. The final order for the tanks to move
out was relayed to them shortly before 1430 by the S-3 of the tank battalion, who
was in the battalion's light aircraft. One tank had had to be abandoned because
of mechanical trouble. Guided from the air, the remaining five moved east
to Tabu-dong and then turned south toward Taegu. The first enemy gun positions
were encountered 1000 yd south of the village, and the last about halfway between
it and Tongmyongwon. The tanks claimed nine enemy weapons destroyed, eight
of them 45-mm AT guns (some with short barrels and some with long), and the
other a 76-mm regimental howitzer. They sustained no casualties, and, coming
on the AT positions from the rear, apparently were not fired on. The entire
124
ORO-T-261
UNCLASSIFIED
$lERn
SURH
UNCLASSIFI ED
move was rapid, for by 1453 the 2d Plat had rejoined friendly units at 1152.5
1458.0. It went into position with Co I, 8th Cav, two tanks of the 3d Plat, Co B,
and one tank of the 4th Plat, Co B, in the vicinity of 1151.8-1459.5. The re
mainder of the 3d Plat was back at 1153.4-1451.3.
It is interesting to compare the tankers' claim of seventeen 45-mm AT
guns destroyed with information furnished by prisoners on the number of weap
ons present at a different times. The AT-gun company of a North Korean regi
ment normally had twelve 45-mm AT guns, and the divisional AT-gun battalion,
eighteen. According to the S-3 of the 19th NK Regt (Item 1, App A) the 19th and
21st NK Regts each had their 12 guns on 31 August, while the 23d NK Regt had
hardly any heavy weapons. At the time of his capture he stated that six 45-mm
AT guns were along the road 4 miles north of Tabu-dong. It would seem, then,
that as of this date the 13th NK Div had perhaps 42 guns, some of them well for
ward. Another prisoner from the 19th NK Regt stated that as of 2400 on 4 Sep
tember four guns of unspecified caliber were along the road south of Tabu-dong.
A prisoner from the AT-gun battalion of the 13th Div stated that as of 6 Septem
ber all the division's 45-mm AT guns were along the roadsouth of Tabu-dong.
The tankers' claim therefore appears very reasonable, and it would appear that
the 2d Plat alone knocked out a large fraction of the 13th Div antitank capability.
Armor Operations along the MSR at Hill Mass 570
Part I inc ludes a description of the action on 6 September in which an
armored task force in a prolonged engagement cleared an enemy fire block
across the MSR at Hill 570. Elements of the 3d and 4th Plats, Co B, and of the
3d Plat, Co C, 70th Tk Bn participated. It is noteworthy that antitank fire was
intense, scored many hits on the tanks, yet caused no serious damage. Appa
rently the only antitank weapon present was the 14.5-mm AT rifle. The 45-mm
AT guns, through which the tanks of Co B's 2d Plat passed several hours later,
evidently were not sited quite far enough to the south to participate in the fire
block engagement.
On 7 September three tanks of the 2d Plat, Co B withdrew to Taegu for
maintenance. The remaining two were moved back to 1153.4-1457.3. On 8
September, with four tanks, the platoon was recommitted, taking up base posi
tions at the highway behind and on either side of the knoll in the valley, at the
coordinates 1153.4-1452.0. From these positions, the platoon made 12 patrols
up the valley before the move forward on 21 September.
Infantry (2 squads from F Co) accompanied the tanks on the first patrol on
the evening of 9 September, advancing to the vicinity of the village of Samsan
dong. No enemy activity or heavy weapons were observed, and only a few enemy
small arms were found. Tanks alone took part in the succeeding patrols. In
addition to the patrols run by the 2d Plat, several were made by the 3d and 4th
Plats of Co B. Except for the incidents described below, on no occasion were
any' enemy' troop.!!, heavy weapons, or activity. observed.
At about noon on 11 September, while the 2d Plat was in position to the
left of the highway at the base of Hill 225, what was considered by armor per
sonnel to be an enemy attack materialized from Hill Mass 570, the enemy troops
coming down from the unnumbered 480-m peak along the finger that leads to the
southwest. This might have been an organized attack of the type in which green
troops are herded into battle. It lasted for four hours, during which at least the
ORO-T-261 125
SEERH
UNCLASSIFIED
UNCLASSIFIED
Hm'T
several platoons of tanks fired and many of the enemy troops attempted to sur
render to the friendly infantry. As recalled by the former leader of the 2d Plat,
prisoners stated that about 150 men were involved, of whom 50 or 60 were hidden
in the village of Kwiam-ni. Tank personnel claimed 100 killed. In the opinion
of the writer, this incident Is the same as that reported by other sources, in
which the intense round of air strikes on Hill 570 on the morning of 11 Septem
ber flushed the enemy troops and caused them to flee toward the highway, there
to be hit by fire of the 99th FA Bn and evidently by tank fire as well.
The incident is included in the following summary of operations by Co B
of the 70th Tk Bn on 11 September:
1500. CO of Company B reports to 8-2 giving the following information on action
of 1st and 2d Platoons today: 1st Platoon was in support of 3d Battalion, 8th Cavalry
from position at apprOximate coordinates 1154.3-1452.6 [southernmost finger of Hill
373J fired into enemy position on Hill 373, coordinates 1155.2-1454.2 [coordinates of
Hill 373 peakJ. 2d Platoon was in support of 2d Battalion, 8th Cavalry Regiment from
position at apprOximate coordinates 1153.2-1452.4 [base of Hill 225J. Fired into enemy
positions in the following locations: coordinates 1154.3-1454.4 [draw north of Kwiam-nlJ,
1154.5-1454.8 [southwest slope of 480-m \'seakJ and 1153.3-1455.7 [village of ChungsanJ.
InfUcted estimated 100 enemy casualties. .
As noted in the above report, the 2d Plat advanced as far as the village of
Chungsan. On this or some other patrol they fired from here at a suspected
enemy observation post in a cave on Hill 263. No enemy were observed, but
from up the highway the tanks received high-velocity flat-trajectory fire judged
to be 76-mm. A brief analysis of this fire appears under Site 4 in the subsec
tion of enemy weapons on Hill Mass 314, given earlier in this memorandum.
On 12 September, during the attack on Hill Mass 314 by the 3d Bn, 7th
Cav, the western slopes of Hill Mass 570 were neutralized by tank patrols made
up the highway by the 2d, 3d, and 4th Plats of Co B, 70th Tk Bn from 1130,
shortly after jump-off time, to 1330. On these patrols the 2d Plat fired only
its coaxial machine guns against suspect areas. From 1500 to 1730 elements
of the 3d and 4th Plats were again active in this way, firing at the trails on
Hill 570, with, apparently, their heavy gun as well as the machine guns. There
seems to have been no patrolling from 1330 to 1500; since mortar casualties
on Hill 314 were heaviest around 1400, this gap may be significant.
Operations of Co B, 70th Tk Bn on 12 September are summarized in the
battalion's journal as follows:
1455. 8-2 received Tp cali from Liaison Officer with 8th Cavalry Regiment. 2d
Platoon, Company B destroyed two MGs, suspected ammo dump at coordinates 1153.3
1456.2 [this was a small, untended pile of ammunitionJ and inflicted estimated 10 casual
ties to enemy at apprOximately 121230 1.
1732. CO radioed CO, Company B requesting results of today's operation. He
reports a little action. Destroyed 3 MGs, ammo dump and inflicted apprOximately 15
enemy casualties.
2100. 8-3 Tp CO, Company B for report of action today. Received follOwing re
port: 3d and 4th Platoons, Company B supported attack of 3d Battalion, 8th Cavalry
Regiment [7th CavJ on attack of Hill 314. Platoons attacked abreast, 3d I:latoon in the
river bed from position along line 1153.2-1453.0 to 1153.1-1455.7; 4th Platoon along
126
ORO-T-261
UNCLASSIFIED
SURE'T
$[ERn
UNCLASSIF!ED
road from 1153.3-1452.8 to 1152.8-1456.4. Time was 121130 I to 121330 I Sep 50.
From 1500 to 1730 elements of 3d and 4th Platoons, Company B supporting 3d Battalion,
8th Cavalry Regiment fired on the south slopes of Hill 314 [must be Hill 570] from along
coordinates 1153.5-1453.0 to 1153.2-1456.4. From 121700 to 121800 I Sep 50, 2d Platoon,
Company B in support of 3d Battalion, 8th Cavalry Regiment fired into south slopes of
Hill 314 [must be Hill 570] from along road 1153.5-1453.0 to 1153.2-1456.4. Destroyed
oile enemy MG (also fired on trails and assembly areas) .1'
The same operations are reported in the division's log:
122030. From 8th Cavalry: Results of the tank show supporting Scrappy [8th Cav]:
Morning s ~ o w : 2 machine gun nests destroyed, 1 ammo dump destroyed, 12 enemy killed,
shot up village. Afternoon show: 1 machine gun nest destroyed, 4 enemy killed, trails
leading up Hill 570 covered with tailk, machlne gun and HE fire. Ammo expended: 180
rounds HE, 12,000 rounds .30 caliber.
1
Armor in Direct SUPP'ort of the Attack on Hill Mass 314
Information in this subsection was obtained primarily from communica
tions from the leader of the 3d Plat, Co C, 70th Tk Bn, and from the battalion's
unit journal. Some additional data were obtained from entries in the personal
log of a corporal ofthe platoon and from personnel of the 3d Bn, 7th Cav.
The five M4 tanks of the 3d Plat, Co C, 70th Tk Bn were attached to the
3d Bn, 7th Cav for the attack on Hill Mass 314. These units of the two arms
had not previously operated together, and until 12 September both armor* and
infantry expected the attack to be on Hill 570. In theory, the tanks could have
furnished valuable support to the assault by advancing around the right side of
the hill mass and firing on the reverse slopes of the objective. This move was
not attempted duringthe battle. Afterward it was discovered that such an ad
vance would have been a physical impossibility because of a deep crater in the
road at a defile (see Fig. 12). Investigation of the poor coordination and bad
feeling between the armor and infantry personnel would therefore be academic.
It is pertinent only to determine the extent to which the tanks actually did sup
port the attack, and to take account of the independent observations of the armor
personnel.
Operations of Co C, 70th Tk Sn on 12 September are reported in the unit
journal as follows:
0830. 8-2 received Tp call from CO Company B, said the whole thing as far as
he is concerned has not Jelled yet, probably It will be later than scheduled. Ask CO
if he knows that Company C is involved.
0834. CO Company C Tp S-2, going to move one platoon to support operation
this morning. 3d Platoon can get up on ridge [to Todok Temple?] to give fire support.
Can give coordinates but will hold these. Right of unit now in position but left of unit
making the operation today. Will support the left unit. 0930 preparation, [attack] 1 1/2
hour after thai.
1005. CO and Col Withers left CP for 8th Cav Regiment to observe the attack
planned for today.
Tentative plans to attach the 3d Plat, Co C to the 3J Dn, 7th Cay were formulated on 9 September.
ORO-T-261
127
UNCLASSIFIED
SiCREI
UNCLASSIFIED
1020. CO Company C reports to CP with following information: 3d Platoon,
Company C will support attack of 3d Battalion, 7th Cavalry Regiment from pOSition at
approximate coordinates 1155.8-1452.7 [southern tip of Hill 314J.
1845. CO Company C reports to CP with following report: 3d Battalion, 7th
Cavalry Regiment has completely taken Hill 314, forward units at 1156.5-1455.5 [Knob
3J. 3d Platoon supported 3d Battalion, 7th Cavalry Regiment, firing 19 rounds of WP,
89 HE. Mark target twice with smoke for aircorps and laid smoke for infantry. He was
used to screen flanks and break up counterattack. 3d Platoon is at coordinates 1156.4
1452.8 [southwest tip of Hill 314J ...."
With the 1845 report quoted above for reference, the former leai:ler of the
3d Plat, Co C, 70th Tk Bn supplied most of the information used to construct
the following account. Some of the movements are given a time reference by
entries in the personal log of a corporal of the 3d Plat. He states in a recent
communication that his watch was always in agreement with the time maintained
by the company CP, and several events mentioned in the log are given as at the
well-established times.
Early movements are timed by the corporal's log as follows:
It is now 0730 and we're all prepared to go on a mission in support of the 7th Cay
Regt ...It's now 0913 and we still haven't moved out but we will shortly... 0930, we're
mOving out... 1000, we just got out of the river hed; we will soon be in touch with the
enemy; right now our infantry is all around us ... 1018, we just caught up with the main
body of infantry (ours) and I guess there will be a waiting period while the Air Corps
finishes its job and then the artillery does their share. Then we move. The Air Corps
is busy right now. 1100, the Air Corps is through. 1110, the artillery is now busy plus
the mortars [actually. mortar fire, but no artillery preparation].
It is evident that the tanks arrivetl at Hill Mass 314 after the infantry, and
therefore could not have marked the target for the air preparation. On arrival,
the five tanks took up a position on the trail at the left side of the hill mass in
the vicinity of the coordinates 1155.7-1453.3 (621820). From here Knob 3 could
be seen, but Knob 2 was partly masked by Knobs 0 and 1. Two of the tanks, one
of them that of the platoon leader, moved back around the hill to the right at the
approximate coordinates 1156.7-1453.7 (630824). The corporal's log reports
this move: "1125; the Lt. just took off to the right side with one other tank."
These two tanks received 120-mm mortar fire which seemed to walk
along with them, and which the armor personnel considered directed at them.
The sites of enemy weapons and observation points are nOw known, and the
tanks could not possibly have been the objects of observed fire. The infantry
received its first mortar fire at 1135, from 120-mm mortars on the line of
departure. Rounds that overshot the ridge would have fallen just where the
tanks were, and as this fire shifted to the assault units advancing in the saddle,
long rounds would have seemed to follow the tanks too. The two tanks withdrew
to their original positions to the left of Hill 314.
The Infantry's report on this movement of the tanks is the only documen
tary information from this source on the supporting armor. The 3d Bn journal
~ ~ ~ : .
121300. Tanks were given the mission of support of Battalion by moving along
road to right and left flanks of Battalion. They moved up the right side and fired into
128 ORO-T-261
UNCLASSIFIED
s(ERn
SUIU
UNCLASSIFIED
the village [evidently Popsong-dong] to the front of position and received some mortar
fire and retired.
S
Early in the attack all electrical communication between Co I and the rear
was lost. Company L still had radios, but the completeness of linkage with the
rear CP at the base of the hill is not known. From their midway position on
the left flank, the tanks were able to serve as a secondary communications link .
and, in fact, relayed requests for ammunition, litters, etc.
By around 1200 the infantry's second assault on Knob 2 had been repulsed.
Infantry sources state that an enemy grouping on the northeast slope of Knob 2
was dispersed by the fire of artillery and 81-mm mortars. The two tanks at
the original position began 'firing at this time on the western slopes. The cor
poral's log states: "1215, we fired 10 rounds [reference is apparently to the
corporal's tank; he is sure some of these were WP], stopped at 1230, now
waiting."
At around 1320 a third assault was stopped, and Co L on the left was in
serious trouble. The CO of the 3d Plat, who was now back at the original posi
tion with all five tanks, could see the bunching of friendly troops on the left,
west slope of Knob 2, and could see enemy mortar fire inflicting casualties
there. A check at this time showed that friendly heavy mortars, as first sus
pected, were not firing there. A large enemy, grouping on the left side of Knob
3 could also be seen, though the number of troops could not be estimated from
such a distance. Via radio, Co L was calling for artillery on this enemy con
centration. Without orders, the tanks fired 25 HE rd at the crest of Knob 3.
The infantry reported them successful, but, not knowing of the participation of
the tanks, attributed this supporting fire to artillery. Observers in the saddle
or on the forward slopes of Knob 1 could not have seen either the enemy group
ing or the tanks firing. The corporal's log reports this second period of tank
fire: "1330, we just finished firing 20 more rounds of HE and 7 of WP."
Tank personnel .reported that the tanks marked the objective twice for
air strikes; this is confirmed by the logged report of WP firing. It is not clear
whether two separate air strikes actually occurred during the afternoon. The
time of the final one of the day is fixed by the corporal's final entry: "Well,
we've been waiting for over 2 hours while the Air Corps got busy again and now
at 1445 our infantry has advanced to the top of the ridge. They don't have it
all yet."
Operations of the 3d Plat, Co C, 70th Tk Bn on 12 September are summa
rized in the battalion's journal as reported above. ,
After the objective had been secured, an enemy counterattack seemed
imminent. The tank platoon was therefore requested to advance and fire into
the villages to the east and north where the enemy was thought to be c o n ~ e n
trating. Two tanks with six supporting infantry troops advanced along the trail
to the right of Hill Mass 314, this time reaching the coordinates 1157.1-1454.2.
Here, in a defile formed by a 30-ft cliff on one side and masses of boulders on
the other, the trail was found impassable because of a large crater made pre
viously by friendly forces. This block could not be bypassed and the tanks had
to back 300 yd to find a place wide enough to turn around. These facts about
the crater were also reported by a member of the 3d Bn, 7th Cav P&A Plat.
Beyond the site of the crater the trail narrowed to jeep width. At a late stage
ORO-T-261
129
SHiEl
UNCLASSIFIED
UNCLASSIFIED
of the battle a jeep-carried 75-mm recoilless rifle got as far on this trail as
Phase Line 2.
On the afternoon of 13 September, after the crater had been filled in,
armor personnel in a jeep proceeded further north on reconnaissance. The
valley appeared to have been deserted by the enemy.
SUPPORT FROM THE AIR
A special study has been made of close air support operations in Korea
during the period July 1950 -January 1951.
27
The comprehensive statistics
presented afford a perspective of the support furnished the 1st Cav Div. The
organization of the tactical.air force at that time is thoroughly described and
its typical operations are discussed. Operational data, however, are restricted
mainly to the air Side, and there is little basis for evaluating physical results
or the ultimate effects of the air effort on the ground situation. Correlation of
particular air strikes with the established picture of the ground situation in the
zone of the 1st Cav Div may be of value. It is attempted below.
Further information on air operations was obtained from: (a) the TACP
log of the 1st Cav Div; (b) air strike mission reports and mosquito reports;*
(c) communications (in one case, an interview) with pilots and observers of
strike-control aircraft who had been furnished with the reference material
cited in (a) and (b); (d) statements by infantry and armor personnel; and (e) a
few entries in the journals of infantry units. No contact was made with former
personnel of the regimental or divisional TACPs, but it is thought that these
did not operate close to the front lines. '
Scale of Close Air Support in Korea
The data under this heading are drawn from the special study.27
The level of air- support in the defense of the Pusan perimeter was sev
eral times higher than in northwestern Europe in World War ll. Then, the 30
to 40 US divisions in the line were supported by an average of about seven
sorties per division per day. The over-all average in Korea was nineteen,
and in September 1950 the 1st Cav Div received a daily average of about thirty.
For Air Force p;issions, the average time from request for a strike to
attack on the target_Ylas 45 min, and the average time that the planes stayed in
the target area was,.30 min. The average distance from friendly troops to the
pOint attacked w a s : ' ~ 3 , to 4 miles. When contact was intense, 17 percent of tar
gets were within 1000 yd of the friendly MLR. It is considered by the writer
that these figures are in error due to lack of complete data in periods of intense
action, such as the subject of this study, when ground troops were often given
very close support.
The ave-rage munitions expenditure per sortie in Korea was 2.68 rockets,
0.083 tons of bombs, and 0.314 tanks of napalm. The total expenditure corres
ponding to the 30 sorties per day in support of the 1st Cav Div would then have
been 80 rockets, 2.5.tons of bombs, and 10 tanks of napalm. Figures are not
available on 50-cal ammunition.
* These were collected and microfilmed by ao ORO field team in the Far Bast Command in 1950 and are
otherwise no longer available.
130 ORO-T-261
UNCLASSIFIED
SUitT
SUREl
UNCLASSiFIED
General Situation
Now, several years after the operations in question, few detailed obser
vations on the effects of air strikes or on other aspects of the ground situation
are forthcoming from air personnel who made or directed ground-support
strikes at that time. It is the result in part of a particular and temporary
situation. Because of the acute emergency that had developed by September
1950, air operations were highly informal. Many men were working under
great pressure in highly specialized jobs to which they had just been assigned
without the training that would have been desirable. Only 14 of the 49 mosquito
pilots and observers had had prior training. Air personnel were not in a posi
tion to be aware of the details of the situation on the ground. They had almost'
no occasion to meet men of the ground forces and often did not know what units,
below diviSion size, they were supporting.
Tactical air operations were controlled from the rear. After the evacua
tion of Taegu on or about 6 September, the 5th Air Force tactical air control
center (Mellow) was located at Pusan, though strike aircraft and mosquito
planes still operated from the Taegu airstrip. Almost all ground support mis
sions were directed by mosquito aircraft, since the regimental TACP was usu
ally located at regimental headquarters. The mosquito pilot was an Air Force
officer and the observer either an Air Force or an Army officer. Uusually one
mosquito handled all missions within the boundaries of a diviSion. When air
traffic was especially heavy, a second might share the load. (In the 1st Cav
Div the mosquito planes were designated Wildwest and Wildwest Special.) On
a busy day, a mosquito pilot and observer would take off two or three times
to direct the prebriefed or on-call flights allocated to a division. One 1st Cav
controller stated that in one 3-hr period he directed a flights of 4 aircraft each
at a particular hill in preparation for a friendly -attack. The number of passes
at the hill is not known. Sometimes a single mosquito directed two separate
missions at the same time; between Hill Mass 570 and Hill Mass 314, for ex
ample, the mosquito could control planes flying a clockwise pattern around Hill
Mass 314 and those flying a counterclockwise pattern around Hill Mass 570.
The US Army units were furnished close air support by the Air Force with
propeller-driven F51's and FaO jet aircraft. Marine F4U's occasionally were
. vectored to targets in Army sectors. The B26's operated singly and usually
at night on bombing and observation missions somewhat to the enemy rear.
When strike aircraft were put under control of a mosquito plane, they
were usually held in stand-by orbits several miles back from the front. (At
tacking troops therefore would not know that aircraft were on hand to back
them up.) Priority in attack was given to the jet aircraft, because of their
limited flying time. For this or other reasons, it was not Uncommon for F51's
to stand by as long as an hour or an hour and a half before attacking. The tar
get was ordinarily marked for the mosquito by white phosphorus bursts fired
by the ground forces. Air-recognition panels put out by the most forward
friendly infantry troops were taken by the pilots of the strike aircraft as the
basis for their aiming points. In September 1950, When the oppOSing forces
were often in very close contact on opposite slopes of a ridge or knob, the
aiming point would be only a few hundred yards beyond the most advanced
friendly elements. It is not surprising that on one occasion troops on Hill 373
reported that they themselves were strafed, even though panels were displayed.
ORO-T-261
131
UNCLASSIFIED
SHREl
Each strike plane would make several passes, the last uSually a swinging
turn to hit enemy gun positions. Once ammunition was expended, the planes
would return at once to the Taegu base. Pilots operating in the zone of the 1st
Cav Div had to do little more than cut the throttle for the return trip. They
sometimes took off again as soon as their planes could be refueled and reloaded.
It would therefore appear that the 1st Cav Div actually or potentially had con
siderable air support, as long as the weather permitted air operations.
Weather exerted general control on air operations through its effect on
the Taegu airfield, which was located in a trough like valley. Sometimes the
ground fog that flowed down the adjoining slopes di, I not clear off until the
afternoon and so prevented all air operations until:his time. Once the planes
took off, they continued to operate despite low clouu strata, rain squalls, and
partially concealed targets.
From the air, Korea appears to consist of hi]' s and ridges in unending
repetition. Targets were located by map coordinates. For strike aircraft
and even perhaps the mosquitoes the terrain in view was not clearly related
to the map 'or to the current ground situation. It was the responsibility of
ground units to authorize air attack and mark the targets. Security demanded
that personnel in the air be told a minimum about the ground units via radio.
Sometimes there was no radio link at all with forward observers on the ground.
It is understandable that in the periodunder study, Hill 373 and Taegu itself were
strafed by friendly aircraft and that in the early fighting on Hill Mass 314 air
support was refused as dangerous to friendly units in such close contact with
the enemy.
As for observation from the air, enemy skill in using brush for camou
flage was generally recognized. Strike aircraft on their attack runs flew low
enough for observation of details but too fast for pilots to form much of a
close-up picture. As a rule, the slow, vulnerable mosquito planes remained
high to avoid the intense small-arms and automatic-weapons fire usually di
rected at them. From such altitudes, personnel and gun pOSitions were very
hard to make out. Sometimes, however, when low clouds forced them so low
as to risk colliding with ridges, enemy and friendly troops could be clearly
seen.
Though it is very difficult to report enemy behavior during particular air
strikes, some reliable generalizations can be made. Enemy troops were usu
ally emplaced around the wooded crowns of hills near the summit rather than
up on the bald ridge. Napalm strikes were therefore usually directed at the
slopes near the top. As a rule, several passes would cause the enemy to re
treat to prepared positions further down the slopes. At this time the enemy
had not learned to minimize casualties by retreating at right angles to the
flight path. The flight path was predictable, since friendly aircraft always
flew from south to north to minimize the danger to friendly infantry.
Air Strikes in Support at Tabu-dong and Hill Mass 570
According to the TACP log," the only air strike in the 8th Cav Regt area
from 1 through 5 September was on enemy artillery northeast of Tabu-dong on
1 September. An unlisted strike against the same target was made by F4U's
on 3 September. Mission reports are lacking for most of the strikes during
132 ORO-T-261
TABLE 11
!il
o
I
AIR STRIKES ON HILL 570 ON 11 SEPTEMBER
...,
I
N
0'>
,..
Flight Squadron
1st Cay Div TACP log Mission reports
Planes, ,I
no. and type
Load IRo:po,t j T"'get
hme
PI I Time I .
I
Target, attack, results anes, MOSqUIto
no. and type JTake-off I Land I
Ratkiller A Unknown 4 F51 Napalm land rockets?] 0830 Hill 570 4 F51 0715 0915 Ballot Contacted Ballot 14, went to pre-
briefed ridge south of Walled
City. Napalm and strafed town
with good results.
... , , r _ 1
Ratkiiier B Unknown 4 napalm tana rOcketSJ 0853 Bin 570 3 F51 0825 0910 Prehriefed to ridge one rp.ile south
of Walled City. Dropped napalm.
Results excellent, strafing results
fair, fired rockets at 20-mm gun
and missed.
Ratkiller C Unknown 4 Napalm [and rockets] 0905 Hill 570 3 F51 0915 0945 Wildwest Napalmed and rocketed ridge
01 Walled City ll55-1456 [Hill


570J. Good hits.
DropkickL 77th RAAF 4 F51 Rockets, 0847 Hill 570 4 FSI 0645 0930 Wildwest Rocketed and strafed ridge one
mile southwest of Walled City
ll55-1456. Results undetermined.
C
One horse and cart destroyed.
C Z

0
Gasmask B 80th F .. 4F8O Rockets, 50'8 llOO Hill 570 4F8O 1030 lIDO Wildwest Strafed troop concentrations on hill
0
r
at 3603N_12833E in Tabu-dong
r
(.1)
area, results unknown. TOT Ill5
C/)
C/)
ll40. 16 rockets. [Note: Coordi-
C/)
-,.,
nates are lU miles ENE of Tabu-
dong.]
rn
rn
" 0
0 Ratkille, D Unknown-------------Not listed --------:----- 4 F51 0925 1015 Wildwest Bombed and strafed ridge east of
highway 3 miles 80uthof Tabu-
dong, results unknown.
,.. ,
'"
'"
UNCLASSIFIED
the period 6-10 September, and the principal source of data is the TACP log.
This was considered by air personnel as the most authoritative source of data
on the identity of targets. Reports are available from the flights of ground
support aircraft that struck Hill 570 on 11 September. In Table 11 they are
listed in parallel with the TACP entries, since the air effort on 11 September
was of critical importance. These mission reports contain neither much infor
mation on the results of the strikes nor additional observations, but they do in
general confirm the TACP log record that a given target was attacked. Accord
ing to available records, no air strikes were directed at Hill 570 on 12 September.
On 6 September, at about 2000, two aircraft hit the ridge of Hill 570. This
was perhaps the first time this hill was the target of friendly air. Around 0700
the next morning seven planes hit the same point. The results observed are
not known, nor are the ultimate effects, since there was little or no contact
between friendly and enemy ground forces on 7 September. Bad weather pre
cluded any air effort at all on 8 September. The 8th Cav therefore had no air
support for its first, unsuccessful attempt to take the hill. Air action was again
meager on 9 September, when four planes dropped napalm at around 0930.
On 10 September air strikes were postponed by bad weather until late
afternoon and then directed principally at the lucrative targets offered by the
enemy build-up around Tabu-dong, though one or two flights attacked Hill 570
itself. Prisoner-of-war reaction to these strikes is reported earlier in this
part in the section on enemy units.
The critical situation of thEt friendly forces holding Hill 373 on 11 Septem
ber was described in Part I, and the 20 or 24 sorties against Hill 570 during
the morning were listed there. The detailed data available on this air effort
are presented in Table 11. The pilot of the Wildwest mosquito in the air at the
time stated that the attacks thoroughly burnt out the ridge and caused enemy
troops to flee down the western s lopes toward the road. The mosquito observer,
in a separate interview, stated that enemy troops also fled to the northeast.
The 2d Plat, Co. B, 70th Tk Bn interpreted the movement of enemy troops to
wards Its positions at this time as an attack. An unknown observer, according
to a logged report, had previously observed enemy troops milling around in the
saddle between the 570-m and 480-m peaks. The former S-3 of the 99th FA Bn
agreed that the movement probably was a rout and added that artillery fire
chased the fleeing troops back up the hill. From these various reports It ap
pears that the enemy on Hill 570 fled in complete disorder.
Air Strikes on Hill Mass 314 on 11 and 12 September
The pilot of a Wildwest mosqUito stated in a recent interview that enemy
troops were concentrated on Hill Mass 570, on Hill Mass 314, in the valley
between them, and in the valley to the north. No distinct groupings were ob
served. He made the significant observations of the fighting on Hill Mass 314
which follow, but was not sure whether the date was 11 or 12 September. Though
the weather appeared clear on the ground, it was not favorable to air operations,
for there was a low, virtually continuous layer of haze (11 Sep: cloud conditions
not known; 12 Sep: 9/10 cloud cover at Taegu at 1040 with base at 2500 ft, tops
at 4500 it). Visibility from the air was hardly impaired in the vertical direction,
but horizontally it was very poor. The mosquito had to fly unusually low, and
enemy and friendly troops therefore could be seen clearly. During the after
134 ORO-T-261
UNCLASSIFIED
Sinn
UNCLASSIFiED
noon, the slow but steady progress of the attack could easily be followed through
a period of several hours by the advance of the yellow (8th Cav?) air-recogni
tion panels up to Knob 2. Friendly troops were not in any definite disposition
but seemed to be in good order. Though they did not have the benefit of cover,
they showed no tendency to disperse down the slopes. The air. strikes were
delivered atop the ridge. Their effect was clearly discernible-the
enemy troops to the rear and on the flanks began to slip back. The main body
of enemy troops was in foxholes on the ridge, and assaulting troops were ex
ceedingly close to forward enemy elements. With contiriuirig air attack, the
enemy began to crumble. All but the troops pinned down at the front line in
contact with US troops began to display restlessness. Finally, the breakpoint
was reached, and the enemy troops abandoned their foxholes and ran to the
rear. It is not known whether or not a thin enemy line still remained to oppose
the friendly attack. Since the details of the attack on 11 September are not
known, this account cannot be verified. It seems likely, however, because of the
two missions recorded for this date in Table 12 and several other facts. As
cited in the table, the mission report states: "Results not Observed, but con
troller informed flight after first pass [that] NK troops started to give up
positions and run."
It would seem that the situation on Hill Mass 314 on the afternoon of 11
September was similar to that on Hill Mass 570 during the morning: enemy
troops fled en masse under US air attack, but some enemy units remained to
hold the high ground. In the evening, the same mosquito pilot observed large,
active groupings of enemy troops on the upper slopes of the Walled City.
According to the TACP log, the air preparation for the attack by the 3d
Bn, 7th Cav on 12 September consisted of strikes by four flights of planes and
lasted approximately an hour and a quarter (1000 to 1115). The mission re
ports attribute control of the first three flights to the mosquito Wildwest and
the last to the Wildwest Special. This attribution may be correct in the case
of the first two flights, but it hardly seems possible inthe case of the third, which
reported to the TACP at 1030, just 20 min before the Wildwest landed at Taegu.
A second mosquito Wildwest had already taken off at 1015 but was not involved;
the pilot in the air at that time reported in a recent communication that though
he worked Hill 570 he never directed strikes against Hill Mass 314. The Wild
west Special could imd may have handled all the flights. It took off at 0930, 20
min before the first flight, Ratkiller Able, reported to the TACP.
No mission reports are available for the third and fourth flights, which
are listed as Ratkiller Charley and Dog. According to a Wildwest mosquito
pilot, flights were deSignated "Ratkiller" instead of by the squadron code name
when the loads included 5-in. rockets. The unlisted reports of the flights Wolf
hound Charley and Dog match in every respect except name and are assumed
to be the ones corresponding with the TACP entries. '
The observer (Army) of the first mosquito Wildwest was supplied with
maps and reference material and interviewed by the writer.' He could supply
no information about Hiil Mass 314, although he did recall the attacks on Hill
570 on' 11 September. The observer (Air Force) of the Wildwest Special could
supply no specific data about Hill Mass 314 except that intense small-arms
and machine-gun fire was received by attacking aircraft from emplacements
at the base of the hill on the northeast, north, and northwest Sides, and from
ORO-T-261
UNCLASSIFIED
SHin
135
0
c::
z
r
;p-
TABLE 12
en ....
C/)
'"
C)
'AIR STRIKES ON HILL MASS 314 ON 11 AND 12 SEPTEMBER
--n
r,-i
0;:.
Flight
Planes,
Squadron
no. and type
1st Cay Div TACP log
Report
Load
time
Target
Mission reports
Planes,
Time
Mosquito Target, attack, results
no. and type
T ake-oU I Land
11 September
Admiration 8 39th F" 4 F51 Rockets, napalm, 50's 1505 Hill 314
1156.5
1455.5
------No report available ------ TACP Log states: "'Bodies
seen blown in air, perfect re
sults. [Note: This remark may
refer to following strike.1
Utah A 8th F" 4 F80 Rockets. 50's 1615 Hill 314
1156.5
1455.5
4 F80 1520 1710 Wildwest Contacted at 1600 and was di-
reeled to Wildwest control,
Taego area. Contacted Mosquito
Wildwest and had a ridge to he
strafed at 3605-12835 [coordi
nates of Knob 2, Hill 314:
3600N-I2835E:] Fired 15 x 5
IIVA rockets in troop positions
and strafed area. Results not
observed, hut controller informed
light after first pass NK troops
started to give up positions and
run. In area from 1600 to 1635.
16 rockets.
12 September
g
?
'"'l
,
""
0>
....
Ratkiller A
Ratkiller B
Unknown
Unknown
4 F51
4 F51
Napalm, rockets, 50's
Napalm, rockets, SO's
0950
1015
Hill 314
Hill 314
4 F51
3 F51
0835
OB55
1030
1045
Wildwest
Wi.ldwest
Bombed bridge [sid 3 miles
southeast of Walled City.
Rocketed and strafed ridge with
good results. 6 napalm, 2 frogs,
14 rockets.
Bombed and strafed ridge 2 miles
southeast of Walled City, results
unknown
g
0
,
WolfhoundC
a
40th F" 4 F51 Napalm, rockets, 50's 1030 Hill 314 4 F51 0910 !l00 Wildwest Strafed ridge of enemy troops.
>-3
,
TOT 1030-1045. 4 napalm,
""
3 frags, 14 rockets.
....
'"
Wolfhound Db 40th F" 4 F51 Napalm, rockets, 50's !l05 Hill 314 4 F51 0930 !l20 Wildwest Bombed and rocketed Tabu-dong.
Special n rockets into Hill 570 [prob.
ably Hill 3141. 4 frags,
15 rockets.
COll,sin B 36th F" 2 F51 Napalm, rockets, 50's 1215 Hill 314, ------No report available ----- TACP Log states: "Area
N side covered. "
Wolfhound II 40th Ftr 3 F51 Rockets and 50's 1330 Hill 314, ------No report available ----- TAep Log states: Area
.,
n
"
Slue covered. "
Cousin W 36th F" 2 F51 Napalm, rockets, 50'1 1410 [hlank] 3 F51 1345- Wildwest Bombed hill 2 miles SW [sid of
1445 Tabu-dong [Hill 314 is 6 miles
southeast of Tabu-dong].
4 napaim, 4 GP. il rockets.
aListed in TACP log as Ratkiller C. When load included 5-in. rockets, term "Ratkiller- was somefimes used instead of squadron code name.
bListed in TACP log as Ratkiller D. See footnote a.
c:
L'
z
Z C"')
C"')
.-
.-
:>
(I)
U)
en
(/)
-n
-n
rr1.
rn O
o
UNCLASSifiED
the ridge (Walled City) further to the north. The pilot of the Wildwest mosquito
of the afternoon of 12 September was the one whose remarks were reported a
bove. He had no additional observations to report.
No reports of the air preparation on Hill Mass 314 are available from 8th
Cav documents. Personnel of the 3d Bn, 7th Cav were not in a position to ob
serve most of the air effort. The CP at the base of the hill was not opened
until 1015, about the time the second flight was beginning its attacks. From
then until about 1100 the battalion was making the long climb up the ridge on
Hill Mass 314, out of v,iew of the objective. Evidently only the last two flights,
Wolfhound Charley and Wolfhound Dog, were observed. The former S-3 of the
3d Bn, who was on the hill, correctly recalled (without reference material) the
napalm bombing, rocket runs, and strafing by these eight F51's (though attri
buting the effort to four FaO's), and estimated seven or more runs as follows:
(a) pass by mosquito plane, (b) four napalm bombs dropped by two planes while
two other planes located target, observed, and covered, (c) rocket run(s), (d)
several strafing passes, and (el several dry passes. The former commander
of Co I reported that when the assault reached Knob 2 for the first time napalm
was still burning there.
The 3d Bn, 7th Cav had ground-support aircraft on call if needed. At 1220,
when the battalion's second assault had been repulsed, the following telephone
message was received at the division's headquarters:
121220. Telephone from Skirmish 3 [7th Cav S-3]: request air to come in on
north slope of hill, stayj,!!lUust north of east-west grid U n ~ 55.5 [1455.5, the Une pass
ing through Knob 3]. Want napalm in draws running north, northeast and northwest.
At 1300 hours, front Unes will mark their positions with panels. We have advanced to
east-west 55 grid Une [Knob 2]. Skirmish 3 wants to know when this is set up, Located
at Scrappy 3.'
Reaction was prompt. As recorded in the preceding subsection on armor
in support; tanks in support of the 3d Bn, 7th Cav marked the objective with
white phosphorus bursts between 1215 and 1230. According to the TACP Log
(Table 12) the flight Cousin Baker checked in at 1215, and soon afterward, pre
sumably, attacked the north side of Hill Mass 31.4 and draws to the north as
requested,
Around 1300, the attacking infantry had again been repulsed, and required
further air support. A logged message states: "121303. From Liaison Officer,
7th Cavalry Regiment: Blue waiting on 55 grid line for airstrike north of ob
't' .1
J ~ ....
Again the reaction was prompt. Before 1330 the tanks had again marked
the objective with white phosphorus, and at 1330 the flight Wolfhound Baker
reported to the TACP that it would attack the north side of Hill Mass 314, pre
sumably soon after,
Assault troops knew of neither of these"strikes. Very likely the reason
is that contrary to general practice, as with the preattack strikes, the aircraft
did not follow a south-to-north course up the ridgeline. With intense enemy
mortar fire falling continually in the saddle, and with friendly artillery and
mortar fire falling on the objective ridge, such an approach would have been
most hazardous. In attacking the north side of the hill mass on these two
strikes, the planes probably, but not necessarily, approached from the south
138
ORO-T-261
UNCLASSifiED
UNCU,SSiFiED
east along the valley north of Hill 660. Their fire would then have been con
centrated on the northeast base and slopes of the hill mass, and on the known
troop and supply concentrations in the nearby villages. Such a flight path
would not have allowed the planes to reach the 82-mm mortar position now
known to have been in defilade on the northern slope. It is presumed that it
was the aircraft of flight Wolfhound Baker (which were armed only with rockets
and machine guns, not napalm) that are referred to tn the observation from the
air quoted below, to the effect that some time prior to 1445 friendly aircraft
were seen "shooting down in the valley."
The only evidence confirming the fact that more than one air strike was
directed at Hill Mass 314 during the afternoon is the followtng afterbattle
report:
121820. From Liaison Officer, 7th Cavalry Regiment: 3d Battalion, 7th Cavalry
jumped off at 1115 hours. after an airstrlke which was extremely effective.... They
called for an airstrlke at 1300 hours which was delayed until 1330 hours. An airstrlke
was given again at 1400 hours ... ,1
The one air strike reported by the infantry was the one that turned the
tide of the battle, when the assault units were still held up at Knob 2. Two
F51's appeared before 1445, perhaps as early as 1410. Whatever the hazards,
they attacked up the ridgeline and at an unusually low altitude. The most for
ward friendly personnel were gathered tn the peculiar semicircular d e p r e s ~
sion called "the shallow," just short of the crest of Knob 2, and the aircraft
began strafing 75 or 100 yd beyond this point. The command officer of Co I
stated that the airwash of the planes blew the helmet off the head of the CO
of Co L, who was standing by his side. The artillery liaison officer had also
moved up to this position. In a recent communication he drew a sketch of the
area attacked and showed straftng along the ridge from just beyond Knob 2 to
Knob 3. Two napalm tanks were dropped. One of them went over the north
east slope and the other landed squarely among a group of enemy soldiers
assembled behind a small knoll on the ridge between Knob 2 and Knob 3 (see
Fig. 3). This square hit with the napalm is mentioned in various infantry
documents as the decisive stroke that -led to the followtng final successful as
sault of the objective.
The fact that this air strike occurred before 1445 is established by an
observation, logged by one of the tank personnel, that at this time the air at
tack was over, and that friendly troops had advanced up onto the ridge but had
not yet captured all of it. Also, the following message was recorded tn an
official log:
121445. From General Allen [Asst Cmdr, 1st Cav Div]: the line of the 3d Battal
ion. 7th Cavalry looked as it was hefore on the 55 line [the grid line 1455, which passes
through Knoh 2]. That is the highest peak there. As I Was flying, the air dropped a
500-pound bomb about 200 yards in front of the panels. The rest of the air was shooting
down in the valley. oUr troops were receiving mortar fire.
1
The photograph Fig. 3, made two and one-hair years after the action, still
clearly shows a crater squarely in the declivity midway between Knob 2 and
Knob 3 on the western side of the ridge, i.e., about 250 yd forward of the potnt
where the panels must have been displayed. This bomb-burst fits all the de
ORO-T-261
UNCLASSIFIED SHRrr
139
UNCLASSIFIED
uatT
scriptions of the successful napalm drop. Friendly observers in the shallow
could not have seen the burst itself, merely the cloud of smoke. The existence
of a bomb crater at the point shown on the picture was confirmed by the CO of
Co I, who described it as a large bomb crater (diameter in photograph about
30 yd) and related how he and several others in it became casualties from a
single mortar round. As to the enemy concentration of troops, ,however, he
reported negatively. As observation of the dip in the ridge from the shallow
was at best limited, it would seem that the reported concentration was
exaggerated.
The identity of the planes that made the last air strike is not definitely
known. They would seem to have been the three F51's of the Cousin William
flight. As shown in Table 12, the TACP log listed the flight as having reported
on station at 1410, but the target is not specified. The corresponding mission
report is available and is also given in the table. Th!J only inconsistency is
the specification of the target as a hill 2 miles southwest of Tabu-dong, which
would have the planes attacking Hill 714, 2 miles to the rear of that critical
point north of Hill 401 known to be under air attack at this time. Although Hill
Mass 314 is 6 miles southeast of Tabu-dong, the well-known gross errors in
the identification of ground locations by fighter pilots could easily account for
the discrepancy. Also, there was no reason to attack anywhere in the vicinity
of Hill 714 on an emergency basis.
The Cousin William flight was in the air for a very short time, just 1 hr
from take-off to landing. If the usual 20 min is allotted for travel from Taegu
airfield to the target area, etc., the flight would have arrived at the target area
at 1405, 5 min before reporting to the TACP. Subtraction of the same interval
from the landing time leaves the flight in the target area until 1425. The air
strikes would then have lasted only 15 min. This short time suggests a special
emergency strike.
It was the personal impression of the 3d Bn personnel that the final air
strike was delayed for refueling. This is the reason given in the 7th Cav Regt
war diary for the delay. As may be seen from Table 12, the final strike was
by planes of that same squadron which had had an air strike on Hill Mass 314
eartier in the afternoon, the flight Cousin Baker, which had checked in with
the TACP at 1215. If these planes stayed in the air the normal 2 hr, they
would have landed at the airfield at about 1345. It is quite possible that they
did in fact land shortly before this time, were refueled and reloaded, and then,
at 1345, took off as flight Cousin William to return to Hill 314.
The load carried by the Cousin William flight agrees with the observa
tions from the ground. The planes carried 4 general-purpose bombs, presum
ably 500-pounders, 4 napalm drums, and 11 rockets. The number and type of
the planes also agree with the ground reports.
At 1635, after the objective had been secured, the 3d Bn, 7th Cav called
for preventive air strikes on the village of Kisong-dong, just northeast of Hill
Mass 314 at 1157.5-1456.0, and on Namch'ang, 3/4 mile to the north of 1156.9
1457.1. These strikes were to have been made if the aircraft were available.
Though they were not made, no enemy counterattack materialized.
140 ORO-T-261
UNCLASSIFIED sUln
SUUT UNCLASSIFIED
Partlll
SYNTHESIS AND INTERPRETATION
UNCLASSIFIED
$fERET
sum
UNCLASSIFIED
SYNTHESIS: THE TACTICAL PATTERN
IN GRAPHIC OUTLINE
This section shows in graphic form the progression
of move and countermove and blow and counterblow In
the period 2 - 16 September, and indicates some of the
essential factors involved. The data on friendly and
enemy forces given earlier In the memorandum are
brought together with some additional data in a space
time setting.
ORO-T-261 143
SECAE.J.
UNCLASSIFIED
UNCLASSIFIED
sum
DAY-BY-DAY DIAGRAMS AND HOUR-BY-HOUR
. ACCOUNTS OF THE ACTION
FOR 2 -16 SEPTEMBER
The representation of terrain is based o.n the map presented as Fig. 2,
which may be consulted for details. In the diagrams, the transcribed contours
have been selected to emphasize significant features. Except for a few breaks
for clarity, all 200-m index contours and all the higher 100-m contours are
shown. The dotted contours at the bottom of some of the diagrams are 100-m
lines. Some intermediate contours also appear dotted to indicate peaks. All
ground over 500-m elevation is shaded.
The only weather data available is the meager amount heading each right
hand column of text. It was often difficult to distinguish between a report and
a forecast. Two conditions were typical: intermittent rain, and fog in the val
leys in the morning. In the absence of lOW-hanging clouds, visibility between
showers was generally good during the day. The effect of weather on friendly
air operations was variable and highly dependent on locality. As a crude gen
eralization of precipitation and visibility, close- or wide-spaced black cross
hatching or none at all appears at the upper right-hand corner of each diagram
to represent, respectively, bad, poor or good conditions for observation from
the air or on the ground. The daily range of temperature was roughly 68 to 85F.
The location of forces is shown by the lines bounding forward units (blue
for the fdendly side, red for the enemy), or by the disposition of specified units.
Roughly equivalent opposing units are plotted, since, with an average strength
of 280 or less, an enemy battalion was about the size of a friendly company. The
primary sources of information on the location of friendly companies were the
available regimental operations overlays of 2,4, 5, 11, 12 (two) and 13 -16 Sep
tember for the 8th Cav Regt and of 13 September for the 7th Cav Regt. Several
overlays at the 77th FA Bn also were of use. With corrections or changes when
appropriate, dispositions are given as of daybreak on the indicated date. Moves
during the day are indicated by arrows. Locations on days not represented by
an overlay have been plotted from miscellaneous information or filled in on the
basis of continuity and simplicity. The sites of regimental and battalion head
quarters are given when known. An enemy regiment only generally located is
represented by the box symbol. The numerical deSignation of enemy units de
notes the battalion (to the left of the slash) and regiment (to the right). Friendly
companies are denoted similarly by letter and parent unit.
Friendly reserves behind the diagrammed area are listed at the head of the
left-hand column of text. These were not always formally attached as such but
were present and could have been used. Several companies of the 8th Engr Comb
Bn were occupied with normal engineer duties during part of the time the Bn was
carried in reserve. Adequate data do not exist on enemy reserves.
Friendly intentions are indicated'by entries "Operations planned" and
"Operations ordered" at the head of the left-hand column of text; unless other
wise indicated, these are from the operations reports of the 8th Cav Regt.
Speciiic enemy intentions are only known for certain actions, and such infor
mation is included in the columnar text.
Tactical moves are summarized in chronological order in the left-hand
column of text. An apparent response to a friendly or enemy move is shown
ORO-T-261 145
UNCLASSIFIED
SECREl
UNCLASSIFIED
by a connecting arrow. In the diagrams, a planned attack or counterattack is
represented by a broad, hollow arrow, all other movements by plain arrows.
Time notations are added when necessary. The diagrams show the front line
in the ROK sector on the right flank and in the 7th Cav Regt and subsequent 5th
Cav Regt sector on the left flank, but the text is restricted to action in the cen
tral sector.
Fire is not covered comprehensively. Friendly air strikes and certain
concentrations of friendly and enemy supporting fires are listed in the right
hand column of text. Where there is an apparent causal relation, the entry is
linked with a tactical move by an arrow. In the diagram, mortar and arti11ery
concentrations of particular note are indicated by small solid Circles, an air
strike (usually by four planes) by a large solid circle. Fire fights and close
combat not associated with an overt attack are shown by zigzag arrows. As an
indication of the available and actual ground-support effort, the right-hand
heading gives the daily total of both plane strikes in the entire 1st Cav Div
and those which were directed at targets in the 8th (and later the 7th as well)
Cav Regt zone. These figures are derived from the 1st Cav Div TACP Log.
They are sometimes considerably greater than official listings of daily sorties
in the 1st Cav Div zone, perhaps because strikes at several targets were some
times made on one sortie.
The intensity' of combat is not indicated but may readily be inferred by
comparing the diagram and text for any day with the comprehensive tabulations
of casualties by company in App D.
For graphic effect certain nonstanda"d symbols are used in the diagrams.
These appear below, among others of note.
Tank group Artillery concentration (vicinity)
1=1

Individual tank Air strike (vicinity)
p:j

.... (four) Fire fight

6666
Artillery or AT gun {four pieces)
{r
Overt attack
146 ORO-T-261
UNCLASSIFIED $[(REf
UNCLASSIFIED
SEERn
The area 0 f the battle
ORO-T-261 147
UNCLASSIFIED
SEEREJ
c
z
C'")
r
:x=
C/.)
C/.)
-,..,
fT1
C
2 SEPTEMBER
o
I
2000
I
4000
I
6000 yd
I
c
z
C'")
r
:x=
C/.)
C/.)
...,.,
fT1
C
Reserves at Taegu: 16th Recon Co (minus organic tanks) Weather observation: Operational for aircraft. In morning, intermittent light
Companies A, C, and D, 8th Engr (Comb) Bn rain and overcast, with low clouds from 1500 to 2500 ft; in afternoon,
5th 1I0K Spl Tng Bn clearing.
Operations planned: "Seize and hold Hill 290 in order to protect right flank plane strikes: 1st Cav Div-52
of the 7th Cavalry Regiment." 8th Cav Regt-O
Operations ordered: "B Company to attack and hold Hill 290. L Company
to attack to the northwest and take Hill 400 0142.5-1464.0)."
F In morning Co B. Co L, and two platoons of the Bv Mort Co leave Tabu
dong area and mo... e to west to support 7th Cav large-scale diversionary
attack.
E F:nemy build-up activity observed north of Co G. F An air strike is called for on enemy build-up but is not delivered.
F Tanks of 4th Plat, Co C, 70th Tk Bn (backing up 7th Cav right flank) F The remaining 3d Plat, Hv Mort Co fires at enemy activity throughout
displace to rear just before 1700. day. The 99th FA nn is continually active.
F At 1700 forward units alerted when word received that prisoner captured
by Co C on night of 1-2 September is S-3 of 19th Regt, 13th NK Div
and states that the division is to attack at dusk on 2 September.
E The 13th Div is scheduled to attack after a lS-min artillery preparation
which begins at 1800. In first drive the division is to advance 6 km
down the road to Taegu.
(E
The artillery preparation opens as scheduled at 1800.
firing are artillery and 120-mm mortars to northeast.
Among pieces
Fire falls all
across 8th Cav front, but only 1 rd every 2 min falls on regimental CPo
E Around 1830 a first enemy thrust straight down the highway is stopped
by Co G.
F Around 2000 an illuminating shell reveals an enemy battalion moving
down ridge toward Co F.
E Second thrust overruns one platoon of Co F. Enemy attacks repeatedly"--+--.-.F Position of 3d Plat, Bv i\!ort Co threatened. Platoon fires until about
in waves. Company F pushed to right, Co G platoon on Hill 356 (right 2300, when ordered to displace to rear.
side of road) driven to left rear. Wedge driven into friendly front line.
E Around 2300, 1st Bn, 14th Regt, 1st NK Div leads attack on HOK force
on llil1558; ROKs repulse attack.
E The 14th NK Regt and 2d Bn, 2d l'iK Regt push through gap, 3d NK negt
moves up.
,
E During night an enemy lank moves toward Tabu-dong to reinforce
penetration.
At 0200, :1 September, enemy attack opens against units of 11th ROK
"
Hegt on Hill 655. Attack still under way at OWO whhout loss of
ground by nOKA.
F At 0320, 3 Septen:ber a special strike of two F4U's hits enemy artillery
northeast of Cav sector.
....
on
o
'\ /
3 SEPTEMBER
t 2000
I
I 4000 ! 6000 yd
I
Reserves at Taegu: 8th Engr (Comb) Un
@
16th Recan Co
o
I
5th HOK Spl Tng lln
>-l
I Operations planned: "Hegiment to hold present positions."
""
,... Operations ordered: "R c.ompany and L Company to button up and hold
'"
their positions for the night."
E The 13th NK Div offensive down the highway loses momentum with
penetration almost to Tabu-dong. Enemy troops dig in on high ground
on both sides of road.
E The 1st Div SP-gun Bn moves toward Tabu-dong in support of the
13th Div.
E During the morning, a new attack by the 1st NK Div drives the lith)
ROK Regt off Hill 55R.
F Units of the 11 th ROK Regt withdraw generally to the right rear by
stages during the day.
E Units of the 1st NK Div advance rapidly toward and through the gap on
the 8th Cay right flank. Advanced enemy elements move onto Hill Mass
570 and drive off a friendly sweep force.
F Regimental infantry reserves and armor move up to establish defense
line east from Tabu-dong. The six medium tanks of the 2d Plat, Co B,
70th Tk Bn move through Tabu-dong to take up positions to the west on
the Waegwan road.
F Company F withdraws to reorganize; Co E is virtually cut off and is
ordered to withdraw at 1000. The 2d Bn takes up new defensive posi
tions on the regiment's left flank.
F To reinforce the new line of defense, Companies A and C of the 8th
Engr (Comb) Bn are attached as rifle companies and occupy the anchor
positions at Tabu-dong. The 8th Cav Svc Co is also committed as a
rifle company.
F Company n, 8th Engr is attached and given the mission of occupying
the Walled City, moves up in the evening.
,...
c.n
,...
Weather obser'l/ation: Operational for aircraft all day. Scattered low clouds
from 2500 to 5000 ft; no precipitation.
Ground-support plane strikes: 1st Cav Div-20
8th Cay Regt-2 (F4U's at 0320,not in TACP
log)
c:
z
(")
r
:l>
en
en
~
IT!
o
Enemy counterbattery fire silences the 99th FA Bn.
Counter-counterbattery fire of 9th FA Bn on enemy artillery in the vicin
ity 1155.6 -1469.8 silences the cGunterbattery fire, permitting the 99th
FA Bn to resume firing.
....
en
""
c:
:z
C"">
r
;l:o
(I)
(I)
::!J
rT1
0
4000
I
6000 yd
I
c:
:z
C"">
r-
G;
(I)
::;:;
~
;
....
Reserves at Taegu; 27th Brit Brig Weather observation: Scattered showers. Night of 4 -5 September rainy.
@
16th Hecan Co
o
Ground-support plane strikes: 1st Cay Div-8
Two provisional battalions, 1st Cay Div
I
8th Cav Regt-4
~
Operations planned: "To regain control of Walled City and high ground
I
en
'"
running from Hill 673 across the road to the high ground in the center of
.... the grid square 1150-1460 and establish a line along Hill 466 and then
along the high ground to the east of the Walled City0"
Operations ordered: "Same as 'Operations planned.' "
F The 5th HOK SpI Tog On is moved up from Taegu to occupy backup
positions in the 8th Cay rear.
E Attack at 0415 on Co A, 8th Engr is repulsed.
E Attack at 0800 pushes Col off high ground east of Tabu-dong.
F Company I counterattacks four times but in the end fails to recapture
the lost ground.
E Enemy elements slip through friendly line to set up a weak fire blOCk)
across the Taegu road.
F The 16th Recon Co clears the road, observing only small scattered
groups of enemy.
F During the morning, Co B rejoins the 1st Bn; Co L and attachments
return from 7th Cay area and rejoin parent units.
c:
z
E The enemy gains an OP on Hill 466. ) E Mortar and artillery fire called down on 3d Bn, 8th Cay.
("')
F The 3d Bn position becomes untenable, battalion withdraws a mile down/
)::0
r
the Taegu road and sets up a perimeter defense.
en
en
E, The enemy attempts to haul artillery to Tabu-dong with tanks. -----:;J--,
~
F Company I knocks out two T34 tanks towing field pieces.
I'T1
o F Company D, 8th Engr moves up toward Walled City.
E The enemy engages Co D, 8th Engr in a fire fight but fails to stop it
from moving into the Walled City.
F Company E is augmented with fiOK troops and is moved by truck to the 3' E A mortar and artillery barrage fails to stop Co E.
southeast base of the Walled City. At 1530, Co E advances. )
E Enemy small-arms and automatic -weapons fire pins down Co E on slope
south of Walled City. Company remains here for the rest of the night.
UNCLASSIFIED
$Inn
( ~
J
154 UNCLASSIFIED
ORO-T-261
SEER"
Reserves at Taegu: Two provisional battalions, 1st Cav Div Weather observation: Operational for aircraft. Scattered showers after 1300.
~ At 1530 recon from air reported hindered by rain, low ceiling, extreme turbu-
Operations planned: "E Company, 8th Cavalry and P Company, 8th Engineer
o
1ence. Visibility bad at Walled City.
I Battalion withdrawn from ',,\'alled City at 1630 hours. Strength of E Company
>-3
""
I
is 3 officers and 99 men." Ground-support plane strikes: 1st Cay DiY-4
8th Cay Regt-O
'"
Operations ordered: "Same as 'Operations planned.' ..
....
F Company D, Bth Engr intensely engaged in Walled City dllTing night of ~
4 -5 September. ~
F At 0700. Co E fights up to the Walled City and Joins Co D. 8th Engr. ) F Attempts to resupply friendly forces in the Walled City by airdrop fail
Both companies have severe ammunition and supply shortages. as drops land in enemy-held territory.
F the 8th Cav Regt is fully committed and there are no regular infantry
units in reserve at faegu. The 1st nn is ordered to pull into a perim
eler defense, with Co II blocking the road to the west, The 2d Bn is
ordered to form a line tying in with the 3d Bn. Through the day, the
16th Hecon Co and the tanks of the 70th Tk Bn patrol the t\.lSIl from the
3d Bn Lack to the rear of the regimental CP.
.,
, F
E The 13th Div eases pressure on friendly units as replacements arrive Tanks of 2d Plat, 7th Tk nn fire throughout day on enemy troops hauling
from pools to the rear, manhandles artillery through Tabu-dong to south. artillery and ammunition through Tabu-dong.
E The 1st Diy continues to send units forward. Troops of the ad nn,
14th Regt fire from 11 ill 570 on Htry A, 99th FA nn.
c::
:z
C?
r
:t:
V>
V>
.."
rn
CI
....
01
01
@
o
';'
",
CO>
.....
7
,
o
4000 6000 yd
I ! ! !
c:
:z
("")
.
.....
en rn
CO>
en
...,
I"T1
o
I
Reserves at Taegu.: Two provisional battalions, lst Cay Div. Weather observation: Some rain. Visibility 15 miles.
@
o Operations planned: "Move vehicles and heavy equipment to new area. Groundsupport-plane strikes: 1st Cav Div-42
I
After vehicles have cleared area, occupy new defensive positions." 8th Cav Regt-IO
>'l
I
Operations ordered: "Same as 'Operations planned.' ..
C>
""
.....
E The 13th Div remains generally unaggressive. A weak attack at 0530 F Around 0600 a night of four FaO's and two flights of two FaD's each
on outposts of the 3d Bn, 8th Cay, is driv.en off. cover area between Walled City and Hill 314.
E The 1st Div becomes more aggressive. Two companies of the 3d Bn,
14th Regt come down from llill 570 and attack Btry C, 99th FA Bn, but
are driven off by machine-gun fire.
E The two companies of the 3d Bo, 14th Regt, possibly with additional)
support, set up a fire block across the Taegu road.
F A large tank-infantry task force moves out during the morning and, after
a prolonged engagement, eliminates the block.
F rn a coordinated divisional withdrawal, the 1st Bn, 8th Cav withdraws
to the west through the 7th Cav sector.
E Advanced units of the 2d negt, 1st Div move onto the northern end Of)
Hill Mass 314.
F Companies of the 5th ROK Tng Bn are placed on the southern part of
Hill Mass 314.
F At 2000 an air strike by 2 F51's hits the top and east side of Hill
~ l a s s - 570.
00
....
'"
)
::,
6000 yd
4000 I
1
__L - ~ 2 0 ~ O O __~ ~ , __~ _ o I
I
Reserves at Taegu: 3d Bn, 7th Cay
Companies A, B. and D, 8th Engr (Comb) Bn
Operations planned: Continue on present mission."
Operations ordered: Continue on present mission,"
E The 13th NK Div remains unaggressive, but the 23d NK Regt advances,
following the withdrawal of the 7th Cay and 1st Bn. 8th Cay.
E Two companies of the 3d Bn, 14th NK Regt which were hard hit in the
fighting on 6 Septemher withdraw from llil1570 to receive replacements.
Tpe 1st Sn of the regiment advances toward Hill 373.
F The 1st Bn, 8th Cay swings around through the 7th Cay sector toward a
new line of defense on the 8th Cay left flank.
c::
:z
("'")
r
:J:o
en
en
-n
I'T1
o
....
to
'"
Weather observation: Frequent showers. Operations at Taegn airport limited
during morning hours.
Ground-su.pport plane strikes: 1st Cay Div-54
8th Cay Hegt-50
F At 0630, TACP net opened; at 0700 five flights hit Hill 570.
F At 0820, artillery fires on troops and mortars observed on Hill 570.
F Artillery fires on enemy troops observed on the northern part of Hill
Mass 314.
E At 1430, enemy artillery fires intensely on I ~ t r y A, 99th FA Un at rate
of 1 rd-eYery 2 sec.
F At 1530, surrender leaflets are dropped on Hill 570. Then two flights
strike the hill mass.
F During the afternoon, the enemy assembly area north of Bill 314 is hit
by an air strike.
F At 1700 a flight of two B26's strikes area between Walled City and Hill
314.
....
o
'"
c:::
(")
..
z
:J;:o
U)
U)
""T"\
I'T'1
C
~
o
.4
I
'"
....
'"
o 2000 4000
I
6000 yd
I
Reserves at Taegu: 8th Engr Bn (entire)
@
3d Bn, 7th Cay
o 2d nn, 7th Cay
I
>-l
I
Operations planned: 'To continue present mission."
0> Operations ordered: "1st Battalion to attack to rear in order to take up new
'"
I-'
defensive position."
E Night of 7 -8 September: enemy force infiltrates from east, blows up
culvert on Taegu road, and mines bypass.
E The 13th Div remains generally unaggressive.
I
F In the early morning. the 3d Bn, 8th Cay attacks lIillS70 and toward its
own rear. One provisional rifle platoon of Co A, 7lst Tk fln attacks
hill simultaneously from the south. The 3d Sn takes two peaks leading
to main objective with little or no opposition.
F Late in the morning, an enemy counterattack splits the forward two
companies of the 3d Bn, causing it to withdraw from Hill Mass 570.
F The 3d Bn, 8th Cav is withdrawn to constitute the regimental reserve.
The 2d Bn then displaces rearward to new defensive positions.
F The] 6th Recon Co moves without opposition onto the southern part of
C
Hill Mass 314.
Z
C? E Northeast of the 8th Cav sector, the 1st NK Div attacks the ROK Army.
.-
:t> F The two ROK police companies on Hill Mass 314 are moved across to
CI')
CI')
1Ii1I 660.
""T"1 F The 1st Bn moves up to new defensive positions on the 8th Cav
I"T1 left flank.
0
E Enemy forces engage Co C in a fire fight but do not stop it from
moving up.
I-'
0>
I-'
Weather observation: Light showers; rain through night of 8 -9 September;
fog. Visibility aloft unlimited, except in showers.
Ground-support plane strikes: 1st Cav Div-O
8th Cav Regt-O
F No ground-support atr actiVIty throughout the day. presumably because
of bad weather.
F The attack is supported by the fires of the 9th and 99th FA Bns and of
tanks of the 70th Tk Bn.
E At 1630. 30 rd of enemy artillery fire fall on the unit attempting to
repair the blown culvert southwest of Hill 373.
....
0'>
'"
c:
Z
C">
r
:t:
en
en
""T'1
rr1
0
i
-
~
o
I
>-l
I
0'>
'"
....
N
I
o
I
2000
I
4000
I
6000 yd
I
c:
z
C">
r
:t:
en
en
""T'1
rr1
0
I
-
Reserves at Taegu:
@
o
I
>-l
Operations planned:
I
'"
""
Operations ordered:
....
8th Engr fln
2d On, 7th Cay
3d- On, 7th Cay
"'Continue on present mission."
Same as 'Operations planned. ..
E Enemy units on Hill 314 engage the 16th Hecan Co in fire fight during
night of 8 - 9 September.
E At 0700, enemy units on Hill 314 attack the 16th Heeon Co, pushing it
back 10 the southern part of the hill mass.
F During the morning, the 3d Bn, 7th Cay moves up from Taegn to occupy
blocking positions across the Taegn road.
E Enemy units on Hill 314 move back to the northern end of the hill mass,)
though out of contact with the 16th Recan Co.
F Around 1130, the 16th Recan Co reoccupies Knob 2 on HilI 314 without
contacting enemy.
F At 1215, Co G is ordered to occupy Hill 373. The company does ~ o
without opposition.
F Late in the afternoon, a tank-infantry patrol moves up the road west of
c: Hills 373 and 570. No enemy activity is observed here or elsewhere
z by the patrol.
CJ
r F During the afternoon, the 16th Hecon Co attempts to take all of HilI
:>
Mass 314, but is driven back from the northern end by enemy fire.
C/)
C/)
E The 1st Bn, 19th NK Hegt moves onto the northern end of Hill Mass S70,
-n
the fir$t time that a unit of the 13th Div is on the hill mass.
rn
F HOK Companies A and C move, back onto the southern part of Hill MaSS)
o
31<1 from Hill 660. '
E An enemy group of battalion size infiltrates to the southern slope of
lIill 660. )
F nOK Co B is moved to a backup position to block the enemy infiltration
on Tlill 660.
Weather observation: Intermittent rain. Walled City covered with clouds at
1335. Visibility 10 to 15 miles, except 3 miles in early-morning ground fog
and 1 to 3 miles in showers.
Grou.nd-support plane strikes: 1st Cav Div-41
8th Cav Regt-4
F At 0930 there is an air strike with napalm on Hill 570.
F There is no more ground-support air activity throughout the day, pre
sumably because the weather closes in.
.....
....
'"
c:
2
C":>
r
:t:
en
en
...,.,
,."
0
~
o
I
>-l
I
~
.....
~
N
j
o
I
. 2000
I
4000 6000 yd
I
c:
2
C":>
r
:t:
en
~
~
,."
t:::7
I
'"
l\eserves at Taegu: 8th Engr Bn,
2d nn, 7th Cav
Operations planned: "To take Hill 3H 0156.3-1453.8) and to prepare to
capture and bold 11 ill 570 (1154.6-1456.1)."
Operations ordered: "To take Hill 314 0156.3-1453.8) and to prepare to
capt ure and ho 1d Iii II 570 (l L,)4.6 - 1456.1). "
E Around 0400, enemy attacks 16th Heeon Co on Hill I\lass 314. Attack
is repulsed.
F Around 1200, the 3d Bn, 8th Cav passes through the 16th Reeon Co and
attacks on lIilll\lass 314.
F Attack on lIilll\lass 314 is stopped by enemy small-arms and supporting
fires.
F The second attempt by the 3d On, 8th Cav to assault the objective on
IIiIl 1\lass 314 is stopped. Aroll,nd 2000, the battalion digs in for the
night.
F The 8th Cav CP displaces 1 mile to the rear.
E At 2024, enemy patrols probe Co G positions on Hill 373. Prisoners
warn of impending enemy attack.
E At 2335, the enemy attacks Co G in battalion strength, pushes

company back.
_____ F
.
F Company G reestablishes original positions by 0210, 11 September.
on
'"
E During night of 10 -11 September, two companies of the 3d Bn, 14th NK
Begt are recommilted to Hill 570 for aggressive patrolling down
Taegu road. First COlllpany is ambushed by friendly guards, remnants
join second. Bemaining enemy force then moves out, is also ambushed
anrl. hard hit.
Weather observation: Occasional rain. Getween 0600 amI 1000, visibility
3 to .s miles, due to haze and ground fog. At 1030, clouds prevent air strike
on Hill 570. In afternoon, visibility unrestricted, except in intermittent
showers.
Grourul-support plane strikes: 1st Cav Div-54
F
F
E E


E
F
F
F
F
F
8th Cav Hegt-17
Short artillery preparation on lIill Mass 314.
At 1227 a B26 strafes and drops fragmentation bombs in vicinity of the
Walled City.
Enemy mortar fire on Hill Mass 314.
Around 1400, friendly artillery fire hits enemy troops defending northern
part of 11 ill 314.
High-velocity artillery fire is delivered for the first time down the
Taegu road, lands in the vicinity of the 8th Cav CPo
At 1630, 30 to 50 rd of mortar fire land on Co C positions, cause no
casualties.
At 1800 a flight of four aircraft and another flight of two planes strike
at a tank south of Tabu-dong.
c:
z
At 1835 Rtill another flight of two planes strikes at this tank and at
C?
artillery in the vicinity.
r
:t>
At 1900, friendly air strike on the ridge of Hill 570.
v:>
v:>
At 1840, a fljght strafes and rockets the road and adjoining slopes from
-n
nest of Hill Mass 570 to Tabu-dong.
rT1
From 1902 to 1930 an F'82 strafes and bombs in the area of the highway o
and south of Tabu-dong.
Fire of the 99th FA On north of Bill 373 helps stall the enemy attack .
....
C>
C>
c::
Z
C"':l
r

(/)
(/)
::::!:!
I'T1
C?
N
I
.. . . . - ~ .
~ I'C/7
c<:\
o 2000 4000
L __ -L __1--___1.. . ....L __-1_-'
c:
z
C"'>
r
::J::o
(/)
~
-n
m
C?
Reserves at Taegu: 8th Engr Bn Weather observation: No precipitation. Visibility 10 to 15 miles, but visi
2d On, 7th Cay bility 1 mile in early-morning fog in low places.
Operations planned: "No change." (See 10 September: "take lIiH 314 and Ground-support plane strlkes: 1st Cay DiY-82
prepare to take and hold IIiIl 570.")
Operations ordered: "No change.' (See 10 September: "take lIiIl 314 and
prepare to take and bold IIill 570. ")
F
E At 0710, the enemy attacks again on aill 373 and overruns Co G
forward positions.
F At 0730, Co I and Co K, 8th Cav, resume the attack on the
part of Hill ]I,']ass 314.
F Co E moves onto Hill 373.
. F
E
'""eO ....., " , "" "'" "'" ,,, ;. "'.. "_."j:
the Taegu road. F
F A.round noon the 16th Recon Co moves up to reinforce Co F. and Co G,
c:
who move out in limited advance to reoccupy the former friendly'--....
z
positions on Hill 373.
n
r
>
CI)
CI)
"'T1
E An enemy counterattack in battalion strength on Hill Mass
rr1 around IS15 drives Co I and Co K back to original positions
o
3d Bn.
E During the afternoon, the 3d nn, 19th NK Hegt moves from the
314 at
of the
rear of
Hill l\iass 570 to reinforce enemy units on HilI Mass 314. This is the
first lime a unit of the 13th Diy is on Hill 314.
E In the early morning hours of 12 September, two companies of the 14th
NK ncgt and the 2d Un of the 2d NK negt further augment enemy forces
on lIill r..lass 314.

F
)0 F
F
F
F
F
8th Cay Hegt-47
At 0645, a flight of 3 F4U's hits Hill 307 on 8th Cav left flank with
napalm and fragmentation bombs and rockets.
Enemy mortar fire falls on lIill 314.
From 0830 to 1140, twenty or more aircraft hit Bill 570.
From 1045 to 1125 a flight of Fao's strikes at two tanks at Tabu-dong
and claims one destroyed.
At 0900, enemy artillery on the road at the northwest end of HilI r\lass
570 shells flOK forces to the east.
Friendly tanks, artillery, and infantry fire on and disperse enemy troops
on west slopes of Hilll\Jass 570.
Enemy mortar fire on !lill373.
From 1405 to 1435 a flight of FSO's strikes at six to eight guns south
of Tabu-dong at 1152.01 -1458.50.
At 1540 a very effective air strike is placed on lIill 314.
At 1640 another air strike hits Hill 314.
At 1710 a flight of F'51's drops two 500-lh bombs on Walled City.
From 1820 to 1915 a flight of F4U's drops napalm and fragmentation
bombs on artillery near Tabu-dong.
From 1920 to 1935 an air strike hits the southwest slope of the Walled
City.
C
.....
0>
<X>
c::
2:
("""")
r

(/)
(/)
..."
I"T'1
C
I
, 7
TH
REGT
aTHREGT
TH ' ,31.::::::..J\ '
9 REOT ' \ \
\\ \\ \\
\ I i""C'" J "
.. \ I V
I I, I I
\
\ I I
\ I I

ii . >
N
I
o
__
I <1000
6000 yd
I
c::
U
:z:
("""")
r

. (/)
(/)
..."
I"T'1
Reserves at Taegu: 8th Engr Bn
1st Bn, 7th Cay Regt
Operations planned: "Sth Cavalry (less 1st Bn) to revert to Division
Reserve in the vicinity of 1153-1442, 1st Battalion to remain in present
position under control and logistical support of 7th Cavalry when objective
one is secured."
Operations ordered: 8th Cavalry: "1st Battalion and 2d Battalion to remain
in present positions, dig in and hold. The 3d Battalion, relieved by elements
of the 7th Cavalry [3d Bn, 7th Cay]. to be placed in assembly area in Regi
mental Heserve!'
c:
z F At 1110. the 3d Bo, 7th Cay crosses the LD in its major attack on BiB
CJ Mass 314. FOn A DETAILED ACCOUNT OF TillS ACTION, SEE
r
TilE NARRATIVE OF TilE KEY ACTION, LATER IN TillS
J>
C/)
F During the morning, the 2d Rn, 7th Cay moves up to relieve fiOK units
C/)
...,
on and near l!ill 660.
fT1
o
F The 3d 13n, 7th Cav assaults objective
time and is successful.
on Hill I\lass 314 for the
E Hemaining enemy
northwest.
troops on lIill ;"!nss 314 flee to the northeast and
Weather observation: No preCIpItation. Visibility unrestricted, except in
early-morning haze. Locally, at 1005. 10/10 cloud coveT; at 1040, 9/10
cloud cover at 2500 ft.
Gro.und-support plane strikes: 1st Cav Div-60
E
E
E
F
E
E
F
F
F
F
F
E
8th -7th Cav Regts-27
Beginning at 0420. enemy artillery walks up the highway from the
regimentai CP at Chiigok to the ironto hits Co G.
At 0607 mortar fire hits Hill 373.
At 0715 mortar fire hits Hill 225, causes many casualties in Co B.
At 0810 an air strike hits artillery pieces two miles south of Tabu-dong.
An SP gun fires down the highway.
At 0855 mortar fire falls at the southern base of Hill 314.
Around 0930 a flight of two F51's drops napalm on the hill east of
Tabu-dong.
From 0930 to llOO, preparatory air strikes hit the northern part of Hill
tl.lass 314, but there is no artillery preparation.
Extremely intense mortar fire and some SP fire fall on Hill Mass 314.
On two occasions between assaults, the 77th FA Bn fires concentra
tions on the objective on Hill Mass 314 and also fires to the rear and
on Hill 570.
The FA Bn fires on enemy mortars but fails to silence them.
During the early afternoon, two air strikes are delivered just north of
Hill r-,lass 314.
At 1430 an unusually accurate and effective air strike is delivered on
the northern part of Hill Mass 314 just ahead of the pinned-down
friendly assault troops.
At 1640, artillery fire hits the 99th FA Bil.
.....
..,
o
c::
z
C"?
r
:x:
U)
U)
.,.,
rn
o
I
-
4000 6000 yd
I
I
Reserves at Taegu: nth F:ngr Bn Weather observation: No precipitation. Visibility 8 to 10 miles. At 1102.
@
1st Bn, 7th Cay Hill .570 clear, Lut at 1::J35 JOe reports weather nonoperational. At 1840
o
3d Bn, 5th Infantry Hcgt (after 1300) winds of 25 knots with gusts to 40 knots reported in Taegu area.
I
>-l
I
Operations planned: Bth Cavalry; "To move two platoons of E Company Ground-support plane strikes: 1st Cav Div-40
forward on Hill 373 with mission of advancing their positions." 3th -7th Cav Regts-2
~
... 7th Cavalry: Apparently none-2d and 3d BDs to remain in newly
occupied positions.
Operations ordered: 8th Cavalry: "1st Battalion and 2d Battalion to remain
in positions, dig in and hold,3d Battalion to keep <?D as Regimental Reserve
with no assigned mission at end of this period of report."
7th Cavalry: Apparently no new orders.
F At 0940 a flight of two F51's hits IJill'570 with napalm and rockets.
F At 1715. friendly patrols on Hill 373 eliminate a small enemy outpost
and contact main enemy force.
E Two enemy companies concentrate on Hilll\Iass 570.
~ F
A battalion-vaHey fired by the 99th FA Bn disperses enemy concentra
tion on Bill Mass 570.
F During the night of 13-14 September, aggressive patrols from the 3d Bn,
7th Cav move out to Hill 570 and to the north.
-
'2 ~ ' - - - , > ~ .
14 SEPTEMBER .' ~ ~
N
1
o 2000 4000 6000 yd
I I I
c:::
:z
C'")
r
:x>
C/)
C/)
...,.,
IT!
o
I
-
-
@
o
I
...,
I
0'>
""
......
Reserves at Taegu: Company n, Bth Engr Bn Weather observation: Intermittent rain. Visibility unlimited, except in rain,

1st Bn, 7th Cay Ilcgt when reduced to 3 to 4 miles.
o .1d Dn, 5th Inf !legt
I Ground-support plane strikes: 1st Cay Div-,l8
>-l Operations planned: 8th Cavalry: (a) 1st Battalion to remain in present 8th -7th Cay Regts-IS
I
positions and maintain contact with the enemy by aggressive patrolling or
'" ,.... by fire. (b) 2d Batlalion to advance under covering support of our air and
'"
artillery to Hill 570, capture and hold the hill. (c) 3d Battalion to remain in
Hegimental Heserve and to he prepared to counterattack any sector of the
Hegimental zone, with emphasis placed on the left flank."
7th Cavalry: Apparently aggressive patrolling; 1st Bn to move up.
Operations ordered: 8th Cavalry: "Same as 'Operations planned.'''
7th Cavalry: i'.pparently nene.
E From 0100 to 0300 unintense harassing fire is delivered on Hill
l'I.Jass 314.
F At 1420 there is an air strike on Hill 570 with napalm, rockets and
strafing; a second strike at 1430.
F Companies E and G and the 16th Recon Co jump off in attack, drive, I
enemy back to 570-m peak. ...........
-
E Enemy troops on Hill l'Ilass 570 withdraw north of 570-m peak; many flee/ F The attack is supported by the fires of the 9th and 99th FA Hns.
from the hill mass.
The 3d Rn, 7th Cay fires with automatic weapons and recoilless rifles
from Hill 314 on enemy troops fleeing from lIill
F At 1500 an air strike hits the \lalled City area.
F At 1630 an air strike hits the Tabu-dong area.
F Battery C, 17th FA Bn fires first mission on Tabu-dong area.
F A patrol from Co I, 8th Cay takes a prisoner of the 3d Bn, 23d NK Regt
on Hill 401.
c
z
(""")
r
:J>
(.I)
(.I)
"
rn
o
....
...,
""
c::
z
a
r

(/)
(/)
""T'1
I'T1
0
@
o
I
.,
I
""
....
'"
)

N
j
o
I
2000 4000
!
6000 yd .
--.J
I
-
Reserves at 1'aegu: 1st Bn, 7th Cay Begt
~ 3d Bn, 5th InI Begt
o
2d Bn, 5th luf Regt (after 1300)
I
>-l
'"
I
Operations planned: 8th Cavalry: "A Company, 715t Tank Battalion [minus
tanks, two platoonsl and A Company. 8th Cavalry to assist in securing Hill
..... 3.50 and lIi1l312. 3d Dattalion will secure Hills 350 and 312. A Company,
'"
7th Cavalry [attached to 8th Cavalry] will Secure gap in main avenue of
approach between B Company, 8th Cavalry and Headquarters Company, 2d
Battalion...
7th Cavalry: Apparently 1st Bo to move up to positions across Taegu
road, 3d Bo to remain on Ilill r-,-Iass 314 and 2d fin to be recommitted to
west of Taegu road.
Operations ordered: 8th Cavalry: "'Same as 'Operations planned.'"
7th Cavalry: If any. not known.
E After midnight 14-15 September, an ene1ny force of 10 to 20 men sets
up a block across the Taegu road, at 0420 attacks the 2d Bn, is driven
off at 0440.
F A combat patrol of the 3d Bn, 7th Cay draws fire and sustains casualties
in a move to the northeast.
Weather observation: No precipitation. V i ~ i b i l i t y 15 miles, except for
early-morning haze in valleys, when 3 to 5 miles.
Ground-support plane strikes: 1st Cay Div-59
8th -7th Cay negts-12
At 1400 enemy mortar and SP fire falls on Hill 570.
Friendly counterbattery fire is ineffective.
c:
:z
C")
r
:J>
en
(,/)
.,.,
rn
Cl
c::
z
C")
. ~
C/)
C/)
.."
fT1
o
\ I
"
... )
I
N
j
,
6000 yd
4000 -L-_-----1 2000 --L __J __
~ - ' - -
Reserves at Taegu: 1st Rn, 5th Cay Regt
~
1st Bn, 5th InC Regt
o
2d Bn, 5th Inf Regt
t
>-3 3d Bn, 5th Iof Regt
t
'"
Operations planned: 8th Cavalry: "1st and 2d Battalions to remain in posi
'"
..... tion and maintain contact by aggressive patrolling or by fire, 3d Battalion
to secure Hill 312 and Hill 350:'
7tti Cavalry: Not determined.
Operations ordered: 8th Cavalry: "Same as 'Operations planned'"
7th Cavalry: Apparently. the 2d Bn to be relieved by HOK forces on
Hill 660 and to displace to west, 1st An to move up to new positions
astride Taegn road, 3d Bo to remain on Hilll\'lass 314.
E A platoon on a "'suicide" mission attacks ROK forces on Hill 660.
I
F During the morning, a platoon from the 8th Bn, 8th Cay advances on
Bill Mass 570.
E The enemy attacks Hill Mass 570 from the northeast.
~
F The friendly platoon on Ilill Mass 570 withdraws to its former positions.
F The 2d Bn,8th Cay withdraws south of the 57Q-m peak on II ill Mass 570.
c:::
Z
(J
r
;t:o
(.1)
(/)
"T1
rt"1
0
.....
..,
..,
Weather observation: Intermittent light rain. Visibility 3 to 6 miles in rain
and haze. Locally, at 1020, ceiling less than 500 f1.
Ground-support plane strikes: 1st Cay Div-3S
Bth-7th Cay R e g t ~ O
~ F
Enemy attack on Hill Mass 570 is dispersed by 100 rd of
Friendly artillery cannot reach site.
c:::
z
(J
r
;t:o
U)
(/)
"T1
rt"1
o
S5ERET
UNCLASSIFIED
INTERPRETATION: THE BATTLE NORTH
OF TAEGU IN SEPTEMBER 1950
This material which follows consists of an analysis of the general aspects
of the battle north of Taegu, viewed on the time-space scale of the preceding
sections of this memorandum, and a summary discussion outlining the operation
of the essential factors considered to have been involved. More detailed aspects
in the case of the key action are taken up in the same way in Part IV.
GENERAL SITUATION AND THE LOCAL SETTING
The small size of the total forces on each side in proportion to the amount
of territory contested determined in large part the character of the campaign.
On the UN side the perimeter defense was linear, thin, and not quite continuous.
Very little strength was held in reserve at divisional and army levels, yet unit
frontages were unusually wide. Through August 1950, the well-known "fire
brigade" method was necessary to meet the enemy thrusts which came now in
one sector and now in another. Improvised task forces and entire regimental
combat teams would be dispatched from currently stable sectors to counter
penetrations outside the areas of their parent units, sometimes far distant.
Though the defense in September was characterized by stable divisional zones,
these emergency methods perSisted.'
The opening attacks of the new general offensive launched by the enemy
at the beginning of September 1950 seem to have been phased to embarrass
the Eighth Army command in shifting unengaged units to threatened pOints.
The offensive began on the night of 31 August - 1 September with massive
drives against the southern and central portions of the Naktong front, but the
thrust toward Taegu from the north was delayed until the evening of 2 Septem
ber. Further east along the northern front; the growing resistance to continu
ing limited friendly attacks only became apparent as a developing offensive on
The 1st Cay Ojv was in a somewhat exceptional situation during the period under study in that none of
its units were detached for special missions. All the battalions were present and, with the division's
organic reconnaissance, engineer. and other companies operating as rifle units, continuously held a contin
uous segment of the front. The 2.3J Infantry's 3d Bn, the 27th Brit Brig, and the 5th ncr, all of which were
attached at one time or another, took up positions on the division's left front. The close {ommlion of 1st
Cay Div units ended on 22 Septemher when the 3d Un, 7th Cav and attachments {Task Force Lynch} broke
out of the perimeter at Tabu.dong to link up several days later and some 170 miles to the northwest with a
force of the 7th Uiv m o v i n ~ down from Seoul.
178 ORO-T-261
UNCLASSIFIED
s&nH
UNCLASSIFIED
KERn
the morning of 3 September or later. EUSAK could not afford to shift units al
ready in the line, and so prepared to commit its few reserves. It was not done
precipitously. Succeeding events demonstrated the wisdom of the delay made
to weigh the danger of existing penetrations against possible emergencies,
for some of the considered reinforcing moves that were made proved to be
imperative.
As time passed, the defense lost resilience, while still under stress.
Reinforcements, including the 3d US Inf Div, were to come from Japan, but
their arrival was not imminent. The theater commander evidently deemed
an amphibious invasion of the enemy rear necessary to relieve pressure on
the perimeter and so facilitate a breakout. For this operation some of the
force currently withstanding that pressure was to be withdrawn. As soon as
it could be released from the line, the 1st Mar Brig was to move back to Pusan,
embark, and lead the 7th US Inf Div in the Inchon landing. By 6 September the
enemy offensive had been halted everywhere along the perimeter, and the Ma
rines were being relieved. Activity, though still general, was gradually slack
ening. United States casualties were still beiog sustained at over half the peak
rate of 2 September.- No forces were left to eliminate the menacing gaps and
pockets which persisted, as all army reserves had been committed. The situa
tion of the 2d Div, recently desperately engaged, was typical. Its commanding
general was informed that after the return of the 3d Bn of its 23d Regt from
the zone of the 1st Cav Div no further relief would be forthcoming, and he
would have to get along with the forces on hand. Casualties in the 2d Div were
still occurring at over half the previous very high peak rate.
The over-all defense had become by the second week of September simply
the sum of a series of separate divisional battles. For brief periods, the fight
ing fragmented further to quasi-independent actions of regiments or battalions.
The burden of tactical direction undoubtely rested heavily on division command
ers. Uncertain that reinforcements would be available, and perhaps operating
minus certain normal elements of their command, they would naturally tend
toward caution io employiog local reserves. Until 13 September there was no
corps echelon to coordinate operations between diviSions, and control within
divisions was greatly impeded by the frequent interruption of electrical com
munication. Conservative maneuver was to be expected.
The enemy was even more fully committed. There were no strategic re
serves below the 38th Parallel, and tactical reserves were very short. Except
for the security forces kept in the Seoul area and those divisions which had
virtually been destroyed, all divisions were in the lioe. DiVisions were often
forced to use all three rifle regiments in attack abreast. As wheeled transport
See Ref 25. Cumulative friendly battle casualties and daily casualty fates are plotted for each US d i v i ~
sion during September 1950. In the 25th Div, the peak rate occurred on 3 September, after which the rate
fell off rapidly to the lowest and longest sustained range. In the 2d Div, decline was gradual from the peak
of 1 September to the minimum of 11 September. The peak of the 1st Cav Div occurred on 3 September, its
minimum on 7 September, with a sharp upswing thereafter while the division was still OIl the defensive.
Data are not given for the 24th Div because of its split commitment. For all US Army combat troops in
Korea, the rate d.eclined gradually and remarkably smoothly fmm a high value at the beginning of the month
to a minimum around 11 September. It had already begun to increase again when the general US offensive
beginning on 16 September accelerated the rise. These variations demonstrate the shift in enemy pressure
to the north and east, and suggest that a new offensive phase was starting when the UN went over to the
offensive.
ORO-T-261
179
SfERn
UNCLASSIFIED
UNCLASSIFIED
nER"
was lacking at the regimental level, large units could be shifted laterally only
by suspending the offensive in a wide area. By September 1950 the enemy had
spent almost a critical portion of his original capital of trained manpower.
Tapping the hastily organized replacement pools of untrained or poorly trained
impressed troops enabled him to fight on, but it further impeded lateral shift
ing of large units. To feed green replacements successfully into the fighting
line, regiments and even battalions had to be kept in place while individual de
pleted subunits were progressively pulled back, padded out, and recommitted.
Until combatworthy divisions could be supplied by the mobilization and train
ing program in North Korea or by foreign intervention, time was working a
gainst the enemy. .
Capture of territory was not a measure of enemy progress so long as the
friendly line remained unbroken and a few key points such as Taegu stayed be
yond his grasp. Advance was of increasing but only potential value to the extent
that these ends were approached. Enemy progress had to satisfy a series of
deadlines, and it was made at the cost of progressive decline of combat effec
tiveness. Shortening the line would increase the density of the friendly defense,
enemy line would not thicken in proportion because of the higher casualty rate
and the continuing loss of irreplaceable trained manpower. Success would not
come from the attrition of balanced opposing forces, but from breaking the
pattern-such apparently was the aim of the phased thrusts toward Pusan, Taegu,
and Yongchon. In the case of Taegu, we know for certain that the concept was
to stab deep, break through, "rush on," and advance 15 to 20 km beyond the city,
before stopping to organize its defense.
At the end of August 1950 Taegu was still a key enemy objective. A pillar
of the over-all UN perimeter (see Fig. 1), it had among the other assets of a
large city the only major airport besides Pusan remaining in friendly hands
in Korea. Though loss of the city would not necessarily have meant disaster
to friendly forces, the consequent abandonment of the perimeter to a far shorter,
if tighter, line would have greatly increased the difficulty of breakout and post
poned the planned strategic counterattack at Inchon. *
Observed enemy activity and reports from agents and prisoners of war
that enemy units were being reinforced led the EUSAK G-2 to estimate that a
renewed drive on 'Taeg., was more likely to come from the north than again
from the northwest along the Kumchon-Waegwan axis, where in August the
enemy had been beaten off in heavy fighting. Enemy strength was apparently
overestimated and its center of gravity not appreciated. Periodic Intelligence
Report Number 42 of the 1st Cav Div (27 Aug 50?) identified the 2d, 10th, and
15th NK Rifle Divisions as on the 1st Cav Div's front, as well as the 1st, 3d,
and 13th NK Divisions, which are now known to have been still in contact there.
In retrospect, it appears that the shift of enemy forces to the north and east
during the lull at the end of August was realized, but the new alignment of the
3d, 13th, and 1st NK Divs across the Tabu-dong - Taegu axis had not yet been
determined. Thus (moving clockwise on the map) the 3d NK Div was still thought
If the seaborne operation had been abandoned, friendly forces could have achieved a fonnidable concen
tration around Pusan, if, of comse, the line divisions withdrew in good order. The Marines would still have
been on hand in a reserve position, and the 7th InC Div could, of course, have been landed at Pusan instead
of Inchon. Much later, at Hamhung, these two units together demonstrated their ability to hold while with
drawing before superior forces. and to finally successfully evacuate a port.
180 ORO-T-261
UNCLASSIFIED
SiERET
SEEMT
UNCLASSIFIED
to be opposite both the thin line of the 5th Cav Regi along the east bank of the
Naktong River and the 7th Cav Regi, the 13th NK Div was also placed opposite
the 7th Cav, and the 1st NK Div was carried as in contact with the 8th Cav Regi.
It now appears definitely established that the 3d NK Div was confronting only
the 7th Cav, the 13th NK Div was opposite only the 8th Cav, and the 1st NK Div
was facing only the 11th and 12th ROK Regis. Very few enemy troops seem to
have remained opposite the 5th Cav. It was subjected neither to the expected
frontal drive nor to any pressure at all until 4 September, when enemy elements
from the north Slipped by the 7th Cav defenses to attack 5th Cav positions on
the northeast slope of Hill 303 (1140.5-1456.0). The enemy attacked here again
on 5 September, but then the action died down, and the 5th Cav did not become
intensely engaged until it relieved the 7th Cav and absorbed the full force of
the 3d NK Div continuing drive from the north. Furthermore, analysis of the
effects of the massive carpet bombing by B29's on 3 September in the vicinity
of 1127 -1444 (west of the Naktong at the latitude of Taegu) indicated that the
presumed concentration of enemy reServes had not been in the target area.
With two of its regiments, the 7th and 8th, squarely lined up against the
two enemy diviSions in its sector, the 1st Cav Div was fairly well situated to
meet a general attack. Having a combined front of about 16,000 yd, the two
regiments were, however, thinly spread, and the third regiment, the 5th Cav,
was of course not free to back them up. The 8th Cav front of 10,000 yd at the
center of enemy concentration may be compared with a normal defensive regi
mental front of 2400 to 4800 yd (two battalions In the line).
Terrain exerted a controlling influence on the development of the battle
north of Taegu. First, it canalized any SUdden mass movement-the basic
condition for breaking the tactical pattern-in the narrow valleys. Whoever
possessed the adjacent high ground obviously could cover the valleys with the
fires of organic infantry weapons. Perhaps more important, the high ground
afforded the forward observation points necessary for calling in supporting
fires and directing suitable tactical maneuvers. To an enemy lacking an air
arm, adequate observation from the ground was indispensable. It was fre
quently also a necessity for the friendly side, when bad weather prevented or
hampered observation from the air. The virtual ineffectiveness of US sound
flash-radar batteries has already been noted, and the character of the terrain
was apparently in large part responsible.
As emphasized in the base map of Part ill, the east-west ridges on both
sides of Tabu-dong dominate the territory to the south. Their capture was a
required preliminary for any enemy mass movement in this direction. As the
diagram shows, the second line of lesser ridges, consisting mainly of Hill Mass
570, Hill Mass 314, and Hill 660, is broken by five north-south valleys, whereas
the major ridge is cut only by the corridor at Tabu-dong. The second line was
virtually the last ground north of Taegu on which a coherent defense with the
existing forces could be based, for behind it to the south the corridors widen
and the hills flatten out. To block the valleys, friendly forces had to hold the
last high peaks; to make a massive penetration to Taegu, the enemy had first
to take them. The stable pattern of the pOSitional fighting that evolved here
was the result of these facts: Hill Mass 570, Hill Mass 314, and Hill 660 are
side-by-side; they are all over 500 m in elevation; the first two have several
peaks affording defensible positions. Fought peak by peak by fully committed
adversaries, the battle here became like a vast game of chess.
ORO-T-261
181
UNCLASSiFiED
SEERn
UNCLASSIFIED
SUREr
COMPOSITION OF THE OPPOSING FORCES
The opposing forces newly confronting each other in the general sector
north and west of Taegu at the end of August 1950 were approximately equal
in numbers. The 1st Cav Div had a strength of 15,867 (including 736 attached
ROK troops) and the attached battalion of South Korean police increased its
strength by about 650. It was opposed by the 3d and 13th NK Divs, which had
a combined strength of about 16,000.* In particular, the 7th and 8th Cavalry
Regts, with a combined strength of about 5500, faced enemy rifle regiments
totaling about 5000. As noted above, there were few if any uncommitted
friendly units in Korea, but reinforcements were on the way. Though the
enemy had no strategic reserves, he did have large organized pools of raw
manpower close to the front.
The rather full data on the 13th NK Div from the time it was first com
mitted to action show that prior to contact with the 8th Cav Regt it had re
mained on the offensive until reduced to approximately 40 percent of TO
strength. In the latter part of August the first infusion of replacements was
received, but when the offensive against the 8th Cav was launched, on 2 Sep
tember, the division's strength could not have been much more than 50 percent
of the TO figure. However, the division had a pool of about 4000 recently Im
pressed and almost untrained men, most of them South Koreans. This reserve
was drawn on in the lull after the offensive bogged down early in September.
Data as complete as this are lacking for the 1st and 3d NK Divs, but frag
mentary information supplied by prisoners of war on unit strengths and on the
circumstances under which they joined their units indicates that the 13th Div
manpower situation was typical.
The US manpower situation is well known. Suddenly committed to battle
from occupation duty in Japan, units were understrength and under equipped.
By the standards of a general war, combat efficiency was probably still not
good at the beginning of this engagement, though as a result of the experience
of the intense battles in August it was much improved over the original level
of training. United States forces, like those of the enemy, included green
Koreans. The ROK troops, integrated at squad level in the 7th and 8th Cav
Regts, had recently been conscripted, and had had only a few days of training.
They made up only 5 percent of the total strength of the 1st Cav Div at the
beginning of September, but increased to 18 percent toward the end of the
month. In the forward units the percentages were higher. On the critical
date of 11 September, integrated ROK troops constituted 25 percent of the
total combined strengths of the nine rifle companies of the 8th Cav Regt and
the attached 16th Recon Co. The battalions of South Korean police attached
as infantry units were at an intermediate level of training and combat
experience.
In proportion to Its strength, the 13th NK Div at the beginning of Septem
ber had adequate supporting artillery, a little armor, and adequate stocks of
ammunition. In these respects it was much better off than the 1st and 3d NK
Divs. Of course, the enemy lack of air support and air observation must not
be forgotten. Supporting artillery remained at a generally adequate level
throughout the month. Ammunition shortage, however, became critical after
the middle of the month.
-Assuming that an additional regiment. elements.of the "15th" NK Div with a strength of 1000, was
attached to the 3d Div (see Part JI).
182 ORO-T-261
UNCLASSIFIED
SEun
SERrT
UNCLASSIFIED
In comparison with the enemy, friendly forces were well supported by
artillery, armOI", and aircraft, though the volume of fire was low by US stand
ards. Perhaps the best index of the amount of support is the average expendi
ture during the first two weeks of September of aboutthree thousand rd pel"
day by all artillery of the 1st Cav Div. The 99th FA Bn, in direct support of
the 8th Cav Regt, contributed approximately five hundred 105-mm rd to this
daily figure. Such limited expenditures reflect ammwlition shortages which
temporarily became serious at about the time that the enemy stocks of am
munition began to run out.
COURSE OF THE BATTLE
Friendly forces were explicitly informed of the location and intentions
of the 13th NK Div when the disaffected S-3 of its 19th Regt gave himself up
to the 8th Cav at 2040 on 1 September, and warned that the 13th Div was to
attack at dusk the following day. His estimates of the strengths of the several
regiments were misleading, in that they appal"ently inc luded large numbers of
green conscripted troops, received by the division and on paper assigned to
the regiments, but not yet equipped and assimilated. At the time, there was
of course no compelling reason for believing anything the prisoner said, ex
cept that some of the weapons positions he pinpOinted on the map had already
been observed. As confirmed by enemy documents captured 3 weeks later,
the objective assigned the 13th Divby the commanding general of the 2d NK
Corps was to "rush on six kilometers" toward Taegu in a first drive rather
than to capture specified terrain. Before the offensive began, there was no
particular information on the plans of the nanking divisions. In retrospect
it is seen that the 1st NK Div was to attack simultaneously, and though defi
nite information is lacking on the 3d NK Div, a similar mission can be inferred.
Since the 7th Cav diversionary attack was ordered because of the grave
situation in the zone of the 2d Div, the EUSAK command had evidently assumed
that the enemy either had strategic reserves near the front or was capable of
rapidly shifting divisions already in the line, It is now known that neither was
the case, and that the effect of the attack against the 3d NK Div on 2 September
could only have been local. The immediate resistance and the prompt general
counterattack on the night of 2 - 3 September suggest that in this sector too
the enemy was prepared to drive ahead. Though the 7th Cav did not succeed
in capturing Hills 518 and 490, and though it never got as far as Hill 346, it
would seem that the attack had significant delaying value and even' ultimate
preventive effect. Instead of moving out to engage the 7th Cav in the vicinity
of Hills 464 and 620, the last rugged ground before the corridors open up to
ward Taegu, the 3d NK Div was drawn into battle 3000 to 4000 yd beyond this
inner ring of ridges and worn down before reaching it.* Also, the fact that
the 3d Bn, 7th Cav saw its battle-hardening first action at Hill 518 must be
credited.
tAn undated enemy document taken from a dead soldier at 1141.3-1459.8 (southern foot of Hill 518) by Co
G, 7th Cav on 3 September states that in an unspecified unit only 126 of the original 609 men were left; 112
had been killed, 268 wounded, and 103 were missing. If these and the accompanying figures on weapons
were compiled in the previous lull, they would indicate that this unit of the 3d NK Div, with only 21 percent
of its original personnel remaining, was in the same situation as 13th Div units, and must have received
replacements before entering the further fighting.
ORO-T-261 183
SEE RET
UNCLASSIFIED
UNCLASSIFIED
SUitT
If the diversionary attack forestalled a penetration between the 7th and
8th Cav Regts, it was probably at the expense of the 8th Cav ability to counter
the penetration of its front and the advance along its right flank. It is not clear
that the shift of the two rifle companies and supporting elements of the 8th Cav
(especially the heavy mortars) to support the attacking 7th Cav was essential,
though covering the 7th Cav right flank evidently seemed necessary. The de
tached units do not appear to have been significantly involved during the day
or two they spent in the new sector. Their departure certainly embarrassed
the 8th Cav Regt, which almost immediately afterward was attacked itself.
The 8th Cav Regt organic heavy-mortar support had been reduced to one-third.
Potential support at a vital point was also lost just before the enemy struck
when the additional platoon of heavy tanks that had been sent to back up the
7th Cav from the 8th Cav left flank fortuitously withdrew. Furthermore, all
the (considerable number of) air strikes within the zone of the 1st Cav Div
during the day were delivered in support of the 7th Cav. It may be concluded
that the diversionary attack amounted to a concentration of friendly force on
the left for a limited, though major, effort at a time when. the enemy had con
centrated his resources (especially artillery and armor) for an all-out effort
on the right.
It is a striking fact that battle casualties in the 7th and 8th Cav Regts
accumulated at almost the same uniform rate during the first week of Septem
ber, despite the initially differing missions and situations of the two regiments."
A glance at the diagrams which precede this section will show that the
critical penetration which led to the progressive deterioration of the over-all
situation of the 1st Cav Div was made by the 1st NK Div through the 1st ROK
Div. The line was breached only after ROK forces had repulsed frontal attacks
at two points; they were then pushed back by a new attack on one of them. The
gap evidently widened because neither the ROK Army nor the 8th Cav Regt had
forces available to bring to bear. Whether or not the attack plan of the 2d NK
Corps had included a feint, US strength was successfullY, and (viewed narrowly)
needlessly, tied up. The enemy command apparently gambled that the 8th Cav
would not counterattack the 13th NK Div and perhaps stationed only the very
weak 21st Regt across the front of the 8th Cav 1st Bn. After all the enemy
probing in this area, no attack ever came. Though it now appears that the 1st
Bn was well situated to make a telling counterattack to the east, there was of
course little current information on the actual situation, and events to the left
and right must have appeared highly disturbing. Here again it can be argued
that had the detachments of 2 September not been made, not only might the gaps
have been more adequately plugged but a counterattack might have been feasible
to restore the friendly line at Tabu-dong in time for an effective defense of the
Walled City .
See Her 25. ~ v e n though each of the battalions of the 7th eav was for a brief time in division reserve
during the period under study, while those of the 8th Cav never were, casualties kept pace in the two regi
ments. Casualties in the 5th eav Regt built up in a different manner, but the cumulative totals for September
on 19 September were almost identical for all three regiments. The regularity suggests a response of the
divisional command, whether fully intentional or not, to a principle of equalized losses. At battalion level
and lower, no such regularity is found. See also Part II with regard to this point.
184
ORO-T-261
.HUET
UNCLASSIFIED
UNCLASSIFIED
SUitEr
The 13th Div penetration to Tabu-dong cannot be called successful when
measured against the enemy intentions. Neither was it very deep; but it did
unhinge the friendly defense. If the 8th Cav front had not been split, the enemy
advance along the right flank would hardly have gone uncontested until the be
lated battle at the Walled City. To the enemy, the cost of driving the wedge to
Tabu-dong was temporary loss of the offensive capability of the 13th Div. Yet,
as indicated by the analysis of friendly casualties in Part II, its casualties were
not much more than double the number it inflicted on the 8th Cav. This perhaps
surprisingly low ratio becomes more credible when it is noted that the Initial
enemy attacks and main gains were made at night, when close combat would be
expected to increase in importance and the effectiveness of supporting fires
to lessen. Also, the enemy had the formidable support of artillery, SP guns,
and heavy mortars, at a time when the 8th Cav had only the support of its bat
talion of 105-mm howitzers, shared the support of an understrength battalion
of 155-mm howitzers, and during a critical period was backed up by only one
third or none at all of its heavy mortars. The casualties that the regiment
was able to inflict on the 13th NK Div were crippling simply because the di
vision's units were already at such a low strength level.
Measures were promptly taken to stem the enemy advance toward Tabu
dong. Tanks as yet uncommitted were sent up to the threatened area with the
regimental reserve as soon as daylight permitted. The companies of the com
bat engineer battalion constituting the main division'al reserve were committed
almost as quickly. With no army reserves likely to reach the area for some
time, plans were rapidly formulated to organize a new divisional reserve from
noncombat troops. These efficient moves successfully countered the known
main threat.
The inefficiency involved in the initial reaction to the enemy offensive and
throughout succeeding operations apparently was due to lack of comprehensive,
detailed, up-to-date tactical G-2 and G-3 information. Why, for example, was
there such a long delay before occupying Hill Mass 570 with even a token force?
A very important contributing factor was failure of electrical communication.
Though this subject has not been analyzed systematically, it is clear from the
contents of the message logs that it was usual for some or many of the main
links to be out. The divisional and regimental commands often lost all contact
with one or several forward battalions, and a battalion often had no information
on what the situation was in the flanking battalions. Information was least avail
able when most needed-during moves and from units intensely engaged-and
reports frequently had to be relayed via a few overloaded roundabout pathways.
The enemy shifted tactics after frontal attack, at cost, had achieved the
limited advances beyond Tabu-dong and the Walled City. By around 5 September
the 13th NK Div was too depleted to mount new frontal attacks for some time.
The 1st NK Div was still capable of massive efforts, but expended them against
the ROK Army (see the prisoner's account of the attack of 10 September in Item
8, App A). To exploit the penetrations of the 8th Cav line, small groups from
the 13th and 1st Divs attempted t.o infiltrate. Their attempts to put artillery
batteries out of action failed, and no fire block across the MSR was maintained
for long. The one determined attempt actually to cut off the 8th Cav battalions
from their supporting units was made by 1st Div troops on,6 September from
ORO-T-261 185
SHR!Y
UNCLASSIFIED
UNCLASSIFIED
their salient on Hill 570. In defeating this and previous blocks, friendly armor
seems to have been invaluable. The enemy did continue to succeed in blowing
culverts and mining bypasses, but as these roadblocks were not covered by fire
they were quickly eliminated by US engineers.
In contrast to the enemy's original intention to "rush on" toward Taegu,
the new phase of the offensive was based on taking high ground. The advantage
of a hill salient must have become apparent when the penetration down the ridge
of Hill Mass 570 enabled harassing the friendly rear at a time when enemy
strength was insufficient to secure the entire hiU mass or fan out across the
friendly MSR. That enemy doctrine by this time was to advance on a hill-by
hill basis is clearly shown by the 3d NK Div attack order of 11 September (App
B). All the specified.objectives are hills, even though the friendly line (here
that of the 5th Cav, on the 8th Cav left flank) was more or less correctly ap
praised as " on a scattered hill basis. , an independent outpost type of de
fense." . It is also of note that all three of the 3d Div rifle regiments were to
attack abreast in taking the assigned hill objectives, i.e., the hills were not to
be taken to cover a following thrust down a valley by a reserve regiment.
Priority in occupying a hill seems to have greatly increased the proba
bility of holding it. Advance enemy elements were first in moving into the
Walled City and onto the northern portions of Hill Masses 570 and 314, and
these points were held in bitter battles, though the enemy was eventually out
flanked or driven off. Conversely, small friendly groups were first on the
southern portions of Hill Mass 570 (that is, Hill 373), Hill Mass 314, and on
Hill 660, and the enemy was never able to push them off. It would seem that
the advantage of first occupation resulted not only from the obvious defensive
value of high ground but also from the fact that the opposing forces were roughly
equal and thinly spread. To capture a hill required assault by at least as many
men as were defending it, and with limited total forces a greater commitment
could hardly be risked. The occupying force was similarly constrained, but
could always risk increasing his strength on any hill just enough more to offset
an increasing commitment by the opposing side to capture it.
The enemy's all-out bid to take all of Hill Mass 570 and gain access to
the Bowling Alley was clearly made on the night of 10 - 11 September and the
morning of 11 September. These attacks mainly involved close combat, and
they almost succeeded. They seem to illustrate the differing tactical signifi
cance of casualties on the two sides. The analysis of friendly casualties in
Part II shows that friendly units were sensitive to casualties. Whether or not
a given engagement would end successfully tended to depend on how many
friendly casualties had been sustained. Whatever the outcome, the integrity
and combat effectiveness of the unit (company) was preserved. Filled out with
replacements, it could continue the battle another day. Enemy units as groups
were less sensitive to casualties because of the military system and the low
concern for the individual. The private soldier was at least as fearful as one
would expect, and he often had to be herded into battle at gun point. Up to the
point that organizational discipline or coercion. held, the unit could survive a
high proportion of losses. Beyond this point, it might break completely. The
enemy units that attacked and reattacked Hill 373 took many casualties without
breaking and were evidently reorganizing on Hill 570 to attack again on the
morning of 11 September. At this crucial time, they were hit by an unusually
ORO-T-261
186
UNCLASSIFIED
SHRn
UNCLASSIFIED
HcRtr
intense round of air strikes-then, they broke. Fleeing, they were mowed down
by artillery and tank fire. The combat effectiveness of 1st Div units on Hill
Mass 570 was destroyed. The shattered units of course did not cease to exist
but, if typical, were reorganized from their pulled-back remnants, padded out
with more green replacements, and after some time sent back into the line,
now more "brittle" than ever.
If it had been intended that the 1st Bn of the 19th Reg!, 13th Div was to
move up from its supporting pOSition on Hill 570 and capture Hill 373, the at
tack was hardly likely to succeed at this time, for friendly forces on it had been
trebled. To the east the enemy was also stalemated. Attacks against the 1st
ROK Div had been repulsed. Though Hill 660, an anchor of the friendly line,
was only lightly held, it was not a promising objective for frontal assault, be
cause of its precipitous .northern slopes. The one large enemy group which had
succeeded in infiltrating to its southwesterly slope had been contained by friendly
forces and presumably dispersed. Between Hill 660 and Hill 373 Hill Mass
314 lay like a keystone, outflanking them both,if captured, leading down into
the Bowling Alley. Until this time, the enemy had merely resisted friendly
attempts to take the whole hill mass, or had reacted with limited counterat
tacks, increasing his local strength in proportion to the growing friendly com
mitment. On 11 September Hill Mass 314 seems to have been designated the
primary enemy objective. During the latter part of that day -a.lld the early hours
of 12 September, reinforcements were sent up from the 19th Reg! On the west,
and from the 2d Reg! of the 1st Div on the east. There is no evidence that an
attack was ordered for 12 September, but it seems likely, for on the morning
of 12 September the 3d NK Div was to launch attacks toward the Bowling Alley
against the 5th Cav Reg!, and enemy artilillry and mortar fire in the 8th Cav
sector reached a new pitch of activity.
In delaying .committing its striking reserve in a major counterattack until
12 September, the 1st Cav Div had again waited until the enemy was recommit
ted. Application of increased force was the price of increased assurance of
enemy capabilities. The 3d Bn, 8th Cav had been attacking on Hill Mass 314
with only one of its companies forward at any time, the third backing up, and
the ROK Tng Bn in reserve, but the 3d Bn, 7th Cav now necessarily attacked
with two assault companies abreast and the third just behind them, the same
ROK battalion backing up, and, initially, the relieved 3d Bn, 8th Cav in reserve.
The attack by the 3d Bn, 7th Cav on Hill Mass 314 was crucial. So to the
enemy was its defense. The fighting consequently reached a new intensity.
The 3d Bn was ordered to hold the gains won early in the action. To do so
necessitated remaining organized and effective while enemy mortars inflicted
casualties at a rate which might otherwise have justified withdrawal. The
enemy casualty rate was somewhat more than double the friendly rate, as in
the enemy attacks at Tabu-dong early in September. With substantially equal
forces initially present, the ultimate result was a shift in the balance of forces
in the friendly favor, yet the enemy did not break and flee until the end.
Sticking out a bad situation was the contest, for close combat was of les
sened importance after the first assault. Small units in the forward positions,
more or less isolated by small irregularities of terrain, and pinned down by
a rain of friendly supporting fire, seemingly had the option of making limited
withdrawals when fire or casualties were deemed excessive. The detailed
ORO-T-261 187
Hun
UNCLASSIFIED
UNCLASSIFIED
analysis of this action in Part IV' shows that it proceeded by the point-by-point
reaction of two forces sensitive to local pressures. Almost to the end, the
outcome was in doubt at anyone time. Stage by stage, the action tended to
progress one way or the other under the control of factors intervening almost
unpredictably. Among the determinants on the friendly side were outstanding
acts of leaderships, certain particularly effective concentrations of mortar or
artillery fire, and an unusually close and effective air strike. Factors which
might have been thought to be of great effect, such as counterbattery fires, or
preparatory or general-support air strikes, were without evident influence.
Though Hill Mass 314 was won with but a small margin of safety, its cap
ture definitely marked the transfer of the initiative to the friendly side. Though
the combat effectiveness of the 3d Bn, 7th Cav was severely reduced, that of the
enemy in this area apparently was lost, for the enemy was not able to make any
aggressive countermove for some days." The pOSitional advantage gained was
made evident on 14 September when the 8th Cav attacked the now partially out
flanked enemy units on Hill Mass 570. After offering only weak to moderate
resistance, the enemy withdrew, retaining only a toehold on the northwestern
end of the hill mass. The 13th Div, however, was in no sense beaten. Bitter
resistance to any further friendly advance was offered, with mortar and artil
lery fire, counterattacks, road blocks, and mines, up to the time the division
was virtually enveloped on 21 September.
The 13th Div command seems to have decided to hold in the 8th and 7th Cav
areas after the defeat on Hill Mass 314, and to use its few free forces in concert
with the 3d Div's continuing attempts to penetrate toward Taegu through the gap
of almost unoccupied high ground between the 5th and 8th Cav Regts. These at
tacks were checked in what must again be called a dynamic equilibrium. Here
too the initiative passed to friendly forces, which took up the offensive on 17
September. The weight of the friendly effort to break out from the perimeter
in the 1st Cav Div zone was increased by the redeployment of an additional
regimental combat team and the arrival of other supporting units. Progress
against tenacious enemy resistance was very slow, even though enemy food
and ammunition shortages were now becoming critical. The maneuver of US
and ROK forces set the stage for the final decisive actions. Their outcome
unqualified victory-must, however, be attributed essentially to the exhaustion
of enemy manpower and materiel.
ESSENTIAL FACTORS DETERMINING THE COURSE OF THE BATTLE
Until the UN forces set up a coherent linear defense, there was no reason
for the North Korean Army to abandon the tactics employed from the start of
. the war: penetrate deeply and rapidly in many places; when opposed, sideslip
and continue to advance along unblocked parallel routes. For the enemy to
realize his general intention of occupying all South Korea, it was not necessary
to overcome all islands of friendly resistance, nor to push them back; mere
isolation from the friendly rear would ensure the fall of territory. The tactics
*There may have been another reason. A prisoner from the 1st Div stated that he had heard that on 13
September the 1st Div was to send three battalions to reinforce the badly beaten 8th and 15th NK Divs.
188 ORO-T-261
UNCLASSIFIED
SURH
UNCLASSIFIED
employed by the UN forces were imperative: resist to delay the enemy as much
as possible but only up to that point at which the integrity of major units was
imperiled.
When compressed into the southeastern corner of Korea, UN forces gained
command of almost all the corridors crossing the perimeter. North of Taegu,
two major avenues leading to the south became one. Here the rapid enemy ad
vance was now predictably canalized. Also, enemy mobility had already been
seriously impaired. US air attack, in particular, had destroyed a large frac
tion of the enemy's tanks and vehicles. Few remained for continued thrusts
straight ahead, and there were not at all enough for rapid, large-scale, lateral
shifts. When the 2d NK Corps reached the mountain barrier north of Tabu-dong,
it was forced to maintain the offensive with the means at hand. Subsequent pro
gress depended on the maneuver of small infantry units.
Three basic factors shaped the battle north of Taegu: (a) the character of
the terrain; (b) the numerical equality of friendly and enemy manpower; and
,(c) the necessity for each side to commit most of its strength in order to cover
the front.
These factors exerted their influence soon after the battle was joined in
the latter part of August 1950. As the battle progressed in the next weeks, the
tactical choices open to each side became ever fewer. At the outset, for ex
ample, the enemy could have used his moderate quantity of armor to support
rather than to lead the attacks, but in September he was forced to relegate his
little remaining armor to a support role. Similarly, after the original free
choice of drives along the highway as a suitable mode of infantry attack, the
only direction for tactical change in the given terrain was to attack through
the' hills and along the ridges. On the friendly side, there was a corresponding
shrinkage in possible courses of action. Whereas on 3 September there was
the possibility of a strong tank-supported counterattack, there remained on 12
September hardly any choice'but to make the all-out attack on Hill Mass 314. '
, As noted above, the setting and the original resources of both sides cir
cumscribed the battle at the outset, and also, therefore, the way in which these
resources were expended. It may be true that the outcome at each stage was
then also in part predetermined. If this is correct, the demonstration may sug
gest some lessons to military students.
The timetable which follows is a summary outline of the battle from be
ginning to end including the action prior to the arrival of the 1st Cav Div. The
moves and countermoves at each stage are sketched, along with notes on the
outcome of each action and its effect on the battle. In the final column the
interaction of what seem to have been the essential factors affecting each out
come is indicated.
One general conclusion is implicit in the various facts determined in this
study: only' a small fraction of any' resource of either side was ever a p p ~
at the currently...IDlmificant point.
ORO-T-261 189
SECRET
UNCLASSIFIED
UNCLASSIFIED
SHI!T
THE COURSE OF AN ENTIRE BATTLE AND SOME ESSENTIAL FACTORS INVOLVED
Date Moves Counters Outcome, effect Interaction of the essential factors
18 Aug Deep enemy penetration of ROKA. Tank-led
motorized force keeps moving down valley
toward Tabu-dong during early hours of
darkness (13th NK Div).
19 - 25
Aug
Enemy infantry deploys on hills along val
ley, masses behind tanks and follows them
in night attacks down road.
2, 3
Sep
3, 4
Sep
With armor almost gone, enemy deploys
more widely across valley or ridges.
Enemy (13th and 1st NKDivs) makes co
ordinated night attacks down main cor
ridor to Tabu-dong and along adjacent
ridges and valleys.
Friendly counterattacks to reestablish
line along high ground across Tabu-dong.
5 Sep Advanced enemy elements (lst NK Div)
occupy dominating high ground (Walled
City) to right front.
5-7
Sep
Enemy infantry on salient high ground
attempts to cut friendly MSR.
8 -11
Sep
12 Sep
Enemy (1st NK Div) and friendly forces
(8th Cav) progressively committed in
attacks to take high ground controlling
corridor (Hill Masses 570 and 314).
Fire support somewhat increased on both
sides. US air concentrates on forward
targets when weather permits.
Enemy shifts primary offensive effort to
Hill Mass 314 by reinforcing units already
there with 13th Div troops.
13 - 20
Sep
Limited friendly advances (Hill Mass 570),
limited enemy counterattacks with effec
tive artillery and mortar fire on occasion
by both sides.
21 Sep Enemy still holding front enveloped by
ROKA from northeast and flanking sweep
of 7th Cav from west.
Strong friendly infantry-tank force makes
opposite advance, sets up massive block
across valley north of Tabu-dong (27th
RCT),
Static defense coordinating infantry,
armor, and artillery. (US air not sig
nificantly involved, since enemy attacks
are made at night.)
Linear defense with reduced armor and
mortar support and no front-line tank
defense (8th Cav). (US air not available
for strikes at observed build-up or to
hit positions taken by enemy.)
Enemykeeps up pressure. attacking to
take high ground.
Friendly infantry task forces attempt
to beat back enemy and hold high ground.
(Not significantly supported by armor or
artillery. Friendly air held off by bad
weather.)
US armored task forces operate along
MSR,
Each side on offensive-defensive.
Friendly local reserves increased (by
two battalions of 7th Cav and support
ing artillery).
Friendly all-out attack on Hill Mass 314.
Each side exerting pressure. Friendly
infantry strength in area reduced (7th
Cav redeployed elsewhere). Lossofclose
support artillery (105-mm) battalion par
tially compensated bY arrival of long
range artillery (8-in., 90-mm). Enemy
organizes active positional defense.
Continued enemy resistance.
Enemy armored spearhead runs up against
defense, is turned back.
Attacks turned back with high enemy troop
losses (10 times friendly?); almost aU
enemy tanks and SP guns knocked out.
Enemy (13th NK Div) temporarily loses
offensive capability.
Initial enemy drive down corridor stopped,
but enerr_y penetrates US pOSitions on high
ground, unhinging defense line and per
mitting advance down valley. Enemy losses
relatively low (2 to 3 times friendly).
Attacks fail to dislodge enemy from the
commanding high ground. With an origi
nally low strength level, the 13th NK Div
again temporarily loses offensive capa
bility because of accumulated casualties,
though casualty rate relatively low (see
above).
Seesaw fighting in which US infantry stands
off enemy ends with enemy in possession
of high ground.
Enemy road blocks eliminated.
Seesaw fighting. Side originally in posses
sion ends by holding position. Units of 1st
NK Div lose offensive capability.
Hill Mass 314 captured by US. Enemy
(1st and 13th NK Divs) again loses offen
sive capability due to casualties.
Stalemate.
Enemy defense crumbles under attack
from all sides. Section of front wide open.
US tanks spread in defense stop armored point of
attack; US artillery and infantry hit main advanc
ing body.
US artillery able to disrupt attacks because enemy
concentrates in predictable area, approaches along
predictable route. US tanks can set up ambush
across certain route of approach, decisively stop
enemy armored spearheads that achieve limited
penetration.
Enemy infantry attacking US troops in rough ter
rain at night. The multiple, diffuse attacks and
confused situation give enemy limited success
as US firepower is not brought fully to bear.
Enemy troops spread out on high ground. Con
fused, flUid situation plus bad communications
prevent friendly infantry, armor, artillery, and
air from being effectively brought to bear.
US troops not significantly supported by fire,
exhaust own supplies and ammunition. Bad
weather foils resupply.
Advanced enemy units do not have significant
support. US tanks are unopposed by enemy
armor or effective AT guns.
Infantry of either side defending high ground can
be sufficiently reinforced from limited local re
serves to hold off attack. Supporting artillery
and mortars can concentrate on geographically
fixed and observable assaulting troops.
Opposing infantry troops extremely vulnerable to
artillery and mortar fire. Outcome depends on
which holds out longest. Unusually close air
strike decides action.
Net reduction in friendly assault capability in area
tends to restore balance of opposing forces. Fur
ther advance of friendly infantry exposes it to well
sited enemy weapons. Operations of US armor hin
dered by road blocks, mines and fire.
Details not determined. Exhaustion of enemy per
sonnel and material probably responsible in large
part.
ORO-T-261
UNCLASSIFIED
!fUEl
191
.&EEI" UNCLASSIFIED
Part IV
THE KEY ACTION
"SEERET
UNCLASSIFIED
sCRET
UNCLASSIFIED
NARRATIVE OF THE CAPTURE OF HILL MASS 314 BY THE
3D BATTALION, 7TH CAVALRY REGIMENT
Information in this section has been obtained from: a brief report on the
action included in the 1st Cav Div command report, parts of which are repro
duced in App E; an unpublished account written by the former S-3 of the 3d Bn,
7th Cav Regt; all available logs of units involved; and, particularly, recent
interviews and communications with officers and enlisted men formerly of the
3d Bn, 7th Cav and the various supporting arms. No one of these sources or
individuals, though providing useful data, afforded an essentially complete or
qualitatively accurate account of the battle. (One of the most accurate summary
reports is reproduced in App E.) In the narrative that follows, these various
independent viewpoints have been reconciled to the fullest extent possible with
each other and with the objective data of the other sections of this memorandum
on friendly casualties, enemy weapons, artillery support, air support, and
armor support. For the assault phases in particular, synthesis was difficult.
Some additional remarks on the method employed here are given at the head
of the section "Assaults on the Objective and Its Capture." From the informa
tion presented here, and from some additional interpretation and assumptions,
a graphic timetable of the attack has been constructed to show the phases of
movements of enemy groups and friendly platoons and friendly and enemy
fires. This reconstruction necessarily remains incomplete and questionable
as to certain details. A fuller and more reliable account might have been
obtained by a comprehensive program of interviews at a time not so long after
the action, and by research on documents no longer available.
BACKGROUND OF THE BATTALION
The 3d Bn, 7th Cav Regt was composed principally of former personnel
of the 30th Inf Regt at Fort Benning, Ga. Though not many of the officers or
enlisted men were veterans of World War IT, a considerable fraction had par
ticipated in i!l.fantry problems as members of the 2d Bn of In.fantry School
troops. To bring the battalion more nearly up to strength for duty in Korea,
noncombat specialists at Fort Benning, such as bakers and truck drivers, were
reassigned as riflemen. The newly constituted 2d Provisional Bn was assigned
to fill out the 7th Cav Regt, which on occupation duty in Japan had had only two
battalions. Tnough scheduled to land in Japan, the battalion did not debark on
arrival at Yokohama; because of the grave situation in Korea it was sent
directly on to Pusan, arriving there on or about 30 Aug 50. Now designated
the 3d Bn, 7th Cav, it was at once committed to battle.
ORO-T-261 195
UNCLASSIFIED
SEERET
UNCLASSIFIED
The battalion's first mission was to participate in the 7th Cav attack on
Hill 518 on 2 September 1950. After continued enemy counterattacks and a
disorganizing night action, the 3d Bn withdrew on 4 September. Though casual
ties had not been high, morale was shaken, and large amounts of materiel had
been lost. In a foot march the battalion moved back 14 miles to Taegu to be
re-equipped. Soon after, the 2d Bn, 7th Cav was also pulled back, and the two
battalions constituted the 1st Cav Div reserve.
Around the end of August 1950, US Army units were augmented with
attached KATUSA troops, or soldiers of the Army of the Republic of Korea
(ROKA). In the 7th and 8th Cav Regts they were "integrated," i.e., distributed
among the individual squads; other methods of augmentation were tried in other
regiments. Most of the ROKA personnel had had only a few days of training,
and could communicate verbally only with a few American Nisei who could
speak Japanese. In spite of these difficult circumstances, some of the attached
Koreans proved to be good soldiers. For tactical control, each of them was
given an English nickname beginning with the letter of the appropriate company,
and this name was stenciled on his helmet.
PREPARATORY PHASES OF THE ATTACK
Planning
On 9 September the 3d Bn, 7th Cav Regt was attached to the 8th Cav Regt
and moved up to occupy blocking positions across the Taegu - Tabu-dong road
7 miles north of Taegu. As shown in Fig. 12., Co L was astride the highway,
Co K to the west on Hill 181, and Co I to the east on Hill 184.
It was planned to use the battalion to take Hill Mass 570 and, if necessary
as a preliminary objective, Hill 373.. However, this mission was postponed
and repostponed when attacks by the 8th Cav failed to dislodge the enemy from
Hill Mass 314. It was apparently not until 11 September that it was decided
to use the 3d Bn, 7th Cav on Hill 314 instead of Hill 570. Much of the existing
'plan of attack of course still applied. Reconnaissance both of the new objective
and for a suitable route of march was necessary but could not be as thorough
as might have been desired. Furthermore, in attacking Hill 314, the 3d Bn,
was to be under the control of its own regiment and supported by its own
Held artillery, rather than, as already arranged, prepared to attack within
the well-established organization of the 8th Cav in this area. Some of the
lack of coordination of the infantry and supporting arms during the attack on
12 September must be attributed to the late change in plans.
The operations order of the 3d En, 7th Cay for the attack on Hill Mass 314
is reproduced in App E. In this complete form, it was most likely written after
the action.
The Move-Up
According to plan, the 3d Bn, 7th Cav departed from its blocking positions
at 0630 on 12 September. The route of the 2-mile march to the assembly area
at the base of Hill Mass 314 is shown in Fig. 12.
196
ORO-T-261
UNCLASSIFIED
HeRrl
UNCLASSIFIED
SEERn
H58
1457
1456
1455
H5i
1453
1452
1153 1154 !l55 1156 1157 1153 1159 1160
Fig. 12-Situation around Hill Moss 314 at time of attock by 3d Bn, 7th COy Regt on 12 Sep 50.
Figure shows initial blocking positions of 3d Bn, 7th Cay, route of march to Hill 314, and concentrations
of enemy fire received. Positions held by 8th Cay units and ROK police battalion are also shown, as are
enemy-held pasi tion 5, enemy bunkers, and enemy observation point. 1000-yd grid 5 are numbered; light I i no 5
indicate lOOQ-m grids.
ORO-T-261 197
UNCLASSIFIED
SEERET
UNCLASSIFIED
SUllO
For maximum concealment of the move from the enemy, the minimum
number of vehicles required for communications and heavy weapons was used;
also, these vehicles were sent up to Hill Mass 314 before daybreak, while
vehicles for resupply and evacuation were held behind the old positions until
after the attack had begun. From the map, it is seen that the move could have
been observed in its initial stages by enemy forces on Hill 570 and at a later
interval from the northern part of Hill Mass 314. The route of march there
fore had been carefully planned to take advantage of concealment afforded by
stream beds, irrigation ditches, etc. Since the battalion was in extended file,
it could have been observed at any exposed point during the 20-min time of
passage.
The only enemy fire drawn during the approach march was from what is
believed to have been an SP gun that had moved from Tongmyongwon down the
highway to a temporary position at Samsan-dong, under the nose of friendly
forces on Hill 225. Perhaps to cover this move, Hill 225 was subjected to
intense 120-mm mortar fire. As shown in Fig. 12, one of the two 76-mm con
centrations straddled the 3d Bn march route. The failure of the SP gun to
inflict casualties can be attributed to the combination of flat-trajectory fire on
flat terrain, the restricted angle of fire due to the confining slopes, the fog that
lay in the valley, and to the battalion's exploitation of what slight protection the
terrain afforded.
It is considered highly significant that the 120-mm mortar fire shifted
from Hill 225 to Hill 373 instead of to the 3d Bn approach march. If observed
friendly movements prompted the SP gun's fire, they must have been inter
preted either as routine traffic such as had been harassed at that point for several
days or as a reinforcing move for Hill 373, the pressure point of the previous
day.
The order of march of the companies of the 3d Bn was: I, L, H3, M, K.
The assault companies, I and L, arrived at the assembly area at the south end
of Hill Mass 314 somewhat ahead of schedule, around 0800, and started the
climb up the ridge toward the line of departure soon after, arriving there,
again somewhat ahead of schedule, around 1030. The assault companies there
fore missed the concentrations of the 120-mm mortar fire plotted in Fig. 12 and
reported as follows: "3d Battalion proceeding as planned. Receiving mortar
fire at 120855."'
The battalion's command party also proceeded up the ridge, and reached
the line of departure at 1015. Here, at the positions held on Knob 0 by Co L,
8th Cav, the battalion observation post was set up. Among the officers who
occupied it were the battalion commander, the S-3, and the S-2. Ten yards
to the right on Knob 0 the artillery observation post was set up, and this was
occupied by the artillery liaison officer and others.
According to the plan, the supporting tanks were to arrive at the base of
Hill Mass 314 before 0600, i.e., before daybreak. Emerging from the river bed
south of Hill 373 at 1000, they very likely were observed by the enemy on Hill
Mass 314. By this time the preparatory air strikes were already in progress,
and the forward movement of the battalion troops along the ridge may have been
observed, though perhaps not in detail because of the residual haze. The air
strikes alone were not especially indicative of friendly intentions, as the
intense round of strikes on Hill 570 the morning before had taken place without
198
ORO-T-261
UNCLASSIFIED
SE(AEl
SUREr
UNCLASSIFIED
concurrent movements of friendly infantry and armor and had turned out to be
purely an active defense measure. The appearance of tanks for the first time
at Hill Mass 314 may now have created suspicion that a major friendly effort
was impending there. At 1018 the tanks caught up with the infantry, most
likely with Co K, the battalion reserve, which had remained at the assembly
area at the base of the hill. Accurate 120-mm mortar fire apparently still
fell on the assembly area. Only one casualty is known to have resulted, and
this was in Co K. The casualty report says "shell fragment" and gives the
time as 1100. If this fire actually fell as late as 1100, the absence of more
casualties would be explained by the fact that most of Co K had already followed
the assault companies up onto the southern ridges of the hill mass.
The 3d Bn, 7th Cav took over the former command post of the 3d Bn, Bth
Cav at the southeast tip of Hill Mass 314. As shown in Fig. 12, the battalion aid
station and supply points were nearby. The aid station opened at 1100 on 12
September. The headquarters of the ROK police battalion were also in the
vicinity. At the same time that the 3d Bn of the 7th Cav was to attack on Hill
Mass 314, its 2d Bn was to relieve the ROK companies on Hill 660, and the
ROK force was then to be attached to the 3d Bn for commitment in an emergency.
The approach march of the 2d Bn to Hill 660 was over the hills from the south
and in no way coincided with that of the 3d Bn described above.
So far as is known, none of the officers and men of the 3d Bn, 7th Cav Regt
had been on Hill Mass 314 before the battalion moved onto it on the morning
of 12 Sep 50. Of the 3d Bn, 8th Cav Regt, which had been fighting on the hill
for two days, only Co L remained in position. It is not known to what extent
the experience of this battalion was passed on to the battalion that relieved
it. Certainly the time for conference on the hill itself was short. One point
of information that should have been passed on was the fact that friendly forces
had blown a large crater at a defile on the trail that skirts the hill on the right,
and thus prevented the contemplated movement of tanks beyond this point.
As discussed in detail in Part II, the 3d Bn, 7th Cav was confronting a
recently strengthened enemy force now totaling about six hundred and fifty men.
It appears that units of the still potent 13th NK Div had, in the previous 24 hr,
reinforced those of the rather exhausted 1st NK Div that had been fighting on
the hill. Perhaps it was not realized that where before the enemy units on the
hill had been backed up with only meager to moderate supporting fires, a formi
dable, if limited, apparatus of communication and supporting fire had now been
brought to bear.
The ridgeline and western side of Hill Mass 314 were wide open to enemy
observation from Hill 570. Figure 3 (given earlier) is a composite panoramic
photograph of Hill Mass 314 taken from the then most forward enemy position
on the unnumbered 4BO-m peak on Hill Mass 570'. (East-west distances across
the ridge on Hill Mass 314 are highly foreshortened.) Though this picture was
taken in springtime 1953, the hill did not look much different in September 1950.
Close-up views taken on the hill after its capture and shown as Figures E 1
through E7 give the same impression-a steeply sloping- undulating hill mass
covered with rocks, low bushes, and scrubby trees.
Comparison of the photograph (Fig. 3) with the map (Fig. 2) reveals some
errors in cartography; in the photo, for example, Knob 3 is lower than Knob 2.
The map with which friendly forces had to work at the time of the action was
even more inaccurate.
ORO-T-261 199
SEERET
UNCLASSIFIED
UNCLASSIFIED
SEEK"
The main enemy force held the ridge from Knob 2 to Knob 3. Shallow
foxholes dotted the bald top of the ridge, some of them perhaps covered with
vegetation or logs. On the nose of Knob 2 there were about a dozen shallow
foxholes; a few can be seen in Figs. E6 and E7. In addition, there were a few
more substantial log-covered bunkers. One was back on the finger northwest
of Knob 3 and may have been an OP or forward command post; another, which
may have been a command post, was on the northwest side of the hill. Though
in general these positions provided scant protection from air bursts, they did
afford personnel concealment, and were characteristic of an enemy skilUul
at camouflage. Perhaps the larger part of the enemy force occupied the wooded
crown around and just below the top of the ridge. Some units may have been
at the base of the hill. Knob 1, several hundred yards from friendly positions
across a dip in the ridgeline, was an enemy outpost.
The number and distribution of enemy weapons on Hill Mass 314 and in
the surrounding area are considered in Part II, and the more important positions
are plotted in Fig. 8. Enemy supporting fires could have come only from the
north and west.
The task of the 3d Bn, 7th Cav was to capture Knob 0, move down a steep
drop into the saddle, then up a long steep incline to assault the main enemy
forward position at Knob 2. This advance through the saddle entailed, at least,
a descent through a vertical distance of 40 m and a subsequent climb of 160 m
through a horizontal distance of 500-600 yd. The actual distance traveled
would be more nearly 700-800 yd. Though the northern slope of the saddle was
not as steep as the reported 60 deg, it can be seen from Fig. 3 that the least
steep climb would be several hundred yards up a slope of over 20 deg. Once
the hurdle of Knob 2 was passed, the assault force would have to advance 450
yd further and take Knob 3 before the objective would be secured.
When the fog lifted on the morning of 12 September, the air was clear
and the temperature warm. Cloud cover at Taegu airport (K2), 8 miles to the
south, was reported as follows: "1005, 10/10; 1040, base 2500 feet, tops
4500 feet."
The footing was still slippery from recent rains, but not enough to make
climbing difficult. Beneath its top layer, the soil was hard and resistant to
digging.
PreRaratory Fires
The preparatory air strikes began after the assault companies had begun
the 2-hr climb up the ridgeline to the line of departure from the assembly
area at the base of Hill Mass 314. Members of the assault and reserve com
panies, and command personnel and observers accompanying them on the way
to the forward command post, could not have observed the main preparation,
and appear not to have found out about it. The attack plan called for air strikes
from 0930 to 1030. The data listed in Table 12 and discussed in Part II show
that between 0950 and 1045 eleven F51's of three missions rocketed and strafed
the objective and dropped napalm. The schedule may have been delayed some
what by the haze. Not until 0930 had it lifted sufficiently to permit the artillery
liaison officer to lay down registration fires of the 77th FA Bn on the objective
and mark it with white phosphorus bursts for the mosquito plane. The artillery
preparation was to have followed the air preparation at 1030 and was to have
200
ORO-T-261
UNCLASSIFIED
SECRET
SHRU
UNCLASSIFIED
lasted until the jump off at 1100; however, it had been canceled, due to ammu
nition shortage, or, rather, restriction. Perhaps as partial compensation,
another flight of four F51's, which had bombed Tabu-dong, was led to the
objective by a second mosquito plane, and rocketed and strafed the objective
unti11100. The preparatory fires of the 3d Sn 81-mm mortars, which had
been scheduled for H-IO, could not begin until 1100. Jump off of the attack
was therefore delayed 10 min, until 1110.
Contrary to some reports, the planes, after they had expended their
ammunition, did not remain in the area to buzz the objective while the assault
troops advanced. The last flight landed at Taegu airport at 1120, and must
have left Hill Mass 314 promptly at 1100. The forward command and observa
tion posts opened at approximately 1015. Their personnel, who had not learned
of the regimental order of 0735 that canceled the artillery preparation, learned
of it at this time via the newly established communication net. They apparently
did not know that two flights had already hit the objective, for they wa\ted
anxiously for the two flights that materialized after 1030. In all, about five
fragmentation bombs (500-lb?), 16 drums of napalm, 51 rockets, and an unknown
number of :50-cal rounds were used in the preparatory air strikes.
For its own preparatory fire, the 3d Bn had the 81-mm mortars of Co M.
As shown in Fig. 12, two of these were located at the base of Hill Mass 314 and
two were situated part way up the ridge. The latter pair with several hundred
rounds of ammunition, was taken over from the 8th Cav in an equal exchange
to avoid needless transport up and down the ridge.
The battalion lacked the planned support of a platoon of heavy mortars
attached from the 8th Cav (their position is shown in Fig.12) .. The only
available information on this point is from the 7th Cav side and may be biased.
The log of the 3d Bn, 7th Cav states:
121130. Units in general support: one platoon of 4.2 mortars of 8th Cavalry Regi
ment; 3d Platoon, Company'C, 70th Tank Battalion in general support. Comments: the
4.2 observer could not be located, when located was found to be ineffective in his job as
an observer and therefore we failed to get cr'itically needed support from this platoon.
6
Though it was not intended that the tanks should participate in the prepar
atory fires, theoretically they might have helped make up for the canceled
artillery fires by firing on forward slopes. However, the tanks only arrived
at about the time of this news, when the battalion commander and several of
his key staff officers were already at the forward command post. Until much
later, the commander of the tanks lacked any orders, let alone revised
instructions.
Four heavy machine guns of Co M were installed at the line of departure,
a section of two on the left side of Knob 0 in support of Co L, and a section of
two on the right side in support of Co I. While the 81-mm mortars were
firing on the objective (from 1100 to 1110), these machine guns laid down
intense fire on the enemy outpost at Knob 1.
The 75-mm recoilless rifles were brought up around the right side of
the hill by jeep to fire on the objective at a late stage of the battle. The 57 -mm
recoilless rifles were up on the hill, but in the absence of exposed enemy
bunkers, it would not appear that they were appropriate weapons for prepar
atory fire.
ORO-T-261 201
SEERET
UNCLASSIFIED
UNCLASSIFIED
SEnn
ASSAULTS ON THE OBJECTIVE AND ITS CAPTURE
There is no existing adequate account of the climactic phase of the action,
and no individual, even of those who were present, appears to have an at all
complete conception of the course of the fighting. From the battalion obser
vation and forward command posts atop Knob 0, the staff officers could have
seen little of the action on Knob 2, and nothing beyond. Furthermore, their
electrical communication with the assault units was crippled early in the
battle. Most of the officers of the assault units were not in a position to
report afterwards. Of the officers of Co L and Co I, only the two commanding
officers and one or two platoon leaders ever got beyond Knob 2. The CO of
Co I was wounded late in the battle; Co L's CO survived unhurt but was killeli
ten days later; one platoon leader was hit at the same time as the Co I CO,
and the other, if he was indeed present, has not been interviewed. Appendix
E includes a report of an unknown interviewer's brief conversation with the
CO of Co L shortly after the battle. A sequence of events outlined by an
unknown staff officer of the 3d Bn and interviews with personnel of Co I were
also prepared at that time, and they are given in App E. As attempts to give
ac<;.ounts of the assaults they are hardly coherent, but they do contain useful
details.
In the narrative which follows, all available scraps of information have
been considered and fitted together as far as possible in chronological order.
Recent interviews with certain officers and enlisted men-most notably, the
former CO of Co I and the former S-3 of the 3d Bn-have been especially valu
able in establishing the continuity of events. Objective checks of the time-space
pattern of the action are afforded by some data on friendly movements and in
casualties and information on the supporting air, artillery and armor presented
in the earlier section "Friendly Support."
Platoons aCted as independent tactical units at several crucial points of
the battle. Since the platoon is not an administrative unit, movement and deploy
ment are not reported in any formal document, nor can they be inferred, from
such information as the circumstances of individual casualties, as is possible
for a company as a whole. In the narrative which follows, platoon experience
is sketched from certain facts mentioned in the appendices, from information
supplied by those interviewed, from reports about identifiable individuals, and
finally, simply from what was possible and logical in the given circumstances.
In the original attack plan, the line of departure was to be the knob to the
rear of Knob 0, and Knob 0 would be Phase Line 1. When on 12 September it
was found that Co L, 8th Cav was still in possession of Knob 0, the LD was
moved up to here, and the phase lines were moved up accordingly to the desig
nations used in this section. Though regimental and all battalion personnel
were notified of the change, the earlier designations seem to persist in certain
reports, notably in the sequence of events and the interviews given in App E,
but without consistent application. Interpretation is therefore required to
decide even such a vital question as whether forward units reported to be on
Phase Line 2 were actually on Knob 2 or Knob 1.
The redesignated phase lines are defined as follows: Phase Line I, the
meter-grid line that passes through Knob I on the map; Phase Line 2, the line
of the edge of the map sheet crOSSing the objective just north of Knob 2; Phase
Line 3, the meter-grid line that crosses the north end of Knob 3.
202 ORO-T-261
SUREf
UNCLASSIFIED
.EERU UNCLASSIFIED
Initial Phases
The assault companies jumped off at about 1110 in parallel columns.
Company L was On the left side of the ridgeline with its 2d Plat leading, followed
in order by its 1st and 3d Plats. Company I was on the right side of the ridge
line with its 3d Plat in the lead, followed by the 2d and 1st Plats.. Except for
observers 100 yd down the ridgeline, each company front was about 50 yd.
It was soon proved that.Knob I was held by 12 to 15 enemy troops equipped
with 2 machine guns and 2 automatic rifles. The intensity of their fire during
the friendly advance of 150 yd from Knob 0 has not been reported with any
certainty, but few if any friendly casualties resulted. Evidence that active
fighting had previously occurred between Knobs 0 and 1 was seen in the form
of several enemy and three American bodies. By 1130, Knob 1 had beenover
run, the enemy troops killed, and their bodies thrown out of the holes. An
exception was an enemy soldier who jumped up during the advance and, without
being shot at, ran back toward Knob 2. Whatever his motive was, he doubtless
gave warning that the attack had begun.
There was a pause to reorganize. In the two assault companies, the two
platoons previously leading became the support platoons and took up temporary
positions on each side of Knob 1. The two following platoons passed through to
take up flanking positions, and the two trailing platoons passed through to speaI'
head the next phase of the attack.
The first enemy reaction occurred at this time, when 120-mm mortar
fire fell on the positions the assault companies had just vacated at the LD.
This fire caused some casualties in Co L, 8th Cav. Company K, 7th Cav had
remained further to the rear just to avoid such hazard. The enemy mortars
had evidently been preaimed at Knob 0, but had received the word to fire rather
late.
Jump off for the first assault on Knob 2 was at about 1135. The deploy
ment of the assault companies is shown in the corresponding diagram in the
succeeding section of this memorandum. As before, the ridgeline was the
boundary between the two companies. Whether by design or because of the
more difficult going off the top of the ridge, the two flanking platoons trailed
somewhat, so that the entire battalion front was like a plowshare extending
symmetrically across the saddle. Within each forward platoon, two squads
were forward on a skirmish line and the third somewhat behind in wedge forma
tion. As the assault platoons advanced, the two support platoons moved down
off Knob 1, and one squad from each moved out to sweep the wooded slope well
down on each side of the hill. The weapons platoons of Co L and Co I moved
up to the rear of Knob 1. Down On a finger off to the right of the saddle, the
sweep squad of Co I discovered an enemy automatic-weapon position and killed
the crew.
Shortly after jump off, advancing troops of Co L noticed inaccurate hostile
rifle fire coming from a group of rocks at the bottom of the saddle off to the
left (see Fig. 3). It had caused no known casualties when silenced by the fire
of a friendly squad, which also caused no casualties. Four enemy soldiers
jumped up. Three of them ran directly to the rear, that is, toward Knob 2, and
seem to have escaped; the fourth chose an illogical course across the saddle
to the right, and was cut down by other friendly fire.
ORO-T-261 203
SECRET
UNCLASSIFIED
UNCLASSIFlE.D
U(RfT
The commanding officers of Co L and Co I moved their men down into the
saddle as rapidly as possible. Reaching its bottom-the line just vacated by
the enemy outposts-at about 1143, friendly units were suddenly taken under
concentrated fire, as if this were the enemy's final protective line. Enemy
weapons involved were 120-mm and 82-mm mortars, 50-mm company mortars,
machine guns, and captured US BARs. The two companies moved about two
hundred yards on up the saddle. So far as is known, the mortar barrage did
not follow then; they were stopped by casualties sustained and by the small
arms and automatic-weapons fire that continued to be directed at them from
the top of Knob 2. This phase of the attack was reported as follows: "From
. 7th Cav. Jumped off on time, moved 300 yards [from Knob 0] and were pinned
down by sniper firing but attack is continUing. n 1
This first, intense, enemy fire had several important effects. Because of
a mortar hit, the command of Co I lost radio contact with the rear at the out
set of the battle, and from undetermined fire it lost all walkie-talkie communi
cation with its forward units. Company L also lost all walkie-talkie communi
cation, though it retained the use of its radio throughout the battle. Communica
tions are discussed more fully later in this section. At the outset of the main
phase of the battle, Col also lost the leader of its 2d Plat. The casualty report
gives the time as 1130. The enemy fire responsible was proved by surgical
removal to be a 0.25-cal Japanese bullet. One Japanese light machine gun was
later found on the objective.
The first move to assault the objective is summarized in the 7th Cav Regi
war diary as fOllows:
.. . The advance continued for 500 yards [measured from the LD-that is, 200 yards
short of Knob 2] under sporadic machine gun and rifle fire and a 120-mm mortar con
centration on the line of departure and reserve company assembly area. At this point,
the entire battalion was pinned down by intense rifle and mortar fire and Simultaneously
observed about 400 enemy to the left preparing for a counterattack. An air strike was
requested but was delayed because of refueling, as the counterattack was beaten off by
infantry, artillery, and mortar fire.
The figure of 400 enemy troops was probably derived from one report
after the battle that this number of enemy had been on the entire hill mass.
Mention of an air strike and artillery support is somewhat premature. Only
supporting mortar fire was involved. The log of the 3d Bn states: "121145.
When it was impossible to obtain any other fire support, the 81mm mortar
platoon delivered very effective fire in large quantities and broke up the first
counterattack.
A rough estimate of the intensity of this mortar fire was recently made
by the former unit administrator of Co M, who was at the battalion CP on the
line of communication to the mortars. Co M brought about three hundred rd
to the hill and took over several hundred from the 8th Cav up on the ridge.
Perhaps four hundred fifty rd were expended during the battle, of which about
one hundred twenty rd in all were fired on three occasions specifically to
break up enemy troop concentrations. On each occasion, each of the four mor
tars fired about ten rounds at maximum rate to give a 40 rd concentration.
If they fired at the standard maximum rate of 30 rd per min the concentration
204
ORO-T-261
UNCLASSIFIED
SEERE'I'
UNCLASSIFIED
would have been laid down in 20 sec. Though about fifteen bearers were carry
ing ammunition up to the ridge, there was a problem in keeping the two forward
mortars supplied.
It is a moot question whether the enemy actually attempted to counter
attack or merely moved troops forward to fire the final protective line after
the preparatory fires had forced them back. As shown in Figs. E6 and E7,
shallow enemy foxholes were at the forward edge of the plateau-like nose of
Knob 2; three machine guns and one automatic rifle are also thought to have
been there. To fire down into the saddle, enemy defensive troops would have
had to occupy these positions. If they delayed fire until the attackers were
within 200 yd, they would have had to expose themselves above the edge to
obtain any field of fire.
The first concentration of al-mm mortar fire on observed enemy troops
was called down directly onto Knob 2. It is considered that this fire chased
the enemy down onto the wooded slopes without causing many casualties.
The First Assualt
Authority to resume the attack rested with the platoon leaders, since
communication with the company commanders at Knob 1 had been lost. At
about 1155 friendly units again moved forward. This time the enemy mortar
fire was distributed over the entire area of the saddle occupied by friendly
troops, and began to fall on the supporting units as soon as they started down
from Knob 1. For the first time, 76-mm artillery fire was received on Knob 1.
The source is considered to have been an SP gun located at the cemetery to
the north of Hill Mass 570. Its fire would necessarily have been restricted to
Knob 1 and the western slopes of Knob 2. An unconfirmed report states that
fire from the objective was again intense, but does not localize the source.
It is surmised that the most forward enemy positions on the nose of the knob
were not reoccupied, for the heavy machine guns of Co M had now moved up
to Knob 1 and covered this assault with continuous overhead fire. Some friendly
troops thought that this fire was enemy and required convincing to the contrary
to move up.
Friendly units, believed to be elements of the 3d Plat of Co L, succeeded
in reaching the enemy positions. From his post on Knob 1, the leader of the
K Co Wpn Plat saw 12 to 15 friendly troops running around on the nose of the
knob, and an occasional enemy soldier darting among the rocks nearby. The
circumstances of a particular casualty at this time were also observed (No.
7, Table E 1). The leader of an L Co MG squad ran forward and picked up
the weapon after the gunner was killed. Finding it jammed, he kneeled right
on the crest of Knob 2 to fix it and was almost cut in half by fire from an
automatic weapon beyond. (The casualty report attributes this death to artil
lery fire, and a second witness thought this was the cause.)
Advance at least as far as the top of Knob 2 was reported in the following
message: "121200. Now on Phase Line 2. Reorganizing and pushing on to
Phase Line 3.'
The attack was turned back. Fire from the dense underbrush on the
transverse ridge which overlooks from the left the shallow draw on the left side
of Knob 2 was at least partly responsible. As the troops of Co L attempted to
ORO-T-261 205
SEEREJ
UNCLASSIFIED
UNCLASSIFIED
SEERfT
move over the crest away from this fire they may also have encountered enemy
troops assembled on the northeast reverse slope. When Co I was advancing on
the right side of the saddle, enemy groups were seen to withdraw toward this
spot. Whether an actual counterattack occurred has not been determined.
Enemy exploitation of the repulse of the first assault was prevented by
friendly supporting fires. The second concentration of 81-mm mortar fire
called for against an enemy grouping may well have been directed at this time
toward the northeast side of Knob 2. In addition to the mortars, the first
supporting artillery concentrations were now fired. As discussed in Part U
and plotted in Fig. 8, the 77th FA Bn fired approximately eighteen rounds of
VT-fuzed 105-mm rounds at Knob 2 and along the ridge to Knob 3. Since few
enemy bodies were found up on the ridge later, it is inferred that any enemy
troops who had moved onto the ridge to counterattack had returned to the
wooded slopes before much of this fire had landed.
The situation forward was reported by runner to the CO of Co I shortly
after 1200: Co L was in trouble and one of its platoons had been almost
annihilated by fire from the underbrush to the left of Knob 2. With the walkie
talkies out and no idea where his 3d (support) Plat was located, the CO went
up to the critical point. There he found the Co L commander crying, "They're
killing all my men I "
Moving forward and into the draw to the left of Knob 2, the two company
commanders found the troops "stunned." They were boxed in between the
mortar fire falling behind them and the fire from the dense underbrush along
the transverse finger ahead and to the left. The only visible evidence of the
enemy was the continual flight of grenades from the bushes. The grenades
were of the offensive, or concussion, type, producing blast but few effective
fragments. * rOnly one grenade casualty is attributable to this incident. See
No.3, Table El. The casualty report states: "SWA. Shrapnel. Fragment
wound (grenade) left shoulder, left forearm, laceration of face and scalp,
1130 hours, 12 Sep ... Entire company was engaged in an attack against the
enemy, when a hand grenade exploded near his position." The position of the
individual at the time is eVident.] To break the men's mental block, the two
COs stood up and started throwing fragmentation grenades into the bushes
over the finger. When each had thrown five or six, the men regained their
initiative and did likewise. The enemy grenades stopped coming. It is
inferred that surviving enemy troops withdrew from the first transverse
finger at the left of Knob 2 back across the bare gully and up onto the wooded
and more defensible finger beyond. (See Fig. 3.)
Action on the Right during the First and Second Assaults
From his original position at Knob 1, the commanding officer of Co I
observed only one of two enemy soldiers, "obviously trying to get out."
Several members of the 2d Plat, which was on the right flank, reported see
ing three or four enemy groups of perhaps a dozen men withdrawing toward
the northeast side of Knob 2 with very little defensive fire. It is inferred that
around the time troops of Co L were being cut down on the nose of Knob 2 by
fire from the underbrush to the left, the 1st Plat of Co I was stopped just short
.These were retained for friendly use, See photograph, Fig. t:1D.
206 ORO-T-261
UNCLASSIFIED
SECRU
nERn
UNCLASSIFIED
of Knob 2 by fire from the underbrush to the right. This situation is suggested
by the circumstances under which the 1st Plat leader (a lieutenant), and then
the assistant platoon leader (a sergeant), became casualties. The platoon leader
was peppered by a grenade at a time thought to be the height of the first assault,
around 1200. (The casualty report states 1300, but the lieutenant, who refused
early evacuation, was only brought back to the aid station twelve or more hours
later.) The assistant platoon leader took command and apparently was wounded
almost at once by "shrapnel." (The medical tag gives the time as 1400, but
this was evidently the time at which the sergeant, who remained in action,
finally collapsed.) Some indication of the action of the platoon in this interval
is afforded by the report recommending the sergeant for an award:
.. . Sgt Lancaster was wounded and his platoon forced to seek cover by the heavy
machine gun and small arms fire of a fierce counterattack. Despite his wounds, Sgt
Lancaster refused to be evacuated and, remaining with his men, moved about exposed
to a hail of enemy fire in order to organize and deploy his platoon in an effective defense
position. So skillfully did he employ his small force, that they were ahle to withstand'
and fight off two enemy banzai attacks. Immediately seizing the initiative, Sgt Lancaster
led his men forward in an attack to seize their objective, during which action he fell un
conscious from loss of blood ....
It is probable that the sergeant remained with his unit until the third
assault at about 1430. An enemy counterattack is also mentioned in App E:
M/Sgt McCollum, the Weapons Platoon leader of Company I, was following the at
tack with his weapons platoon at the moment the enemy launched his fanatical counter
attack. M/Sgt McCollum rushed forward leading his men after he had organized them
as riflemen and came to the assistance of the hard pressed rifle platoons ....
According to the medical tag, the leader of the weapons platoon was
wounded, by mortar fire, at 1645, Le., when he was up on the objective.
Whether the reported counterattacks were more than groupings of enemy
troops and concentrations of enemy fire is also questionable. When the CO
of Co I moved up to his company's forward positions during the second assault,
he found the 1st and 2d Plats,which he had sent around the right of Knob 2, .
held up without visible opposition. The two platoons were deployed along a
ditch that strangely ran down from the right side of Knob 2 along the top of a
finger. This ditch or gully was about six feet deep and eight feet wide at the
top, and about fifteen feet wide and fifteen feet deep further down the slope.
On the friendly side there was little or no cover; on the enemy side there was
dense scrub growth. Enemy fire was being received; according to one of the
sergeants, this panicked the men so that they would fall back "like a wave"
into the mortar fire, and would have to be urged forward. Company I's CO,
who was forward at this time could not say whether small-arms fire was being
generally received, but he saw dust spurt from the back of the jacket of a man
who was hit "like a stone wall" and killed. On the right side of the saddle
the situation was very much like that on the left.
The CO of Co L, a captain, was the senor officer of the two assault
companies. After the repulse of the first assault, he radioed the battalion
commander that his company was disorganized. He was told to hold up the
attack in anticipation of an air strike, and to have the men dig in. Because of
the hardness of the ground, the latter could not be done.
ORO-T-261 207
SEERH
UNCLASSIFIED
SEERn
UNCLASSIFIED
Seeing the repulse of the 3d Plat, Co L, and lacking communication with
his commanding officer, the leader of the 3d Plat, Co 1 had moved his platoon
from the right side of the saddle forward and to the left. When the platoon now
reached the forward location of the two company commanders, it was attached
to Co L and directed to move to the left to retrieve the more deteriorated
situation there. The presence of the two 3d Plats at about the same spot at
about 1215 is established by the following event involving the two platoon
leaders: while standing, the leader of the 3d Plat, Co L was hit by a mortar
burst. The leader of the 3d Plat, Co I, who was his friend, ran over, saw that
he was dead, turned away, and dropped his hands-at this moment, he was
himself killed by the next round.
Since enemy troops, as it seems, were no longer in the most forward
defensive positions, the enemy mortar fire walked right up onto the assault
units, stopping just short of the crest of Knob 2. It was no longer spread out
over the saddle but concentrated at this time in the two areas on the northern
side of the saddle designated as "1" and "3" in the graphic timetable of this
action which follows. Some 76-mm fire was also received as shown; it is thought
to have come from the SP gun plotted in Fig. 8. The location of this gun and the
flat trajectory of its fire allowed it to reach only a small fraction of the pinned
down friendly troops.
There was little cover. The right side of the saddle was naturally less
exposed than the left, but on the left there was the shallow draw to offer some
concealment, if little protection. From Hill 570, men in the draw could be seen
only while standing. In Co L SOme men moved down the wooded slopes to avoid
the fire, while the 3d Sqd of its 1st Plat moved over to the Co 1 sector.
As shown in Part II, the message requesting the on-call air strike was
received at division at 1220. Though the assault troops may have hoped for
closer support, this request-presumably in this form from the regimental
level-called for attack north of Knob 3. This attack was delivered promptly
and as -specified, and apparently without the knowledge of friendly forward units.
Though enemy reinforcements on the northeast slopes of the hill mass and in
the nearby villages may have been hard hit, it is difficult to see how such a
strike could have reduced either the opposition of. the enemy troops confronting
the assault units or the mortar fire falling on them. The tanks, however, which
marked the objective with white phosphorus bursts for the air strike, also
fired onto the western slopes of the objective around 1215, with what might have
been telling effect.
The Second Assault
At about this time supporting artillery was augmented with some of the
155-mm howitizers of the 9th FA Bn. The artillery air observer reported that
the enemy mortars had been located, and these were taken under counterbattery
fire. The site reported from the air is not known, but the mortars and the
counterbattery concentrations are plotted in Fig. 8. No evident diminution of
enemy mortar fire was noted, however, until late in the battle.
After reorganization down the slope, toward the right, the second assault
was ordered, at about 1230. The two 1st Plats were to advance up the backbone
of the saddle to establish a base of fire while the 2d Plat of Co L worked around
the left flank and the 2d Plat of Co 1 worked around the right flank. Whatever
208
ORO-T-261
SfUET
UNCLASSIFIED
5-!CRET
UNCLASSIFIED
its tactical design, this maneuver would tend to take the companies out of the
mortar-fire concentrations. The action to the right of the saddle-how the
1st and 2d Plats of Co I were stopped by enemy fire from concealment and
p'erhaps by actual counterattacks-has already been sketched.
The movement of the 2d Plat, Co L in the second assault is reported in
App E, on the basis of information supplied by the CO of Co L, as follows:
Captain Walker ordered Sgt Mitchell [the platoon leader] around to the left of the
1st Platoon with the mission of attacking and moving over the crest of Hill 314. As the
2d Platoon moved over the crest, it came under very heavy time fire, small arms fire
and grenades, which caused very heavy casualties among personnel of the platoon and
pinning it to the slope of Hill 314....
The so-called "time" fire has been otherwise reported as 76-mm direct
fire, impact-fuzed. It was plotted by the artillery liaison officer as concen
trated in the area through which the 2d Plat passed.. Additional information
on the advance of the 2d Plat is to be had from affidavits describing the
circumstances under which its leader was lost. One of these states:
During the hottest part of the fighting when the enemy mortar and small arms fire
was the heaviest, I saw M/Sgt Mitchell moving around among the platoon when everyone
else was down in position. He moved up to each man and told him where to fire and to
move forward. He remained exposed to this fire all the way up the steep slopes of the
hill. As we neared the top of the hill, someone said that Sgt Mitchell had been wounded,
and I saw him and he was wounded in the chest and was bleeding very much, but kept
the platoon and himself going up to the top of the hill. He was one of the first to reach
the top of the hill despite his condition, and the rest o ~ the men went with him. When the
enemy had been driven from the hill {much later], Sgt Mitchell asked for the medic to
get treatment, and I saw him start down the trail. A little later he was killed by an enemy
mortar round.
A second affidavit also states specifically that the unit had been in a
covered position and that the platoon leader was first wounded by small-arms
fire when he was almost at the top of the hill.
It would seem that the 2d Plat moved onto and up the wooded finger that
runs up to Knob 2 to the left of the draw, the commanding position from which
enemy opposition had come during the first assault. It may be supposed that
the enemy had moved back to a line on and beyond the higher transverse finger
just across the gully (see Fig. 3) and repeated the earlier tactic of resisting
with grenades and small-arms fire and then withdrawing to let the mortar
fire move back still further. Though the platoon evidently was hard hit and
could not go on, this move placed friendly forces in possession of Knob 2 up
to a transverse line running across its crest. Simultaneously with this flank
ing action, two squads of the 1st Plat of Co L moved up the draw to occupy
the left side of the nose, while the 3d Sqd moved around onto the right side.
Command and observer personnel were thus enabled to move up the draw to
where it ended in a curious depression called "the shallow" on top of the knob
just short of the crest. The commanding officers of Co L and Co I, the Co L
radio operator, an artillery liaison officer, one artillery forward observer,
the two 81-mm mortar forward observers, and some messengers moved into
it.
ORO-T-261 209
S!UU-
UNCLASSIFIED
SEUr-I'
UNCLIi.SSIFIED
The shallow was the one good observation point on Knob 2. Perhaps a
Korean burial site, it was a semicircular dishof earth about twenty feet in diam
eter and enclosed, rather like a revetment, with four-foot walls. From here
for the first time the ridge beyond the crest could be observed, but the reverse
slopes and declivities along the ridge, the flanks of the hill mass, and even
much of the sides of the ridge, were still out of sight. Except for three or four
wounded whose torsos projected above shallow foxholes, no enemy were in
sight. These "sitting ducks" were fired at.
An incident at this time illustrates how much observation depended on
position. A sergeant of the 2d Sqd, 1st Plat, just on the left side of the shallow,
saw one of the men firing over Knob 2. When he asked if the man were sure
the soldiers were enemy, he was told to come over to the right and see. On the
reverse, northeast slope of the Knob were fifteen or twenty enemy. This obser
vation of enemy troops on the right side fits in with the situation of Co I already
reconstructed. The first man was hit by mortar fire at 1500, according to the
casualty report.
After the second, partially successful assault, the artillery again worked
over the objective ridge, this time with an estimated 36 rd of VT-fuzed 105-mm
shells. In Fig. 11, given earlier, the hypothetical transverse distribution of
this fire across Phase Line 2 is graphed. Very little of it would have reached
enemy troops on the finger to the left of Knob 2, but it may have fallen some
what to their rear at the steeper waist of the objective, prompting them to with
draw from the hill mass. It is assumed that the 81-mm mortars were contin
ually active. Their second and third concentrations on observed enemy group
ings may well have been on the northeast side of Knob 2 in support of Co I.
The following report appears in the division log: "1310. Blue at original
[previous] positions waiting for airstrike." 1 Shortly after 1330 another air
strike was directed at the northern or northeastern slopes of the hill mass,
apparently again without the knowledge of the assault troops. With the issue
of the battle still in doubt, it would seem that aircraft readily available on call
had now been used.
With the inferred withdrawal of enE<my forces on the left side of the objec
tive, enemy mortar fire moved somewhat further up on the nose of Knob 2. As
plotted by the artillery liaison officer (see the graphic timetable which follows
this section), who was in the shallow at the time, the mortar fire fell in four
distinct, adjacent areas, two on the right side of the ridgeline and two on the
left. The fire on the right was thought to be 120 mm. The caliber of that on
the left was not known. Most mortar casualties were thought to have occurred
on the left.
The location of the mortar concentrations is confirmed by the circum
stances of several casualties. A sergeant of the 2<1 Sqd, 1st Plat, Co L was
hit at a time he thought was about 1400 (medical tag says 1500) just to the left
of the shallow. A sergeant of the 3d Sqd, 1st Plat Co L was hit while pinned
down by mortar fire in a small crater about twenty-five yards to the right of
the shallow at a time he estimated to be 1330 (medical tag says 1400). He
described the mortar fire as comir ~ in salvos of three.
210 ORO-T-261
UNCLASSIFIED
SEElEY
UNCLASSIFIED
It is quite possible that each of the two concentrations on the right side of
the ridgeline was fired by three of the battery of six 120-mm mortars north
west of Hill Mass 570, and that each of the two concentrations left of the ridge
line was from a pair of the group of four 82-mm mortars at the north end of
Hill Mass 314.
The over-all intensity of mortar fire was judged to be less than during
the first assault. From the battalion OP on Knob 0, it was noted as coming in
sporadic sharp bursts, and some 120-mm fire fell on Knob 0 itself. The fire
of three 120-mm mortars may well have been diverted for a time from Knob
2 to other targets.
There was a very long wait for aircraft to appear in close support. The
pinned-down assault troops sustained casualties from the continuing mortar
fire at an average rate o{ one every two minutes.
The men in the shallow fired at anything that moved within their limited
field of observation. One incident is notable. Atop the ridge was a rectangular
boulder about four feet high and eight or ten feet long. From behind it, an
enemy soldier would periodically pop up his head, look around and then disappear
again. The CO of Co I estimated the range as 300 yd but said he could easily
discern the cap the man was wearing and the whites of his eyes. Though he
always stuck his head up in the same spot, and though about a dozen men were
firing at him, the enemy soldier was not hit until after about "fifteen minutes"
when he fled to the rear and fell down from a shot in the leg. This was the
fourth such incident known to have occurred during the battle. On this occaSion,
at least, the enemy soldier was very likely a forward observer, and of course
he had to report to the rear in person. The contour of the ridgeline (see Fig:
3) suggests that the pOSition was about one hundred and fifty yards from the
shallow just forward of the decliVity midway between Knobs 2 and 3. A body
of enemy troops was apparently not far behind.
There was doubt that the final objective would be taken. The following
delayed entry in the division log probably represents the 3d Bn situation around
1415:
121545. From 7th Cavalry (Colonel Nist) [the regiment's commanding officer):
I talked to Colonel Lynch [the battalion's commanding officer] about a half-hour ago.
He was getting ready to jump off for his third phase line. He doesn't know whether or
not he can make it. I am going to stop them at 1630 because he must get dug in for the
night. He is having quite a few casualties. The ROK's with the 8th Cavalry are tied in
with my 2d Battalion across the valley, but not with the 3d Battalion. They are on the
best ground there.
The ROK police battalion which had been relieved on Hill 660 by the 2d Bn,
,
7th Cav evidently was not yet in position to back up the 3d Bn.
The urgency of the situation is reflected in orders thought to be from the
division command. As reported in the 3d Bn journal (which is not a running
log), the following message was received from the 8th Cav area, where divisional
and Eighth Army observers had gone some time before 1445 to observe the
attack:
ORO-T-261 211
senEY
UNCLASSIFIED
UNCLASSIFIEf'
SECAEi
121445: ... Call from Scrappy 6 [headquarters 8th Cav Regtj. Must hold objective.
If unit stops, men must dig in. They must hold the present position even though they are
going to get casualties. If you lose any ground, you have to retake it in the morning.
Plan all prearranged fire of organic attack or supporting weapons. Wanted to know num
ber of casualties. We have the use of the ROK Training Battalion but must treat them
with care.
6
The Third Assault
The long-awaited close air support finally appeared shortly after 1410.
Until about 1425, three F51 's dropped napalm and general-purpose bombs,
rocketed, and strafed. This time the planes flew up the ridge over the heads
of friendly troops and necessarily through the enemy mortar fire. They flew
unusually low, to begin strafing the ridge only 75 or 100 yd beyond the shallow,
where the Co L commander had his helmet blown off by the airwash. At least
two tanks of napalm were dropped. One went over the northeast end of the
hill; the other landed just over a little knoll about two-thirds of the way to
Knob 3. It is thought that the scar showing at this point in the photograph taken
two and one-half years later (Fig. 3) was burnt out by this hit. The spot was
also plotted at about this point by the artillery liaison officer, who, from the
shallow, watched the smoke cloud rise.
Various sources state that the napalm landed squarely amid a group of
enemy troops, but actual observations are not reported. The CO of Co L
subsequently did see enemy troops, but the Co I CO did not. It can be assumed
that they had indeed been taking cover behind the knoll and were flushed to the
left and rear. Firsthand reports on casualties caused by the bomb are lacking,
but circumstantial evidence allows them to have occurred; though many enemy
bodies were later found on top of the ridge, very few were found between Knob
2 and the point where the napalm landed.
Immediately after the air strike, the assault units were ordered to move
up, but lacking their original platoon leaders, they did not respond. The CO
of Co L then led the charge himself. He was accompanied by the CO of Co I
and about a dozen others. The Co L commander is quoted as urging, "Come
on up here where you can see them. There are lots of the bastards and you
can kill therid" The troops at Knob 2 then moved up over the crest in a general
advance.
It was a moment of emotional release. The yelling of the advancing
friendly troops could be heard back at Knob 1. The only enemy soldiers close
at hand were the wounded in the holes, and these were now killed. The Co I
CO separated from the CO of Co L and moved over to the right hunting for the
enemy, but found only about five enemy bodies. However, an intact prisoner
was brought in by one of the Co I men, who was crying for permission to kill
him. The prisoner was ordered taken to the rear. His interrogation appears
as Item 10, App A; the following extract reports his observations on the battle:
121100. 2d Battalion shelled by US artillery and mortars. Strafed and bombed by
US aircraft. Heaviest casualties from artillery. mortars. Number unknown. Air attack,
least effective. POW, unarmed, hid between boulders during engagement. 2d Battalion
units forced into disorganized retreat as a result of heavy US artillery and mortar shell
ing and small arms fire. POW captured immediately thereafter by US Army units.
212
ORO-T-261
UNCLASSIFIED
SECRET
UNCLASSIFIED
SEun
The disorganized retreat mentioned by the prisoner does not seem to have
occurred until Co I had finally broken enemy resistance and advanced in the
third assault. Not having been hit by napalm,' the enemy down on the slope in
Co I sector appears not to have been panicked by the final air strike, as witness
the prisoners' statement that the air attack was less effective than the other
supporting fires. While it was in progress, at least two men of Co I's 2d Plat,
which was as far as 150 yd down from the ridgeline, were hit by small-arms
fire. Furthermore, the enemy evidently offered active opposition, as noted in
the following incident recorded in App E:
Sgt Manuel of Company I charged a fanatical North Korean who was holding up two
squads of the 1st Platoon. Sgt Manuel killed the enemy soldier with a grenade, thus allow
ing the platoon to continue its advance. In this action Sgt Manuel received a wound in his
right leg and right hlp.
This actio!) evidently occurred during the third assault, for the sergeant's
medical tag states that he was wounded by mortar fire at 1500.
Friendly units continued to advance along the ridge at a moderate, steady
pace. Company L moved toward the right with Co I to avoid the 76-mm artillery
and (now, for the first time) automatic-weapons fire from Hill 570. Only meager
information is available about Co L. Midway along the ridge (the forward slope
of the little knoll is the probable site) the Co L CO and two others killed an
,
enemy soldier armed with an automatic weapon. As for the general enemy
reaction, the Co L CO later told the Co K CO that he saw the enemy slipping
down the slopes in all directions.' This observation must have been made from
Knob 3. It is circumstantial evidence for the effect of the napalm hit. The
Co I CO never saw any number of enemy because he never reached Knob 3.
Friendly forward elements were on Knob 3 at about 1530. Soon after, the
following message was received: "121545. From Artillery Liaison Officer
with Skirmish Blue: Objective hill (1156.5-1455.5) [Knob 3] ours." 1
For a period offifteen or twenty minutes during the final advance, continual
heavy explosions were heard from the northern or eastern slopes of the hill mass.
No cause has been reported. The explosions might have resulted from fire by the
jeep-mounted 75-mm recoilless rifles, which were moved along the trail at the
right of the hill mass to a point midway between Phase Line 2 and Phase Line 3
late in the battle, and then brought up- onto the objective when i ~ had been taken.
Enemy demolitions may just as well have been responsible.
When Co L and Co I advanced from Knob 2, the mortar fire did not follow
them but continued to fall in concentrations on the southeast and southwest
slopes of Knob 2. Moving up through this barrage to reinforce the advancing
assault companies, two platoons of Co K were hard hit. A sergeant of the
P&A Plat of the Bn Hq Co passed through the mortar fire at a late stage of the
battle to take ammunition to the forward units. As he moved along the ridge,.
he could hear enemy mortars beyond the north end of the hill firing regularly.
On the next trip up they were no longer to be heard. Later reconnaissance
showed that these had been 82-mm mortars (See Part n). '
When personnel of Co I moved up onto the crest of Knob 2, they saw napalm from the preattack air
strike still smoldering around several charred enemy bodies.
ORO-T-261 213
SEERET
UNCLASSIFIED
UNCLASSIFIED
SURtT
The 76-mm direct fire did follow the final advance, and at an undetermined
time the 120-mm mortars shifted their fire to cover the entire ridge from Knob
2 to Knob 3 with sp<;lradic fire. One notable burst of a round thought to be
smaller than 120-mm wounded the CO of Co I and an officer of Co K, killed
one enlisted man, and wounded two more. They were standing just nQrth of
the dip in the ridge between Knob 2 an!1 Knob 3. It is thought that the mortar
barrage on Knob 2 lifted after Co K had passed through it, probably shortly
before 1515. As noted above, the 52-mm mortars stopped firing around this
time and were later found to have been removed. Of the six 120-mm mortars
at Site 4 (Fig. 8) which up to now had been firing at Knob 2, three now distrib
uted their fire between Knob 2 and Knob 3, and the other three apparently sought
targets elsewhere, for the 8th Cav reported the following observation; ~ 1 2 1 5 1 5 .
Scrappy Red 5 [Ex 0, 1st Bn, 8th Cav 1reports that all forward observers
report 120mm mortars are located at 1152.6-1457.5. Further reports that all
movement draws fire from these mortars."'
As soon as the objective was secured, a check was made on the person
nel of Co L and Co I. Only 80 men were present. Others then straggled In,
a squad at a time, but it was an hour or mare before all remaining effectives
showed up.
Movement of the Support Comp'any and Battalion OP
When Knob 1 was captured at the outset of the battle, the Co K 1st Plat
moved up to this location, and the rest of the company, 8tm strung out down
the ridge in tile, advanced accordingly. Although the Wpn Plat leader went
forward to Knob 1, the platoon itself, which had been assigned the GO-mm
mortars of Co L and Co I in addition to Its own, presumably remained to the
rear of Knob 0 to use without further transport the large stock of aO-mm
mortar shells left by the 8th Cav.
During the later stages of the battle, the trallihg platoons of Co K were
brought further forward. After Co L and Co I had advanced from Knob 2 an
the third assault, Co K was ordered to reinforce them on the objective. Its
2d Plat moved down and up the saddle to the left of the ridgeline , skirted Knob
2 to the left, and finally went up onto the ridge midway between Knobs 2 and 3.
AdvanCing abreast, the lst Plat moved rather far down from the ridgeline on
the right, skirted the right side of Knob 2, and joined the 2d Plat up on the
objective. The 3d Plat followed later.
Though the course of the two platoons took them somewhat out of the
continuing mortar concentrations, and though they moved through the barrage
as quickly as possible, casualties were high. As shown in the later subsectionon
friendly casualties in this attack, In terms of the number of men exposed, the
rate was 50 or 60 percent of that of Co L and Co I, although the durations of
exposure were in no way comparable. The reason is thought to be that most
men were hit while standing up to move rather than when pinned to the ground.
With the advance of Co L and Co I In the third assault, the battallon OP
was moved up to a position on the left side of the nose of Knob 2. It never
actually was opened here, but, with the continued advance of forward elements,
almost at once was moved further up onto the right side of the objective.
214
ORO-T-261
B(Rf.T
UNCLASSIFIED
UNCLASSIFIED
COMMUNICATION DURING COMBAT
No special study has been made of the system of electrical communica
tion connecting the various units of the 3d Bn, 7th Cav Regt and linking it with
its supporting units. This system is assumed to have been normal except for
impairment sustained during the battle. In this section, information from
various sources is brought together on the functioning of the communications
system during the assault phases.
The initial enemy barrage at 1140 crippled the electrical communic;ation
of the assault companies. The Co I radio operator was wounded, and the com
pany lost both its radio link with the rear and its walkie-talkie. communication
with its forward elements. Company L also lost all walkie-talkie linkage at
this time, but retained radio communication throughout the battle. Loss of
contact with forward units first made communication by runner necessary,
and later, when direct control of the troops became essential, required the
physical presence of the two company commanders at the front with the assault
troops. Some observation and liaison personnel also had to move up.
At an unknown time early in the battle, Co M also lost radio contact with
its forward observers. Calls for 81-mm mortar fire and target information
had to be passed along a circuit consisting of three or four links.
Figure 13 is a diagram of available information on communication links.
As indicated, the critical breaks were the connections with the most forward
elements, as only they could have knowledge of enemy movements beyond the
crest of Knob 2. Theoretically, the remaining open circuits to the several
forward observers could have taken over, but as personnel in the saddle were
pinned down by intense mortar fire, communication from group to group was
presumably at a minimum. Furthermore, the intermingling of units fairly
early in the battle, and the casualties sustained, confused both the lines of
communication and the situation to be reported.
For fire support the battalion had to rely chiefly on its own 81-mm mor
tars and secondarily on the 105-mm howitzers of the 77th FA Bn. Ground
support aircraft were also to be on call. Technical or human failure to com
municate information from the forward pOSitions rendered the attached heavy
mortars of the 8th Cav Regt ineffective. The massed 60-mm mortars. of the
attacking battalion were far enough forward (to the rear of Knob 1) to operate
effectively with whatever channels of communication were open to them, and,
by direct observation, even without them.
Why the Co M radio went out is not known. In place of direct radio con
tact with its forward observers, the relay circuit of one radio and two telephone
links, as shown in Fig. 13, had to be used. Use of the telephone between the
battalion OP and CP doubtless contributed to the overloading of this line.
Locating one section of 81-mm mortars up on the ridge may have introduced
still another telephone connection.
So far as is known, artillery communications remained unimpaired. The
physical proximity of the artillery and battalion OPs permitted the artillery
liaison officer to keep in touch with the battalion command and perhaps to
inform it about forward infantry elements via the artillery observers. Since
the liaison officer called in the fire of the 9th FA Bn as well as that of his own
77th FA Bn, it is assumed that both a direct link and one to division artillery
ORO-T-261
215
SEEREJ
UNCLASSIFIED
c::
:z:
C")
r
::t:
V)
V)
...,
rn
o
SADOLE KNOB I KNOB 0 BASE OF HILL TO-THE REAR

(Counterbattery)
t
99th FA Bn Div Arty
8thCay Regt/
*Radio aI-rom Mort fwd Sec
l
__ ___ 17" FABn
-n. Mort (0;"" '""'''')
jP,,""'" ReaJSec

/


Bn OP TeleptKlne Bn CP
*"


Aid"Sia,'etc.
tpelson31
6O-mm Mort Tk Psn
*?_41.;...rt
CoSC1:K
8th Cay Regt
*Communication losl
X Comll1lnicalion.ineffeclive with dispersed,
pinnerl-down 1l00ps
Fig. 13-Cammunicatian on Hill Mass 314 during assaults on Knob 2 (linkage to reOr incompletely shown).
@
o
I
..,
.,
I
'"
.....
Co LCP
t
UNCLASSIFIED
stCRET
were involved. It cannot be assumed, however, that a channel was held open
to the 99th FA Bn, which was at the time more or less active in support of the
8th Cav. Clearance with the 99th FA Bn was required for the 77th FA Bn to
fire in the former's zone. It is known that after some delay this clearance
was obtained, and that the 77th FA Bn did fire at targets on Hill 570. Whether
roundabout communication was responsible for some of the delay can only be
considered possible.
The request for an air strike had to go over the interbattalion telephone
line. Some delay may have occurred here in addition to the inevitable series
of delays in approving, arranging and dispatching a flight.
Even if most of the messages relayed by the tanks were routine calls for
ammunition, litters, etc., it is clear that diversion of some of the message
load from the battalion telephones must have expedited such emergency calls
as requests for protective mortar fire at specified coordinates.
Reporting of the capture of Hill Mass 314 to the higher echelons was neither
complete nor accurate. Broken lines of communication and the absence of a
corps command level were doubtless contributing factors. While there is no
reason to suspect lack of full information within the 1st Cav Div, it is curious
that in all the entries about the 3d Bn, 7th Cav in the division journal
1
on 12
September, Hill 314 is never mentioned, nOr is a north-south coordinate given
which would identify the battalion's objective. The misinformation of Army
headquarters may have begun with the following message received at EUSAK,
G-3 Sec (now at Pusan): "0630 [12 Sep]. [from] 1st Cav. 7th Cav, 3d Battalion
will attack at 1030, airstrike 0920. Objective to seize and hold area: 1153.9
1456.9, 1154.6-1456.0, "21
These coordinates define the enemy-held portion of Hill Mass 570, which
the 3d Bn, 7th Cav was scheduled to attack (until plans were changed in the
early hours of 12 September).
The next report to the Army G-3 section apparently represents the message
from the artillery reproduced in App E. It states: "1620. From 1st Cav Div...
121615. We believe that the attack by the 3d Battalion, 7th Cavalry is successful.
Reported by FO through arty channels. No confirmation from the ground as yet. "21
The first of the two following reports is late, the second again wrongly
identifies the hill mass.
1750... 1700. 3d Battalion. 7th Cavalry attack on Hill 314 commenced at 121100
reached a general line, grid line 55, at 1300. A counterattack held them up in this
position.
1845. From 1st Cav... at 1845. The 3d Battalion. 7th Cavalry took its objective
apprOximately 1600l. Ammo and supplies are being sent up to them and they are holding
well. Note: the objective is 1154-1457. 1155-1456. 1155-1455.
21
No later reports were logged by the Army 0-3 section. Repetition of the
wrong coordinates of Hill Mass 314 appears with other errors ("moderate
enemy resistance") in the summary of the operations in the period 1200-2400,
12 September given in EUSAK Periodic Operations Report No. 186. The formal
summary prepared for the briefing of the commanding general at 0800, 13
September repeats the errors. No mention at all is made of the 7th or 8th Cav
Regt in the Eighth Army War Diary for 12 Sep 50.
ORO-T-261 217
S!UET UNCLASSIFIED
SECREf
UNCLASSIFIED
SYNTHESIS: GRAPHIC TIMETABLE OF ACTION
AT THE OBJECTIVE ON HILL MASS 314
This timetable presents a consistent picture at
the platoon level of action at the objective throughout
the battle. The fragmentary and often ambiguous evi
dence considered in the preceding narrative of the
action has been joined in a pattern with the aid of the
simplest likely assumptions that would account forthe
continuity of events. In detail, the table therefore
represents hypothesis and interpretation as well as
established fact. In general outline, its objectivity
and reliability are much higher.
ORO-T-261 219
SEun
UNCLASSIFIED
SECRET
UNCLASSIFIED
MlNUTE-BY-MINUTE ACCOUNTS
OF THE ACTION FOR 1135 - 1530
The action is shown as the clear-cut sequence of phases that appears to
have unfolded. The estimated time at which each phase began heads each sec
tion of the table, and a diagram is given that shows the approximate positions
of friendly and enemy units at this time. Subsequent important movements are
pointed out in the left-hand column of text and indicated in the diagram by a
black arrow with each unit. Friendly units are circled in the diagrams, and
given with the platoon to the left of a slanted line, the company to the right.
The right-hand column of text lists important concentratioris of fire, with
occasional notations of the casualties they produced. Enemy fires are shown
in red in the diagram, friendly in blue. Concentrations are indicated by cross
hatched rectangles, with the type of fire designated by the caliber. For the
120-mm and 82-mm mortars each rectangle represents the 90 percent disper
sion pattern of rounds aimed at its center. Concerted small-arms fires and
the throwing of grenades are symbolized by zigzag arrows. Scattered or spo
radic fire and concentrations without a known definite target do not appear, even
though these may have been important or had a significant cumulative effect.
The intensity of fire is considered only qualitatively here. The symbols for
the friendly platoons may be given quantitative meaning by the company strength
and casualty figures presented elsewhere, but the enemy groupings shown imply
no information whatsoever as to the number of men at any position.
The deployment of units sketched on tpe diagram and the areas covered
by the indicated concentrations of fire are meant to be roughly to scale. The
diagram is derived from the map used throughout this study (Fig. 2). Additional
terrain features made known by various sources and important to the action
have been taken into account. Some of these on the west side of the hill mass
are visible in Fig. 3, but it must be remembered that east-west dimensions
appear highly foreshortened. Close-up photographs of the east side of the hill
around Knob 2 are given in App E. As small irregularities and variations in
terrain were highly important, the axes of ridges and fingers have been drawn
in, and these prominences and intervening draws are suggested by the shading
of the contour lines, as if the sun were at the enemy observation post on Hill
570. The scale of the original map is 1:50,000, and the contour lines have
lost their relative accuracy as a result of the great enlargement. This lack of
topographic precision is a weakness of the present analysis. It should also be
noted that only the battleground at the crown of the objective is included within
the frame of the diagrams. Enemy units possibly down .on the side or rear
slopes do not appear, nor are the friendly air strikes on the northeast side of
the hill during the battle represented.
In the table the tiine estimates previously derived by reconciling all
available information have been checked in addition against the speeds of move
ment possible. Except for the intentionally rapid moves down into the saddle
from Knob 1, at a speed here taken to be 2 mph, no advance or withdrawal
plotted exceeds an average speed of 1 mph. The excessive back-and-forth
movement noted in some reports and the frequently mentioried final rout and
flight of the enemy hav" been ruled out as impossible, for this as well as other
reasons.
ORO-T-261
221
-SEEKEY
UNCLASSIFIED
UNCLASSIfiED
-SUitT
The dotted arrows between the columns of text point out the effect of
particular friendly or enemy moves in stimulating opposing fire; solid arrows
point out moves following directly from fire received.
222 ORO-T-261
UNCLASSIFIED
@
1135
o
I
...,
Situation: Preparatory fires, napalm hits have cleared enemy troops from nose
I
and crest of Knob 2. After capture of Knob I, Co L and Co I reorganize there for
."
'"
main assault. Though subjected to aI-mm mortar fire distributed over objective,
.....
enemy units on slopes prepare for defense.
Friendly units jump off. Forward Friendly overhead covering fire from
platoons move down into saddle as light machine guns. PossiblyaI-mm
rapidly as possible. and 60-mm mortar fire on objective.
Enemy units move up on reverse
slopes of fingers on either side of
northern slope of saddle.
Enemy outposts amid rocks in sad
dle deliver scattered rifle fire.
seAL E IN YARDS
o 100 200 300 400
I I I I
~ 1138
o
I
~ Friendly units advance. Deployment
~ in forward platoons: two squads for
Q) ward on skirmish line, third squad
I-l to rear in wedge focmatian. Attack
front is 300 yd.
Enemy opens up with concentrated
fire as friendly units reach bottom
of saddle, as if this were final pro
tective line-82-mm and 120-mm
mortar fire. rifle and automatic
weapon fire. Radio and walkie
talkie communication lost in Co I;
walkie'-talkie communication lost
in Co L.
Friendly units move forward rapidly
to get 011t of mortar concentrations.
I
c
z
c C'?
r-
Z
::r::
C'?
en
r-
en
en
""
"'T1
::r::
en en
""
rro
"'T1
0
rro
0
@
1145
o
I
>-i
I
'"
'" ....
Friendly lead platoons, already
somewhat disorganized by mortar/
barrage, now are stopped, move OVer
toward the right to regroup.
Large enemy group moves to Knob 2
from reverse slope of finger "to left. ",
Enemy group dispersed. Counter /
attack forestalled.
Enemy groups firing from underbrush
atop fingers on left and right of
saddle pull back slightly to reverse
slopes, losing fields of fire.
c::
z
CJ
r

en
en
"'T'1
I"T'1
o
Enemy small-arms fire intensifies
as range to friendly assault units
Closes. Leader of 2d Plat, Co I a
casualty to enemy machine gun.
'\, Friendly BI-rom mortar concentration
down on enemy grouping.
r'dove had been observed from
Knob 1.
c::
z
CJ
r

en
en
-n
I"T'1
o
SCALE IN YARDS
o 100 200 300 400
@
1150
o
I
>-l Friendly attack resumed on initia
I
tive of leaders of forward platoons.
~
.... Enemy mortar concentration follows
first friendly moves, this time is
spread accurately OVer and within
area of saddle crossed by attack,
falls on support units moving down
from Knob 1.
Enemy 76-mm direct fire from SP
gun falls on west side of Knob l.
/I
Friendly heavy machine gun fire
from Knob 1 for first time directed
i
aterests of fingers aheadol assault.
Enemy group on left pinned down on
reverse slope of finger.
Enemy group on finger at right seen
withdrawing piecemeal toward north
east side of Knob 2 by Co I flank
(2d) platoon.
c::
2:
C"?
r-
):,.
en
en
:::!J
,...,
o
I I I I I
1155
o
Enemy group moves up onto north
east side of Knob 2 from slope on\.
e right side. \. \
Friendly spearhead (3d Plat, L Co) small.arms fire from close
moves into gap in center and onto range,possible hand-te-hand combat.
nose of Knob 2. Leader of 1st Plat, Co I a casualty. i
Enemy fire from dense underbrush
on finger to left.
Friendly assault units driven back Enemy on finger to left throws con
disorganized. / cussion grenades from underbrush.
Platoons to. left pinned down,
"stunned,"
The 3d Plat of Co I moves over to
back up disorganized 3d Plat of
L Co.
The Wpn Plat, Co I moves up on
right to back up withdrawing 1st
Plat, Co I.
The COs of Co L and Co I move up
from Knob 1 to units pinned down
at left.
81-mm mortar fire brought
. / down on northeaRt side of Knob 2.
c
z
Enemy group pushed back down
C"?
slope from northeast side of Knob
r
:>
2. Casualties unknown.
en
en
Friendly IOS-mm artillery fire hits /'
""T1
/ objective ridge for first time. CO/y
Enemy units discouraged from high-explosive fire from o
/-x
m
jng back onto ridge and Knob 2. tanks on western slopes of ridge
north of Knob 2.
@
o
,
>-3
,
""
.....
'"
=
:z:
C?
r
::t=
en
en
..."
rn
o
1215
Friendly troops .
follow example down on left
mentation gr , throw frag
enades .
to left. mto underbrush
/
Enemy <:rco f left with dr t> up on inger to
aws to next f"mger.
\
\
\
':.uEnemy malar f'
'k"II I Ire moves d
I seader of 3d up sa die
leader of 3d PI Plat, Co Land
at Co I
companies s . if en of both
omewhat . k
.. " to fall back int h pamc ed, tend
nendly units urged Caward 0 t e saddle, must 'e
/
m move forward "
orlar fire re" out of ,j "
2d Plat of' C attack. The r
flank; 1st Pla
o
, "' moves around left
Co I" I
the right flank' 10 earl around
SCALE IN YARDS
o
I
100
I
200
I
300
c::
:z:
C?
r
:x:.
en
en
..."
rn
o
1230
Friendly second assault on Knob 2.
The 2d Plat, Co L moves up across,
finger to the left. '\
,
Enemy 76-mm direct fire from SP
gun concentrated on north slope of
finger crossed by 2d Plat.
Enemy troops on woooeri. higher fin
ger along Phase Line 2 to the left,
make limited move toward Knob 2. ""
"-
Enemy small-arms and grenade fire.
The 2d Plat of Co L stopped short/
of nose of Knob 2 on left side.
Enemy group chased down draw be
yond Phase Line 2 to left and out
of the battle. Casualties unknown.
The 1st Plat of Co L moves up on
nose of Knob 2 without opposition.
The COs of Co L and Co I and other
personnel move into "the shallow,"
just short of the crest of Knob 2.
From here for the first time there is
observat ion of the top of the
beyond Knob 2, but not much more.
Friendly units on right side advance
to finger along Line 2. "
The 1st and 2d Plats of Co I pinnefl /
down. No cover on south side of
nitch.
Friendly 8I-mm mortar concentration
/ beyond Phase Line 2 on left side.
Friendly lO5-mm artillery works over
objective ridge.
"
fire from dense underbrush
across ditch that runs down axis of
finger. Leader of 2d Plat, Co L hit
by smallarms fire, keeps fighting.
c
z
C"?
r
):>
(/)
(/)
""T1
",
o
Friendly assault units pinned down.
Expect and wait for close air sup
/
port. Troops in draws on left side
not visible from II ill 570 when down
on ground, can be seen when stand
ing to move.
1300
Enemy mortar fire moves further up
hut stays just short of crest of Knob
2 at Phase Line 2. Fire concentrated
on nose in four areas: the two on
the right receive 120mm fire in 3-rd
salvos each; the two on the left
thought to receive 82mm fire. Fire
sporadic, responds to friendly move
ments, takes casualties at average
rate of one every two minutes.
Leade, of 2d Plat, L Co, a!ceady
wounded, killed by mortar fire as
he walks back for treatment.
Friendly rifle fire by dozen men in
shallow at the few visible targets.
Enemy observer down ridge hit while
trying to get back to rear.
SCALE IN YARDS
200 300 400
I I
I
c
z
c-:>
r-
;J>
en
en
...,
!T1
o
1410
j
l
Friendly close-support air strike for
first time on top of objective. Three
F51's rocket, strafe, drop napalm
and general-purpose bombs. Fly low
and from south to north, start stra(
ing 75 to 100 yd beyond friendly per
sonnel in shallow. One napalm load
goes over ridge, effect unknown;
other napalm lands amid group of
enemy behind knoH atop ridge.
Enemy units if present on far t:;!ast
slope may have been flushed by
long napalm drop.
Enemy troops on top of ridge dis
persed by simultaneous s..econd
napalm drop to rear and down left
slopes.
Enemy sporadic small-arms fire from
underbrush down right slope' of
Knob 2 causes some casualties in
2d Plat, Co I.
Enemy mortar fire continues on con
centrations as before. Some 120-mm
fire lengthens in range, hits some
men already wounded and walking
back down evacuation trail.
SCALE IN YARDS
o 100 200 300 400
@ 1430
o
.!,
I Friendly third assault ordered by
t>:I CO. Co L; troops do not respond.
e Situation: units sJYead out of sight
of each other. walkie-talkies out,
company and platoon boundaries
blurred anclelements somewhat inter
mingled, four or five of original pla
toon leaders lost, many men of units
now casualties, enemy mortar fire
responds to movement. The COs of
Companies L and I, with about a
dozen men from around shallow.
charge over crest of Knob 2 (Phase
Line 2), Example and urging inspire
general body of assault troops, who,
greatly excited, follow leaders.
Enemy fire (?) from automatic weap
on halfway to Knob 3 atop ridge.
Enemy small-arms fire or grenades
offer some opposition to advancing
1st and 2d Plats, Co I.
Enemy mortar fire keeps up barrage
c:
on nose of Knob 2. Casualties in
Z
a
flicted during advance of
.
units not .
:J>
(/)
The 1st and 2d Plats of Co K move
(/)
out from Knob 1 to support advancing
."
assault companies .
.,.,
C1
I I I I I
1500

o
Friendly units advance along ridge
I at steady. moderate rate (average,\
forward movement about mph). "
" Enemy 76-mm fire from SP gun falls
/' along left side of ridge.
Advancing units of Co L move automatic-weapons fire te
to right side of ridge. ceived for first time from Hill 570.
CO of Co I joins his units on right
side.
Company I captures enemy soldier,
unarmed and hiding among rocks.
1st and 2d Plats of Co K move
around flanks of Knob 2 as rapidly"
,
as pOSSible.
'\
\ Enemy mortars maintain barrage on
nose of Knob 2. The 82-mrn mortars
at base of BiB 314 to the north are
still firing. !\Iortar fire inflicts rela
tively high casualties on Co K.
c: Continual heavy explosions down on
z northeastern slope of hill mass.
C")
r
Cause not determined. Might be ef

fect of friendly jeep-mounted 7S-mm
en
en
"
recoilless rifles that had
around trail on right side of
gone
hill.
rr1
o
SCALE IN YARDS
o 100 200 300 400
@
1530
o
~
I Friendly advance elements reach
I:\:) Knob 3.
0>
.....
Enemy troops withdraw toward CP
.on finger northwest of Knob 3 and
beyond. The CO of Co L,at Knob 3,
sees enemy soldiers "slipping down
slopes in all directions." Roughly
40 percent of enemy troops originally
on hill withdraw, with weapons in
good order.
Enemy 82-mrn mortars (four) at base
of hilI on north evacuated. About
600rd of ammunition left at position.'"
"
~ E n e m y 82-mm mortar fire ceases.
Enemy 120-mm mortar fire leaves
Knob 2, is now distributed along
eastern side of ridge.
Enemy 60-mm (US) mortar fire re
ceived sporadically from Bill 570.
One round lands on left side of
ridge in drop midway between Knob
2 and Knob 3; CO of Co I and four
others casuahies.
Companies L. I and K organize per Friendly IOS-mm artillery fire be
imeter defense. Count shows 80 ef yond hill mass on retreating enemy.
fectives in Co L and I together.
Squad-size groups straggle in from
left s lopes in next hour.
I I I I I
SEERtT"
UNCLASSIFIED
FRIENDLY CASUALTIES AND THE EFFECTIVENESS
OF ENEMY MORTAR FIRE IN THE CAPTURE OF HILL MASS 314
The simplicity of the enemy weapons system and the open exposure of
friendly assault troops to enemy observation and fire invite an attempt to link
friendly casualties with the weapons responsible. This section consists of a
discussion and some relation of these tWo aspects of the battle on Hill.Mass 314.
FRIENDLY CASUALTIES
The Master CasuallY List
It was originally intended to use the data reported in a previous study"
in establishing a master list of casualties sustained in the capture of Hill Mass
314. These data were based on a compilation run off by the Machine Records
Unit in Tokyo in February 1951. Further research showed, however, that only
about one-third of the total casualties were included. As a check On the record
system, a new run was made two years later (February 1953) from the casualty
cards on file in Washington, D. C. By using the results from both runs, a
fairly complete list (about 90 percent, see below) of the casualties that occurred
on 12 September in the fight for Hill 314 was obtained.' Information on the
ultimate status of casualties was obtained from the new list.
Under the casualty-reporting system in use at the time of the action, the
primary source of information on individual casualties was the morning report
of each company.t . Therefore, in the present study, a master list of casualties
in the 3d Bn, 7th Cav on 12 September had to be based on entries in the morn
ing reports of all its companies in a period including this date. The task was
somewhat difficult because the battalion had been in intense combat and had
suffered many casualties. The status of a number of men was established only
SOme days after the battle, and was perhaps incompletely and inaccurately re
ported because of the loss of trained clerical personnel. Information was re
ported as it became known. For the first mention of casualties that occurred
on 12 September one must consult morning reports dated as late as 28 Septem
ber. Yet the final accounting seems substantially complete, since only a few
names on the final master list are not to be found in some morning report.
*The following over-all comparison may be made for the 7th Cay negt in t h ~ period 1-23 Sep 50: machine
run of February 1951. total casualties 77.0; machine run of February 1953. total casualties 1012.
tFrom mid-September 1952 to the end of the fighting in 1953, detailed casualty reports by eyewitnesses
were used on a comprehensive trial basis in addition to the morning reports.
ORO-T-261 247
SEUET
UNCLASSIFIED
UNCLASSIFIED
After Hill Mass 314 had been captured, an enemy counterattack was con
sidered imminent. There was initially little opportunity to search for the bodies
of many of the KIA. The counterattack did not materialize, however, and by
15 September all bodies had been recovered and no men were any longer miss
ing from the action. For lack of definite Information, some men whose bodies
had been picked up on 13, 14, or 15 September were listed as casualties on
13 September, and so carried in all official records. Each day after 12 Septem
ber, the 3d Bn did sustain some casualties, almost all of them from mortar
fire, though one man was killed by an enemy soldier who had been in hiding
on the hill. Since little patrolling was done, it is likely that all the casualties
during this period occurred In defensive positions and were observed by sur
viving eyewitnesses. For compiling the master list, the following rules were
therefore arbitrarily adopted with regard to casualties reported for 13 September
when there was no evidence for this or another date: (a) all WIA carried, as
reported, on 13 September; (b) all KIA attributed to the attack on 12 September.
Examination of detailed casualty records necessitated some further cor
rections. One or two cases were simple mistakes, as in the transcription of
the date from an original to a supplemental medical tag. More were the result
of incorrect dating of what was obviously the battle on Hill Mass 314 by the
casualty himself when, after the medical tag had apparently been lost, a cas
ualty report was made out from his statements to medical personnel at a rear
hospital.
No list was available of casualties among ROKA personnel integrated into
the various companies of the 3d Bn. The morning reports of the 3d Bn Hq and
Hq Co (Co H3) do list the K numbers of those Koreans assigned to the 3d Bn and
those who subsequently became casualties. They do not, however, mention the
companies to which these troops were assigned, nor, with several isolated
exceptions, do the morning reports of the several companies. Additional re
search therefore had to be done on the medical records of the 1st Cav Dlv. At
the aid station during the battle tags were made out for Korean personnel just
as for US casualties. Though the tags were not available, data from them had
been entered on the A&D sheets of the clearing company (15th Med Bn). In the
few cases where the company affiliation was not given, it could be deduced from
the K number, as described below.
Table E1 is the resulting master list of the 150 casualties considered to
have been sustained by the 3d Bn in the capture of Hill 314. Name, serial
number, ra...n.k, a..nd addition::ll casualty data are given for each man. The indi
viduals are grouped by company; within each company they are ordered first
chronologically for those casualties occurring at a known time and then alpha
betically for those occurring at an unknown time. Casualties among integrated
Korean personnel are Included in the same way , to give a comprehensive table
for ihe entire battalion.
Of the 129 US casualties listed in Table E1, 13 are not to be found in the
later machine tabulation (February 1953) for any date In September 1950 (except
one man listed a casualty on 2 September, apparently for a previous wound).
All are mentioned In the morning reports. The order numbers of these individ
uals in Table El are: 8, 15,48, 50, 52, 75, 78, 88, 103, 109, 114, 129, and 148.
Three of them were killed and one died of wounds. Medical data available on
some of the remaining nine cases indicate that these too were not, in general,
248 ORO-T-261
SKIH
UNCLASSIFIED
SKRfT
UNCLASSIFIED
trivial cases. One of the men (No. 148) was interviewed. He said he had been
rather badly wounded in the battle for Hill Mass 314.
One man who was killed (No. 62) was induded in the later machine run
but not in any morning report. One or two more cases of omission are sus
pected. Also, one mention in a morning report was found to be a typographical
or other error, for no similar name and similar serial number could be found
in Army records. As company rosters for August and September 1950 are
missing from AG records, this case could not be corrected and included. If
feasible, a check of US personnel on the master list against the A&D sheets of
the 15th Med Bn Clr Co and the casualty cards of the 1st Cav Div, as made for
the far smaller sample of attached Korean personnel, would have been desirable.
Only one Korean casualty (K 1107187) listed in the A&D sheets of the clearing
company is not mentioned in the morning reports. This soldier was hit in the ear
by a mortar fragment at 1400 on 12 September. Since the disability may not have
been incapacitating, even though the individual was sent to the clearing company
for treatment, this casualty has not been included in the master list.
Table E 1 is considered a nearly complete list of men who really became
noneffective on the battlefield as a result of enemy action. The former surgeon
of the 3d Bn, 7th Cav stated in a recent communication that an effort was made
to report as casualties all men entitled to the Purple Heart. The analysis
below indicates that all the reported casualties were treated at the aid station.
A survey of the diagnoses reveals very few cases that would seem from the
description to be trivial, as a finger wounded by a shell fragment. Self-treated
wounds and those requiring only first aid therefore are not represented in any
of the casualty data. According to the medic'allog of the 3d Bn, many men with
wounds of a minor nature did not turn in for attention until after the battle.
These additional unreported casualties could account for several estimates
that place total casualties in the battalion as high as 190. The neuropsychiatric
cases could also have contributed to these estimates.
The medical log states that in the 30-hr period after 1100 on 12 September
a total of 120 casualties was treated. On the master list 113 US and ROKA
personnel of the 3d Bn are carried as WIA after this hour on 12 September.
Of these, a few, say 5, came back through the 8th Cav aid station, leaving
about 108 as the number treated at the aid station of the 3d Sn, 7th Cav. The
following additional casualties were sustained by the 3d Bn, during the 30-hr
period: one man was hit by mortar fire on thE' 12th while reporting sick at
the aid station on 13 September, one man was killed, and five men were wounded.
In addition, three medical aidmen were wounded during the battle and one
more was killed. From these figures, the total number of casualties treated.
at the 3d Sn aid station would be 122.
Casualties, Strengths, and Casualty Rates
Table 13 gives a breakdown of casualties in the capture of Hill 314 by
company, nationality (US or attached ROKA personnel), and type (KIA or WIA).
The distinction between SWA (seriously wounded in action) and LWA (slightly
wounded in action) is not very meaningful from the operational standpoint and
has been abandoned here in favor of the general term WIA (wounded in action).
ORO-T-261 249
SEERET
UNCLASSIFIED
UNCLASSifiED
SHRH
It is of interest that in the battalion as a whole 25 percent (23 percent, if
DOWs-dled of wounds-are excluded) of total casualties were KIA. This per
centage is somewhat higher than the prevailing average figure. The former
surgeon of the 3d Bn reported that some men who could not be evacuated from
TABLE 13
CASUALTIES IN THE 3D BN, 7TH CA V REGT IN THE CAPTURE
OF HILL MASS 314, 12 SEP 50
Unit
US personnel
Integrated ROKA
personnel
Entire company
KIAa I WIA I Total KIA I VIlA I Total KIA
I \VIA I Total I %KIA
Co L 21 37 58 0 4 4 21 41 62 34
Co I 9 41 50 5 6 10 46 56 16
Co K 2 14 16 3 4 7 5 18 23 22
Co M 0 4 4 0 0 0 0 4 4 0
Co 113 0 I 2 2 4 2 3 5 40
Entire 3d Bn 32 97 129
b
6 15 21 38 112 150 25
aThree cases of DOW included with KIA.
bAdditional data, 129 US casualties:
W1A KIA (UOVl)
Not listed in machine run
10 3
Listed in machine run 87 29
Of those listed:
Eventually returned to duty 53(61%)
Evacuated 34(39%)
the hill were wounded again where they lay (one officer lay wounded 12 hr and
was wounded a second time by mortar fire). Also, badly wounded men who
might otherwise have been saved died during the long trip down the trail to the
aid station. The percentage of fatalities therefore appears to be a significant
indication of the validity of the casualty data. In interpreting the battle, the
high percentage of KIA among the casualties of Co L should be considered.
TABLE 14
FRIENDLY STRENGTHS, CASUALTIES, AND CASUALTY RATES IN THE CAPTURE
OF HILL MASS 314 BY THE 3D BN, 7TH CA V REGT, 12 SEP 50
Entire company US personnel Integrated ROKA personnel
Unit
Strength ICasualties I Cas% Strengtha Icasualties I Cas % Strength ICasualties! Cas %
rate, rate, " rate, "
Co L 144 58 40.0 36 4 11.0 180 62 34.0
Co I 142 50 35.0 29 6 21.0 17l 56 33.0
Co K 148 16 11.0 33 7 21.0 181 '23 13.0
Co M 113 4 3.5 12 0 0.0 125 4 3.0
Co 1I3 93 I 1.0 35 4 11.0 12B 5 4.0
Entire 3d On 640 129 20.2 145 21 14.5 785 ISO 19.1
apresent for duty at close of 11 September, as stated in morning reports covering that date.
250 ORO-T-261
UNCLASSifiED &EERE-T
UNCLASSIFIED
SECRET
TABLE 15
BALANCE SHEET OF ROKA PERSONNEL ATTACHED TO THE 3D BN,
7TH CAY REGT
a
Day covered
Losses Balance, end of day
by morning
report
Additions
K number I Company 1
casua ty
P'''"ot I I f d In bosp Cum KIA
or uty
7 Sep K 1105101 to
K 1105200 100 0 0
9 Sep K 1107151 to
K l107200
b
K 1107196
Me
NBC 149 0
II Sep K 1105155 K SWA
K 1105173 K SWA
K 1105193 I LWA
K 1107190 M LWA 145 5 0
12 Sep K 1105117 H3 LWA
K 1105135 113 LWA
K llOSI37 K LWA
K 1105146 K LWA
K 1105164 K LWA
K llO5167 K LWA
K 1105175 I LWA
K IIOSl82 I LWA
K 1105185 I LWA
K 1105187 I LWA
K 1105198 I LWA
K 1107171 L' LWA
K n07ln L' LWA
K 1107185 L' LWA
K 1107186 L LWA 130 20 0
13 Sep K 1105125 113 KIA
K 1105127 H3 SWA
K 1105128 H3 SWA
K 1105140 K KIA
K 1105147 K' NBC
K 1105148 K KIA
K 1105163 K KIA
K 1105180 I LIA
K 1105197 I KIA
K 1107181 113 KIA 120 24 6
14 Sep K 1107196 Kl105122 H3
c
KIA
(Returned to
duty) K 1105188 I' KIA
Kll05148 K' NBC 118 24 8
15 Sep K llO5122(?)
I(?)C
KIA
K 1107163 L' NBC
K 1107176 L' NBC 115 26 9
16 Sep K 1121151 to
K 1121200 165 26 9
19Sep K 1123351 to
K 1123400 2lS
d
26
d
9
aData. mainly trom morning reports of Hq and Hq Co, 3d On. Dates of casualty not changed hut
several K numbers corrected.
bDate of attachment not specified, must have been on or before 9 September.
cCompany deduced from K number.
dCheck of balance: Figures of 215 present for duty and 26 in hospital on 19 September check
IIlOl'ning report statement that as of this date 241 all enlisted men, were attached to the
battalion.
ORO-T-261
251
UNCLASSiFIED
SHAn
UNCLASSIFIED
SEEREJ
The numbers of US personnel present for duty in each company on 12
September are readily available from the morning reports. The figures are
given in Table 14. Since the 3d Bn had been in reserve, there is no reason to
doubt the accuracy of the figures for the close of 11 September, and these are
taken to represent strengths prior to the attack on Hill Mass 314. The 640 US
personnel represent 92 percent of the battalion's assigned strength, the remain
der being on the sick list or missing. The percentages for the several compa
nies are about the same.
It is more difficult to establish the exact number of integrated ROKA
personnel present on 12 September. Table 15 is a balance sheet on these troops
constructed from various morning reports of the 3d Bn Hq and Hq Co. On or
about 7 September, 150 ROKA enlisted men were attached, and later in the month
100 more followed. The daily attrition of these personnel is shown in the table
and agrees with the count of 241 men remaining with the battalion on 19 Septem
ber. The numbers of men present for duty and in the Korean hospital near
Taegu on each day have been computed from the daily entries in the reports.
As for US personnel, the dating is doubtful. No KIA are listed for 12 September.
Under the rule stated above, all of the KIA listed for 13 September have been
attributed to 12 September.
The ROKA personnel attached to the 8th Cav Regt were assigned directly
to the companies according to their K numbers, usually in blocs of 5, 10, or
25 men, but sometimes not in round-number groups. Table 16 is an example
of the distribution. It is to be noted that heavy weapons and headquarters
companies were augmented, as. well as the rifle companies.
TABLE 16
EXAMPLES OF ASSIGNMENT OF ATTACHED ROKA
PERSONNEL IN THE 7TH AND 8TH CAY REGTS
K numhers Assignment
llO5101-1105200 3d Bn, 7th Cay
llO5201-1105250 Co E, 8th Cay
llO5251-1105275 Co H3, 8th Cay
ll05276 -llO5300 Co E, 8th Cay
llO7151-1107200 3d BQ, 7th Cay
llO7201-1107225 Co F, 8th Cay
ll07226 -llO7250
Co H, 8th Cay
llO7251-1107260 Co H3, 8th Cay
llO7261-1107270
Co I, 8tb Cay
llO7271-1107280 Co K, 8th Cay
llO7281-1107290 Co L, 8th Cay
IlO7291-1107300 Co M, 8th Cay
Within the 7th Cav Regt, Korean personnel apparently were assigned in
blocs to each battalion headquarters and maintained under battalion accounting
after further assignment to the various companies. In a personal interview
the former S-1 of the 3d Bn, 7th Cav stated that the Korean troops were sepa
rated into groups for the several companies according to how they happened
252
ORO-T-261
UNCLASSIFIED
SEEITU
SEElEY
UNCLASSIFIED
to be standing in the field. Perhaps fortuitously, this grouping nevertheless
seems to have been in blocs according to the K numbers on the individuals'
dog tags. Such regularity has been assumed for attributing to particular com
panies several Korean casualties of unknown company affiliation and for esti
mating the number of ROKA personnel attached to each company. Anyadjust
ments of the data are indicated in Table 15; the assumed grouping in blocs is
shown in Table 17. The resulting distribution has 30 to 37 Koreans assigned
to each of the rifle companies. Several officers of the 3d Bn recalled that
30 to 35 ROK troops were attached to each rifle company.
TABLE 17
ROUGH OUTLINE OF THE NUMBER OF US AND ATTACHED ROKA
PERSONNEL IN THE 3D BN, 7TH CAV REGT
Company
IEnti
Personnel
L K M H3 3d Bn
United States
8
Present for duty at
close of:
11 Sep 144 142 148 ll3 93 640
12 Sep 148 145 163 130 106 692
13 Sep 149 14<1 165 131 104 693
14 Sep 123 131 150 128 102 634
15 Sep 151 158 168 126 103 706
Integrated ROKAb
Assumed assignment K llO7l51 K llO517l K llO5136 K l107l8S- K llO5101
(c)
in blocs K llO7l87 K llO5200 K ll05170 K 1107200 K 1105135
No. assigned 37 30 35 13 35 150
No. present, close
II Sep 36 29 33 12 35 145
aData from morning reports. Entries for 12-14 September do Dot fully reflect casualties on 12 September.
bData from morning reports Hq and Hq Co. 3d Bo.
eSee Table 16.
Casualty rates for the Hill Mass 314 action have been computed from the
strength and casualty figures, and entered in Table 14. The most striking
feature of the table lies in the high and' almost equal rates of Co L and Co I,
which attacked abreast and became somewhat intermingled during the battle.
Each company lost about one-third of its total personnel.
The casualty rates must be interpreted in the light of the number of men
actually under fire at different stages of the battle. Often two unwounded men
would carry one casualty to the rear. Various officers made a determined
effort to stop the noncasualties and send them back forward. Nevertheless, a
count immediately after the capture of the objective showed that there was a
total of only 80 effectives in Co L and Co I together. As Table 14 shows, the
initial combined strength of the two companies was 351, and their casualties
ORO-T-261
253
UNCLASSIFIED
UNCLASSIFIED
numbered 118. Therefore, at least 153 men who were not casualties were not
on the objective when the count was made. The companies' two 60-mm mortar
squads are thought to have been held back at Knob 0, and so far as is known
the six flame-thrower teams (the equivalent of two squads) were never com
mitted. Perhaps miscellaneous personnel equivalent to another squad from
each company were legitimately absent on special duties. Thus about 90 men
of the two assault companies were unaccountably missing when the objective
was first secured.
It was reported that men continued to straggle in from the left, a squad at
a time, for an hour after the objective was captured, yet later in the evening
41 men were reported missing. This number was far greater than the number
of the bodies to be picked up in the next days, and eventually no men present at
the outset of the battle were carried as MIA. Therefore it seems clear that
during the battle a considerable number of men moved away from the scene 'of
active combat. One man formerly in Co L stated that men sought to escape
from the intense mortar fire by moving down the wooded slopes to the left. A
former squad leader of Co I said that two ROKA soldiers attached to his squad
disappeared at the start of the attack. That desertion was not the rule with the
attached Korean troops is shown by their casualty rates and by the direct obser
vation that the bodies of several ROK troops were on top of Knob 2. The five
cases of combat shock also contributed to the missing unwounded.
If unwounded personnel did leave the center of action during the battle, the
real casualty rates were higher than the values computed. For instance, if
troops scattered at a constant rate amounting eventually to one-quarter of
original strengih, then the real casualty rate would be closer to 40 percent than
to the calculated 33 percent. The upper limit of the real rate follows directly
from the number of men who survived the entire battle intact, 80, and the num
ber who became casualties, 118, and is equivalent to a rate of 60 percent. As a
reasonable guess at the real casualty rate, perhaps 50 percent of the men of the
assault companies which were exposed to enemy fire for more than a small
interval of the battle were hit.
Sources of Detailed CasuallY. Data
Under the reporting system in use until September 1952, the basic source
of detailed data on an individual casualty was the medical tag made out at the
battalion aid station. When known, the time of the casualty. and the missile and
weapon responsible would be recorded along with the medical d a t a ~ The former
surgeon of the 3d Bn, 7th Cav Regt stated that especial care was taken to make
out the tags fully and properly and, in general, to keep accurate medical records.
This assertion agrees with the frequency with which detailed data on the casual
ties in the Hill Mass 314 action were found available, and is supported by both
the accurate data maintained on the status of the attached ROKA personnel and
the particular effort after the action to find and account for the large number
of missing men.
In the previous investigation," the individual casualty files of one-third
of the US casualties in the battle for Hill Mass 314 were scrutinized at the
Casualty Branch, AGO, FECOM. As stated above, the remaining files were
not covered because of the incomplete index of names supplied by the machine
records unit. The medical tags were not included in these files, but data from
ORO-T-261
254
UNCLASSIFIED
HEll"
UNCLASSIFIED
them did appear often on the individual casualty and interment forms sent in
from the division. Individual casualty reports from hospitals, but no clinical
records, were also in these files. In consequence, the desired casualty data
. were available for a large part of the limited sample covered.
In 1953, further research was done to obtain recorded casualty informa
tion on all US casualties in the capture of Hill Mass 314. Personal ("201")
files were obtained for (a) any man in the sample previously covered (one
third of the total) for whom data on the time of casualty and missile and wea
pon responsible had not already been discovered, and (b) all additional men
on the new master casualty list (Table El).
When the individuals concerned were in FECOM, their "201" files were
kept within the 1st Cav Dlv. By 1953 they had all been either returned to one of
several locations in Washington, D. C. or retired to the Demobilized Personnel
Records Branch of the AGO at St. Louis.
The "201" files were often found to contain the original medical tags, .
usually made out in pencil and sometimes authenticated by bloodstains. It was
the practice at the aid station of the 3d Bn, 7th Cav to type a new card, includ
ing any additional data, at the first opportunity. This fair copy was often found
stapled to the original tag. Whether or not either was present, a medical tag
made out at a field hospital was sometimes in the file. Data from these supple
mental tags were used when necessary but regarded as less reliable on the
circumstances of a casualty.
All material in the "201" file was scanned for casualty data. When the
medical tag and field casualty report were lacking a useful notation might be
found in the casualty section of the Service Record or Separation Form, as,
for example, "12 Sep 50, SFW (0)" (shell fragment wound, grenade). Letters,
affidavits, or transcripts of courtroom proceedings occasionally furnished
bits of information. In general, however, the clinical records were the most
valuable secondary source; these might include a description by the patient
of how he was wounded or a surgeon's notation on the kind of missile he
removed. Some further casualty datawere obtained in the course of the inter
views conducted for this study, sometimes from the man himself, sometimes
from former fellow members of his unit. The various sources of detailed
casualty data are indicated in Table E1; miscellaneous information from these
sources is reported in the "Remarks" column of that table.
For the ROKA personnel attached to the 7th Cav Regt, data from the medi
cal tags were recorded at the 15th Med Bn Clr Co in A&D sheets and a log book.
Research on these records was done by current personnel of the 7th Cav Regt
at the request of the writer.
A!m!:v.sis of the Detailed CasuallY. Data
Table 18 shows the frequency of available data on time, type of missile,
and type of weapon for casualties sustained by the several companies of the 3d
Bn, 7th Cav Regt in the capture of Hill Mass 314.
The approximate time a man was hit is known for 107 out of the total of
150 casualties. In 96 of these cases the time was given by a medical tag or by
a document quoting it. In the other 11 cases the time was estimated from infor
mation supplied by eyewitnesses.. They could relate the time a certain man was
hit either to some established event in the course of the battle, as the end of the
ORO-T-261
255
UNCLASSIFIED
SECRU

SECRET
second air strike, or to the occurrence of another casualty (perhaps from the
same shellburstl for which the time had been reported. Of the 96 cases, 2
required some correction as the result of such testiinony. It appears that a
comprehensive program of interviews shortly after the action could have estab
lished the time of almost all casualties.
TABLE 18
FREQUENCY OF AVAILABLE DATA ON TIME AND TYPE OF MISSILE
FOR CASUALTIES IN THE 3D BN, 7TH CAV REGT IN THE
CAPTURE OF HILL MASS 314, 12 SEP 50
Time or type
of missile
Entire
known or unknown
3d Bn
Time known
Time unknown
Time. total
Missile type known
Fragments
Mortar
Artillery
Grenade
Unknown
Total
Bullets
Rifle or SA
Machine gun
Unknown
Total
Missile type unknown
Total. missile type
107
43
150
112
73
2
3
19.5
97.5
II
I
2.5
14.5
38
150
Number of casualties
Company
L K M
I
H3
I I I
41 46 14
21 10 9
62 56 23
44 44 18
25.5
8
30 1I.5
b
2 0 0
2 I 0
8 6.S
c
5
37.5 37.5 16.5
4.5
8
S.Sc I
0 I 0
2 0 0.5
b
6.5 6.5 0.5
18 12 5
62 56 23
3
I
4
3
2
5
3 3
3
0
0
0
3
0
0
0
3 3
0
0
0
0
0
0
0 0
I 2
4 5
a()[)e case both small-arms Bnd mortar fragment.
hOne case both unknown bullet and mortar fragment.
cOne case both small-anns and unknown fragment.
As it is unlikely that aidmen up on the hill accompanied casualties to the
aid station, the times recorded on the medical tags were presumably based on
the time a wounded man arrived there, and on his statements as to where on the
hill he had been hit and how long he had lain there. Since the route of evacuation
down to the aid station was long and tortuous (see Fig. Ell, the resulting esti
mates could not be very accurate, and could also just as well be too early as too
late. Therefore,.the trend in the accumulation of casualties with time, rather
than the number that might happen to be reported at a given hour, is significant.
256 ORO-T-261
UNCLASSIFIED
SECRET
UNCLASSIFIED
SECRET
Accumulation curves are given in Fig. 14 for two groups, the assault units
(Co L and Co I) and the supporting units (Co K and elements of Co M and Co H3).
The combined curve for the entire battalion also appears. The data from which
the curves were constructed are given by individual companies in Table 19, and
the number of casualties at each reported or estimated time is stated. It would
seem that at the aid station times were estimated to the nearest quarter or half
hour.
160
Securing 1st 2d Pi nned 3d
w
'"
I
..J
.,
'"
.,
:::>
u
w
..J
I
I
.,
'"
UJ
~
I
.,
..J
:::>
:::>
'"
u
1100 1200 1300 1400. 1500 1600 1700 1800 1900 2000 Unknown
140
120
100
80
60
40
20
Assau It Down Assault Objective
Entire 3d Bn
---
Assault Units (Co L and Co I)
Supporting Units (Co K, Co M,
and Co H3)
TIME
Fig. 14-Accumulation of casualties in the 3d Bn, 7th Cav Regt
during the action on Hill Mass 314, 12 Sep 50.
The breakdown shown in Fig. 14 of the casualty data for the several phases
of the battle follows the reconstruction of the battle as given in the two preceding
sections of this paper. Grossly, the curves show that casualties increased in
a rather regular way. In the initial phase of the battle, when friendly units were
moving out toward the main enemy positions on Knob 2, casualties were sus
tained at an increasing rate. A maximum rate was then reached, and mairitained
for several hours. In absolute numbers, casualties then occurred at a decreas
ing rate. It is felt that this drop-off does not represent a slackening in the
intenSity of the action but rather a reduction in the number of troops exposed, as
discussed earlier in this section. The probability that a surviving individual
ORO-T-261 257
SECRET
UNCLASSIFIED
UNCLASSIFIED
sum
would become a casualty probably remained .at least as high as it had been until
the break in the action around 1530. At this time assault units at last moved
up onto Knob 3.
The relation of friendly casualties to the character of combat at different
stages of the battle is discussed more fully in the succeeding interpretation of
the action. It suffices here to point out that the accumulation curves do corre
late with the graphic timetable of the battle. A possible bias must also be
mentioned: the time the man was hit is known for very few of the KIA (5 out
TABLE 19
CASUALTIES IN THE 3D BN, TfH CAY REGT IN THE CAPTURE
OF HILL MASS 314, 12 SEP 50, BY TIME AND COMPANY
Time
Company
Casualties at reported time Cumulative casualties
L I
I
I KIM I 113
L & I I K, M, & II3
llOO 2 1 2 1
ll30 1 1 4
ll45 1 5
1200 5 1 10 2
1215 1 12
1230 1 14
1300 8 9 1 31 3
1315 1 32
1330 3 1 36
1350 37
1400 7 13 3 2 57 8
1430 3 1 1 1 60 II
1435 61
1500 5 4 2 70 13
1515 1 2 3 73 16
1530 1 74
1600 3 5 2 82 18
1645 1 83
1700 1 84
1800 1 1 85 19
2000 1 86 20
2030 87
Total, time known 41 46 14 3 3 87 20
Time unknown 21 10 9 1 2 31 12
Total casualties 62 56 23 4 5 ll8 32
of 35). If men were more likely to be killed in close combat than when approach
ing through a barrage of shellfire or when pinned down by it, then the effect of
such combat is not fully represented in the curves.
The attribution of casualties to different types of missiles and weapons
has been described in previous reports (such as Ref. 25). In this study the
former definitions and distinction between fragment and bullet ("F" and "B" in
Table E1) have been retained. For the Hill Mass 314 action the object of this
258 ORO-T-261
UNCLASSIFIED
SKI"
stERn
UNCLASSIFIED
classification is the resolution of two essential points: (a) the incidence of
casualties in close combat (from enemy bullets and grenades) vs that of casual
ties from enemy supporting weapons, and (b) concerning casualties from the
supporting weapons, the number from mortars vs the number from artillery.
The second point will be considered first.
The former surgeon to the 3d Bn, 7th Cav Regt stated that when questioned
at the aid station as to the type of missile that caused their wounds the men
were rather definite in their replies. Their basis for differentiating artillery
from mortar fire was the whistling or screaming of the artillery shell prior to
exploding as compared to the complete lack of warning of the mortar shell.
When the writer interviewed some of these men in 1953, they again volunteered
the same reason. Not all of them would assign a cause for their wounds, even
when offered a choice; several stated that they had been hit by fragments but
simply didn't know from what type of shell these had come. Others confidently
asserted their ability to distinguish mortar shells of different caliber from the
sound of the explosion. Several individuals cited as evidence that a near burst
had been a small-caliber mortar shell the fact that they had suffered blast
effects, and even scorching, without other injury. Documentary evidence on
this point was found in a number of clinical records and is reported in the
"Remarks" column of Table E1. The writer believes, to the contrary, that
prominence of blast effects would be evidence of large-caliber shells, and
prefers the quite different explanation given in the later subsection on the
effectiveness of enemy mortar fire. Aside from these fine points, the definite
consensus of all personnel who had been on Hill Mass 314 was that most
friendly casualties had been caused by mortars, few by artillery.
An entry in the 3d Bn medical log states that 80 percent of the WIA had,
been hit by mortar fire. The former surgeon of the battalion recently repeated
this statement but added that he felt that among the fatalities a higher incidence
was due to bullet and grenade wounds sustained in close combat than to wounds
by shell fragments. On the other hand, the former S-l of the battalion, who had
occasion to view all the bodies brought in on the day of the battle, said it was
his impression that most of these too had been hit by shell fragments.
The statement in the log is probably only a careful estimate. Also, It
represents the few casualties outside the 3d Bn listed above. Nevertheless, the
figure may be checked fairly closely by analysis of the WlA data in Table E1.
Of 115 WlA (including 3 DOW), 15 were from missiles of unknown type. In the
case of the WlA there is no reason to suspect a bias in favor either of fragments
or bullets. Therefore, these 15 cases are distributed in the ratio of the knowns,
that is, 87 fragments to 13 bullets, giving a total of 100 fragment casualties and
15 bullet casualties. Again from the ratios of the knowns-67.5 mortar, 3 gre
nade, 1 artillery-the 100 fragment cases are further distributed as follows:
mortar 94.4, grenade 4.2, artillery 1.4. Of the 115 WIA, the 94.4 mortar
casualties amount to 82 percent, thus confirming rather closely the SO-percent
figure given in the medical log.
Of 35 KIA (excluding 3 DOW), 2 ~ were hit by missiles of unknown type.
The remaining 10 cases were distributed as follows: mortars 5.5, fragments
of unknown type 3, artillery fragments 1, and bullets 0.5. This distribution is
much the same as that of the WlA.
ORO-T-261
259
5Eun
UNCLASSiFIED
UNClfl.SSIFlED
SEERfT
The fatalities from unknown causes may be distributed in two ways, the
first to provide a lower limit of the incidence in close combat, the second a
probable upper limit: (a) the unknowns are arbitrarily distributed in the same
ratios as the knowns, and (b) possible numbers of casualties in close combat
are arbitrarily assigned on the basis of knowledge of the tactical movements
of the units.
For the lower limit, the 25 KIA resulting from unknown causes are divided
in the ratio of 9.5 fragments to 0.5 bullets, giving 1.3 additional bullet casual
ties and 23.7 additional fragment cases. As 3 cases are already listed as
killed by fragments of unknown type, there are 26.7 fragment fatalities to be
distributed in the ratio of 5.5 mortar to 1 artillery, yielding 22.6 additional
mortar casualties and 4.1 additional artillery casualties.
For the upper limit, additional individuals killed by bullets are estimated
from the figures for the separate companies. Of the 14 KIA that resulted from
unknown causes in Co L many, say 9, may have been due to enemy bullets fired at
close quarters in the first assault. Fatalities in the grenade-throwing incident
which followed are not to be attributed to the enemy grenades, because these
grenades were of the concussion type and were not very likely to cause death.
The ineffectiveness of the grenades was reported by the commander of Co I,
who was present with his 3d Plat in this local fight. Perhaps four of the six
KIA from unknown causes in Co I may also be assigned to bullets. Since Co K
was never engaged in close combat during the battle, its three men killed from
unknown missiles were almost certainly not hit by bullets, and since the elements
of Hq Co that were on the hill were usually in protected positions, or at least
behind the assault troops, the twoHq Co Korean personnel killed by unknown
missiles were probably not hit by bullets. As the probable maximum, then,
14 additional KIA are credited to bullets, increasing the total bullet fatalities to
14.5. The remaining 11 cases, with the 3 deaths already listed as due to frag
ments of unknown type, are divided between mortars and artillery in the ratio
of 5,5 to 1, giving 11.8 additional mortar casualties and 2.2 additional artillery
casualties.
Casualties in close combat may now be computed from the several break
downs above. At the lower limit, total casualties are 15.0 WIA by bullets, 4.2
WIA by grenades, and 1.8 KIA by bullets-a total of 21.0, or 14 percent of all
150 casualties. At the upper limit, total casualties are 15.0 WIA by bullets, 4.2
WIA by grenades, and 14.5 KIA by bullets-a total of 33.7, or 22.5 percent of
all 150 casualties.
Casualties from mortar fire alone can also be computed. At the upper
limit for bullets and grenades,total mortar casualties are 94.4 WIA and 17.3
KIA (15.5 percent of those hit KIA)-a total of 111. 7, or 74 percent of all 150
casualties. At the lower limit for bullets and grenades, total mortar casualties
are 94.4 WIA and 28.1 KIA (23 percent of those hit KIA)-a total of 122.5, or
82 percent of all 150 casualties.
Consideration of the calculated percentages of those men hit by mortar
fire who died leads to the conclusion that the actual situation was probably
about midway between the upper and lower limiting cases. In the medical
log the battalion surgeon noted that most of the wounds were from mortar fire
and were not serious. (But 39 percent of.the wounded did not return to duty.
See Table 13.) On the other hand, men who had been hit at all seriously were
260 ORO-T-261
UNClfl.SSIFIED nUEl
SEEREY
UNCLASSIFIED
immobilized on the spot, likely to be hit again, and once evacuated were very
likely to die before reaching the aid station. Therefore, a fatality figure
between 15 and 23 percent seems appropriate. Such a choice would lead to a
reasonable, though high, fatality percentage for bullet hits. The following
figures are therefore considered as most likely to represent the causal per
centages of all casualties in the battle for Hill 314: 19 percent in close com
bat (bullets and grenades); 81 percent from sUPP'orting..!!!1llpons (artillery,
mortars); 79 percent from mortars.
TAlJLE 20
COMPARISON OF ACCUMULATION OF CASUALTIES FROM ENEMY SUPPORTING
WEAPONS WITH ACCUMULATION IN CLOSE COMBAT FROM
BULLETS AND GRENADES, 3D BN, 12 SEP 50
Time
Casualties from
mortar, artillery and
unknown fragments
Casualties from bullets
and grenades
Number I Cumulative Number
I
Cumulative
!l00 2 2 I I
!l30 2 3
1200 5 7
1215 2 9
1230 I 10 I 4
1300 14 24 3 7
1315 I 8
1330 13 27 I 9
1400 21 48 4 13
1430 5 53 I 14
1435 I 54
1500 10 64 2 16
ISIS 6 70
1530 I 71
1600 7 78
1645 I 79
1700 1 80
1800 2 82
2000 1 83
2030 1 84
Total, time known 84 84 16 16
Time unknown 12 12 3 3
Total 96 96 19 19
aTotal (115) casualties here greater than total (1l2) in Table 18 because 3 men were
hit by 2 different types of missile. .
For casualties from mortar fragments and fragments of unknown type an
analysis has been made of the distribution of hits over the various regions of
the body. The results of this analysis are presented in Table 23.
261
ORO-T-261
SEUEY
UNCLASSIFIED
UNCLf\SSIFIED
SECRET
Casualties with a reported cause have been separated into two groups,
those from bullets and grenades and those from mortars and artillery, for the
construction of curves of accumulation with time. These are given in Fig. 15,
and the data from which they were plotted are given in Table 20. With such
a small statistical sample, the details of the curve for bullets and grenades may
100
'"
80
!!!
0
J
<
"
60
'"
<
u
w
::
0
<
40
J
"
'"
"
u
20
0
1st 2d Pinned 3d Securing
Assault Down Assault Objective
1L/
...
/Vcasuaities from Mortar, Artillery, and
Unknown Fragments
/'
Casualties from Bullets and Grenades
d
..
,
...
I.
,
I
,
1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 Unknown
TrME
Fig. -15-Comparison of accumulation of casualties from enemy supporting
weapons with accumulation in close combat from bullets and grenades, 3d
Bn, 12 Sep 50.
not be significant, but it is of interest that the curve drops off sooner than that
for the supporting weapons. It certainly is significant that no casualties from
bullets or grenades were reported after 1500. If the defense had not already
been broken by the time Knob 3 was finally overrun, at around 1530, one would
expect casualties from these causes to rise to a maximum at this time. Since
the assault was successful, any men hit would have been reported.
As mortars were responsible for almost all casualties inflicted by sup
porting weapons, separated accumulation curves for mortars have been con
structed in Fig. 16. Fragments from unknown missiles are included here as
probably almost entirely mortar fragments. The data from which the curves
were constructed are given in Table 21.
The data on mortar casualties are analyzed more fully in the later sub
section on the effectiveness of enemy mortar fire, where they are related to
information on enemy weapons and fire. A point worth noting here is that the
maximum slope of the curve for the supporting units is about 40 percent of
that of the two assault companies, while perhaps 80 percent as many men were
involved. Though they advanced some distance behind the assault companies
and, according to the former commander of Co K, tried to move through the
mortar barrage as fast as possible, the supporting units were apparently hit
man for man about half as hard as the assault units, despite the seemingly
262
ORO-T-261
SECRET
uNCLf\SSIFIED
UNCLASSIFIED
SECRET
great difference in exposure. To put it somewhat differently, Co K, with two
platoons moving through the mortar barrage as fast as possible, had 11. 5
mortar casualties (Table 18), whereas Co I and Co L, with three or four platoons
exposed, had, respectively, 25.5 and 30. This result suggests that the vuiner
ability of men standing up to move was an important factor.
Fig. 16-Accumulotion of casualties from mortar fragments and fragments of unknown
type in the 3d Bn, 7th Cay Regtin the capture of Hill Mass 314, 12 Sep 50.
74% of sample consists of known mortar casualties, 26% of casualties from fragments
of unknown type. Includes integrated ROKA personnel.
Casualties vs Rank
Table 22 gives a breakdown of casualties in the three rifle companies by
rank. The distributions are seen to be similar in the two assault companies in
the classes of private and corporal and different in the classes of sergeant and
lieutenant. Company L initially had two fewer lieutenants than Co I. Its former
executive officer had become the CO of Co I, and one of its platoons was led by
a sergeant.
If officers and sergeants are regarded as leaders, corporals and privates
as nonleaders, the following comparison may be made: leader casualties-Co
L, 13, and Co I, 12; nonieader casualties-Co L, 49, and Co 1,43. If leaders
are more exposed than the men they lead, one would expect a higher percent
age of officers and sergeants to become casualties than corporals and privates.
In the absence of a roster giving the composition of the battalion before the
battle, this comparison cannot be made, but a somewhat different approach
"i. possible.
In a battle in which assaults and combat occurred initially in quick suc
cession, to be followed (at 1300) by a long pinned-down period during which
many mortar casualties were sustained, one would expect that casualties
among leaders' would occur somewhat earlier, on the average, than casualties
ORO-T-261 263
SECRET
UNCLASSIFIED
UNCLASSIFIED
TABLE 21
ACCUMULATION WITH TIME OF CASUALTIES FROM MORTAR FRAGMENTS AND
FRAGMENTS OF UNKNOWN TYPE IN THE COMPANIES OF THE 3D BN.
7TH CAY REGT IN THE CAPTURE OF HILL MASS 314. 12 SEP SO
Cause of casualty
and time,
known or unknown
Caaualties at reported time
Company
L I I I KIM I H3
Cumulative casualties
Co L & Co K. Co M.
Co I & Co H3
Entire
3d Bn
Time known
llOO
1 1 1 1 2
1200 3 1 4 2 6
1215 1 1 6 8
1230 1 7 9
1300 7 6 1 20 3 23
1330 2 1 23 26
1400 6 11 2 2 40 7 47
1430 2 1 1 1 42 10 52
1435 1 43 53
1500 4 4 2 51 12 63
1515 1 2 3 54 15 69
1530 1 5S 70
1600 4 2 59 17 76
1645 1 60 77
1700 1 61 78
1800 1 1 62 18 80
2000 1 63 81
2030 64 82
Total 28 36 12 3 3 64 18 82
Time unknown 6 1 5 0 0 7 5 12
Total, time 34 37 17 3 3 71 23 94"
Frag, mortar 26 30 12 3 3 56 18 74
Frag, unknown 8 7 5 0 0 15 5 20
Total, mortar and
unknown frag 34 37 17 3 3 71 23 94"
I:lGrand total of 94 is l.!i r,eater than figure in Table 18 because three casualties to two causes are here
rated as a full, instead of hl f. casualty from each cause,
264 ORO-T-261
UNCLASSIFIED
SiUH
UNCLASSIFIED
SEun
100
w
~
I
...
..J
80
:::l
:::l
'"
U
~ .
60
!!!
I
..J
...
:::l
~
...
40
u
u.
0
I
Z
w 20
0:
W
..
U
0
JUmp off, 1st and Pinned
2d Assaults Down
x- 24 Officers and Sergeants

.-63 Corporals and Privates
1100 1200 1300 1400 1500 1600 1700 1800 1900 2000
HOUR, 12 SEP so
Fig. 17-Cumulative casualties differentiated as to rank, Co L and Co I.
among nonleaders. Figure 17 gives the results of the appropriate test; curves
for Co L and Co I together, show the percentage distribution versus time of
casualty for 24 lieutenants and sergeants and 63 corporals and privates. The
difference appears to be significant: by 1300, 50percent of all eventual casual
ties among leaders had been sustained, as compared to only 30 percent of those
among nonleaders. .
TABLE 22
CASUALTIES BY RANK IN THE CAPTURE OF HILL MASS 314
Rank
Company, 3d Bn, 7th Cay Regt
L K
US privates
34 27 8
ROKA pri vates 4 6 7
Corporals II 10 3
Sergeants 12 7 5
Lieutenants 1 5 0
Unknown
0 1 0
ORO-T-261
265
SECRET
UNCLASSIFIED
UNCLASSIFIED
S E E R ~ T
THE EFFECTIVENESS OF ENEMY MORTAR FffiE
As concluded above, almost all casualties from enemy supporting weapons
are attributable to mortars; also, as worked out in Part n, only two mortar
batteries were significantly involved: the six 120-mm mortars northwest of
Hill Mass 570 at Tongmyongwon and the four 82-mm mortars at the foot of
Hill Mass 314 to the north.
The 120-mm mortar battery unquestionably belonged to the 19th Regt of
the 13th NK Div and was in telephonic contact with the observation post on the
570-m peak In the area of the 1st Bn, 19th Regt. Information on enemy prac
tices in siting 120-mm mortars, as supplied by a prisoner from the 15th NK
Div, is given in App C, Item 1.
There is no definite information on the organizational affiliation of the
82-mm mortar battery; it could have belonged either to the lst or 13th NK
Div. It is considered most likely that these mortars too belonged to the 19th
Regt, since (a) there are no data to Indicate another position for the 19th Regt's
four mortars of this caliber (unless the mortars observed far forward on
Hill 570 and considered to be 60-mm are taken to be 82-mm), and (b) the
north end of Hill Mass 314 was the only suitable site. The latter point requires
brief analysis.
The 19th Regt's battalion-support 82-mm mortars would certainly be
required to reach the most forward friendly positions on Hill Mass 570, i.e.,
Hill 373 and its reverse slopes, and probably would also be required to cover
Hill Mass 314 from Knob 2 to at least as far back as Knob O. The only possible
pOSitions with proper defilade, etc., would be the northern foot of Hill Mass 570
and the northern foot of Hill Mass 314. For each position the required ranges
of fire are:
Distance, yd,
from north end of:
Hill Mass
Hill 570 Hill 314
570: 373-m peak and reverse slopes 3000-3700 2000-2600
314: Knob 2 to Knob 0 3200-3700 800-1600
As given by the enemy firing table; the maximum range of the 82-mm
mor.tar (with six increment charges) is 3324 yd. It is therefore concluded that
the four 82-mm mortars of the 19th Regt were at Hill Mass 314, Installed there
when a battalion of this regiment made the presumed move onto the hill on 11
September to reinforce 1st Div units. .
Telephone wire led from the 82-mm battery, but in which direction is not
known. It is possible that It linked the position with a forward observer on
Knob 2. If so, some other line of communication must be postulated to account
for the continuing accuracy of fire after the enemy was driven from the knob.
Since, at the outset of the battle, the 82-mm mortars fired the apparent final
protective line simultaneously with the l20-mm mortars, and since both fires
then moved back toward Knob 2, unified control is indicated. A possible and
reasonable path of communication Is suggested by the location of the few enemy
ORO-T-261
266
UNCLASSIFIED
SEERU
UNCLASSIfIED
SECRET
dugouts (Fig. 11), as follows: from the 82-mm mortar battery to the dugout
(OP?) on the northwest finger ofHill Mass 314, from there to the dugout (CP?)
at the northwest foot of the hill, and from there across the narrow valley up to
the 570-m peak. For the 120-mm mortars the line would be: from the 570-m
peak to the CP of the 1st Bn, 19th Regt on the north end of the ridge, and from,
there to Tongmyongwon. The total length of all these lines, about 6 km, is just
that of the amount of telephone wire usually carried in a 120-mm regimental
mortar company.
The responsiveness of the enemy mortar control system to friendly moves
up to and on Hill Mass 314 is pointed out in the narrative of the action. It is
concluded that the 120-mm mortars were active and In search of targets on the
morning of 12 Septe'mber, but the observers were unaware of, the move-up for
the attack, probably because of the haze, until it began.
Accuracy of Fire
A complete Soviet firing table is available for the 82-mm mortar;8 but
not for the 120-mm mortar. For 82-mm fire on Knob 2 at a range of about
eight hundred yd, one to six increment charges could be used. With few
the effectiveness of fire would suffer because of the too-acute angle of impact
on reverse slopes; with a greater number of charges, some accuracy would be
lost and the time of flight would be increased unnecessarily. Charge No.3 is
taken as the likely compromise. The following data apply:
Range 800 m
Angle of elevation 77 49'
Angle of fall 80
Maximum ordinate 976 m
Time of flight 29 sec
Probable error in range 13 m
Probable error in azimuth 7.1 m
The quoted dispersion errors give the bracket of 50 percent of all shots.
From the normal distribution curve it is then calculated that 90 percent of
shots would fall within a rectangle measuring 34 m (range) by 19 m (azimuth)
if fire Is upon a horizontal plane.
According to the contours of Hill Mass 314 on the map, the ridgeline
through the northern part of the saddle has an average inclination of about 27
deg. In Fig. 3 an apparent inclination of about 21 deg is seen. If 24 deg is
taken as the likely value, the' angle of fall upon this section of the ridgeUne is
reduced from 80 to 56 deg and the dispersion in range is lengthened accord
ingly. The 90 percent rectangle is then calculated to measure 39 by 19 m on
the slope and 37 by 19 m (40 by 21 yd) in horizontal projection.
In the absence of data on the error of 120-mm fire the data for the 82-mm
mortar is scaled up, using available angle and range data for the 120-mm
mortar.
29

-Insofar as the tube affects accuracy, one would expect the 82-mm mortar to be the more accurate. With
shells of each caliber the ratio of over-all length to the span in contact with the tube (most forward ring on
hody to end of lin) is very nearly the same; however, in the 82-mm tube this span travels in contact with the
tube wall a considerably greater distance in proportion to its length than in the 12Q-mm tube.
ORO-T-261 267
5EERH
UNCLASSifiED
UNCLASSI FI ED 5EEREJ
With six increment charges, giving a muzzle velocity of 810 fps, and an
elevation angle of 61 0 3 0 ~ the 120-mm mortar attains the required range of 4483
yd from Tongmyongwon to Hill Mass 314.
With six increment charges, giving a muzzle velocity of 691 fps, and an
elevation angle of 60
0
30', the 82-mm mortar attains a range of 2734 yd, with
the range error 45 m and azimuth error 9.9 m. Extrapolating the curve of
range versus muzzle velocity at constant angle of elevation to a velocity of
810 fps gives a range of 3320 yd, with the range error 67 m, and azimuth error
11.3 m. Applying further proportional corrections to account for the range
that the 120-mm shell would have attained at this elevation angle and muzzle
velocity gives a range error of 90.5 m and azimuth error of 15.3 m. Using
the normal curve with these 50 percent dispersions gives a 90 percent rec
tangle measuring 260 yd (range) by 44 yd (azimuth).
Data on angle of fall are not available for 120-mm shells. It is estimated
that with an elevation angle of 61
0
39' the angle of fall would be about 65 deg.
On the northeastern side of the saddle the contour lines are roughly normal to
the line of fire and the slope inclines about 23 deg. The actual angle of impact
here would then be about 42 deg. For fire aimed at the ridgeline, dispersion
In range would then correspond to a 90 percent rectangle bounded on the right
side by a line 144 m down the slope, or, in horizontal projection, 145 yd from
the rldgeline.
For both the 82-mm and 120-mm mortars the 90 percent rectangles cal
culated above have been plotted In the diagrams of the graphic timetable of the
Hill Mass 314 action. In all cases they represent single shots or volleys aimed
at a single point near the center of the rectangle.
Slope and the Effectiveness of Mortar Fire
Enemy mortar shells were impact-fuzed. Test data on US fuzes of this
type show that the mortar shell explodes while still oriented substantially at
the angle of Impact and before penetrating more than a few inches into the
earth. US ballistic tests on captured Soviet 120-mm mortar shells'" show that
the fragments are concentrated in an exceedingly narrow spray. This property
maximizes the lethality of the shell when the impact angle is near 90 deg, at
the expense of effectiveness when Impact is at acute angles.
During the battle for Hill Mass 314 most 120-mm mortar shells landed
at very acute Impact angles. As estimated above, the angle of fall was 65 deg,
and the east slope of Knob 2 was inclined about 23 deg; impact was then at an
angle of about 42 deg. In Fig. 18 this slope and the fall of a 120-mm shell upon
it are plotted in a vertical cross section that includes the line of fire. The
density of fragments versus angle of elevation around the point of burst is
shown by the polar lobes superimposed on the plot. The dotted lobes represent
fire actually absorbed by the earth.
The fragmentation data used in the construction of Fig. 18 were obtained
by correcting the data of the static test-firings for velocity of fall. A velocity
of 546 fps is consistent with the muzzle velocity and the values of range, time
of flight, and angles of fall already assumed. Fragments are expelled in a
120-mm burst at an Initial velocity of 3200 fps. The velocity of fall then flat
tens out the fragmentation cone so that each lobe is moved about 10 deg toward
a straight line through the point of burst. This correction was made in the
construction of Fig. 18.
268 ORO-T-261
UNCLASSIFIED S[EREJ
UNCLASSIFIED
SEEREt
It is obvious from the figure that the lethal area was greatly reduced.
Most fragments either went directly into the earth or harmlessly into the air
over the heads of men down the slope. Blast, of course, was not affected.
Fig. l8-Relative density of 1 2 0 - m ~ mortar fragments in vertical plane
of trojectory foiling on eost side of Knob-2.
Polen plot; density of fragments in any direction from point of burst proportional
to radial di stance to curve.
A surprising number of men suffered blast effects with little or no frag
ment injury. Figure 18 indicates how close to a burst they could have been
to be knocked down without being hit. Six pertinent cases are to be found in
Table E 1. These cases just happened to be reported in clinical records avail
able for research and in no way represent complete coverage of the 3d Bn,
7th Cav Regt. The data on each incident with the order number of the casualty
in Table El are:
No. 2: Mortar shell landed 6 ft away. Man knocked down a minute, then walked
to aid station.
No. 31: Contusion sustained when knocked down by blast from mortar sbell.
No. 91: Standing when mortar blast occurred, lost consciousness, came to further
down the hill.
No. 98: Lost memory for four days from mortar blast.
No. 99: Standing on east slope of Knob 2. Mortar blast knocked man 10 -12 ft
down hill.
No. 105: Knocked 20 - 30 ft down hill by mortar blast and unconscious for a few
minutes.
Most, if not all, of these cases occurred on the right side of the saddle.
Figures E 1 through E 11 show the steepness of the slopes down which these men
rolled when blown off balance.
The commanding officer of Co I stated that an enemy mortar round fell
amid the men of his mortar squad without inflicting any injury except scorching.
He considered this ineffectiveness as evidence of a small-caliber round; this
indeed may have been the case, and the explanation may be as given for the
reduced effectiveness of 120-mm rounds.
ORO-T-261 269
SEEREI UNCLASSIFIED
UNCLASSifiED
SECRET
If the typical situation was as shown in Fig. 18, one would predict more
frequent wounding of the head and less frequent wounding of the lower extrem
ities than would occur if men were oriented at random with respect to the flight
of mortar fragments. Such Is indeed the case, as is shown by the data of
Table 23.
TABLE 23
BODILY DISTRIBUTION OF HITS BY MORTAR FRAGMENTS
AND FRAGMENTS OF UNKNOWN TYPEa
Unit
Body region hit
IChest, ba,k, I Buttocks,
Total
areas bit
No. of
cases
Ratio:
areas hitl
case
Head I
Shoulders
& neck & anns abdomen legs, feet
Company
L 5 10 6 4 25 17 1.5
I 6 12 6 8 32 24 1.3
K 3 2 4 5 14 10 1.4
M, H3 0 4 0 3 7 5 1.4
Totals 14 28 16 20 78 56 1.4
BSample consist!;! of 56 cases with data on body area out of total of 93. Included are 3 KIA cases.
TABLI': 24
RELATION OF BODILY DISTRIBUTION OF HITS BY MORTAR FRAGMENTS AND
FRAGMENTS OF UNKNOWN TYPE TO TOTAL BODY SURFACE AND TOTAL HITS
Body region hit
56 cases with 78 hite
R
Wound survey,
Korea
31
% cases with
hit in area
% body
surface hit
%total
areas hit
Ratio: %
body surface
b;t/% total
areas hit
%387
mortar
hits
Ratio
Head and neck
25 12 18 1.5 17.9 1.5
Shoulders and anna 50 22 36 1.6 30.0 1.4
Chest, abdomen,
and back 29 27 21 0.8 10.8 0.5
Buttocks, general,
legs, and feet 36 39 26 0.7 41.4 1.1
Percent totals
140 100 100 100
aSee Table 23.
Because of the miscellaneous sources of the detailed casualty data, as
indicated in Table E1, data on the parts of the body affected were not available
for each case although this information was recorded somewhere for all casu
alties as a matter of routine. Of 93 men considered to have been wounded in
the battle on Hill Mass 314, by mortar fragments or fragments of unknown
270 ORO-T-261
UNCLASSifiED
SEEtn
UNCLASSIFIED
SURK
type (and most likely also mortar fragments), data on the body areas affected
are at hand for 56 men, including 3 KIA. The hits have been classified accord
ing to a standard medical delimitation of body regions. Multiple hits within
a region are not counted, but separate regions hit by fragments from one burst
are. On the average, 1.4 regions were affected per casualty. Even though the
total sample is small, it is apparent from the breakdowns for the individual
companies that the distribution was remarkably uniform, and that as a rule the
shoulders and arms were most frequently hit.
In Table 24, the percentage distribution of all hits in the battalion is
compared with the relative areas of the various regions of the body. It is seen
that the general region of head, neck, shoulders, and arms received more than
its share of hits; the trunk and lower extremities, less. The ratio of imbalance
between these two portions of the body was 2.0. other studies have shown that
in general the distribution of hits is not in proportion to body areas; the point
of this comparison is to show the extent of the deviation in a fairly well
established tactical situation.
As another test of this result, parallel data are given in Table 24 on the
bodily distribution of 381 mortar hits in Korea as determined by a medical
survey team.
31
The two sets of data evidently agree in that head and arms
receive more than their share of wounds. They disagree in that the medical
survey does not show a reduced frequency of wounding of the lower extremities. *
, Since the survey sample without doubt represents a variety of tactical Situations,
the low incidence of wounds of the lower extremities in the battle on Hill Mass
314 !!illY reflect the known particular circumstances there, such as the steep
slopes.
Volume of Fire
Basic loads of ammunition maintained in September 1950 are indicated by
the statements of the 13th Div CofS as to the number of rounds on hand in the
division on 25 August in preparation for a new offensive phase, as follows:
82-mm: 120 rd for each of 9 mortars
120-mm: 80 rd for each of 7 mortars
122-mm: 80 rd for each of 9 howitzers
76-mm: 120 rd for each of 20 pieces
Rifle: 200 rd per man
According to Soviet doctrine, the basic load is 120 rd per 82-mm mortar
and 60 rd per 120-mm mortar, and one unit of fire is two-thirds the basiC load.
According to the prisoner's report in Item 1, App C, in normal practice
about 20 rd per 120-mm mortar were stacked near the battery, while 70 or 80
rd were left in trucks 200 m to the rear. As the 19th Reg! had no vehicular
transport in September and its mortars were emplaced at a road junction, it
is likely that the entire basic load was stacked near the battery. Since the
regiment was about to go actively on the offensive again, the quantity of 80 rd
per piece is assumed. The battery at Tongmyongwon would then have had a
total supply of 480 rd On the morning of 12 September.
*The same survey showed with a large statistical sample that wounds of the lower extremities from frag
ments of unknown type also occurred with greater frequency than on Hill Mass 314. Out of 2876 fragment
hits, 44.1 percent occurred in the lower extremities, in contrast to the 26 percent on Hill Mass 314.
ORO-T-261
271
UNCLASSIFltD
SfERET
UNCLASSIFIED
stERn
In view of the importance of Hill Mass 314, it is assumed that from one
half to two-thirds the basic load, or 240 to 320 rd of 120-mm ammunition, were
fired at the hill during the battle on 12 September. This is a very high expend
iture, but men of the 3d Bn 7th Cav Regt said that in the succeeding nine months
of action, including intense engagements with the Chinese army, they never
experienced such intense mortar fire.
Again according to the prisoner (App C), as of 11 August the average daily
expenditure had been 20 to 30 rd per piece, or 120 to 180 rd per battery, and
no restriction had ever been placed on the amount of ammunition expended.
Yet in his 6-mortar company the sustained rate of fire had exceeded 1 rd per
min only on one occasion, when 120 rd were fired in 1 hr. The maximum rate
per tube was then only 1/3 rd per min, even though the same prisoner stated
that the most efficient rate of fire was 7 to 8 rd per min, presumably per
piece. This was a trained unit.
In an interview several years after the battle, a former sergeant of the
P&A Plat of the 3d Bn, 7th Cav said that after the battle he found about 600
rd of ammunition stacked in small piles around the battery of 82-mm mortars
on the northern slope of Hill Mass 314. This seems to be contradicted by the
documentary report after the battle that the P&A Plat on 16 September blew up
one hundred twenty five 82-mm mortar rd at the coordinates 1156.5-1455.6
(north of Knob 3).
If the total recovered was 125 rd, and the basic load was 480 rd, then 355
rd must have been expended. If 600 rd remained at the end of the battle, it is
evident that the 82-mm mortar battery had had an extraordinary supply.
Toward the end of the battle, the same US sergeant cited above heard the 82-mm
mortars firing regularly, so the quantity expended must have been high.
The extent to which 19th Regt units on Hill 570 supported the defense of
Hill Mass 314 is indicated by information supplied by prisoners from the mor
tar company of the 1st Bn, 19th Regt (App A, Item 4). Between 11 and 14
September, a period when there was nO infantry action on Hill Mass 570, one
third of a supply of 200 rd of ammunition for the one 60-mm (US) mortar of one
platoon was expended, presumably at Hill 314. Such fire at maximum range Is
considered to have caused few casualties.
Distribution of 120-mm Fire
The single battery of six 120-mm mortars had to cover such key points
in the 8th Cav area as Hills 373 and 225, as well as 314. While the battle was
still In progress on Hill Mass 314, 8th Cav observers reported that any move
ment elsewhere would draw fire from this battery. It therefore appears that
the enemy split the battery in half, and only three mortars fired on Hill Mass
314.
According to the prisoner's statements in App C, normally only one of
the six tubes was used as the base of fire. Also, only the company commander,
who was at the OP, had a firing table; the only other individual capable of com
puting fire was the leader of the Cmd Sqd, who was also at the OP. Perhaps
this was the reason why the company was said to be incapable of delivering
"massed fires" (volleys?).
272 ORO-T-261
UNCLASSIFIED
'!CREI
SECRn
UNCLASSIFIED
With a general but not immediate ammunition shortage, and with the battery
split as assumed, a reasonable and efficient firing procedure would be: 'in each
group of three mortars one piece used as reference; this mortar fires first and
the other two mortars follow in an invariable order; each burst is observed
before the next mortar fires.
The necessity for accurate adjustment and readjustment of fire is shown
by the dispersion in range estimated above: if the center of impact of each
mortar's fire was always o ~ the aiming point, the scattered bursts would have
been just limited to the desired target area.
Observations of fire received on Hill Mass 314 fit in with such a procedure.
At any time, l20-mm rounds were observed to fall at about the same range. At
the outset of battle, fire overshot the hill mass, and to the tanks advancing on
the right side of the hill seemed to walk along with them. Until the end of the
battle, l20-mm fire always fell along the ridgeline or just beyond on the east
ern slopes. During the different phases of the battle, however, it was spread
out or concentrated. The relatively small dispersion in azimuth (as calculated
above) allowed a choice as to whether fire should be spread out or concentrated
in a narrow strip.
A US sergeant (No. 31, Table E 1) said that fire at Knob 2 seemed to come
in 3-rd salvos, and he reported an incident that suggests the time interval
between bursts. While he and an aidman were occupying a shallow crater to
the right of "the shallow" a man nearby was hit by mortar fire. The medic
started to go to his aid, but then the next 120.-mm round burst, killing the medic
and knocking down the sergeant with its blast. An earlier incident suggests a
similar interval. As described in the narrative of the key action, the leader
of the 3d Plat, Co L was killed around 1215 by a mortar burst, and when the
leader of the 3d Plat, Co I ran over to the spot he was killed by the next burst.
According to the mode of operation assumed above, the salvo interval is
the sum of the time of flight of the projectile, and (since one man must do all
these things) the time to observe the burst, calculate the correction to be .
applied, and transmit the new firing data for that mortar's round in the next
salvo. For the 4500-yd range to Knob 2, the time of flight is estimated to be
39 sec. If an additional 21 sec is allowed for these operations, the interval is
then 1 min. If a round was on target, of course, the next could have been fired
in about 45 sec. An interval of 45 sec to 1 min agrees roughly with the two
incidents cited above.
A 3-mortar rate of 1 rd per min, or 1/3 rd per tube per min, is just the
maximum sustained rate reported by the prisoner from his experience in
August. Although the ammunition shortage had become critical in September,
so had the military Situation, and this high rate of expenditure may well have
been considered justified.
The only friendly assault troops on the northern slopes of the saddle
visible to enemy observation from Hill 570 were those on top of the ridge and
on the we st side. For concealment they gathered in the shallow draw that runs
down to the left from Knob 2. Here they were visible only when erect. The
enemy was then presented with two situations: (a) friendly troops out of sight
and deployed anywhere in the saddle, and (b) assaulting in line as indicated by
troops standing to move on the left side.
ORO-T-261 273
SHAn
UNCLASSIFIED
UNCLASSIFIED
In pOint of fact, the 120-mm fire into the saddle seems to have fallen
mostly on the reverse, northeastern slope. The clue as to when it should be
fired there was supplied by actions of friendly troops on the left side. The
visible advances of Co L units in the first and second assaults were obvious
occasions for intensified fire. With continual prior adjustment of individual
mortars, as assumed above, there Is no reason why they could not have then
fired 3-rd volleys at the rate of 1 per min. Such is taken to be the case.
TABLE 25
ESTIMATED ENEMY EXPENDITURE OF l20-MM MORTAR ROUNDS
ON H1L'L MASS 314 ON 12 SEP 50
Phase of battle
Interval
3-Mort
rate of
fire, rd/min
s
Rd
fired H,
I
Min
Jump off 1135'-1l38 3 0 0
Advance from bottom
of saddle (enemy FPL) 1138-1145 7 3 21
Reorganization 1145'-U50 5 I 5
Assault 1 1150-1155 5 3 15
Close combat 1155'-1215 20 0 0
Assault 2 1215'-1230 15 3 45
Close -contact 1230-1300 30 0 0
Pinned down 1300-1430 90 I 90
Assault 3, move over
Knob 2 1430-1435 5 3 15
Advance toward Knob 3 1435'-1500 25 25
Co K moves over Knob 2 lSOO-I510 10 3 20
Knob 3 occupied 1510-1530 20 I 20
Total, min &rd fired 235 266
aAverage rate of fire 0.38 rd/Mort/min.
From the graphic timetable given earlier, and from the assumptions
above, the volumes of fire in various phases of the battle can be estimated,
with the results given In Table 25.
Calculated Effectiveness of 120-mm Fire
The lethal area of a mortar shell against men lying down and taking advan
tage of the cover afforded by slight irregularities of terrain may be as low as
one-sixth the value against erect men. As shown above and as calculated below,
the effect of slope In reducing the lethal area Is also great. In a very crude
way these combined effects may be computed from data on the battle for Hill
Mass 314. It Is necessary simply to multiply the numper of rounds fired by
the standard accepted value of the lethal area of the 120-mm mortar shell,
(1400 sq yd*) , and by the average density of troops, and then compare the
resulting number of casualties with the number known to have occurred.
-The figure is based on the ss.lb criterion of wounding and a vulnerable area of 41.2 sq ft. Compare it
to the values of -effective radius of fragmentation- of Soviet mortar shells given in the 1948 edition of the
Tactical.Technical Reference Book of the Frunze Academy as follows: 120-mm shell-55 m against men
standing, 25 m against men prone; 82-.mm shell-30 m against men standing, 18 m against men prone. The
lethal area corresponding to the 25-m radius is approximately 2300 yd.
274
ORO-T-261
SEEREJ
UNCLASSIFIED
UNCLASSIFIED
S'ERET
Most of the 120-mm fire fell on the far, eastern slopes of the saddle, i.e.,
in the Co I area. To be sure, some elements of Co I had moved over to the
left side, but elements of Co L made compensating moves to the right. For a
rough calculation It is adequate to assume Co I casualties as the number that
occurred on the right side. As Table 19 shows, from 1130 through 1435 this
number was 30. These casualties apparently were not caused by 82-mm fire,
since, so far as is known, this fire was restricted to the left side of the saddle.
A plot of the deployment of the assault companies at jump off from Knob 1
drawn by the former S-3 of the 3d Bn, 7th Cav and included in the diagrams of
the graphic timetable, shows that two platoons, of, say, 80 men, initially occu
pied an area of about 10,000 sq yd. The density, then, was about 0.008 men per
sq yd. By the time of the third assault, about 40 men (half of the two-company
total of 80 counted on the objective) occupied an area of about 10,000 sq yd, so
the density was about 0.004 men per sq yd. An average value of 0.006 is
assumed.
As shown in Table 25, 176 rd are estimated to have fallen on the saddle
between 1135 and 1430. The number of casualties among troops erect upon a
flat surface is then computed as: 176 x 1400 x 0.006 = 1478 casualties. The
actual number of casu3lties, 30, is 2.0 percent of this fantastic number.
Another very crude calculation can be made upon a different basis to allow
for the effect of slope in reducing lethal area. If the situation shown in Fig. 18
is assumed, the lethal area would be a parabolic section. Along a line at right
angles to the trajectory and through the point o.f burst the standard lethal radius
of 20 yd would obtaln. Up the slope the area of effect would suddenly fall to
near zero. Down the slope it would be reduced by the inclination of the lobes.
With no fragments expelled in a vertical angle of about 16 deg above the slope,
a semierect man with a projected height of 4 ft would have to be within 4.7 yd
of the burst to be hit. The lethal area would then have the following simplified
shape:

T
4.7 yd

The magnitude of this lethal area is 124 sq yd. Use of this value of lethal
area yields the following number of predicted casualties: 176 x 124 x 0.006 =
130 casualties.
This value for erect men is only about four times the actual number of
thirty casualties. Since the men on the right side of the saddle were prone
most of the time, the magnitude of this discrepancy is what one would expect.
ORO-T-261 275
SEun
UNCLASSIFIED
UNCLASSIFIED S,E'AH
INTERPRETATION: THE CAPTURE OF HILL MASS 314
The discussion which follows carries the inductive process of reconstruct
ing the battle from the fragmentary basic data still further in an effort to reveal
the fundamental factors that determined the course and outcome of the action.
At this stage of generalization, hindsight and speculation are necessarily
prominent.
THE SITUATION AND THE SETTING
The general situation around Hill Mass 314 on 12 Sep 50 is discussed in
the interpretation given earlier, in Part ill. When the 3d Bn, 7th Cav Regt took
over the mission of capturing the hill mass on 12 September, the enemy had
just strengthened his forces on the northern, dominating portion and apparently
was himself about to attack.
The course and outcome of the action were determined in large part by
the combined effects of terrain, the particular enemy Situation, and the char
acter of enemy units. These aspects of the battle are outlined here for consid
eration in a later discussion of the course of the fighting.
Terrain
It would be expected that terrain influenced the fighting by reason of such
gross and fine terrain features as the following:
A..gparate, simple, well-defined topgg!:.[lphic feature, Hill Mass 314, was
the focus of observation and fire from both sides.
The s t e e p ~ p e s limited the maneuvering ground to the region of the
ridgetop, and this in turn limited the number of men that could be committed.
Because of the long climbs up along the ridgelines, reinforcements from the
rear could not be effectively committed during an active engagement. Assault
troops were greatly hindered in attacking high pOints. The physical effective
ness of ground-burst mortar fire was greatiy reduced (see preceding section).
Combined with the steepness, the horizontal and vertical convexity of the
slopes tended to isolate the units of an assault line and destroy the coordination
of the attack.
The me!!{;er cover and concealment along the ridgetop diminished the
value of dominating high points by exposing the occupying force to observation
and support fires. The hardness of the ground and absence of large timber
denied the possibility of quickly constructing fortifications. Combined, their
factors would appear to make for fluid combat progressing slowly, in ripples,
rather than rapidly, in tides.
276
ORO-T-261
UNCLASSIfiED
SiElET
SUREr
UNCLASSIFIED
Enemy Situation
The enemy situation on the hill mass on the morning of 12 September can
be surmised. Much of current enemy strength had been present and involved
in the back-and-forth fighting on the central portion for several days. The
enemy must have learned'that a prepared defense could not be set up on what
otherwise would have been the best position from which to repel assault-the
nose of Knob 2 and top of the ridge-and that his men could afford to move up
here only when friendly attacking troops had drawn so near that friendly sup
porting fires could no longer be delivered. Until that time, It would be reason
able for the enemy to disperse his forward troops on the adjacent wooded slopes
and plan for a mobile defense. This would require continual information from
outposts, and control of the troops of the main line. The very low caliber of a
large fraction of enemy strength would probably permit controlled aggressive
maneuvers only in the early stages of an engagement, before the disorganizing
effects of battle were felt.
Character of Enemy Units
Replacements arriving after 31 August made up almost two-thirds of the
strength of the 19th Regt on 12 September, and its units were on Hill Mass 314.
It is also known that in one of the 2d Regt companies on the hill two-thirds of
the men were recent replacements. Most replacements at this time were
virtually untrained, and an enemy device frequently used was that of plaCing
them in forward pOSitions where they could be driven into battle at the gun
points of veterans. * The consequent "brittleness" of enemy units has already
been discussed. In the conditions of fighting as visualized on Hill Mass 314
this would have been a Significant characteristic, though it was of lesser im
portance in enemy attacks. So iong as enemy troops remained 'in place and
offered any resistance, they would have to be rooted out by friendly troops,
which would have to move through the curtain of enemy supporting fire against
the handicap of the slopes. Friendly supporting fire would be effective only if
accurately brought down on the small, separate, enemy groups; dispersed fire
would be of little effect. Eliminating enemy resistance would require breaking
many small, brittle groups, that is, using some but not much fire in the right
place at the right time. Precision of fire and coordination with observation
and current tactical information would then have been especially important.
It would be expected that artillery fire would be more effective than diffuse
strafing, and fire during the battle more effective than preparatory fire.
The net effect of the various situational factors on the limited friendly
offenSive power can be taken as a requirement for nice application and coordi
nation of the several arms and forces involved.
COMPOSITION OF THE OPPOSING FORCES
The identity and strength of enemy units considered to be on the hill mass
are tabulated in Part II in the section on enemy forces. The enemy weapons
on the hill mass are also tabulated in that part, under "Enemy Support,' and
plotted in Fig. 8, as are the enemy supporting weapons in the surrounding area.
*There is concrete evidence that this practice was in fact followen on Hill Mass 314, for the one prisoner
captured during the hattIe on 12 September was not even armed. He had been instructed to pick up a rifle
from a casualty, but instead hid among the rocks until captured.
ORO-T-261
277
nun
UNCLASSIFIED
UNClP-SSIFlED
SURH
Enemy strength on the hill was about 645 men; approximately the same
number of friendly troops fought against them. Friendly forces had a consider
able advantage in immediately available tactical reserves. Friendly light
artillery and medium and heavy mortars were matched by about the same number
of enemy counterparts. The platoon of heavy tanks supporting the friendly
attack was hardly matched by the one enemy SP gun. Friendly air observation
and air support were quite unmatched. There was a shortage of artillery
ammunition on each side, but each had plenty of ammunition for medium and
heavy mortars. It is perhaps significant that direct support of the attack by
the 3d Bn, 7th Cav was to be furnished by the 77th FA Bn (which had not fired
on Hill Mass 314 before), instead of by the 99th FA Bn, (which for several days
had supported the 8th Cav in the fighting on the hill mass). There were no
known recent changes in the dispositions and responsibilites of enemy support
ing weapons, except that additional medium mortars may have been brought up.
COURSE OF THE ACTION
The preparatory air strikes seem to have been mainly preventive: for
ward enemy units were kept from moving up to the best defensive pOSitions,
and perhaps were demoralized to some degree. During the battle friendly as
sault troops found napalm still burning and several incinerated enemy soldiers
on the nose of Knob 2. Few additional bodies, however, were found atop the
ridge. An artillery preparation then, was, not necessary to disorganize a set
defense. For this reason, at least, its cancellation was justified, as no reason
able expenditure could have covered all territory possibly occupied by the
enemy.
In the conditions assumed and outlined above, the first phase of the battle
must have been a race between defending and attacking forces, each trying to
arrive at Knob 2 just before the other, in order to benefit by its commanding
terrain in close combat. Moving up, each was very vulnerable to the supporting
fires of the other Side, but the advantage was all with the enemy, for friendly
assault troops had to move a much greater distance and up far steeper slopes.
The initial deployment adopted by the 3d Bn, 7th Cav was evidently best: a
long thin line closing with the enemy as rapidly as possible. Unfortunately, the
fires of the enemy's final protective line were intense enough to hold up and
momentarily disorganize the attack. It is doubtful that enemy automatic weap
ons were effective here, for, to deliver grazing fire, enemy troops would have
had to move over the topographic crest and expose themselves against the sky.
IntenSifying small-arms fire from scattered outposts down the slopes may have
contributed to the total effect.
The fact that his fires were so successful in stopping the attack 200 or 250
yd short of Knob 2, instead of merely slowing it, apparently caused the enemy
to blunder. He reacted too soon. The large enemy group approaching the knob
from the left was observed, and, because of the still conSiderable separation of
the opposing forces, could be fired on by friendly 81-mm mortars. The group
was dispersed, and the first counterattack was aborted.
Other groups of enemy were moving up on the right side, and they were
not observed. This was the time for a curtain of friendly artillery fire, if it had
been on call.
278
ORO-T-261
UNCLASSIFIED
SERH
SUREY
UNCLASSIFI ED
The enemy reaction to the first assault on Knob 2 was timed correctly.
Scattered enemy groups which had occupied forward positions on the side slopes
of the saddle retreated to the northeast side of Knob 2 before Co I's advance
on the right side. These troops then were in just the right spot to cut down
Co L' s men moving through fire from the left and coming up over the crest of
the knob. It is doubtful that the assault was turned back in hand-to-hand combat.
Enemy troops remaining back from the knob and somewhat off the ridge would
have their fire reinforced by the grazing fire of automatic weapons at Knob 3.
The uncorrected graph of friendly casualties vs time (Fig. 14) does not
adequately represent this first close combat of the battle because it fails to
include the many men of Co L killed in the encounter and reported as casual
ties at an unknown time and under unknown circumstances. In the preceding
section the estimate of nine is made. The correction is discussed below.
If Co L's move in the first assault had been accompanied by a similar
advance on the right side by units of Co I, the assault might have succeeded.
The presence of Co I short of the nose of the knob however, may, have fore
stalled an enemy attempt to rout Co L's repulsed elements. The counterattack
might then have failed to be decisive for the same reason the attack so failed.
The opportunity for enemy troops on the northeast side of the knob to exploit
the situation was removed by the first friendly artillery and by mortar fire.
Enemy units concealed in the underbrush along the finger to the left were so
far forward that they still had every opportunity to attack the pinned -down and
stunned remnants of the elements that had made the assault. Perhaps the con
cussion grenades which these enemy threw were a prelude to such an effort.
If so, this opportunity was also taken away when the two company commanders,
by their own example, aroused the men to make a decisive reply with fragmen
tation grenades.
These several encounters around the time of the first assault illustrate
a feature characteristic of the battle after its opening: friendly: and enemy
groups reacted as much, or more, to local pressures as in accordance with a
consistent general plan.
Mter the first assault, infantry move and countermove decreased in im
portance, and supporting fire became the dominating influence. Though still
in fairly close contact, the opposing troops were hardly visible to each other.
Some men reported gOing through the entire battle in forward elements without
ever seeing more than a couple of enemy soldiers at a time. When the enemy
troops fell back somewhat, the enemy mortar fire moved up accordingly; The
direct effect was to make friendly troops seek the almost nonexistent cover
and such concealment as was afforded by underbrush and irregularities of
terrain. The indirect effect was to tend to immobilize them in a dangerous
situation, for they could not dig in.
Figures 14, 15, and 16 show the accumulation of casualties in the several
friendly units and from different causes through the various phases of the battle.
In interpreting the curves, the reservations discussed in the preceding section
of this memorandum must be kept in mind. To correct the significant omission
of many KIAs because of lack of knowledge as to time, perhaps nine men of
Co L may be taken as killed in the first assault, as noted above. Then, by
1300, after the second assault, Companies L and I would have lost 30 men as
casualties, or about 9 percent of their initial strength. By the end of the period
of inaction, at around 1430, casualties would then have increased to 69, or
ORO-T-261
279
SEERET
UNCLA,SSIFlfD
UNCLASSIFIED
SEEKn
about 20 percent of initial strength. The average casualty rate of those squads
still under fire probably was approaching 50 percent.
When the correction for the first assault is made, the over-all cumulative
casualtY. curve for the two assault companies becomes nearly' straight during
that period of time which runs from the beginning of the first assault to the
of the third. The corrected curve appears dotted in Fig. 14.
As casualties mounted under the enemy mortar fire, the problem of
maintaining control and initiative increased. Platoon leaders had been loat,
electrical communications were crippled, some elements were dispersed, and
others were intermingled. Reports that the troops at times were panicked are
quite understandable. For the individual, participation in a positive move
meant leaving a spot where he had a fair chance of surviving the fire even
though he was pinned down, and accepting a temporarily increased chance of
being hit by intensified fire.
Leadership was vital. There were some old, veteran sergeants as squad
leader s who urged their men forward whenever they tended to slip back into
the fire, and there were platoon leaders, and later assistant platoon leaders,
who reorganized their men and sometimes directed positive moves on their
own The first occasion when the two company commanders acted
to break the hypnotic inaction of the pinned-down men has been noted. They
repeated this service again to get the third assault started. The presence of
the battalion commander in a rather vulnerable forward position at the OP on
Knob 0 also was a positive morale factor.
Without significant addition to friendly striking power it is doubtful that
.the objective would ever have been taken, for after the second assault the
attack was definitely stalled and the inertia of the assault troops had become
considerable. At regimental headquarters, the order to hold up the attack for
continuation the next day was being considered. The 3d Bn, 8th Cav had stopped
short of Knob 2 late on 10 September, only to be hit harder and forced back the
next day. In view of the manpower situation of both sides, it seems very unlikely
that the 3d Bn, 7th Cav would have been any more successful against the now
greater enemy force. The battalion was already nearly crippled by casualties
and would have gone on taking them at Knob 2. Although, with the almost intact
Co K and the support of the ROK training battalion, it may well have been able
to hold at Knob I or Knob 0, it seems unlikely that the battalion would have been
able to again regain impetus at a later time to push up to and over the objective.
The enemy infantry had also been hard hit. But the enemy had the ability to
replace his losses fairly readily with more low-grade manpower. Given the
advantages of situation and terrain and called on to make only limited maneuvers
in defense of the objective, such troops might very well again have stood off
friendly attack.
The necessary stimulus was provided by the unusually effective final air
strike, the first close air support during the battle. Though the detailed circum
stances surrounding this strike remain unknown, it must be assumed that the
presence of the group of enemy flushed by napalm from behind a knoll down the
ridge was not known to the company commanders and fire observers in "the
shallow. Otherwise, artillery could just as well have been called in to do the
same job. The dispersal of these enemy demonstrated that there were no
longer any enemy troops on a commanding terrain feature north of Knob 2.
280
ORO-T-26l
UNCLASSIFIED
SIEREr
UNCLA.sSIFIED
-'ECREf
(Up to this time friendly advances had always been resisted from some com
manding feature beyond.) For the first time, the way for a real advance was
therefore shown clear. Yet another act of leadership was also necessary to
restore momentum to the attack, and this was again supplied by the company
comman?ers when they led the initial charge of a small group of men. Since
this act was essential, it seems certain that if a mortar round had happened
to fall in the shallow whEln enemy mortar fire was inflicting casualties just to
the left and right, the third assault and capture of the objective would not have
taken place, even with the aid of the air strike.
Throughout the battle, the enemy moved back reluctantly and just to the
extent of friendly advance. By the late stages of the battle, whole units seem
to have been eliminated, and shortages of some types of ammunition may have
become critical. The enemy command appears to have prepared to desert the
hill if sufficiently pressed, but in the same piecemeal, reluctant fashion, in
accordance with the pressure at each point of contact. Atop the ridge, the
napalm hit of the final air strike dispersed the spearhead of the last enemy
defense. The support of automatic weapons, as considered below, appears
already to have been lost, the heavy machine guns evacuated. Down on the
right Side, the units which had so long confronted Co I appear to have withdrawn
suddenly, for sometime during the third assault the 120-mm mortar fire from
beyond Hill Mass 570 finally shifted north of Knob 2 to cover the right side of
the ridge, and word of withdrawal here must have been received. The continual
heavy explosions occurring at the base of the hill at this time are unexplained
and may have been enemy demolitions. However, at the base of the hill to the
north, the 82-mm mortars continued to fire at Knob 2 while friendly forward
elements were approaching Knob 3. The shift of the 120-mm fire, the evacu
ation of the mortars (with the exception of one baseplate), and the abandonment
of some six hundred rounds of scarce ammunition combine to suggest a sudden
and late but deliberate withdrawal here.
According to a prisoner of war, the enemy had five M1910 Maxim heavy
machine guns, and all were retained by the machine gun company. The most
likely location for these weapons is Knob 3. They must have been evacuated
well before friendly units reached there around 1530, for these guns are heavy
(145 Ib, with tripod), and awkward to handle. They were supplied with only one
box of ammunition per gun. If one belt of 250 rd was meant by the prisoner, the
guns may well have run out of ammunition in repelling the first and second
assaults, for the "practical" rate of fire is 250 to 300 rd per min, the cyclical
rate 500 rd per min. Aside from these machine guns, friendly units were not
opposed with effective fire in the third assault. It is hard to see how the CO
of Co L and two others, who were in the lead, could have killed an enemy soldier
who had an automatic weapon (the one DP light macr..ine gun that was captlJ.red?)
in a position halfway to Knob 3 along the ridgeline if this gun had been firing.
The number of enemy casualties in the battle on 12 September is not known.
In the subsection on casualties on Hill Mass 314 under "Enemy Units" in Part
II, a rough estimate is made of total casualties in the fighting from 8 through
i2 September. During this period both sides built up strength at about the same
rate, and they remained in tactical balance until almost the end of the final
battle. There is, then, reason to assume that the ratio of the friendly and enemy
casualty rates remained constant. Of total enemy casualties, the fraction sus
ORO-T-261 281
SEEREl
UNCLASSIFIED
UNClA.SSIFIO
SEERn
tained on 12 September can then be computed from the known fraction for
friendly casualties. Table D2 shows that 113 friendly casualties occurred
before 12 September and that 153 occurred in the 3d Bn, 7th Cav on 12 Septem
ber. (A few IIA are included here.) Applying the factor 153/(153 + 113) to the
enemy totals gives the breakdown shown below. A comparison may be made of
over-all strengths and casualties of the two sides:
Enemy 3d Bn. 7th Cay Regt
645 men on hill 610 men forward on hill
a
95 KIA
47 WlA. then KIA
142 killed 38 killed
237 WIA. escape from hill 115 WIA. 11A
379 casualties on hill 153 casualties on hill
266 escape from hill unhit
60% of initial strength casualties 25% of initial strength casualties
48% of remaining strength wounded
8Three rifle companies, one platoon each (rom Companies M and H3.
The ratio of enemy to friendly casualties, 2.5, is close to the figure for
the first phase of the offensive at Tabu-dong.
As noted earlier, in the section on enemy units, enemy wounded at this
stage of the war r e m ~ i n e d with their units for primitive treatmenL If this
was not adequate, they died. It is unlikely that seriously wounded enemy sol
diers could have withdrawn with their units to the rear. If, as calculated, 77
percent of the enemy force escaped from the hill, and if 48 percent of these
troops were wounded, attrition may well account for the failure of the enemy
to counterattack.
If, as discussed above, the enemy did decide to withdraw late in the battle,
this decision was made when over half the total initial enemy troops had become
casualties.
In the preceding section of this memorandum it is concluded that the
causes of friendly casualties break down as follows: 19 percent in close combat
(bullets and grenades); 81 percent from enemy supporting weapons (artillery
and mortars), of which, 2 percent of total casualties from artillery, and 79
percent of total casualties from mortars.
This breakdown includes a correction for men killed in the first assault.
The rather low proportion of easualties in close combat is not surprising in.
view of the outline of the action given in the graphic timetable. After the first
assault, there was little overt movement until the third assault, and neither
side presented many targets. The enemy ammunition shortage may also have
been involved. Individual soldiers went into the battle with 60 rd of rifle
ammunition, an amount typical of this time, whereas in late August the amount
was 200 rd per man. The same 7.62-mm rd were used by the light machine
282
ORO-T-261
UNClA.SSIFIED
SEEREf
UNCLASS1FIEf'l
SEEREl
guns, but no estimate can be made of the quantity per gun. The heavy machine
guns, as pointed out above, had an inadequate ammunition supply.
Why did enemy artillery play such a negligible role? Information from
various sources has located the Arty Regt of the 13th Div along the road north
of Tabu-dong and about 8500 m from the saddle on Hill Mass 314. It is likely
that the pieces were not emplaced to fire in this narrow corridor but were
merely located here in anticipation of a general enemy advance. Theoretically,
the 76-mm guns, with a maximum range of 12,500 m were able to fire on Hill
Mass 314, though not with low-angle direct fire, as the intervening ridge sub
tended an angle of over 30 deg. The fire of the 122-mm howitzers would not,
of course, have been restricted by terrain, but these weapons, with a maximum
range of 8500 m, would have been firing at their utmost reach. Since only the
122-mm howitzer of all enemy artillery had time-fuzed ammunition (see App
C) and there were a few reports of time fire received on Hill Mass 314, it is
possible that a few 122-mm rounds were'fired during the battle.
The supporting artille,y of the 19th Regt has been located rather definitely
at Site 2 in Fig. 8. Defiladed by Hill Mass 570 they were less exposed than the
13th Div Arty to friendly observation, counterbattery fire, and air attack.
Since the 76-mm howitzers have a maximum range of 8500 m and Hill Mass
314 was little more than 4000 m distant, these weapons could very well have
fired. The analysis of friendly casualties agrees with the consensus of
observers' reports that they did not. Possible reasons for the apparent
failure to fire are: (a) The howitzers had been moved into position only
recently and were not yet ready to fire, perhaps because the telephonic link
with the observation post had not yet been established. (According to the
enemy source reported in App C, unobserved fire was very seldom used.)
(b) There was a general ammunition shortage, and 76-mm ammunition was
allotted to the two SP guns for direct fire from temporary forward positions,
one to the west and one to the north of Hill Mass 570. (c) Friendly counter
battery fire was in this case effective.
None of these assumed reasons is very convincing. The CofS of the 13th
Div stated that ammunition shortage rather than weapon shortage (except small
arms) was a critical problem, and several captured enemy documents seem to
support this assertion. Yet when the positions north of Tabu-dong were over
run around 21 September large amounts of ammunition were found with the
artillery pieces along the road. There might, of course, have been very recent
resupply.
ORO-T-261 283
UNCLASSIFIED
SEEREI
UNCLASSIFIED
stCRn
REFERENCES
UNCLASSIFIED
SECRn
UNCLASSIFIED
stfRET
REFERENCES
1. G-3 Journal, 1st Cavalry Division, September 1950.
2. War Diary, 1st Cavalry Division, September 1950.
3. Journal, 8th Cavalry Regiment, September 1950.
4. War Diary, 8th Cavalry Regiment, September 1950.
B. War Diary, 7th Cavalry Regiment, September 1950.
6. Journal, 3d Battalion, 7th Cavalry Regiment, September 1950.
7. Unit Historical Report (including diaries and journals), 5th Cavalry Regiment,
September 1950.
B. War Diary and Journal, Bth Engineer Combat Battalion, September 1950.
9. Unit Historical Report, 16th Reconnalssance Company, 1-30 Sep 50.
10. War Diary, 1st Cavalry Division Artillery,September 1950.
11. War Diary, 99th Field Artillery Battalion, September 1950.
12. War Diary and Journal, 77th Field Artillery Battalion, September 1950.
13. War Diary and Journal, 1st Artillery Observation Battalion, 25 Jun-30 Sep
50.
14. War Diary and Journal, 17th Field Artillery Battalion, September 1950.
15. Journal, 70th Heavy Tank Battalion, September 1950.
16. War Diary, Company A, 715t Heavy Tank Battalion, 3 Jul-20 Sep 50.
17. TACP Log, 1st Cavalry Division, Aug-14 Oct 50.
18. Mission Reports, 5th Air Force, 1950, Unpublished, collected by Operations
Research Office.
19. Daily War Diaries (collections of miscellaneous reports), Eighth United
States Army Korea (EUSAK), September 1950.
20. Summary War Diary, Eighth United States Army Korea (EUSAK) , September
1950.
21. Log, 0-3 Section, EUSAK.
22. War Diary, I United States Corps, 2 Aug-30 Sep 50.
23. Afteraction Report, 27th Infantry Regiment, August 1950.
24. Afteraction Report No.1, Company C, 73d Tank Battalion, covering oper
ations for August 1950.
25. Operations Research Office, "A Study of Battle Casualties Among Equiva
lent Opposing Forces, Korea, September 1950," ORO-T-23(FEC), 14 Apr
52.
26. Operations Research Office, "The Employment of Armor in Korea,"
ORO-R-1(FEC), 15 May 51. .
27. Operations Research Office, 'Close Air Support Operations in Korea,"
ORO-R-3(FEC), 1 Feb 51.
ORO-T-261 287
UNCLASSIFIED
SEERET'
UNCLASSIFIED
SEnn
28. Ordnance Intelligence, Ordnance Corps, DA, "Foreign Firing Tables for
Soviet Mortar, 82-mm, Models 1936, 1937, 1941, and 1943," March 1951.
29. Ordnance Intelligence, Ordnance Corps, DA, "Soviet 120-mm Mortar,
1943 and 1938 Models,' August 1950.
30. Ballistics Research Laboratory Technical Note No. 293, "Analysis of
Fragmentation Tests of Shell, Mortar, 120-mm HE (North Korean),"
October 1950.
31. Surgeon General's Office, DA, Medical Research and Development Board,
"Wound Ballistics Survey, Korea, 15 November 1950-5 May 1951."
288 ORO-T-261
SEERrI'
UNCLASSI FI ED
UNCLASSIFIED
SUDK
n ...
Appendix A
13TH AND 1ST NK DIV POW INTERROGATIONS
ORO-T-261
289
UNCLASSIFIED
SfCREl
UNCLASSIFIED
SURH
EXTRACTS FROM REPORTS OF INTERROGATION OF PRISONERS
FROM THE 13TH AND 1ST NORTH KOREAN DIVISIONS (11 ITEMS)
ITEM 1
ATIS Interrogation Report No. 895
Field Report (ADVATIS 0700)
5 Sap 50
3 Sep 50
Name: Kim, Song Jun
Rank: Major (promoted 20 Aug 50 from Sr Lt)
Age: 25
Organization: 13th Dlv 19th Regt Hq
Duties with Unit: Regtmental S-3
Education: Middle school for two years
Occupation: Farmer
Address: Hamyong Pukto, Hoeryong-gun, Hoeryong-myon, Il-dong #212
Date and Place Captured: Surrendered 2040, 1 Sep 50, 5 km SE Indong (1148.3-1464.8)
Interrogator: Sgt Hayashida (ATIS)
Evaluation:
Subject was very cooperative and is of average intelligence. Subject surrendered
because he was scheduled to be courtmartialed, and is very bitter about his superiors.
S u b j ~ c t seems to be very disgusted with the turn of affairs since he held a relatively high
position with his unit. Reliability-good.
Military History:
Date Inducted: Volunteered 20 Jun 46 at Chinnampo
Movements:
20 Apr 50-15 Jun 50: Asgd 19th Regt, 13th Dlv. Enter Dlv stationed in Sinuiju. Strength
approx 12,000. 13th Dlv organized Mar 50 of recent conscripts.
15 Jun 50: Entire Dlv departed Sinuiju by rail.
23 Jun 50: Entire Dlv assembled Kumchon (940-1710) and marched east to Yongchon.
Crossed 38 Parallel 26 Jun 50.
10 Jui 50: Entire Dlv arrived Seoui. No Incidents en route.
10 Jul-17 Jul 50: Dlv engaged In search for South Korean officials and military pollee.
No one was apprehended. 17 Jul 50, entire 13th Dlv departed Seoul, marched southeast
to Chungju.
ORO-T-261
291
SECRET UNCLASSIFIED
UNCLASSIFIED
$lEMT
1 Aug 50: 13th Div arrived Chungju. No Incidents en route. Upon arrival at City, subject
observed rear elements of 1st Div leaving. 13th Div followed 1st Div to Sangju in an un
known direction on or about 10 Aug 50. 13tb Div departed Sangju on or about 10 Aug
50 and arrived Naktong-ni on or about 15 2000 Aug 50. 13th Dlv had not been committed
to battle until this date. Between Mungyong and Naktong-ni the Division sustained dally
air attacks. Suffered losses of apprOximately 2,000 men. Strength of Dlv on 15 Aug 50,
approx 10,000 men. 13th Dlv started to cross Naktong River aI Naktong-ni 15 2000 Aug
50. Crossing spearheaded by 21st Regiment. 21st Regiment less artillery crossed under
heavy artillery and small arms fire, sustained casualties of about 700 men. During Initial
phsse of crossing, 13th Dlv Arly prOvided fire support but was soon forced to cease fire
after disclosing Its position. 2400 15 Aug 50, 19th Regt less artillery, crossed Naktong
River. 23d Regtment and Div Arty remained in the hills north and south of Naktong-ni.
0400 16 Aug 50, last elements of 19th Regiment crossed river. Immediately after cross
Ing river. POW met a few stragglers from the 15th Dlv wbo were seektng their unit.
Locallon of 15th Dlv aI this time, unknown. After crossing river, 21st Regiment advanced
south In column on Sangju-Taegu Highway, followed by 19th Regtment apprOximately 2,500
men. One company of 1st Baltalion, 21st Regiment marched 500 meters abead of tbe main
body of troops. At point 1 kilometer southeast of Togae-dong (1130-1480) 2400 16 Aug,
leading company drew fire from ROK forces. The company left the road and occupied
high ground Immediately west of road, where it was JOined by otber elements of 1st Bat
talion (21st Regtment). 2d and 3d Battalions, 19tb Regiment occupied valley between 21st
and 19th Regtment. (23d Regiment and Div Arty still on west bank of Naktong River.)
1st Battalion fought south and occupied adjoining ridge at 0400 17 Aug 50. 2d and 3d Bat
talions, 21st Regiment advanced south to the bottom of a narrow valley. ROK forces held
ridge Immediately south of 2d and 3d Battalions, 21st Regiment. 1st Battalion, 19th Regi
ment bypassed main body 21st Regiment on left flank and advanced to the foot of the ridge
facing 2d and 3d Battalions, 21st Regiment. 2d and 3d Battalions, 19th Regiment made no
progress. At daybreak 17 Aug 50, all advances made on night of 16 Aug 50 were lost
because of beavy artillery fire and air altacks. Consequently, tbe units withdrew to their
original positions and took up defensive tactics. Until 2000, 17 Aug 50, the units were
without artillery support and were subjected to constant artillery shelling and air attacks.
At 2000 17 Aug 50, the units with Dlv Arty support went from the defensive to the altack.
The units simultaneously advanced south and overcame the ROK forces. The engagement
ended at 2300 17 Aug 50. In the engagement of 16 and 17 Aug 50, 21st Regiment suffered
800 casualties, 19th Regiment lost 700 men. After the engagement, the units marched
south, led by 21st Regiment.
18 Aug 50: 23d Regiment joined 21st and 19th Regiments. At 0500, the Dlv engaged UN
forces at point 5 kilometers southeast of Naksong-dong. Engagement lasted until 2100.
21st Regiment lost 400 men; 19th Regiment, 300 men; 23d Regiment, 400 men. Strength
21st Regiment, apprOximately 600 men; 19th Regiment, 1,500 men; 23d Reglment,2,100
men. This was the first engagement for 23d Regiment.
19 Aug 50: 0001, 23d Regiment and 21st Regiment engaged UN forces. 19th Regiment
held in reserve. Skirmish ended 0400. Total losses, approximately 100 meD. 0400
1
23d Regiment deployed northern slope of Yuhak-san at a point 1 kilometer southwest of
principal road Junction, 21st Regiment occupied high ground 1 kilometer southeast of
road junction, and 19th Regiment was held In reserve In Sajang-dong [sic, Hajang-dong?]
area.
292
ORO-T-261
UNCLASSIFIED
SEERn
S.E(RET
20 Aug 50: 21st Regiment withdrew because of heavy artillery shelling to southeast
slope of Yuhak-san. 23d Regiment remained in position. 19th Regiment deployed 1 kilo
meter east of road junction east of Yuhak-san. 0400 20 Aug 50, 45th Regiment, 15th Div
arrived Yuhak-san and was deployed 1 kilometer west of 21st Regiment position. 13th
Div Arty situated Sajang-dong.
21 Aug 50: Position unchanged. 2000 hours, 13th Div reinforced by 45th Regt, 15th Div
started offensive. Objective, Tabu-dong. All units except 19th Regiment jumped off at
2000 hours. POW assumed command of 19th Regiment on that day. Also 2d Battalion,
19th Regiment had become disorganized under an art!llery shelling and POW was st!ll
in process of reorganizing the battalion when the attack o r d ~ r came. Consequently, 19th
Regiment started attack ai 2130 hours. Objective of 19th Regtment was to neutraiize
art!llery position 4 kilometers north-northeast of Tabu-dor.g. Because 19th Regiment
was i hour and 30 minutes late, the division attack became disorganized and the units
were forced to fall back to their original position. Attack supported by 5 x T-34/85 tanks.
4 destroyed.
24 Aug 50: 13th Div received 800 replacements. 400 brought their own weapons. Remain
ing 400 were issued weapons when assigned to regiment. 19th Regiment received 182 men,
remaining 600 men assigned to 21st and 23d Regiments. Div strength with replacements,
approJdmately 3,000 men. Subject investigated for failure to obey orders on 21 Aug 50.
24 Aug 50 - 31 Aug 50: 13th Div went on defensive. Position unchanged except for absence
of 45th Regiment, which POW heard was deployed in Kunwi area. Units ordered to have
4 days rations on hand.
31 Aug 50: 4,000 replacements arrived 13th Div. Only 2,000 of the men arrived with
weapons. Remaining 2,000 men equipped upon assignment to regiments. By 31 Aug 50,
division strength was approximately 9,000 men. Replacements, stragglers, and wounded
personnel discharged from field hospitals were absorbed. General preparation for offen
sive made. Each man issued 300 rounds ammunition. Each Regimental Artillery piece
issued 120 rounds ammunition. Unit had in stock enough ration for 4 days. Subject
heard 1st Div deployed on left flank at Kasan-myon. Subject heard a general offensive
was to start. Details, unknown. Objective, Taegu. Division was expecting the arrival
of an undetermined number of heavy tanks, type unknown.
31 Aug 50: Subject deserted because of the 21 Aug incident. He was accused of sabotag
ing the division's effort and was scheduled to be tried.
Enemy Intentions:
13th Div (other divisions' unknown) objective is Taegu. Route of march would be
via Taegu-Sangju Highway. Subject heard that objective nf 2d Corps, of whtch the 13th
Div was a part, would start an offensive with Taegu as the objective: No Information on
plans.
Unit Capabilities: (13th Div)
a. Original Strength Present St:rength (31 Aug)
19th Regiment: 2,500 Men 1,800 Men
27 x 82mm Mortars 4 x 82mm Mortars
ORO-T-261 293
SEUU
UNCLASSIFIED
SlEln
UNCLASSIFIED
4 x 76mm Howitzers 4 x 76mm HOwitzers
6 x 120mm Mortars 6 x 120mm Mortars
12 x 45mm AT Guns 12 x 45mm AT Guns
21st Regiment: 2,500 Men 1,800 Men
Equipment identical to 19th Regt Same as 19th Regt
23d Regtment: 2,500 1,800
No details Hardly any heavy weapons
No details
Div Arty: Strength, unknown. 31 Aug 50, 2Q guns remaining. 10 x 122mm Howitzers and
10 x 76mm Guns.
b. Supply: Div unknown. 19th Regt was resupplied by 15 trucks. Div Supply trucks
hauled food and ammUnition to regtmental supply dump (31 Aug 50, situated Songyo-dong
under a 2.5 meter long bridge on main highway) which was in turn tranaported by 15
regimental vehicles to battalions. Amount and frequency depended upon situation. The
Regtment was never short of ammUnition. Standard Unit of fire carried for each artillery
piece was 60 rounds. 50% of regimental rations came from Div Supply, 50% was obtained
locally.
c. Morale: Morale was high. Orders obeyed without question. Replacements were
eager to enter battle. There was no friction among the officers.
d. Replacements: 13th Div replacements came from Seoul. Majority of troops had
at least one month training. Usually, 50% of troops came armed With rifles. Replacement
Depot, unknown.
Armor:
Observed 21 Aug 50, 5 x T-34/85 tanks. 4 destroyed on 21 Aug 50. One tank re
mained. Location, road Junction east of Yuhak-san. 31 Aug 50-Hearsay, 13th Div expec
ting arrival of undetermined number of heavy tanke, type unknown.
ITEM 2
eONF WEN lIAL
Headquarters 1st Cavalry DiviSion
Office of the ACofS, G-2
Interrogation of Prisoner of War Report
Report No.: 0085
12 Sep 50
Map: Tactical Map of Korea, 1:50,000 HT
1. Prisoner and Unit Identification
Personal Details Unit Place of Capture
Name: Cheh, Song Sam 2d Plat, 3d Co, Captured by F Co.
Rank: Pvt 1st Bn, 19th Regt, 8th Cav in vic
Length of Sv: Inducted 19 Aug 13th Div (1153.9-1453.8) on
Duty: Rifleman 12 Sep 50
Temp. POW No.: CD-2l6
2. Organization, Strength and Diaposition
A. Own Unit: PW stated that he was inducted on 19 Aug, and was told that he would
be returned home after a 40 day tratning period at Hason. However, after three
days of training PW was sent to the front with approx. 1,000 men. The 1,000 men
294 ORO-T-261
SEERE'
UNCLASSIFIED
UNCLASSIFIED
traveled by rail to Seoul, arr there on or about 25 Aug. At Seoul the group was
Joined by 1,000 trained troops from Chlnanpo and marched towards the front. On
or about 28 Aug, the two groups arr ai Wonju 'and here the groups separaied again.
However, where the other group went is unknown to POW. On 28 Aug, the group
of 1,000 continued Its march and on 3 Sep arr at pt 2 hills [sic, must be miles]
N of Kasan (Hill 901) (TN: Believed to be Hill 742). There POW was assigned to
the 3d Co and on the same day the entire Bn left for Mt Kasan. POW stated he
was told that the 3d Co had approx 70 men when i ~ left. However, he did not know
how the 1,000 repls were distributed. POW stated on or abt 9 Sep the 1st Bn arr
at the Wsllad City and continued on to hill a little lower In height than Mt Kasan
localed to S of Mt Kasan. (TN: Believed to be N of Hill 570.2). POW stated that
his unit arr here on 9 Sep with an estimated strength of 200. However, POW did
not know whether this was the Bn or Regt'l strength. POW stated during the day
of 10 Sep the unit received heavy arty shelhng and aircraft strafing. Therefore,
on the night of 10 Sep he and POW CD-213 left tbe unit position and started down
the mountain to 'escape death. Upon getting to the toot of the mountain, due to a
fierce rifle firing in the vic he and his buddy went into a bouse in a nearby village.
WhIle at the village they noticed NK troops In the vic engaged in a fire fight with
the en. Therefore, they remained hid in the house until after the NK troops with
drew and on the morning of 12 Sep he and his buddy walked toward the American
line unarmed with intentions of surrendering. Upon the American line POW is
not certain who had shot, but his buddy was shot through the chest. POW stated
thai 1st Bn HQ Is located on hill at (1154.3-1456.7).
B. Other Units: POW stated that while he was biding in village at (1154.1-1453.8)
an NK soldier of the 1st Dlv found them and asked bim what unit he was from.
Therefore he told him he was from the 13th Div. POW stated that he saw approx
30 1st Dlv troops [Hill 570] withdrawing for the village into hill to the N on evening
of 11 Sep. POW could not identify the regt to which they belonged or the location
of their main force.
POW stated that his sqd Idr imormed them that the 13th and the 1st Dlv Were
engaged In a coordinated operation.
3. Mission: POW stated that the men were not told what their mission was. All he had
heard was that they were advancing to Taegn.
4. Enemy Supplies and Equipment
A. Supply
1. Ration: POW stated that no food is supplied for the rear and It is not out of
ordinary to go without food for 2 and 3 days. POW further stated that they
are able to get barley in the nearby villages. However, If this is exhlll1sted it
will mean going without food or forage fr distant villages.
2. Ammo: POW stated that each man had approx 50 rds of rifle ammo on the
average and two hand grenades. POW stated there are some individusis with
5 or 6 hand grenades but on the whole each man carries two.
B. Eqnipment
1. Weapons: POW ,stated that each Co had 1 LMG and 2 HMGs and the rest had
bolt action rifles. POW stated bis Co did not have any Sub MGs. POW stated
that he had never seen any mortars or arty pieces.
2. Communication: POW ststed that telephone wires were laid between Co and
Bn, and communication done by wire.
3. Armor: POW stated that he observed approx 10 tks (TN: believed to be T -34)
travelling S while at Wonju. However, POW had not seen any tanks after
coming to the front.
ORO-T-261
295
S.(Rn
UNCLASSIFIED
UNCLASSIFIED
SiEHl
4. Transportation: POW stated that he saw numerous trucks transporting ration,
ammo and other supply as far S as the road fork at (1151.9-1464.0). However,
he had not seen trucks S of this pt due to heavy arty shelling throughout the
area. Fr this pt, supplies are transported by civilian and troop carrier parties.
5. Enemy Losses
A. Personnel: POW stated that lus untt suffered heavy casualties through enemy artil
lery but was unable to gIVe percentage or number of casualties due to repis received.
B. Supplies: No inlormatlOn.
C. Equipment: POW stated he had not seen any losses in organizational equipment.
6. Replacement of Personnel and Materiel
A. POW stated that upon arr at his unit, due to small amount of troops he asked the
reason for it and was told that more recruits like himselt would be arriving.
However, POW stated he had not seen any further arr ot repls. POW stated that
although he was given his uniform upon induction into the service he dld not receive
his rifle until arrival at Kunwi.
B. Weapon: POW stated that no repl of weapon was made.
7. Miscellaneous
A. Enemy Methods: POW stated that due to shortage of oflicer repls, Sqd Ldrs are
made Plt and Co Commanders, Co commanders made Bn commanders, and so on.
The same is true for the troops and the majority of the troops consists of men who
were thrown into the line without training.
B. Enemy Intentions: No into.
C. Morale: POW stated that due to the shortage of tood he believes the morale of the
men to be very low. He further stated that there were many men reported missing
in action but he believes these were all deserters, POW stated the officers are
well fed, therefore of high morale.
D. Personalities! POW stated they were not told the names ot the commanders.
8. Remarks: POW, although his military knowledge was nil, was of average intelligence
and very cooperative. Theretore, information given by him 15 considered frurly reliable.
Further interrogation due to bis recent departure fr North Korea is recommended.
S. Yamada
Capt Inf
ITEM 3
ATIS Interrogation Report No. 1158 17 Sep 50
Field Report (164-LDI-086B) 15 Sep 50
1. Personal Detalls:
POW Name: Che, Song Sam
POW Number: 63 NK 926
Rank: Pvt
Age: 30
Duties: Rifleman
Organization: 13th Div, 19th Regt, 1st Bn, 3d Co, 2d Plat, 3d Sqd
296 ORO-T-261
UNCLASSIF1G"
SKMT
UNCLASSIFIED
SERU
Education: Primary school (6 years)
Occupation: Farmer
Date and Place Captured: 12 0900 K [120800 I) Sep 50. Approximately 16 kID NW of
Taegu ( 1153.9-1153.8) by "F" Compaoy, 8th Cavalry Regiment
2. Assessment: POW was cooperative. Information limited due to short period of mili
tary service. However~ it seems reliableQ
He was a deserter, stating he felt safer being a US Prisoner thao with NK Army.
3. Tactical Information:
TO&E of POW's Unit:
13th Div, 19th Regt, 1st Bn, 3d Co, 2d Plat, 3d Sqd
3d Platoon [2d Platoon, see above and Item 2)
A. Personnel
1. 3 squads
2. 5 men per squad
3. Total of 17 men
B. Equipment and Ammunition
1. 17 x M-1891/30 rifles, 50 rounds per rifle
2. 2 x grenades per man
3d Compaoy
A. Personnel
1. 3 platoons
2. Total of apprOximately 70 men
B. Equipment
1. 1 x HMG (ammo unIa1own)
2. 2 x SMG (ammo unIa1own)
NOTE: Above mentioned information as of 3 Sep 50. No further Information.
Disposition of above Unit:
3d Bn Hq, 19th Regt, 13th Division, was located on a mountain apprOximately
18 kID NW of Taegu at the time of his capture on 12 Sep 50. Disposition of
others units was unknown.
Estimated strength of above Units: (as of 12 Sap 50) [10 Sep)
Unit: 3d Platoon
Original Strength: 17 men
Present Strength: 10 men
Unit: 3d Co 19th Regt
Original Strength: 70 men
Present Strength: Unknown
Higher echelon unIa1own.
Personal and Unit History:
The 3d Battalion advaoced southward from Kunwi and arrived on a mountain
(location unIa1own) on 10 Sep.
On 11 Sep 50, POW with another man deserted and was captured by US troops
approximately 16 km NW of Taegu on 12 Sep 50
POW had only 10 rounds of ammo left when captured on 12 Sep 50.
ORO-T-261
297
UNCLASSlnL'
SECRET
UNCLASSIFIED
Hun
ITEM 4
Headquarters 1st Cavalry Division
Office of the ACofS, G-2
Interrogation of Prisoner of Wax Report
Report No.: 0088 15 Sep 50
Map: Tactical Map of Korea, 1:50,000 MW
1. Prisoner and Unit Identification
Personal Detalls Unit Place of Capture
Name: Ko, Yoan Gook 1st Plat, Mortar Co Captured at Hill 570
Rank: Pvt 1st Bn, 19th Reg!, (1154.6-1456.1 by G Co,
Temp POW No.: CD-242 13th Div 8th Cavan 14 Sep
Name: Wan, Yong Sun Ditto Ditto-
Rank: Pvt
Temp POW No.: CD-240 (sic) [24111
Name: Ahn, Un Soon Ditto Ditto
Rank: Pvt
Temp POW No.: CD-240
2. Organization, 'Strength and Disposition
A. Own Unit: POW" stated that they were forced to join the NK army on or abt 22
Aug 50, at Wonju (SK). On or abt 29 Aug 50, POWs group of 1,000 men were told
they were to haul ammo and supplies in vic of Andong but they were deceived and
brought to vic ofSangyo-dong(1l49.5-1468.5), which they heard was the Regt'l Hq
for the 19th Regt, on 7 Sep 50. POWs stated they were repls for the 19th Regt.
Out of the 1,000 men, 300 were slated to join the 1st Bn, 19th Regt as repls and
separated fr the group and went to Hill 570 arr there at sunset, 10 Sep. The men
were assigned to various companies, but the POWs with 27 others were assigned
to Mortar Co, bringing the strength of the Co to 60 men. The Mortar Co had 3
Plats and POW plat was the 1st Plat with 1 mortar and 20 men. After being as
signed to 1st Plat, they took positions on Hill 570 and the 2d and 3d Plat took the
left and right Hank positions abt 100 m apart.
POWs stated that the 1st, 2d and 3d Bns of the 19th Regt were in the vic of Hill 570
and the strength was about 1,000 men including the 300 replacements. POWs stated
that they worked for 2 days cleaning the dust off 200 mortar shells for their plat.
POWs stated that Hill 570 was shelled by Arty on 141600 I Sep and the 19th Regt
retreated to village 1 km N of Hill 570. POWs took shelter in a trench and were
captured. Prior to capture, POWs hauled 2 WlA to a mine 4 km N of Hill 570.
B. Other Units: No information.
3. Mission: POWs stated that the unit's mission was to push toward Taegu.
4. Enemy Supply and Equipment
A. Supply:
1. Ration: POWs stated that they were fed only wheat and ate irregularly, whereas
Officers and NCOs ate rice and 3 meals a day. They further stated that food
was foraged fr nearby villages and the civilians cooked and hauled the food for
the troops.
2. Ammo: POWs' 1st Plat was equipped w/l mortar and 200 rds of mortar ammo,
of which I/S was utilized before the shelUng on 14 Sep. Each man carried 3-4
hand grenades, but never had the chance to use them.
298 ORO-T-261
UNCLASSIFIED
SKRn
UNCLASSIfiED
POWs stated that an ammo dump is located in vic of abt 12 km N of Hill 570.
The ammo is transported by civilians on their back.
B. Equipment
1. 1st Plat had only 1 mortar (60mm).
2. Comm: The Plat had 1 portable telephone
5. Enemy Losses
A. Personnel: POWs stated the 30 KIA and undetermined number of WlA on 14 Sep
by Arty. POWs stated the loss was very heavy.
6. Replacement of Personnel and Equipment
A. Personnel: POWs stated that the 1,000 men came to the 19th Regt as repls hut
what happened to the rematning 700 men is unk to POWs. The 1,000 men were
all untrained and had no intentions of fighting this war.
B. Equipment: No information
7. Miscellaneous
A. Enemy Methods: POWs stated that the sqd ldrs encourage the troops by saying
that the US force will retreat if NK soldiers yell and rush forward.
POWs staled that the civilian people prepare the food for the troops and haul the
food to the soldiers. Furthermore the civilians are used to haul ammo.
B. Morale: POWs stated that morale of the draftees is low because they believe that
they have been deceived. Also the fact that rations are not sufficient further lowers
the morale.
8. Remarks: POWs knowledge of military matters was small but [they] were very co
operative. Information contained herein is considered to be fairly reliable.
S. Yamada
Capt Inf
ITEMS
ATIS Interrogation Report No. 1418 27 Sep 50
Field Report (ADVATIS-0888) 23 Sep 50
1. Personal Details
POW Name: Psk Uk Tong
POW Number: 63 NK 3694
Rank: Pvt
Age: 28
Duty: Ammunition Carrier
Unit: 1st Div, 14th Regt, 1st Bn, 1st Co, 2d Plat, 2d Sqd
Educalion: 6 years
Place of Capture: Nrunsan-<iong [1162-1458], 16 km N of Taegu
Date of Capture: 19 Sep 50.
2. Assessment: PhYSical condition of POW was good. POW's military informallon was
limited. He Is not recommended for further interrogation.
3. Tactical Information:
12 Sep 50: 1st and 2d Co's advanced to a point 16 km north of Taegu and took up posi
tions on a high hill. The hill was attacked by US troops with MG and mortar fire, and
the two (2) companies fled from the hill towards the north. POW deserted and hid in
a village at the foot of the hill for a period of a week.
19 Sep 50: POW surrendered to ROK forces.
ORO-T-261 299
SEERET
UNCLASSIFIED
UNCLASSIFIED
ITEM 6
ATIS Interrogation Report No. 1155 17 Sep 50
Field Report (164-LDI 0 0865) 15 Sep 50
1. Personal Details
POW Name: Kim, Han Tae
POW Number: 63 NK 936
Rank: PYi
Age: 19
Duties: Rifleman
Organization: 1st Div, 14th Regt, 3d Bn, 7th Co, 3d Plat, 2d Sqd
Education: 3 years
Occupation: Farmer
Date and Place of Capture: 12 1830 K [121730 I] Sep 50, 4 km S of Sunsang-san (1155.2
1454.5) [i.e., at 1155.5-1454.5, 4 km S of Sansong]
2. Assessment: POW was cooperative. Information limited due to limited service.
3. Estimated Strength (as of 12 Sep)
Unit
2d Plat
(1) 4 squads
(2) 6 men per squad
(3) total of 25 men
Equipment and Ammunition
1 x HMG, two ammo boxes
1 x LMG, two magazines
4 x PPsh SMGs, two magazines per gun
20 x M 1891/30 rifles, 20-25 rounds
7th Company
a. Personnel
(1) Total of 75 men
b. Equipment and ammunition
(1) 2 x HMGs, 2 ammo boxes per gun
(2) 3 x LMGs, 2 magazines per gun
(3) 12 x PPsh SMGs, 2 magazines per gun
4. Personal and Unit History: [Marched from Kunwi on 11 Sep 50 to Sunsan-san on
12 Sep] ...
POW surrendered voluntarily on 12 1830 K Sep 50 at Sunsan-san (1155.2-1454.5)
[Says above 4 kID south of this]
6. Food and Ammo Supplies:
On 11 Sep, POW and 50 men received M1891/30 rifles, 20-25 rounds per man which
had been taken from dead or wounded 1 km N of Sunsan-san.
6. Enemy Intentions: Objective of 1st Division was Taegu via Chilgok.
ITEM 7
ATIS Interrogation Report No. 1149 17 Sep 50
Field Report (ADVATIS-0787) 16 Sep 50
1. Personal Details
POW Name: 1m, Chin Sun
300 ORO-T-261
S&ER!'T
UNCLASSI FlED
POW Number: None
Rank: Pvt
Age: 17
Duty: Rifleman
Unit: 1st Div, 1st Regt, [redesignated the 14th Regt] 3d Bn, 1st Co, 1st Plat, 1st Sqd
Education: None
Occupation: Farmer
Place of Capture: 4 km S of Kasan Castle (1150.9-1451.4) [coordinates of Hill 402]
Date of Capture: 14 Sep 50 (sic) [date given below as 12th]
Home Address: Hwanghae-do, Yonpack-gun, Kumsan-myon, Taeha-r! 113009
Interrogator: Sgt Hayashida (ATIS)
2. Assessment
Subject was cooperative but is extremely unintelligent. Subject is illiterate and can
not carry on an intelligent conversation. Subject has a slight wound on his forehead
and it is possible that it has affected his mental capacity. Not recommended for fur
ther interrogation. Reliability poor.
3. Tactical Information:
Chronology
1 Aug 50. Inducted into NKPA at Sariwon. Approx 1,000 men were inducted
2 Aug 50. 700 men issued uniforms. Departed Sariwon by ratl. Disposition of
300 men uDknown. The 700 men were unarmed.
5 Aug 50: Enroute, at Namch'on o/a 056000 Aug 50, the train was strafed by 8 US
aircraft and about 100 meri were killed or wounded. The attack occured about 2 km
NW of Namch'on. During the attack subject observed destruction of an ammunition
dump situated in the hills approx 2 km SW of the Namch'on Station. Amount of
ammunition destroyed unknown.
9 Aug 50: 600 men arrived at Seoul, detrained, and marched south .. The men were
unarmed and were escorted by 5 armed NKPA troops. POW did not see any NKPA
troops in Seoul. The troops did not carry any rations or other supplies.
o/a 10 Sep' 50: Approx 500 men arrived 2 km N of the Kasan Castle where they
were all assigned to the 1st Regt, 1st Div. Enroute from Seoul, the group did not
sustain any air attacks but suffered a loss of about 80 men through desertion. POW
does not remember route of march from Seoul but vaguely remembers the place
names Chungju, Mungyong and Hamchang. Upon assignment to the 1st Regt, the
men were issued used but serviceable weapons. The entire 1st Regt was assembled
at the above locations. The strength of the Regt including the 500 replacements
was approx 800 men.
o/a 11 Sep 50: 1st and 2d Companies, 3d Bn were ordered to march South, objective
unknown. Remaining units of the 1st Regt remained at the above location. The 1st
and 2d Companies marched South through the Castle enclosure to a ridge approx
2 km south of the castle. At this point at about 2000 hrs, the 2d Company was
ordered to proceed further south while the 1st Co remained at the position. The
2d Co proceeded south to a point approx 5 km S of the castle and was ambushed by
UN forces. In the ensuing batlle the strength of the 2d Co was reduced from 60 to
14 men. The remaining 14 men retreated to the position of the 1st Co. At about
2200 hrs, the 1st Co, with the 14.men from the 2d Co, proceeded southeast from
its position to a point about 5 km SSE of the castle, where it was also ambushed.
The engagement lasted for two hours until about 2400 hrs. At the end of the engage
ment, the combined strength of the 1st and 2d Co was about 50 men. Prior to the
"'Dates from here on are given as August, but are obviously September. Thus corrected, the chronology
agrees fairly well with that of other prisoners in the same movement.
ORO-T-261 301
UNCLASSIFiED
lIiCRU
UNCLASSIFIED
s'(ERfT
engagement, the strength of the 1st Co was approx 70 men. When the engagement
ended, subject remained concealed. He then hecame separated from his unit.
12 (sic) Sep 50: Subject taken prisoner by Americans.
Morale:
Morale of the troops was very low. The men were aware that the majority of them
were untrained and that the units were understrength. Consequently, tbe men were
very reluctant to become involved in aggressive warfare. Some of the men refused
to advance while the unit was under fire and subject saw three men of his plat being
killed by his platoon leader for this reason. The unit commander has much difficulty
in controlling the troops under fire because the inexperienced men were panicky.
Equipment:
Subject did not see any tanks, artillery or mortars. The only equipment POW has
seen during his servtce wtth the 1st Regt were rifle, SMG's, LMG's and HMG's.
Supply:
No information is avallable on source or method of supply. Upon assignment to the
1st Regt, each man was issued. 2 grenades and 70 rounds. Much of the ammunition
was taken from casualties.
Currency loformation:
Extent of usage: The people of N Korea had unquestioned faith in the value of their
currency and it was used as the standard medium of exchange. Some of the prices
as of 1 Sep 50 were as follows:
1 kilogram of beef - 90 won
1 pair of cotton trousers - 900-1000 won
For the Commanding Officer:
Weeldreyer
Executive
ITEMS
ATIS interrogation Report No. 1103 16 Sep 50
Field Report (ADVATIS-0762) 14 Sep 50
1. Personal Details:
POW Name: Kim Yong Gi
POW Number: 63 NK 2775
Rank: Pvt
Age: 26
Unlt: 1st Div, 2d Regt, 2d Bn, 6th Co, 3d Plat, 2d Sqd
Duty: Rifleman
Occupation: Farmer
Education: None
Place of Capture: P'a!gong-san (1160-1450) [stated coordinates gtve general area of
capture: actual coordinates of P'algong-san: 1167-14571
Date of Capture: 10 Sep 50.
2. Assessmen.t: Suhject w::t1'!. cooperative but not very intelligente Subject was in good
physical condition. His short period of servtce prevented him from knowtng more
about his unit. Subject is not a communist. Reliability falr.
3. Tactical Information:
302 ORO-T-261
UNCLASSIFIED
SEERET
HERn
UNCLASSIFIED
Chronology
8 Sep 50: 200 men arrived on hill approx 8 km NNW of P'algong-san and joined the
2d Regt, 1st Div. The strength of the 2d Regt was shout 1,000 men. The replace
ments Were all issued arms at this pOint. The entire Regt rested until 091000 Sep
50.
10 Sep 50: Entire 2d Regt marched south and occupied the summit of P'a1gong-san
at 100500 Sep 50. Strength of 2d Regt was approx 1,200 men. At shout 0530 firsi
contact with ROK forces was made. 100 of the 200 replacements were ordered to
spearhead the advance with a suicidal frontal attack. The men were told that they
would he shot if they did not make the attack. Consequently the 100 men, without
anyone to lead them, made a wild nharge toward the ROK positions on the southern
summits of P'algong-san. 50 NKPA.regulars followed the 100. Conscripts at a
distance of 50 meters carrying PPsh submacbine guns. The ROK forces met the
attack with effective small arms fire. By 0900, the fight was over and about one
hundred men of the 2d Regt ( ~ were either killed or wounded. The remaining
50 men [of the total 150 attaCking) deserted or were taken prisoners. POW was
among those taken prisoner.
Strength:
Present sirength of the 2d Regt is approx 1,100 men, of which 70 percent were re
placements. Only 30 percent of the men were origtnal memhers of the regtment.
With the exception of the 200 replacements which the Regt received on 8 Sep 50,
.all of the troops were from N Korea. The majority of the men had no tralntng
hefore being committed to battle.
Equipment:
Replacements were issued rifles taken from casualties. The rifles were in good
condition. Subject heard guns and mortars so assumes that the regtment siill had
some pieces.
Morale:
The morale of the troops from N Korea was extremely high but the replacements
from S Korea had no will to fight bec811se they had been discriminated against since
they joined the untt. Subject believes that had they received better treatment, they
would have fought better.
Vehicles: ..
Supplies:
10 Sep 50, each rifleman had an average of 120 rounds of ammunttion.
ITEM 9
G<WYlENTI 1\ 1
Headquarters 1st Cavalry Division
Office of the ACofS, G-2
Interrogation of Prisoner of War Report
Report'No. 0086
Map: Tactical Map of Korea, 1:50,000
13 Sep 50
MW
1. Prisoner and Untt Identificatton:
Personal Details
Name: Cheh, Yong Sum
Rank: Pvt
Unit
2d Plat, 5th Co,
2d Bn, 2d Regt,
. Place of Capture
Captured by I Co,
7th Cav. in vic 314
ORO-T-261
303
SE(Ril.,
UNCLASSIFIED
UNCLASSIFIED
ueHI
Duty: Rifleman 1st Div (1156.0-1454.2) on
Length of Sv: Drafted 10 Aug 50 12 Sep
Temp POW No.: CD-214
2. Organization, Strength and Disposition:
A. Own Unit: POW was one of the 560 drafted at Shlniji [SinuiJu] on 10 Aug 50. The
same day, the 560 men went to Heijo by train and marched S, arr Hill (1156.5
1455.4) [coordinales of knob 3, Hill Mass 314]110400 I Sep. POW took the roule
via Seoul, Wongju [Wonju], Yongju and Andong on the main highway marching during
the night only. After arr at hill, the 560 men were assigned to various plats of the
2d Regt. POW thought the 2d Regt had about 2,400 troops on 10 Sep but how many
Bns in the 2d Regt is unk to POW. Afler sunset, 11 Sep, POW's Co of approx 85
men advanced S to Hill 314 (1156.0-1454.2) [coordinates of Knob 1, Hill Mass 314]
where they dug fox boles. At sunrise, 12 Sep [i.e. when the fog had dissipated
1100 hours], the 5th Co was hit by arty, mortar and MG fire and also straied by
enemy aircraft. POW took cover between two big rocks and observed his CO,
3 Jr Lts and 2 EM escaping to the rear. POW surrendered willingly when he
observed US force.
B. Other Units: POW stated that when the 5th Co advanced fwd on 11 Sep the 2d Regt
was still at (1156.5-1455.4) [Knob 3 on coordinates of Hill Mass 314]
POW observed unidentified units at pt which he believes to be 1 km E fr place of
capture. [Kisong-dong]
3. Mission: POW stated that the mission of the unit was to push toward Taegu and take
the city within 3 days at any cost.
4. Enemy Supply and Equipment
A. Supply:
1. POW stated he ate only one meal made up of wheat after he joined his unit on
11 Sep.
2. Ammo: No info.
B. Equipment: POW staled that the 5th Co was equipped with 10 Sub MGs (300 rds
each), 55 rifles (100 rds ea), 2 mortars (25 rds ea) and 2 HMG (unk quantity of
ammo).
5. Enemy Losses:
A. Personnel: POW's 5th Co losses as of [after] sunrise, 12 Sep, as observed by
POW was 20 KlA, 7 WlA. Losses sustalned by en arty, strafing and MG fire.
B. Supplies: No info.
C. Equipment: POW observed 2 mortars demolished, 1 HMG and many rifles damaged.
6. Replacement of Personnel and Maleriel:
A. Personnel: POW who was one of tbe 560 drafted (Between ages of 18-37) at Shlnuiju
[Sinuiju], were all repls for the 2d Regt, 1st Div. Tbey did not receive any tralnfng.
POW stated 20 men of the 560 escaped while enroute to the Regt. POW heard at
Sbinuiju that if repls are insufficient to make up the shortage the NK will start
drafting men up to age of 55.
7. Miscellaneous
A. Enemy Methods: POW stated the Regt stops at one place and sends out just one
Co to the front line. If the Co is wiped out another Co is sent out. Whenever a Co
is about to be annthilaled a runner informs the Bn CO.
B. Morale: POW stated that morale is low because rumor has it tbat if they have
a cbance to escape, they'll escape. POW stated NK will fight a little longer, but
if the weapons and food get more scarce they.'ll retreat. He stated that UN shonld
drop leaflets because many of them want to surrender but have never seen the
safe conduct leaflets. 0
304 ORO-T-261
UNCLASSIFIED
SEER'f1
UNCLASSiFIED
8. Remarks: POW was cooperative and sincere. Therefore the information obtained
is considered fairly reliable. Because of the fact that he has recently travelled
through NK on the way to his unit, further investigation may prove fruitful.
Dist: S. Yamada
G-2 - 2 Captlnf
File - 1
ITEM 10
ATlS Interrogation Report No. 1142 17 Sep 50
Field Report (ADVATlS-0780) 16 Sep 50
1. Personal Detalls:
POW Name: Ch'oe, Yong Sam
POW Number: 63 NK 933
Rank: Pvt
Age: 22
Duty: . Rtfleman
Unit: 1st Div, 2d lnf Regt, 2d Bn, 5th Co, 2d Plat, 2d Sqd
Education: Primary school (6 years)
Occupation: Liaison Employee, Democratic Party, Sinuiju
Place of Capture: Wooded Hill 14 kID WNW of Taegu [actually Hill 314, 14 kID N of
Taegu]
Date of Capture: 121400 K [1213001] Sep 50
Place of Birth: P'yongan-pukto, Sakju-gun, Sakju-myon, So Bu-ni 41164
Home Address: Same as place of birth
Interrogator: Sgt Hamasaka (ATIS)
2. Assessment:
POW appeared to be fairly intelligent. He was cooperative, however ignorant on
military terms and units due to short term of military servtce. POW considered
fatrly reliable by interrogator.
3. Tacticallnformation:
Chronology
10 Aug 50: Conscripted into NKPA, South Sinuiju. 450 conscripts (mostly merchants
ages 17-37; average age 25) received 9 days basic training.
19 Aug 50: 450 conscripts desiguated as replacements. Activated into 1st Bn
comprised of four (4) companies. POW assigued to 3d Co. 1st Bn left South Sinniju
by rail, unarmed.
20 Aug 50: 1st Bn arrived at Sinuiju. Travel only during night. Left Sinuiju
immediately.
20-23 Aug 50: 1st Bn continued in southerly direction by rall and foot.
23 Aug 50: 1st Bn arrived at P'yongyang. Left P'yongyang immediately.
25 Aug 50: 1st Bn arrived at Kaesong. Travel by rail and foot. Left Kaesong on
foot.
29 Aug 50: 1st Bn arrived at Seoul.
30 Aug 50: 1st Bn left Seoul on foot, still unarmed.
8 Sep 50: 1st Bn arrived at north bank of Naktong. River at point south of Andong
via Suwon, Wonju, Yongju,"Andong. Crossed river same night over repaired con
crete bridge (see" Bridges). No opposition encountered during crossing.
ORO-T-261 305
SEUU
UNCLASSIFIED
UNCLASSIFIED
SiERn
8-11 Sep 50: 1st Bn continued march south from Naktong River over mountain road.
Travel by night only. 1st Bn still unarmed.
110400 Sep 50: 1st Bn arrived In 1st Dlv Area In wooded bill 8 km ENE of Taegu
(location uncertain). Of 560 replacements In 1st Bn 30 troops deserted unit during
travel from 81nuiju to 1st Div Area. 1st Bn replacements assigned to various units,
1st Dlv. POW and 2 assigned to 2d Plat, 5th Co, 2d Bn, 2d Inf Regt. M1891/30 rifle,
100 rounds available for only 2 or 3 replacements assigned to 2d Plat. POW unarmed.
2d Bn remained deployed In bill remainder of day. Other 1st Dlv Unit disposition,
unknown.
111800 Sep 50: 2d Bn left bill positions (8 km ENE of Taegu) for wooded mountain
approximately 4 km E (12 km ENE of Taegu).
120500 Sep 50: 2d Bn arrived at wooded mountain. [Tbis was Hill 314]. Deployed in
wide area.
121100 Sep 50: 2d Bn shelled by US artillery and mortars. Strafed and bombed by
US aircraft. Heaviest casualties from artillery, mortars. Number unimown. Air
attack, least effective. POW unarmed, bid between bonlders during engagement.
2d Bn units forced Into disorganized retreat as a result of heavy US artillery and
mortar shelling and small arms fire. POW captured Immediately thereafter by
US Army units.
Weapons and Personnel:
1st Rifle Dlv-Strength unimown.
2d Inf Regt-Strength unimown.
2d Bn (3 x Inf companies, mortar company, HMG Co). Strength 100. [?]
5th Co-Strength 6. [?]
Remarks: During engagements 2 x mortars similar to 82mm M1937 Medium Mortars
distributed to each company. Returned to mortar company after engagement. In
5th Co 1 x mortar destroyed, remaining mortars, unimown (believed destroyed).
Unit Morale:
During 1st Bn (replacement unit) march from Slnuiju to 1st Dlv area (8 km ENE of
Taegu). POW noticed low morale among troops In unit. Troops (formerly merchants
in N Korea) were bitter over compulsory conscriptIon and possessed no desire to
fight for North Korean Government.
Although troops were warned by unit officers that NKPA troops would be killed by
US Army if they surrendered, rumors by S Koreans in occupied areas circulated
among troops that NKPA POWs In hands of US Army would receive good treatment.
POW believes moat North Korean replacement troops will surrender to US Army
at fir st opportunity.
Sociological:
POW believes at least 80% of North Korean population dislike their government
because of strict control and regulation of Individuals.
Economical:
In Sobu, 240 km SE of Sinuiju, 8.Aug 50, follOwing prices of commodities prevailed:
rice, 1 sung-40 won
work trousers, 1 pair-SOO won
work sbirt-1 to 5 to 600 won
Semi -skilled workers in nearby carbide plant received 40 won plus three meals a
day.
306 ORO-T-261
UNCLASSIFIED
SEUEI
S5QET UNCLASSIFIED
In Seoul, 29-30 Aug 50, S Korean currency was still being used. Exchange rate at
bank was 100 N Korean won for 800 S Korean won.
Followtng price prevailed:
Leather dress shoes, a palr-2,OOO S Korean won.
For the Commanding Officer:
Weeldreyer
Executive
[All this, to the end, is still under Tactical Information]
ITEM 11
HEADQUARTERS EUSAK
Office of the Assistant Chief of Staff, G-2
(164th Mil Intel Suc Det)
APO 301
AMS L 551 1:250,000 164-MISDI-0890
18 Sep 1950
Prisoner of War Preliminary Interrogation Report
POW Name: Lee, Chong 0
Rank: Pvt
POW Number: 63 NK 4040
Organization: 14th Diy, 2d Regt, 2d Bn, 4th Co, 1st Plat, 1st Sqd [probably 1st DiY,
2d Regt, 2d Bn, etc.]
Duties: Rifleman
Educstion: Middle school, 3 years
Age: 21
Occupation: Student
Address: Hwanghae-do, Changyon-gun, Nakto-myon, Samchon-ni
Date and Place of Capture: 140700 K [140600 I] Sep 50 (1150-1450) by Co "1", 7th Cay.
Interrogator: T. H.
Interrogator's Assessment: POW was very cooperative; however. information seems
limited due to his limited service.
TO&E and Strength of: 4th Co, 2d Bn, 2d Regt, 14th Diy (as of 14 Sep 50). The 4th
Company composed of three (3) Inf Plaloons. Each Inf Platoon composed of four (4)
squads.
1st Platoon
a. Personnel Origtnal [11 Sep1] Preseni [14 Sep]
(1) EM 40 15
(2) Officers 1 1
(3) Total 41 16
ORO-T-261 307
UNCLASSI FI ED
SEeM'
UNCLASSIFIED Siun
b. Equipment and ammo:
(1) M1891/30 rifles 36 (60 rds per) 13
(2) PPsm SMG 4 2
(3) Grenades 2 per man Unknown
4th Company
a. Personnel
(1) EM 120 45
(2) Officers 4 4
(3) Total 124 49
b. Equipment and ammo:
(1) M1891/30 rifles 108 (60 rds per) 39
(2) PPsmSMG 12 6
(3) Grenades 2 per man Unknown
HMG Company
a. Personnel
(1) EM 40 17
(2) Officers 1 1
(3) Total 41 18
b. Equipment:
(1) MAXIM HMG 5 (1 ammo box per) 5
Location and Activities of Units:
2d Bn, 2d Regt, 14th Dlv was on offensive (location unknown)
Personal and Unit History:
POW and 1200 men were inducted on 8 Jul at Sariwon. He worked in a mess hall
as KP until 15 Aug 50. 1200 men were trained as riflemen with wooden rifles
until 15 Aug 50. POW and 1200 men, unarmed, left Sariwon on 15 Aug.
POW and 250 men assigned to 2d Bn, 2d Regt, 14th Div on 11 Sep at vicinity of
Kasan. POW and 80 men who were assigned to 4th Co, received old rifles and 60
rounds per men, which had been taken from dead bodies and wounded men.
After being attacked by US forces, POW voluntarily surrendered to Co "r", 7th Cav
at 1150-1450 on 14 Sep 50.
Morale:
Morale of the 4th Company was low because they lacked sufficient training.
308 ORO-T-261
UNCLASSIFIED
SEERfT
neftT UNCLA.SS\flED
Appendix B
3D NK DIY ATTACK ORDER
ORO-T-261
309
UNCLASSIFIED
SEEln
ATTACK ORDER OF 11 SEP 50 FOR THE 3D NORTH KOREAN DIVISION
(OPPOSING THE 5TH US CAVALRY REGIMENT)
Note: This document is probably typical of the operations of the adjacent 13th NK Dlv,
which Is mentioned explicitly once, as well as of the subject 3d Dlv.
Source: Regimental History, 5th Cav Regl. Document believed captured at 1146.4-1447.5.
"TRANSLATION OF A CAPTURED ORDER. THE NKs WILL PROBABLY
CONTINUE TO FOLLOW THIS SCHEME OF MANEUVER"
Battle Order No. Cbangwon-bong
3d lnf Div Staff Section 1150 11 Sep 50 Map 5000
(Note: Hill 371
at 1145-1453)
~ . The enemy Intends to defend the Taegu border along the line Taepyong-dong (1145.0
1446.9), Hill 188 (1146.4-1447.4), Hill 246.6 (1148.9-1448.3), Hill 260 (1150.2-1448.9).
That defense is on a scattered hill basis. It Is an Independent outpost type of defense.
The main defense pOints are at the unnamed hill (1146-1450), Hill 203 (1146-1451), Hill
174 (1147-1451), Hill 232 (1146-1453).
2. The division will attack these partial defenses and as the first assignment will secure
the line from the south slope of Hill 188, Hill 246.6 and Hill 260. When this line is se
cured, continue attack toward Taegu area.
3. The motorized artillery regl will attack the right flank. The 13th Div will attack the
left flank.
4. Continue to pursue the enemy, who oppose in groups, and attack in the direction of
Taegu.
5. The 7th Regl will attack from the present pOSition (HIll 371, 1145-1453) In the ,direc
tion of Hill 174 (1147-1452) and Hill 203 (1146-1451). The first objective will be totake
Hill 203, then to secure Hill 246 (1148.9-1448.3) by 0500 hours tomorrow morning.
The 8th Reg! will attack on the right flank. The sector line will be from Hill 371
(1144-1451) to Hill 174 (1147-1451) to Sang-dong (1148.7-1449.1) to Hill 246.6 (1148.9
1448.3). All the points along the sector line will be the responslblllty of the 7th Reg!.
6. The 8th Reg! will attack toward the line from Hill 203 (1146-1451) to Ha-dong (1147.8
1450.1) to unnamed hill (1147-1448). This line will be secured by 0500 hours In the
morntng.
ORO-T-261 311
SEERU UNCLASSIF:L;
UNCLASSIFIED
SfCRET
The 9th Regt will attack on the right flank. The sector border line (between the 8th
and 9th Regts) will be from Songmun-dong(1146.5-1451.5) to unnamed hill 1147-1448. The
above-named places are the responsibility of the 8th Regt.
7. The 9th Regt will attack In the direction of Hill 203 (1146-1451) by Todang-dong
(1145.3-1451.0), SIn-dong (1146.1-1449.3), Hill 188 (1146.4-1447.4). The south slope of
Hill 188 will be secured by 0500 hours tomorrow. The 9th Regt will be responsible for
the security of Its right flank.
8. The artillery will be the Corps Arty and Regtl Arty Bn, and they will lay down 10
minutes of fire.
Preparation fire: Destroy the observation points In front of the enemy's defensive
front. For attack suppori and final support: Protect the Infantry with continuous fire
until the end of an attack.
9. As for defense against tanks and aircraft, the present weapons will be organized. In
order to fortify the secured pOSition, dig trenches and completely camonflage. The Div
ision CP Is at Changwon-bong (1144.9-1453.2). Future move of Division CP will be in
direction of the movement of 9th Regt.
10. Reports will be:
For completion of preparation - by telephone or by Ln officer.
When the attack commences - by telephone or wireless.
.After the attack - by phone or WT on the bour.
When Important pOSition is secured - by flare or WT
When pursuing next objective - written message, phone or WT.
11. Signals will be as deSignated before.
12. Alternate commander - 1st Chief of Staff.
Lee Yon Ho (signed)
Dlv Commander
Chu Teh Sen (signed)
Chief of Staff
312 ORO-T-261
UNCLASSIFIED
SECREJ
UNCLASSIFIED
SEERH
Appendix C
ON
PRISONER-OF-WAR INFORMATION
ENEMY USE OF MORTARS AND ARTILLERY
ITEM 1:
ITEM 2:
ENEMY USE OF 120-MM MORTARS
ENEMY ARTILLERY
CONTENTS
Page
315
316
ORO-T-261
313
UNCLASSI FI ED
SEERU
-SEERn UNCLASStFIED
ITEM 1: ENEMY USE OF 120-mm MORTARS
Extracts from ADVATlS Interrogation Report No. 0755 and Supplement No.1
POW Name: Kim, Chong Sop
POW No.: 63 NK 2576
Rank: Pvt
Age: 19
Duty: Rifleman (formerly mortarman)
Untt: 15th Div, 45th Regt, 3d Bn, 9th Co, 2d Plat, 1st Sqd (formerly 45th Regt, 120mm
Mortar Company)
Place of Capture: Imkodong (10 Ian N of Yongch'on) (1190-1450) hy ROK Army
Date of Capture: 7 Sep 50
POW and 23 others were tranaferred to the infantry because of large losses in the
45th Regt. All NCOs remained in untt [120mm Mortar Company], supposedly to train
new crews from replacements ...
]!;'l!!!pment of Mortar Company
6 x 120mm Mortars M1943. These mortars were manufactured In 1948, 1949 or
1950. The dates were stamped on the mortar tubes. Two binoculars, power unknown.
One instrument similar to BC scope used for observation of fire. 4 x TAM field tele
phones. 6 kIn of field wire with reels.
Mortars were drawn by GAZ-51 trucks on a carriage similar to the 120mm tubular
type carriage. Ammunttion was fuzed with GVM-2 impact fuze. One type fuze has a
triangular fuze setting which may be used to obtain time fire, hearsay. POW never saw
this type fuze. POW knew how to set fuze for impact or delay but his untt always used
impact setting in combat. POW stated minimum elevation of piece was 45
0
, maximum
elevation 80. Range 5,700 meters. Maximum rate of fire 12-13 rounds per minute.
Most efficient rate of fire was 7 to 8 rounds per minute.
General Tactics, Preparation of Position
During displacements by vehicles the command squad rode In the first vehicle. The
entire untt could be and was tranaported In organic transportation. POW does not know
when or by whom the firing positions were selected. Mortars were generally emplaced
at the base of a reverse slope. Company front was straight or slightly curved, covering
a distance of approx 60 meters and permitting of 10 meters distance between pieces.
Pieces were never staggered appreciably. The base was dug In from 12 to 15 inches.
Ammunttlon (approx 20 rounds per mortar) was piled on top of the ground 10 meters to
the rear of the emplacement. Personnel and ammunition was never dug in. The remain
der of the ammunttlon supply (approx 70-80 rounds per piece, according to POW) was laft
in the trucks, which, depending on the terrain, are normally placed approx 200 meters
to the rear (never left or right) of the mortar positions. If possible, trucks were separ
ORO-T-261
315
nun
UNCLASSIFIED
UNCLASSIFIED
SECRE-T
ated by at least 50 meters. Vehicles were always camouflaged with grass and/or tree
branches and placed under cover of trees to lessen possibility of observation from the
air. POW's unit carried camouflage nets about the size of a truck. Generally camouflage
was laid on top of nets and not intertwined. These nets could be lowered during firing.
The mortars are usually emplaced less than 4 km from the front lines. For this
reason, the unit bas no organic security weapons such as MGs, rocket launchers, AT
guns or mines.
The Command Squad lays wire from the CP to the OP. One TAM field telephone
[of total of four) is used to each location. The remaining two telephones were kept as
reserve. Unit was originally supplied with wire of Russian manufacture. Number of
steel and copper strands unknown. At Ch'unchon ihe unit captured some US wire and
discarded the RUssian wire. A telephone operator told POW that US wire is superior
but POW doesn't know why. Only one line is laid between CP and OP. This line is con
stantly patrolled by members of ihe Command Squad. If this line becomes inoperative,
no other means of communication is available. Lines are not buried at the mortar
position.
The Co CO occupies tbe OP with members of the Command Squad, who act as
assistants and telephone operators. In case of emergency, the Command Squad leader
can adjust fire. Initial data for direction is telephoned to the mortar positions as a base
azimuth. Range is converted to elevation setiing by a firing table dept by the CO. Accord
ing to POW, ihe Co CO is the only man who carries a firing table. CO adjusts fire and
gives commands based on calculations made with the mil relation formula. This adjust
ment is made with ihe base pier, i.e. the piece located on the left flank of the mortar
positions. All pieces follow commands and are brought in on fire for effect. Companies
of mortars cannot give massed fires. POW heard that his CO effected lialson wiih sup
ported units but does not know meihods employed. fn the mortar position fire commands
are relayed to the pieces by the telephone operator. Duty of platoon leaders is to check
operations of crew and to keep records on ammunition expended. No restriction was
ever placed on firing. Ammunition was plentiful until the Han River crossing operation,
after which resupply was sporadic. Average daily expenditure was 20 to 30 rounds per
piece. At time of POW's transfer to ihe infantry his unit had 25 rounds per piece.
Rate of fire exceeded 1 round per minute on one occasion only. a/a 11 Aug 50,
POW's unit fired sustalned fire for 1 hour (1600-1700) to support a Naktong River cross
ing operation. His crew expended 120 rounds.
POW's unit had no smoke or Incendiary shells. POW had never heard of gas shells.
POW's unit never fired specific counterbattery missions to his knowledge. HoweverJ
he heard ihat oiher units had done so.
POW's unit received counterbattery fire on 3 or 4 occasions, but it is uncertain
whether it was artillery or mortar. POW's unit displaced afler 8 to 10 rounds of counter
battery fire. Displacements were always lateral, usually about 400 meters. No prior
reconnaissance was made for selection of alternate or dummy positions on initial
occupation. POW's unit sustained 1 killed in action and no W / A from counterbatiery
fire.
POW's unit never engaged in night firing.
ITEM 2: ENEMY ARTILLERY
Extracts from ADVATIS Interrogation Report No. 0958
POW Name: Lee, Yon Gun
Rank: Maj
316
ORO-T-261
SECRET
UNCLASSIFIED
UNCLASSIFIED
nUET
Age: 27
Duty: Ass't Regt'l Commander
Unit: 15th Div, 45th Regt
Place of Capture: 8 kIn N of Kigye (1210-1460)
Date of Capture: 27 Sep 50
NKP A Artillen' Information
76mm AT Gun (M1942) used in Div Arty. Used as an all-purpose gun. Maximum
range: 12.5 kIn. Impact fuze (only type of fuze used).
76mm Howitzer (M1927) used for Regimental support. Maximum range 8,500 meters.
Impact fuze only.
122mm Howitzer (Ml938) used In Dlv Arty. Maximum range 8,500 meters. Semi
fixed projectiles used. Fuze: time fuze and Impact fuze.
" 45mm AT Gun (M1942) used In the Regtment and In the Division Independent AT
Gun Batialion. Maximum range: effective range against tank 400 meters. Penetrates
steel plate 2 centimeters thick at 400 meters. This piece was often used as a field gun.
Generally for machine gun positions against personnel. Projectiles and fuze: (a) armor
piercing by heat (a lead head on the projectile which creates heat up to 3000
0
C upon im
pact), (b) impact fuze.
Artille!)' Communication E'I1&Pment
Four x wireless telephone to each Battalion. Wireless telephone used only when
telephone Unes could not be used between OP and battery positions. The lofantry Bat
talion kept liaison with the Artillery by radio for artillery support. Predesignated flare
signals were used by respective Battalions to ask for more Artillery support or for
cease fire request.
Miscellaneous Artillery Information
Guns were emplaced during the night at least 2 kilometers to the rear of a hill.
Pits 2 meters deep were dug for the pieces and camouflaged by vegetation. The OP was
placed on the hilltop and was required to find the range. Maps of 1/50,000 with 1000-meter
grid were the only one used. 1/25,OOO-grid maps were scarce and never used. Mass
fire by Battalion was conducted but mass fire by Division Artillery was seldom used.
Usually such mass fire was just to harass the enemy and its effectiveness is not known.
In the 122mm semi-fixed round the cartridge already had a fixed charge, but to
increase the range 3 to 4 additional charges were added. The 122mm projectile used
Impact fuze and time fuze. PW had never heard of a proximity fuze. He did not even know
of Its existence.
Unobserved firing was mainly conducted by Division Artillery but seldom by Battal
Ion. A Survey Recon Platoon of 15 men acted as advance observation party and computed'
direction, range and prearranged positions for gun batteries. Grid sheets were used.
However, more often than not there was no time for unobserved fire. No aerial photo
graphs were available. Ricochet artillery firing was practiced In training but never used
in actual battle. The NKP A troops did not know how to properly set the fuze. Besides,
Korean terrain was not suited for such firing.
ORO-T-261 317
SEEREf. UNCLASSIFIED
UNCLASSIFIED
SEERH
Appendix C
ON
PRISONER-OF-WAR INFORMATION
ENEMY USE OF MORTARS AND ARTILLERY
ITEM 1:
ITEM 2:
ENEMY USE OF 120-MM MORTARS
ENEMY ARTILLERY
CONTENTS
Page
315
316
ORO-T-261
313
UNCLASSI FI ED
SEERU
-SEERn UNCLASStFIED
ITEM 1: ENEMY USE OF 120-mm MORTARS
Extracts from ADVATlS Interrogation Report No. 0755 and Supplement No.1
POW Name: Kim, Chong Sop
POW No.: 63 NK 2576
Rank: Pvt
Age: 19
Duty: Rifleman (formerly mortarman)
Untt: 15th Div, 45th Regt, 3d Bn, 9th Co, 2d Plat, 1st Sqd (formerly 45th Regt, 120mm
Mortar Company)
Place of Capture: Imkodong (10 Ian N of Yongch'on) (1190-1450) hy ROK Army
Date of Capture: 7 Sep 50
POW and 23 others were tranaferred to the infantry because of large losses in the
45th Regt. All NCOs remained in untt [120mm Mortar Company], supposedly to train
new crews from replacements ...
]!;'l!!!pment of Mortar Company
6 x 120mm Mortars M1943. These mortars were manufactured In 1948, 1949 or
1950. The dates were stamped on the mortar tubes. Two binoculars, power unknown.
One instrument similar to BC scope used for observation of fire. 4 x TAM field tele
phones. 6 kIn of field wire with reels.
Mortars were drawn by GAZ-51 trucks on a carriage similar to the 120mm tubular
type carriage. Ammunttion was fuzed with GVM-2 impact fuze. One type fuze has a
triangular fuze setting which may be used to obtain time fire, hearsay. POW never saw
this type fuze. POW knew how to set fuze for impact or delay but his untt always used
impact setting in combat. POW stated minimum elevation of piece was 45
0
, maximum
elevation 80. Range 5,700 meters. Maximum rate of fire 12-13 rounds per minute.
Most efficient rate of fire was 7 to 8 rounds per minute.
General Tactics, Preparation of Position
During displacements by vehicles the command squad rode In the first vehicle. The
entire untt could be and was tranaported In organic transportation. POW does not know
when or by whom the firing positions were selected. Mortars were generally emplaced
at the base of a reverse slope. Company front was straight or slightly curved, covering
a distance of approx 60 meters and permitting of 10 meters distance between pieces.
Pieces were never staggered appreciably. The base was dug In from 12 to 15 inches.
Ammunttlon (approx 20 rounds per mortar) was piled on top of the ground 10 meters to
the rear of the emplacement. Personnel and ammunition was never dug in. The remain
der of the ammunttlon supply (approx 70-80 rounds per piece, according to POW) was laft
in the trucks, which, depending on the terrain, are normally placed approx 200 meters
to the rear (never left or right) of the mortar positions. If possible, trucks were separ
ORO-T-261
315
nun
UNCLASSIFIED
UNCLASSIFIED
SECRE-T
ated by at least 50 meters. Vehicles were always camouflaged with grass and/or tree
branches and placed under cover of trees to lessen possibility of observation from the
air. POW's unit carried camouflage nets about the size of a truck. Generally camouflage
was laid on top of nets and not intertwined. These nets could be lowered during firing.
The mortars are usually emplaced less than 4 km from the front lines. For this
reason, the unit bas no organic security weapons such as MGs, rocket launchers, AT
guns or mines.
The Command Squad lays wire from the CP to the OP. One TAM field telephone
[of total of four) is used to each location. The remaining two telephones were kept as
reserve. Unit was originally supplied with wire of Russian manufacture. Number of
steel and copper strands unknown. At Ch'unchon ihe unit captured some US wire and
discarded the RUssian wire. A telephone operator told POW that US wire is superior
but POW doesn't know why. Only one line is laid between CP and OP. This line is con
stantly patrolled by members of ihe Command Squad. If this line becomes inoperative,
no other means of communication is available. Lines are not buried at the mortar
position.
The Co CO occupies tbe OP with members of the Command Squad, who act as
assistants and telephone operators. In case of emergency, the Command Squad leader
can adjust fire. Initial data for direction is telephoned to the mortar positions as a base
azimuth. Range is converted to elevation setiing by a firing table dept by the CO. Accord
ing to POW, ihe Co CO is the only man who carries a firing table. CO adjusts fire and
gives commands based on calculations made with the mil relation formula. This adjust
ment is made with ihe base pier, i.e. the piece located on the left flank of the mortar
positions. All pieces follow commands and are brought in on fire for effect. Companies
of mortars cannot give massed fires. POW heard that his CO effected lialson wiih sup
ported units but does not know meihods employed. fn the mortar position fire commands
are relayed to the pieces by the telephone operator. Duty of platoon leaders is to check
operations of crew and to keep records on ammunition expended. No restriction was
ever placed on firing. Ammunition was plentiful until the Han River crossing operation,
after which resupply was sporadic. Average daily expenditure was 20 to 30 rounds per
piece. At time of POW's transfer to ihe infantry his unit had 25 rounds per piece.
Rate of fire exceeded 1 round per minute on one occasion only. a/a 11 Aug 50,
POW's unit fired sustalned fire for 1 hour (1600-1700) to support a Naktong River cross
ing operation. His crew expended 120 rounds.
POW's unit had no smoke or Incendiary shells. POW had never heard of gas shells.
POW's unit never fired specific counterbattery missions to his knowledge. HoweverJ
he heard ihat oiher units had done so.
POW's unit received counterbattery fire on 3 or 4 occasions, but it is uncertain
whether it was artillery or mortar. POW's unit displaced afler 8 to 10 rounds of counter
battery fire. Displacements were always lateral, usually about 400 meters. No prior
reconnaissance was made for selection of alternate or dummy positions on initial
occupation. POW's unit sustained 1 killed in action and no W / A from counterbatiery
fire.
POW's unit never engaged in night firing.
ITEM 2: ENEMY ARTILLERY
Extracts from ADVATIS Interrogation Report No. 0958
POW Name: Lee, Yon Gun
Rank: Maj
316
ORO-T-261
SECRET
UNCLASSIFIED
UNCLASSIFIED
nUET
Age: 27
Duty: Ass't Regt'l Commander
Unit: 15th Div, 45th Regt
Place of Capture: 8 kIn N of Kigye (1210-1460)
Date of Capture: 27 Sep 50
NKP A Artillen' Information
76mm AT Gun (M1942) used in Div Arty. Used as an all-purpose gun. Maximum
range: 12.5 kIn. Impact fuze (only type of fuze used).
76mm Howitzer (M1927) used for Regimental support. Maximum range 8,500 meters.
Impact fuze only.
122mm Howitzer (Ml938) used In Dlv Arty. Maximum range 8,500 meters. Semi
fixed projectiles used. Fuze: time fuze and Impact fuze.
" 45mm AT Gun (M1942) used In the Regtment and In the Division Independent AT
Gun Batialion. Maximum range: effective range against tank 400 meters. Penetrates
steel plate 2 centimeters thick at 400 meters. This piece was often used as a field gun.
Generally for machine gun positions against personnel. Projectiles and fuze: (a) armor
piercing by heat (a lead head on the projectile which creates heat up to 3000
0
C upon im
pact), (b) impact fuze.
Artille!)' Communication E'I1&Pment
Four x wireless telephone to each Battalion. Wireless telephone used only when
telephone Unes could not be used between OP and battery positions. The lofantry Bat
talion kept liaison with the Artillery by radio for artillery support. Predesignated flare
signals were used by respective Battalions to ask for more Artillery support or for
cease fire request.
Miscellaneous Artillery Information
Guns were emplaced during the night at least 2 kilometers to the rear of a hill.
Pits 2 meters deep were dug for the pieces and camouflaged by vegetation. The OP was
placed on the hilltop and was required to find the range. Maps of 1/50,000 with 1000-meter
grid were the only one used. 1/25,OOO-grid maps were scarce and never used. Mass
fire by Battalion was conducted but mass fire by Division Artillery was seldom used.
Usually such mass fire was just to harass the enemy and its effectiveness is not known.
In the 122mm semi-fixed round the cartridge already had a fixed charge, but to
increase the range 3 to 4 additional charges were added. The 122mm projectile used
Impact fuze and time fuze. PW had never heard of a proximity fuze. He did not even know
of Its existence.
Unobserved firing was mainly conducted by Division Artillery but seldom by Battal
Ion. A Survey Recon Platoon of 15 men acted as advance observation party and computed'
direction, range and prearranged positions for gun batteries. Grid sheets were used.
However, more often than not there was no time for unobserved fire. No aerial photo
graphs were available. Ricochet artillery firing was practiced In training but never used
in actual battle. The NKP A troops did not know how to properly set the fuze. Besides,
Korean terrain was not suited for such firing.
ORO-T-261 317
SEEREf. UNCLASSIFIED
SHRff
UNCLASSIFIED
Appendix 0
FRIENDLY STRENGTHS AND BATTLE CASUALTIES
IN THE SECTOR NORTH OF TAEGU
CONTENTS
Page
TABLES
01. STRENGTHS 2 SEP 50 AND CASUALTIES 1-9 SEP 50 IN THE 8TH CAv
BEGT AND ATTACHED ENGINEER UNITS FIGHTING AS INFANTRY 321
02. STATISTICAL BREAKDOWN OF FRIENDLY STRENGTHS AND BATTLE
CAsUALTIES IN THE SECTOR NORTH OF TAEGU. 6--16 SEP 50 322
ORO-T-261 319
UNCLASSIFIED
SfERU
SEERH
TABLE 01
STRENGTHS 2 SEP 50 AND CASUALTIES 1-9 SEP 50 IN THE 8TH CAY REGT
AND ATTACHED ENGINEER UNITS FIGHTING AS INFANTRY'
Unit
Strengths, 2 Sep
Casualties among US and attached ROK troops
Uncludes KIA, WlA, end MIA), br darb
US
I
ROK
I
To talc
1,1 21
3
,1 4,15,1 6.1 7,1
8
,1 9
Sep S ~ p Sep Sep Sep Sep Sep Sep Sep
8th Cay
1st So
Co A 1.42 25 167 1 3 0 1 1 3 2 3 1
Co B 147 25 172 1 0 2 0 2 1 0 0 1
Co C 143 25 168 0 1 0 1 0 0 0 I 0
Co D 129 0 129 0 0 1 0 0 3 0 0 0
Co HI 97 0 97 0 0 2 0 0 0 I I 0
2d Bn
Co E 133 22 155 4 5 13 10 28
d
IS" 6 I 1
Co F 146 14 160 1 93d 10 7
!
4 I
Co G 136 24 160 3 15 199 3 1 0 0 0 0
Co H 136 11 147 1 10 3 2 2 2 I I I
Co H2 95 0 95 0 2 3 6 I 0 1 0 0
3d Bn
Co I 166 0 166 1 0 3 2 3 I 0 10 0
Co K 162 10 172 0 0 1 4 0 3 4 9 1
Co L 165 10 175 0 I I 2 3 5 3 16 0
Co M 124 9 133 0 0 2 7 2 0 0 0 0
Co H3 93 II 104 0 0 0 3 0 4 0 0 0
Regtl Units
HM Co 141 0 141 0 2 0 0 0 2 I 2 0
Svc Co 133 0 133 0 0 0 1 1 8 0 0 0
Med Co 144 0 144 0 2 4 2 2 3 1 1 1
Hq& Hq 193 0 193 0 0 3 2
lOh 1 2 1 0
8tb Engr Bn
Co A ~ 175 0 0 8 2 0 0 1 I
Co C
'" 175 0 0 9 1 0 0 1 0 0
Co D 167 0 0 0 241 12 0 0 0 0
Inf Cas
thought in fl.
by 13tb NK
Div, total 12 50 100 65 35 36 10 5
aData from machine run of January 1951. Table 02 gives later, corrected (but almost. identical) data for
6-9 September.
bInstances of more than two MIA per co,&any per day are indicated by footnotes. Casualties considered
sustained by friendly units in combat with t e 13th NK Div are italicized.
cTbe grand total for all companies is thought to be 75 percent complete, but the figure for anyone com
p a n ~ on anyone day may be less complete or in error. '
Includes 7 MIA.
eIncludes 3 MIA.
Includes 20 MIA.
gIncludes 6 MIA.
hlncludes 9 ROKs WIA.
i Includes 14 MIA.
ORO-T-261 321
nCRn
c
..

:z:
(.J
0
J>
UJ
TABLE 02
UJ
..,.,
STATISTICAL BREAKDOWN OF FRIENDLY STRENGTHS AND BATTLE CASUALTIES
IT1
o
IN THE SECTOR NORTH OF TAEGU, 6-16 SEP 50
Unit
US personnel Integrated ROKA personnel Entire unit
Strength
Casualties
Strength
Casualties
Strength
Casualties
KIA I~ : ~ : g g ~ IMIA ITotal
IWIA, OOW I I
KIA IIA; DOl MIA Total
IMA. OOW I I
KIA IIA, 001 MIA Total
6 September
8th Cav
1st lin
Co A
Co B
Co C
Co D
Co HI
179
179
170
141
97
2
0
I
0
0
2
2
0
3
1
0
0
0
0
0
4
2
I
3
1
25
25
24
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
204
204
194
141
97
2
0
I
0
0
2
2
0
3
1
0
0
0
0
0
4
2
3
I
Total 766 3 8 0 II 74 0 0 0 0 840 3 8 0 II
2d B.
Co E
Co F
Co G
Co H
Co H2
169
131
130
137
93
1
0
0
0
0
13
1
0
1
0
2
0
0
1
0
16
1
0
2
0
14
12
19
II
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
183
143
149
148
93
I
0
0
0
0
13
1
0
1
0
2
0
0
I
0
16
1
0
2
0
Total 660 1 15 3 19 56 0 0 0 0 716 I 15 3 19
@
0
I
o-,l
I
'"
'"
....
3d B.
Co I
Co K
Co L
Co M
Co H3
Total
166
159
167
145
86
723
1
I
I
0
I
4
0
5
7
I
7
20
0
0
0
0
0
0
1
6
8
I
8
24
10
5
8
8
34
65
0
0
0
0
0
0
0
3
0
0
2
5
0
0
0
0
0
0
0
3
0
0
2
5
176
164
175
153
120
788
I
1
1
0
I
4
0
8
7
1
9
25
0
0
0
0
0
0
1
9
8
1
10
29
~
,
>-3
,
'"
en
.....
Regt! units
HM Co
Svc Co
Med Co
Hq& HqCo
8th CaY, total
16th RecoD Co
8th Cov & 16th
Recon, total
7 September
8th Cay
1st Bn
Co A
Co B
Co C
Co D
Co HI
Total
2d On
Co E
Co F
Co G
Co H
Co H2
Tota)
3d On
Co I
Co K
Co L
Co M
Co H3
Total
135
141
141
200
2766
136
2902
175
178
170
141
97
761
146
ll8
130
136
91
621
188
159
167
164
86
764
I
I
o
o
10
I
II
I
o
o
o
o
I
2
o
o
o
I
3
2
o
2
o
o
4
I
7
2
I
54
3
57
I
o
o
o
I
2
3
1
o
o
o
4
o
3
3
o
o
6
o
o
I
o
4
o
4
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
2
8
3
68
4
72
2
o
o
o
I
3
5
I
o
o
I
7
2
3
5
o
'0
10
o
o
o
69
264
50
314
25
25
24
o
o
74
14
37
18
11
o
80
20
15
8
18
44
105
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
5
o
5
o
o
o
o
o
o
o
0
o
o
-0
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
5
o
5
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
135
141
141
269
3030
186
3210
200
203
194
141
97
835
160
155
148
147
91
701
208
174
175
182
130
869
I
I
o
o
10
I
II
I
o
o
o
o
I
2
o
o
o
I
3
2
o
2
o
o
4
I
7
2
I
59
3
62
I
o
o
o
I
2
3
I
o
o
o
4
o
3
3
o
o
6
o
o
I
o
4
o
4
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
2
8
3
I
73
4
77
2
o
o
o
I
3
5
I
o
o
I
7
2
3
5
o
o
to
c::
2'
CJ
r

C.'?
(..')
-"
rq
CJ
TABLE 02 (continued)
Unit
US personnel Integrated ROKA personnel Entire unit
Strength
Casualties
Strength
Casualties
Strength
Casualties
IWIA, DOW I I
KIA IIA, DOl MIA Total
IWIA, DOW I I
KIA IIA, DOl MIA Tolal I~ A , DOW I I KIA IIA, DOl MIA To 1.1
7 September (continued).
Regt! units
HM Co 137 0 I 0 I 0 0 0 0 0 137 0 I 0 I
Svc 134 0 0 0 0 0 0 0 0 0 134 0 0 0 0
Med 140 I 0 0 I 0 0 0 0 0 140 I 0 0 I
IIq & IIq Co :1iJO 0 0 I 69 0 0 0 0 269 0 0 0 0
8th Cav, total 'l:I57 9 14 0 23 328 0 0 0 0 3085 9 14 0 23
16th Reean Co 136 0 0 0 0 50 0 0 0 0 186 0 0 0 0
8th Cay & 16th
Recan, total 2893 9 14 0 23 378 0 0 0 0 3'l:11 9 14 0 23
I
""'t
8 September
8th Cay
1st Sn
Co A 175 0 3 0 3 25 0 0 0 0 200 0 3 0 3
Co B 178 0 I 0 I 25 0 0 0 0 :1iJ3 0 I 0 I
Co.C 170 I 0 0 I 23 0 I 0 I 193 I I 0 2
Co D 141 0 0 0 0 0 0 0 0 0 141 0 0 0 0
Co III 98 0 I 0 I 0 0 0 0 0 98 0 0 I
Total 762 I 5 0 6 73 0 I 0 I 835 I 6 0 7
2d Bn
Co E 139 0 i 0 I 14 0 0 0 0 153 0 I 0 I
0
::c Co F 122 0 3 0 3 37 0 0 0 0 159 0 3 0 3
0
Co G 130 0 0 0 0 18 0 0 0 0 148 0 0 0 0
I
>-i Co II 134 0 I 0 I II 0 0 0 0 145 0 1 0 I
I
Co 112 88 0 I 0 I 0 0 0 0 0 88 0 I 0 I
""
... 6 0 6
'" Total 613 0 6 0 6 80' 0 0 0 0 693 0
c:
z
()
~
en
en
"T1
-
m
c
3d Bn
Co I
Co K
Co L
Co M
Co H3
Total
Hegtl units
11M Co
Svc
Med
Hq & Hq Co
8th Cay, total
16th Recon Co
8th Cay & 16th
Recon, total
9th September
8th Cay
1st Bn
Co A
Co B
Co C
Co 0
Co HI
Total
2d Bn
Co E
Co F
Co G
Co II
Co 112
Total
188
159
161
163
86
757
14()
134
14()
!DO
2746
136
2882
175
178
170
141
98
762
115
116
134
134
87
586
o
1
2
o
o
3
o
o
o
o
4
o
4
o
o
o
o
o
o
o
o
o
o
1
14
14
24
2
o
54
2
o
1
1
69
1
70
2
o
o
o
o
2
2
1
o
o
o
3
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
14
15
26
2
o
57
2
o
1
1
73
1
74
2
o
o
o
o
2
2
1
o
1
o
4
20
15
17
18
43
113
o
o
o
69
335
50
385
25
25
23
o
o
73
83
37
18
11
o
149
o
o
1
o
o
1
o
o
o
o
1
o
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
1
1
o
o
o
o
2
o
2
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
1
o
1
2
o
o
o
o
3
o
3
o
o
o
o
o
o
o
o
o
o
o
o
!D8
174
178
181
129
870
14()
134
14()
269
3081
186
3267
!DO
203
193
141
98
835
198
153
152
145
87
735
o
1
3
o
o
4
o
o
o
o
5
o
5
o
o
o
o
o
o
o
o
o
1
o
1
14
14
24
2
1
55
2
o
1
1
71
1
72
2
o
o
o
o
2
2
1
o
o
o
3
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
14
15
27
2
1
59
2
o
1
1
76
1
77
2
o
o
o
o
2
2
o
1
o
4
c:
z
()
~
en
en
"T1
-
m
c
c
:z:
,
C'?
;J>
(./)
(./)
.....,
..,
'"
'"
rT1
0
9 September (continued)
3d B.
Co I 172
Co K 159
Co L 161
Co M 163
Co H3 86
Total 741
Regtl units
HM Co 136
Svc 134
Med 140
Hq & Hq Co 2J)0
8th Cav, total 2699
16th Recon Co 137
8th Cay & 16th
Recon, total 2836
10 September
8th Cay
lst Bo
Co A 174
Co B 175
~
0 Co C 170
I
..., Co 0 140
..,
I
Co HI 97
..... Total 756
'"
TABLE 02 (continued)
US personnel Integrated ROKA personnel Entire unit
Casualties Casualties Casualties
Unit
Strength
1MA, DOW 1 ,I
Strength Strength
1MA, IlOW 1 l 1 ,I MA,DOwl 1
KIA !lA, DOl MIA Total KIA !lA, 001 MIA Total KIA !lA, DOl MIA Total
0
0
0
0
0
0
0
0
0
0
1
3
4
0
0
I
0
0
I
I
I
I
0
4
0
0
0
0
9
22
31
2
0
0
I
2
5
0
0
0
0
0
0
0
0
I
0
1
0
1
0
0
0
0
0
0
I
I
I
0
4
0
0
1
0
II
25
36
2
0
I
I
2
6
2J)
15
17
18
43
113
0
0
0
0
335
50
385
25
25
23
0
0
73
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 7
0 7
0 0
0 0
0 0
0 0
0 0
0 0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
7
0
0
0
0
0
0
192
174
178
181
129
854
136
134
140
2DO
3034
187
3221
199
200
193
140
97
829
0
0
0
0
0
0
0
0
0
0
1
3
4
0
0
1
0
0
I
I
I
I
0
4
0
0
0
0
0
0
I
I
I
I
0
4
0
0
0
0
9
0
0
I
0
1
0
0
0
11
29 0 32
38 1 43
2
0
0
1
2
0
0
0
0
0
2
0
I
1
2
5 0 6
o
~
~
I
0>
'"
....
2d Bn
Co E
Co F
Co G
Co II
Co 112
Total
3d Bn
Co I
Co K
Co L
Co M
Co 113
Total
Regt! units
11M Co
Svc
M.d
IIq & IIq Co
8th CaY, total
16th Recon
8th Cay & 16th
Re<:on. total
7th Cay
3d Bn
Co I
Co K
Co L
Co M
Co 113
Total
Grand Total
112
116
145
134
86
593
172
156
161
163
86
738
133
134
148
:!J9
271J
121
2832
142
151
154
113
90
650
3482
o
o
o
o
o
o
1
1
o
o
o
2
o
o
o
o
3
6
9
o
o
o
o
o
o
9
4
2
2
3
o
II
5
16
2
o
24
o
o
1
o
41
7
48
3
o
o
o
o
3
51
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
4
2
2
3
o
1J
6
17
1
2
o
26
o
o
1
o
44
13
57
3
o
o
o
o
3
60
83
37
19
36
o
177
20
15
17
18
43
113
o
o
o
o
363
43
406
30
35
36
13
35
149
555
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
2
2
o
o
o
o
o
o
2
o
o
o
o
o
o
o
1
o
o
o
1
o
o
o
o
1
1
2
o
o
o
o
o
o
2
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
4
4
o
o
o
o
o
o
4
o
o
o
o
o
o
o
1
o
o
o
1
o
o
o
o
1
7
8
o
o
o
o
o
o
8
195
153
164
170
86
770
192
171
178
181
129
851
133
134
148
209
3074
164
3238
172
186
190
126
125
799
4037
o
o
o
o
o
o
1
1
o
o
o
2
o
o
o
o
3
8
II
o
o
o
o
o
o
II
4
2
2
3
o
11
5
17
1
2
o
25
o
o
1
o
42
8
50
3
o
o
o
o
3
50
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
4
4
o
o
o
o
o
o
4

2
2
3
o
11
6
18
1
2
o
27
o
o
1
o
45
20
65
3
o
o
o
o
3
68
TABLE 02 (continued)
Unit
Strength
US personnel
Casualties
IMA, DOW I I
KIA IIA, DOl MIA Total
Strength
Integrated ROKA personnel
Casualties
IMA, DOW I I
KIA IIA, DOl MIA Total
Strength
Entire unit
Casualties
IMA, DOwll
KIA IIA, 001 MIA Total
11 September
8th Cay
lst Sn
Co A
Co B
Co C
Co 0
Co HI
174
176
170
140
97
1
o
o
o
o
2
I
2
I
o
o
o
o
o
o
3
I
2
I
o
41
39
38
20
8
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
215
215
208
160
105
1
o
o
o
o
2
I
2
I
o
o
o
o
o
o
3
2
I
o
Total 757 I 6 o 7 146 o o o o 903 I 6 o 7
2d Bn
Co E
Co F
Co G
Co H
Co H2
llO
ll8
145
135
86
3
o
1
o
I
9
7
29
7
2
o
o
4
o
o
12
7
34
7
3
74
36
20
33.
o
I
o
o
o
o
9
2
2
4
o
o
o
o
o
o
10
2
2
4
o
184
154
165
168
86
4
o
I
o
I
18
9
31
II
2
o
o
4
o
o
22
9
36
II
3
Total 594 5 54 4 63 163 1 17 o 18 757 6 71 4 81
3d Bn
Co I
Co K
Co L
Co M
Co H3
172
ll2
161
163
86
2
3
o
I
o
25
II
o
6
o
o
o
o
o
o
27
14
o
7
o
25
18
22
18
47
o
o
o
o
o
o
2
o
o
1
o
o
o
o
o
o
2
o
o
I
197
130
183
181
133
2
3
o
I
o
25
13
o
6
1
o
o
o
o
o
27
16
o
7
1
o
~
~
I
'"
'"
....
Total
8th Cay
1st Bn
HM Co
Svc
Med
694
134
135
169
6
o
o
o
42
o
o
4
o
o
o
o
48
o
o
4
130
o
o
o
o
o
o
o
3
o
o
o
o
o
o
o
3
o
o
o
824
134
135
169
6
o
o
o
45
o
o
4
o
o
o
o
51
o
o
4
Hq & Hq Co
209 o 3 o 3 o o o o o 209 0 3 o 3
Total
2692 12 109 4 125 439 I 20 o 21 3131 13 129 4 146
16th Recon Co 101 o o o o 36 o o o o 137 o o o o
Bth Cay & 16th
Recon, total 2793 12 109 4 125 475 I o 21 3268 13 129 4 146
7th Cay
3rd Bn
Co [ 145 o 2 o 2 30 o I o I 175 o 3 o 3
Co K 148 o o o o 35 o 2 o 2 lB3 o 2 o 2
CoL 144 o o o o 36 o o o o 180 o o o o
Co M 113 o o o o 13 o I o I 126 o I o I
Co H3 93 o 1 o I 35 o o o o 12B o I o I
Total 643 o 2 o 2 149 o 4 o 4 792 o 6 o 6
Grand Total 3436 12 III 4 127 624 I 24 o 25 4060 13 135 4 152
12 September
8th Cay
1st Bn
Co A 175 o o o o 41 o o o o 216 o o o o
Co B 177 4 13 o 17 as o I o I 215 4 14 o 18
Co C 170 o o o o as o o o o 298 o o o o
Co D 140 o o o o 20 o o o o 140 o o o o
Co HI 96 o 3 o 3 8 o o o o 104 o 3 o 3
, 903
Total 758 4 16 o 20 145 o I o I 4 17 o 21
2d Bo
Co E 136 . o 2 o 2 74 o o o o 210 o 2 o 2
Co F III o o o o 36 o o o o 147 o o o o
Co G 133 2 2 o 4 20. o o o o 153 2 2 o 4
Co H 131 o 1 o I 33 o o o o 164 o 1 o I
Co H2 B6 n 1 o I o o o o o 86 o 1 o I
Total 597 2 6 o 8 163 o o o o 760 2 6 o 8
c
:z
o
c r
:z ::x>
(.1)
o
C/)
r
::x> ..,.,
C/) TABLE 02 (continued)
C/) '"
Unit
Strength
US personnel
Casualties
IWIA, DOW I I
KIA IIA DOl MIA Total
Integrated ROKA personnel
Casualties
Strength
IWIA, DOW I I
KIA llA DOl MIA Total
Strength
Entire unit
Casualties
IWIA, DOW I I
KIA 1IA. DOl MIA Total
rn
o
'"
o ..,.,
rn
o
12 September (continued)
3d Bn
Co I 172 3 I 0 4 25 0 0 0 0 197 3 I 0 4
Co K Jl2 0 2 0 2 18 0 0 0 0 130 0 2 0 2
Co L 161 0 0 0 0 22 0 0 0 0 183 0 0 0 0
Co M 163 0 I 0 I 18 0 0 0 0 181 0 I 0 I
Co H3 86 0 0 0 0 45 2 0 0 2 135 2 0 0 2
Total 694 3 4 0 7 128 2 0 0 2 822 5 4 0 9
Regt units
lIM Co 135 0 I 0 0 0 0 0 0 135 0 I 0 I
Svc 138 0 I 0 I 0 0 0 0 0 138 0 I 0 I
Med 170 I I 0 2 0 0 0 0 0 170 I I 0 2
IIq & llq Co 209 0 I 0 0 0 0 0 0 209 0 I 0 I
I
....
8th Cav, total 2701 10 30 0 40 436 2 0 3 3137 12 31 0 43
16th Recon Co 98 0 0 0 0 36 0 I 0 I 134 0 0 I
8th Cay & 16th
Recon, total 2799 10 30 0 40 472 2 2 0 4 3271 12 32 0 44
7th Cay
3d Bn
Co I 142 9 43 0 52 29 I 5 0 6 17l 10 48 0 58
Co K 148 2 14 0 16 33 3 4 0 7 181 5 18 0 23
Co L 144 21 37 0 58 36 0 4' 0 4 180 21 41 0 62
Co M 113 0 5 0 5 12 0 0 0 0 125 0 5 0 5
0
Co 1I3 93 0 0 I 35 2 2 0 4 128 2 3 0 5
13
Total 640 32 99 0 132 145 6 IS 0 21 785 38 114 0 153
~
,
..,
Grand Total 3439 42 129 0 172 617 8 17 0 25 4056 50 146 0 197
.....
'"
13' September
8th Cav
1st Bn
Co A
Co B
Co C
Co D
Co HI
Total
2d B.
Co E
Co F
Co G
Co H
Co H2
Total
3d B.
Co I
Co K
Co L
Co M
Co H3
Total
Regtl units
HM Co
Svo
Med
Hq & Ilq Co
8th Cav, total
16th Recon Co
8th Cav & 16th
Recon, total
175
177
170
140
95
757
156
104
100
122
85
567
172
114
160
163
83
692
134
137
181
208
2676
97
2773
o
3
o
o
o
3
o
I
o
o
o
I
o
o
o
I
o
I
o
o
o
o
5
o
5
o
7
o
o
o
7
o
4
I
o
I
6
2
o
o
o
o
2
o
o
I
3
19
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
10
o
o
o
10
o
5
I
o
I
7
2
o
o
I
o
3
o
o
I
3
24
o
24
41
38
38
20
8
145
74
34
20
32
o
160
25
18
22
18
44
127
o
o
o
o
432
38
470
o
o
o
o
o
o
o
2
o
I
o
3
o
o
o
o
o
o
o
o
o
o
3
o
3
I
I
o
o
o
2
I
o
o
o
o
I
o
o
o
o
2
2
o
o
o
o
5
o
5
o
o
o
o
o
'0
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
I
I
o
o
o
2
I
2
o
I
o
4
o
o
o
o
2
2
o
o
o
o
8
o
8
216
215
208
160
103
902
230
138
120
154
85
727
197
132
182
181
83
819
134
137
181
208
3108
135
3243
o
3
o
o
o
3
o
3
o
I
o
4
o
o
o
I
o
I
o
o
.0
o
8
o
8
1
8
o
o
o
9
I
4
I
o
I
7
2
o
o
o
2
4
o
o
I
3
24
o
24
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
I
II
o
o
o
12
I
7
I
I
I
II
2
o
o
I
2
5
o
o
I
3
32
o
32
TABLE 02 (continued)
Unit
US personnel Integrated ROKA personnel Entire unit
Strength
Casualties
Strength
Casualties
Strength
Casualties
IMA, DOW I I
KIA IIA, DOl MIA Total
IMA, DOW I
KIA IIA, DOl MIA
I
Total IM ~ DOW II KIA IIA', DOl MIA Total
13 September (continued)
7th Cay
3d Bn
Co I 145 o o o o 23 o I o I 168 o I o I
Co K 163 o 2 o 2 26 o o o o 189 o 2 o 2
Co L 148 I I o 2 32 o o o o 180 I I o 2
Co M 130 o o o o 12 o o o o 142 o o o o
Co 113 106 o o o o 31 o 2 o 2 137 o 2 o 2
Total 692 I 3 o 4 124 o 3 o 3 716 I 6 o 7
Grand Totai 3465 6 22 o 28 594 3 8 o 11 4059 9 30 o 39
14 September
8th Cay
1st Dn
Co A 151 o 1 o I 41 o o o o 192 o I o I
m
m Co IJ 142 o o o o 38 o o o o 180 o o o o
Co C 173 o o o o 38 o o o o 211 o o o o c
c
Co D 140 o o o o 20 o o o o 160 o o o o
Co HI 89 o I o I 8 o o o o 97 o I o I
Total 695 o 2 o 2 145 o o o o 840 o 2 o 2
2d En
Co E 158 o 9 o 9 74 o o o o 232 o 9 o 9
. I
Co F 98 o I o I 34 o o o o 132 o o I
Co G 100 o o o o 20 o o o o 120 o o o o
Co H 121 o 2 o 2 32 o o o o 153 o 2 o 2
Co H2 82 o o o o o o o o o 82 o o o o
Total 559 o 12 o 12 160 o o o o 719 o 12 o 12
c:
z
()
~
en
en
"T1
~
m
c
~
,
..
>'l
,
......
'"
3d Dn
Co [
Co K
Co L
Co M
Co H3
Total
Regtl units
HM Co
Svo
Med
Hq & Hq Co
8th Cay, total
16th Recon Co
8th Cay & 16th
Recon, total
7th Cay
3d Bn
Co I
Co K
Co L
Co M
Co 113
Total
Grand Total
15 September
8th Cay
1st Bo
Co A
Co B
Co C
Co 0
144
84
140
128
81
577
134
138
174
211
2488
98
2586
144
165
149
131
104
693
3279
154
140
173
140
4
1
o
o
o
5
o
o
o
o
5
o
5
o
o
o
o
o
o
5
o
o
1
o
1
1
o
o
o
2
o
o
1
o
17
5
22
o
o
o
o
o
o
22
1
o
1
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
5
2
o
o
o
7
o
o
1
o
22
5
27
o
o
o
o
o
o
27
1
o
2
1
24
18
22
18
44
126
o
o
o
o
431
38
469
22
25
32
12
29
120
589
50
50
43
20
o
o
o
o
o
o
o
o
o
o
o
o
o
1
o
o
o
1
2
2
o
o
o
o
o
o
o
o
o
o
o
o
o
1
o
o
o
o
o
o
1
2
o
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
1
o
o
o
o
1
o
o
o
o
I
o
1
1
o
o
o
1
2
3
2
o
1
o
168
102
162
146
125
703
134
138
174
211
2919
136
3055
166
190
181
143
133
712
3868
204
190
216
160
4
1
o
o
o
5
o
o
o
o
5
o
5
1
o
o
o
1
2
7
o
o
1
o
2
I
o
o
o
3
o
o
1
o
18
5
23
o
o
o
o
o
o
23
3
o
2
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
6
2
o
o
o
8
o
o
1
o
2 ~
5
28
1
o
o
o
1
2
30
3
o
3
1
c:
z
()
~ ~
~ ~
"T1
-
m
c
c
z
C?
r
J>
en
en
TABLE D2 (continued)
-n
rr1
C
Unit
Strength
15 September (continued)
Co III 90
US personnel
Casualtie&
KIA 1WIA, DOW 1MIA 1
IIA, DOl
Total
Integrated ROKA personnel
Casualties
Strength
.1 WIA, DOW I l I
KIA IIA, DOl MIA Total
Strength
Entire units
Casualties
KIA IWIA, DOW1MIA I
IIA, DOl '
Total
o o o o 8 o o o o 98 o o o o
Total 697 I 3 o 4 17J o 3 o 3 H68 6 o 7
2d Bo
Co E 143 2 14 o 16 84 o o o o 227 2 14 o 16
Co F 99 I 8 o 9 44 o o o o 143 I 8 o 9
Co G 100 o o o o 20 o o o o 120 o o o o
Co H 119 o 2 o 2 32 o o o o 151 o 2 o 2
Co H2 84 o I o o o o o o 84 o o I
Total 545 3 25 o 28 180 o o o o 725 3 25 o 28
3d B.
Co I 133 2 o 3 31 o o o o 164 I 2 o 3
Co K 84 o 4 o 4 25 o o o o 119 o 4 o 4
Co L 110 2 15 I 18 22 o o o o 132 2 15 I 18
Co M 128 I o 2 18 o o o o 146 I I o 2
Co H3 81 o o o o 44 o o o o 125 o o o o
Total 536 4 22 I 27 140 o o o o 676 4 22 I 27
Regtl units
HM Co 134 o 2 o 2 o o o o o 134 o 2 o 2
Svc 138 o o o o o o o o o 138 o o o o
Med 174 2 5 o 7 o o o o o 174 2 5 o 7
Hq & IIq Co 211 o o o o o o o o o 211 o o o o
@ 8th CaY, total 2435 10 57 68 491 o 3 o 3 2926 10 60 I 71
o
..,
I
16th Recon 99 o 5 o 5 38 o o o o 137 o 5 o 5
I
0>
'" 8th Cay & 16th
....
Recon, total 1534 10 62 73 529 o 3 o 3 3063 10 65 I 76
g
o
I
>'I
I
'"
....
'"
7th Cav
3d B.
Co I
Co K
Co L
Co M
Co H3
Total
Grand Total
16 September
8th Cay
1st Bo
Co A
Co B
Co C
Co D
Co HI
Total
2d Bn
Co E
Co F
Co G
. Co H
Co H2
Total
3d B.
Co I
Co K
Co L
Co M
Co H3
Total
131
150
123
128
102
634
3168
151
140
172
140
89
692
149
98
102
114
86
549
131
84
100
12B
82
525
I
o
2
o
o
3
13
o
o
o
o
o
o
o
3
o
o
o
3
1
o
1
I
o
3
o
o
5
o
o
5
67
o
o
2
1
10
13
4
3
5
3
o
15
7
I
3
I
o
12
o
o
o
o
o
o
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
1
o
7
o
o
8
81
o
o
2
1
10
13
4
6
5
3
o
18
8
I
4
2
o
15
21
24
32
12
2B
!l7
646
50
49
43
20
8
170
84
44
20
32
o
180
31
25
29
18
44
147
I
o
o
o
o
I
1
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
3
o
1
o
o
o
I
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
I
o
o
o
o
4
o
I
o
o
o
I
o
o
o
o
o
o
o
o
o
o
o
o
152
174
155
140
130
751
3814
201
189
215
16{)
97
862
233
142
122
146
86
729
162
109
129
146
126
672
2
o
2
o
o
4
14
o
o
o
o
o
o
0
3
o
o
.0
3
I
o
I
1
o
3
o
o
5
o
o
5
70
o
1
2
1
10
14
4
3
5
3
o
15
7
I
3
I
o
12
o
o
o
o
o
o
I
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
2
o
7
o
o
9
85
o
1
2
1
10
14
4
6
5
3
o
18
8
I
4
2
o
15
c::
z
C"'>
r
:>
(/)
~
....,
rT1
o TABLE 02 (continued)
Units
US personnel Integrated ROKA personnel Entire units
Strength
Casualties
Strength
Casualties
Strength
KIA
Casualties
ImA, DOW I I
llA, DOl MIA Total
IWIA, DOW I I
KIA llA, DOl MIA Total
ImA, DOW I I
KIA llA, DOl MIA Total
16 September (contmued)
Regtl units
HM Co 133 I 0 0 I 0 0 0 0 0 133 I 0 0 I

Svc 138 0 0 0 0 0 0 0 0 0 1M 0 0 0 0
Med 174 0 I 0 I 0 0 0 0 0 174 0 I 0 I
Hq & Hq Co 213 0 I 0 I 0 0 0 0 0 213 0 I 0 I
8th CaY, total 2424 7 42 0 49 497 0 I 0 2921 7 43 0 50
: ~
16th Recan Co 96 0 I 0 I 39 0 0 0 0 135 0 I 0 I
!
8th Cay & 16th
~
Recon, total 2520 7 43 0 50 536 0 0 I 3056 7 44 0 51
7th Cay
3d Bn
Co I 158 0 0 0 0 20 0 0 0 0 178 0 0 0 0
Co;; K 168 0 0 0 0 24 0 0 0 0 192 0 0 0 0
Co L 151 0 0 0 0 30 0 0 0 0 181 0 0 0 0
Co M 126 0 0 0 0 12 0 0 0 0 138 0 0 0 0
Co H3 103 0 0 0 0 28 0 0 0 0 131 0 0 0 0
Total 706 0 0 0 0 114 0 0 0 0 820 0 0 0 0
~
Grand Total 3226 7 44 0 51 650 0 I 0 I 3876 7 45 0 52
0
I
'"'l
I
a>
""
....
SfCllET
UNCLASSIFIED
Appendix E
SUPPLEMENTARY FRIENDLY DOCUMENTATION
ON THE CAPTURE OF HILL MASS 314
CONTENTS
Page
OPERATIONS ORDER FOR THE ATTACK ON THE NORTHERN PORTION
OF HILL MASS 314 BY THE 3D BATTALION, 7TH CAVALRY REGIMENT 339
AFTER'ACTION REPORTS 341
SUMMARY ACCOUNT OF THE BATTLE REPORTED IMMEDIATELY AFTERWARD,
12 SEP 50-SEQUENCE OF EVENTS IN THE CAPTURE AS GIVEN IN A PARTIAL
OUTLINE PREPARED SHORTLY AFTER THE ACTION-INTERVIEW WITH THE
COMMANDING OFFICER OF COMPANY L RECORDED SHORTLY AFTER THE
CAPTURE-INTERVIEWS WITH PERSONNEL OF COMPANY I REPORTED SHORTLY
AFTER THE CAPTURE
AFTER-ACTION PHOTOGR'APHS 345
FIGURES
El. VIEW AN ENEMY OUTPosr MIGHT HAVE HAD FROM POINT NEAR BOTTOM
OF SADDLE ON NORTHEAST SLOPE 345
E2. EAST SLOPE OF KNOB 2 346
E3. CLOSER VIEW OF EAST SLOPE OF KNOB 2 346
E4. PHASE LINE 347
E5. WEST ALONG PHASE LINE 2 347
E6.E7. Two VIEWS OF THE NOSE OF KNOB 2 348
E8. HILL 660 AND HILLS BEYOND 349
E9. VALLEY EAST OF HILL 314 AND NORTH OF HILL 660 349
EIO. THE CO, Co L, 3D BN, 7TH CAV REGT 350
Ell. THE CO, 3D BN, 7TH CAV REGT 350
TABLE
El. MASTER LIST OF ALL CASUALTIES IN THE 3D BN, 7TH CAV REGT
CONSIDERED TO HAVE OCCURRED IN THE CAPTURE OF HILL MASS 314 351
ORO-T-261
337
SEun
UNCLASSIFIED
SEUEY UNCLASSIFIED
OPERATIONS ORDER FOR THE ATTACK ON THE NORTHERN PORTION OF
lULL MASS 314 BY THE 3D BATTALION, 7TH CAVALRY REGIMENT
Note: Except for annotations In brackets, document appears In Its original form, which
was undated and apparently written after the action.
1. A. Elements of the 13th North Korean Division hold Hill 570 and Mountain Peak
500 [northern portion of Hill Mass 314]. Enemy units have Infiltrated onto Hill 660.
Enemy strenglh on H!ll Mass 314 [the southern portion of the hill mass] and Mountain
Peak 500 beyond Is estimated at 700.
B. 7th Cavalry Regiment attacks 12 Sep 1950 to relieve ROK units on Hill 660
with 2d Bn and to seize H!ll Mass 314 and high ground beyond with 3d Bn. 8th Cavalry
Regiment defends from present position on west. 1st ROK Division attacks towards
Walled City In the east.
2. A. 3d Battalion, 7th Cavalry Regiment (Inf) attacks at 121100 September 1950
from positions presently held by 3d Battalion, 8th Cavalry Regiment (lnf) to seize H!ll
Mass 314 and the high ground beyond and to defend same on seizure.
Attachments: 3d Plat Co C, 70th Tank Bn
1 Squad, Co C, Engr Bn
Support: Airstrike
77th FA Bn
HM Co [8th Cavalry, one platoon]
Co L, 8th Cav
B. See overlay. [contents Included In Fig. 12]
3. A. Company "1":
Attached: 1st Sect HMG Plat, Co M
1 Squad, Co C, Engr Bn
FO party 77 FA Bn
FO Party HM Co, 8th Cav .
FO Party 8lmm Plat Co M
3 Flame Thrower Teams
P&A Plat, Hq Co attacks on right t21100 Sep 50 to
seize portion of Bn objective in zone and to pre
pare perimeter defense from 1200 to 0400 [o'clock
angle), upon seizure. (See overlay)
B. Company "L":
Attached: 2d Sect HMG .Plat, Co M
FO Party 77 FA Bn
FO Party HM; Co, 8tb Cay
FO Party 81mm Plat Co M
3 Flame Thrower Teams
ORO-T-261
339
SEERET
UNCLASSIFIED
UNCLASSIFIED SEUff
P&A Plat, Hq Co attacks on left at 121100 Sep to
seize portion of Bn objective in zone and to pre
pare perimeter defense from 0800 to 1200 [o'clock
angle] on seizure. (See overlay.) Be prepared
to be passed tbrougb by Co UK" on Pbase Line .
2 if so ordered.
C. Company UK": Battalion reserve initially located on soutbeast portion of
Hill Mass 314 near the LD. Be prepared to counterattack or move to
successive phase lines on order. (See overlay.) Be prepared to pass
tbrough Company L on Phase Line 2 to continue attack and seize and
defend Company L's portion of the battalion's objective.
D. Company "M" (-HMG Plat): Support attack by fire. I-repare to move over
all 75mm RR to tbe axis of attack on order. Coordinated 81mm fires of
3d Bn, 8th Cav in support of tbe attack.
E. Battalion Headquarters Company: P&A Plat to man 3 flame tbrowers for
Co I and 3 for Co L, during and subsequent to tbe attack. Remaining
HQ personnel furnish logistical and administrative support. P&A Plat
be prepared to install chain demolition and barbed wire on seiziug tbe
objective.
F. 3d Plat, Co C, 70th Tank Bn: initially in reserve positions in rear of Hill
Mass 814, prepared to move up either draw on east or draw on west. to
support attack by fire or to protect against counterattack.
G. (1) Air preparation from 0930 to 1030 hours 12 Sep 1950. Arty preparation
from 1045 to .1100 hours 12 Sep 1950. Display red panels.
(2) HM Co, 8tb Cav (- 1 Plai) in general support of 3d Battalion, 7tb Cav
alry until relieved by HM Co, 7tb Cav.
(3) Flame tbrower to be included in the final defense.
(4) 3d Bn, 7tb Cav (Co L) responsible for draw to the west of Hill Mass 314.
(5) Arty protective fires to be registered immediately on securiug position.
(6) Bn final objective and Hill 660 constitute the key to defense of Taegu
and must be held at all cost.
(7) Troops depart present area 120630 Sep 50 in order I-L-Hq-M-K, 400
yard inierval between companies. 3d Bn, 8tb Cav guide will be fur
nished to attack positions.
4. A. Supply will be pushed up to the front by Battalion.
B. initial ASP base of Hill 314
C. Sufficient ratiOns, waier and ammo to hold final objective will be placed
on final objective.
D. 100 South Korean laborers pass to Bn control at Chilgok, coord ( )
[blank. Coordinstes are 1153.4-1448] at 120900 Sep 50. Each company to
receive 20 laborers ai base of Hill 314 prior to time of attack.
E. Evacuation down ridge line to Fwd Supply Point, tben by vehicle to Aid Station
and ASP as applicabie.
F. MSR-road to east of final objective between Hill 314 and Hill 660.
5. A. Bn CP ai base of Hill 314 opens 121000 Sep 50 at coord ( ) [blank].
B. Transfer to 7tb Cav radio channels and call signs at 120600 Sep 50.
C. Red smoke signifies friendly fire fa11iug short.
D. Purple smoke and violet streamers to mark front line elements.
E. Carry and display red panels for air identification. Panel crews to be desig
nated by each company.
OFFICIAL
LYNCH
WEBEL (S-3)
Lt Col
ANNEX: 1 Opn Overlay
340 ORO-T-261
UNCLASSIFIED
~ E E A [ J
uun
UNCLASSIFIED
AFTERACTION REPORTS
SUMMARY ACCOUNT OF THE BATTLE REPORTED IMMEDIATELY
AFTERWARD, 12 SEP 50
Note: The following entry in the log of the 77th FA Bn is the only known outline of the
battle as a whole that gives the facts as they were known just afterwards. It is
considered that only this report correctly states several important pOints, though
there are also inaccuracies.
"121820. From Liaison Officer, 7th Cavalry Regiment: 3d Battalion, 7th Cavalry
Regiment jumped off ai 1115 hours, after an alrstrike which was extremely effective.
The baltalion had advanced about 500 yards by 1230 hours, Love Company on the left
and Item Company on tbe right and K Company in reserve. At about 1230 hours the
baltalion was pinned down by intense rifle and mortar fire. At this time they also observed
a concentration of North Koreans assembling to their left for a counteraltack. They
called for an alrstrike at 1300 hours, which was delayed until 1330 hours. An airstrike
was given again ai 1400 hours. The battalion then assaulted the hill mass twice and was
driven off by intense 60 [sic] and 82mm mortar fire, then by 120mm followed by artillery
fire. On the third altempt our troops stormed over the hill, driving the enemy down the
north slope. The enemy was then taken under artillery fire with good results. The
estimaied number of enemy troops on the hill was 500-800. The character of the defense
was caves with machine guns emplaced in rocks and crevices and sniper fire from among
jagged peaks and ravines. Our casualties are estimaied to be about 35 percent of the
two assaulting companies. The 57 and 75mm recoilless rifles were very effective. The
hill was secured ai 1545 hours. Coordination is being made to tie in with units on our
right and left flanks.'
SEQUENCE OF EVENTS IN THE CAPTURE AS GIVEN IN fA PARTIAL OUTLINE
PREPARED SHORTLY AFTER THE ACTION
Note: This outline is given as prepared hy an unknown staff officer of the 3d Bn, 7th
Cay, except for the annotations in ,brackets.
Troops moved out on time at 120630 Sep 50, in proper order by covered route at
extended interval. Due to advanced position of Co L, 8th Cav, our LD was changed to
Phase Line 1 and all companies and Regt were notified. Co L was moved up parallel to
Co I on Hill Mass 314 in view of this advanced LD.- The guide for Co K got lost and
another guide was arranged. Co's L and I were in position at 1030 ready to attack. Air
strike started approximately 1000 and lasted until 1100 due to cancelation of arty prep
aration on Regt's order at 0735. Attack jumped off on time against moderated heavy small
arms fire and under mortar fire. Phase Line 2 [phase Line 1, a ~ redefined with the
ORO-T-261 341
UNCLASSIFiED
$fUEl..
UNCLASSIFIED
55ERET
change in LD] was seized and reorganization was completed at 1245. At 1315 a counter
attack of 250 enemy was launched against Co I and was beaten off by that unit. At 1345
an additional air strike forward of Phase Line 2 [this designation correci by the present
reckontng] furnished support for resumption of the attack at 1415 by Co's I and L.
During the advance from Phase Line 2 to Phase Line 3 [1 to 2, or as given ?], heavy
mortar and arty fire was received, as well as small arms and grenade fire. Final objec
tive was seized at 1530 hours and, due to the loss of Co I commander, was organized
jointly by Co's L and I under command of Capt Walker, Co L CO.
Co K, reserve Co, was moved to original LD at 1030 due to advanced location of
Co L, 8th Cav. At 1100 hours Co K was moved to Phase Line 2 [phase Line 1] and
[-much later-] began organization of rear portion of Bn objective.
Resupply of ammo, water and rations was instituted Immediately, and ammo bearers
functioned as litter bearers on the return trip. Evacuation of wounded extended from
121130 Sep to 131200 Sep 50. Evacuation of US dead began at 130600 and extended through
131700 Sep.
Evacuation of captured ammo, arms and equipment, as well as salvage items took
place beglnntng 131200 Sep 50 and extended to 171700 Sep 50.
Enemy dead were burled beglnntng at 140600 Sep 50 in shallow graves in place,
using lime as a sanitation measure. ApprOximately 200 enemy dead. on position.
Bn supported tbe attack of 2d Bn, 8th Cav starting at 141400 against Hill 570,
Inflicting heavy casualties on enemy units 75 yards north of Hill 570 on east slope.
Direci hits with 75mm RR obtained on troop concentrations. Support included 1-75mm
RR, 1 of HMGs and 1 seciion of LMGs. Hlll 570 was secured by 2d Bn, 8th Cav ai
141645 hours. CO, 2d Bn stated our support was -extremely effective."
At 171000 Co L was moved to defensive positions vicinity of Phase Line 1 [the 12
Sep LD], due to necessity for movement of Co A, 1st ROK special Training Bn to relieve
2d Bn, 7th Cav on Hill 660. Co I assumed forward positions of Co L and Co K extended
to fill in rear positions of Co L.
[Some further material Is not reproduced here.]
INTERVIEW WITH THE COMMANDING OFFICER OF COMPANY L RECORDED
SHORTLY AFTER THE CAPTURE
Notes: Interview and report by an unknown staff officer of the 3d Bn, 7th Cav.
Except for annotations in brackets, document appears in its original form.
Conversation with Commanding Officer of Company L (Captain Walker)
1st and 3d Platoons were leading the assault, 3d Platoon on the right and 2d Plaioon
in reserve. Company L moved out in the attack ai 1100. 1st Platoon was commanded by
Lt Holder, 3d Platoon by Lt Sheridan and 2d Platoon by M/Sgi Mitchell. Capt Walker
ordered Sgt Mitchell to move 2d Platoon Into draw behind the 1st Platoon, whicb was on
the left. Phase Line 1 and Phase Line 2 [Line of Departure and Phase Line 1, respec
tively, as redefined before the attack] were occupied with negligible enemy opposition.
Upon approaching Phase Line 3 [2], the 1st and 3d Platoons moved rapidly up the crest
of Hlll 314 in the face of Intense and effeciive enemy small arms and mortar fire, which
began to produce exceptionally beavy casualties among personnel of the company. The
individual soldiers of the company continued to advance in the face of this fire even
though units were intermingled and leaders had become casualties, until Phase Line 3
[2] was secured at about 1230 hours.
342
ORO-T-261
UNCLASSIFIED
.s&ERH
UNCLASSIFIED
Captain Walker ordered Sgt Mitchell around to the left of the 1st Platoon with the
mission of attacking and moving over the crest of Hill 314. As the 2d Platoon moved
over the crest, it came under very heavy time fire, small arms fire and grenades, which
caused very heavy casualties among personnel of the platoon and pinning It to the slope
of Hill 314. Captain Walker, Company Commander of Company L, sized up the situation
and requested an air strike, which was on call, that he felt was necessary to regain the
momentum of his company's attack. The air strike was delivered and was very effective
and in some cases seventy five to one hundred yards in front of Company L. At this
time, the desperate and fanatical enemy launched a "banzai" attack which caused the
men to waver as Sgt Mitchell and Lt Sheridan were killed. Company I and Company L
were intermingled. Therefore, Captain Walker and Lt Fields (Company Commander of
Company I) organized the remnants of both companies while exposed to the continuous
enemy small arms, grenades and mortar fire and launched a counterattack which caused
the enemy to break, discard his weapons and flee in utter riot.
INTERVIEWS WITH PERSONNEL OF COMPANY I REPORTED SHORTLY AFTER
THE CAPTURE
Notes: Interviews and report by an unknown. staff officer of the 3d Bn, 7tb Cav.,
Except for annotations in brackets, document appears in its original form.
Conversation with 1st Sgt and other EM of Company I, as no officers were available who
took part in the attack.
Company I aitacked tbe rigbt portion of tbe battalion objective in tbe following
formation: 2d Plat on the right, 1st Plat on the left, 3d Plat in reserve. Almost at the
outset of the attack, the platoon leaders of the 1st and 2d Platoons were wounded in action,
and the units became intermingled. However, the individual soldiers of the two platoons
conttnued on their own initiative in the face of heavy small arms, grenades, and mortar
fire, which caused heavy casualties among personnel of the platoon toward the objective.
Upon lifting of the air strike, the fanatical enemy launched a -banzai" attaCk, which
caused the men to waver and elements of Company L to be overrun. Lt Haynes Company
I [leader of tbe 3d Platoon] noted this situation and organized a pickup group of men
from Company I and Company L while in the face of the continuous enemy fire, which
restored the situation just at the moment Lt Haynes was killed. Lt Fields, Company
Commander of Company I, organized the remnants of his company aild counterattacked
tbe enemy up the crest of tbe hill, Which was an estimated 60-degree slope. Lt Fields
and his men continued the attack until the enemy discarded his weapons and fled in utter
confusion. At approximately the time the objective was secured, Lt Fields was wounded
in action, but he refused to be evacuated until the completion of the reorganization of
his company on their portion of the battalion objective.
M/Sgt McCollum, tbe weapons platoon leader of Co I, was following tbe attack witb
his weapons platoon at tbe moment tbe enemy launched his fanatical counterattack. M/Sgt
McCollum rushed forward leading his men after he had organized them as riflemen and
came to the assistance of the bard pressed rifle platoons. M/Sgt McCollum continued
to lead his men even though he was wounded four times. M/Sgt McCollum was wounded
in both shoulders, his right arm and finally received a head wound before be allowed
hiIhself to be evacuated.
ORO-T-261
343
S"-RET
UNCLASSIFIED
UNCLASSIFIED
SEERE-r
Sgt Manuel of Company I charged a fanatical North Korean who was holding up two
squads of the first platoon. Sgt Manuel killed the enemy soldier with a grenade, thus
allowing the platoon to continue its advance. In this action Sgt Manuel received a wound
in his right leg and right hip.
Sgt Lancaster, Assistant Platoon Sgt, 1st Platoon, Company I, took command of his
platoon when the platoon leader became a casualty and started leading his men toward
the objective. Sgt Lancaster was wounded by shell fragments and continued to lead his
platoon until he falnted from exhaustion and loss of blood.
344 ORO-T-261
5!CR!I
UNCLASSIFIED

:nnn
AFTERACTION PHOTOGRAPHS
The photographs in this section were taken by the commanding
officer of the 3d Bn, 7th Cav Regt on 13 September. The objective had
necessarily already been cleared of much of the litter of battle, but
these pictures do illustrate details of terrain, cover, enemy positions,
etc. Since most of them were taken on the east side of the hill mass,
they complement the photographs of the west side taken from Hill 570
and shown in panorama as Fig. 3. Unfortunately, no close-ups are avail
able of the southern and western slopes of Knob 2, where the most intense
action took place.
The pictures are arranged in the order in which the views might
have unfolded to members of Co I during the battle,i.e., from south to
north and from bottom to top.
. ! .' - , . .
'J ."
. '
.;" .:..
- '.
-, ,
,'.,:.
. .':'"
Fig. El-View an enemy outpost might have had from point near bottom of saddle on north
east slope looking southeast at finger that runs east from bottom of saddle. Sweep squad,
Co I overran outpost AW position in this vicinity. Underbrush is typical of east side of
hill mass. Trail was friendly supply-evacuation route during battle. A party of Korean
bearers appears here. In lower right corner it is possible (with the aid of a glass) to make
out a Korean body.
ORO-T-261 345
S[(R"
UNCLASSIFIED
UNCLASSIFIED
SEERff
Fig. E2"':'East slope of Knob 2 short of Phase Line 2 and just below nose.
Taken from point on southeast slope, looking due north. The Walled City
is in background. The 1st Plat, Co I had this view .
. Fig. E3-Closer view of east slope of Knob 2, showing rocky surface
and I ack of cover.
346
OR0-T-261
UNCLASSIFIED
UCRn
HERM
UNCLASSIFIED
Fig. E4-Moving around the eastern slope of Knob 2 to Phase Line 2,
members of Co I would noW have been able to see Knob 3 and the dip in
the ridgeline between Knob 2 and Knob 3. Early in the battle, enemy
troops were seen withdrawing around this slope.
Fig. E5-Moving west along Phase Line 2, a soldier of Co I would have
caught a glimpse of the crest of Knob 2 as he advanced within a few feet
of the nose.
ORO-T-261 347
SEER.El UNCLASSIFIED
UNCLASSIFIED
fig. E6(top) ond fig. E7(bottom)- Two views of the nose of Knob 2 look
ing south across saddle. Knob 1 and Knob 0 apporently merge in the back
ground. 'The lower ridge much beyond is also visible. Enemy materiel
and cat holes can be seen on the almost bare nose of the knob. The
neatly cut out bank of earth at the left faces south or east, origin unknown.
During the bottle, enemy rifles were stuck in the ground to mark the loca
tion of friendly wounded, etc.
348 ORO-T-261
UNCLASSIFIED
SECRET
SEER!T
UNCLfi.SS\f\ED
Fig. E8-Hill 660 and hills beyond as seen from east slope of Knob 2.
Fig. E9-View, from SOme position as in Fig. Ea, of volley east of Hill
314 and north of Hill 660. This picture may be tied in with the previous
one by the light colored patch at the foot of Hill 660.
ORO-T-261 349
SEERn
UNCLASSIFIED
UNCLASSIFIED SEEREf
Fig. E 10-The CO, Co L, 3d Bn, 7th Cay Regt. Enemy concussion
grenades ready for use.
Fig. E11-The CO, 3d Bn, 7th Cay Regt
ORO-T-261
350
UNCLASSIFIED
TABLE EI
o
~
,
MASTER LIST OF ALL CASUALTIES IN THE 3D BN, 7TH CAY REGT
...,
,
CONSIDERED TO HA YE OCCURRED IN THE CAPTURE OF HILL MASS 314
'"
(Integrated ROKA personnel included)
.....
'"
Order
No.
Co Rank Name Serial No. Type Cas Time
Type of missile/
causative weapon
Source Cas dataS
Remarks
CaSt IMed IOthe,
Rept tag
I L PIc McCurry, S. R. RA 14322>17 \VIA 1100 ll/RiOe X
2 L SFC Rowland, J. H. RA 34868113 IVIA 1100 F/Mortar X X Cas reports state 13 Sep but subject's description in
clinicai records indicates 12 Sep action. SheH
landed about six feet away; knocked out a minute,
then wal,ked to aid station.
~
3 L Cpl Joyce, R. H. RA 13320244 \VIA 1130 F /Grenade X
~
4 L PIc Young, C. T. RA 13332751 IVIA 1145 -/- X
5 L PIc Boucher, J. N. RA 11199488 IVIA 1200 F/Mortar X
i
i
6 L Sgt Dent, F. J. RA 15205646 \VIA 1200 F/Mortar X
7 L Sgt jones, D. C. RA 45018350 KiA 1200 . F; Ariillery X Time est. from account of tbis casualty hy EM
eyewitness.
8 L PIc Martin, E. RA 13397311 IVIA 1200 -/- X
9 L PIc Wonn, J. T. RA 13330961 WIA 1200 F/Mortar X
.......
10 L 1st Lt Sheridan. R. B. 0963356 KIA 1215 F /Mortar X Time and cause from interviews with EM

Z
eyewitnesses. c->
c::: r
11 L Pfc Slussear, A. M. A 13315887 WIA 1230 F/Mortar X Type of fragment from EM eyewitnesses. ",.
:z
(/)
c->
r-
12 L PIc Donoghue, O. J. RA 12342502 \VIA 1300 F/Mortar X
~
)::0
13 L PIc Entrekin, H. E. RA 24518670 IVIA 1300 F/_ X
(/) ::!!
U)
14 L Cpl Grassmyer, R. P. RA 33760985 IVIA 1300 B/- X .31-cal bullet. I'T1
..., 0
15 L PIc Herdman, D. J. RA 13317444 \VIA 1300 F /Mortar X
rr1
16 L SFC Runnels, L. L. RA 7009241 WIA 1300 F /Mortar X
0
'"
en 17 L Sgt Ryan, ll. J. RA 16303418 IVIA 1300 F/Mortar X
.....
c
z
c-:>
,
:x:
C/)
C/)
..."
I"T'1
o
18 L PIc Throckmorton, R. K. RA 13305518 WIA
19 L PIc Zemar, Z. RA 15278564 WIA
20 L PIc Edwards, W. E. I\A 43056840 WIA
21 L PIc Hutcheson, J. L. RA 14334749 WIA
22 L Cpl O'Brien, James O. RA 13316676 WIA
23 L PIc Sundo, R. RA 13333835 WIA
24 L PIc Fairchild, L. G. RA 12324561 .WIA
25 L SFC Howell, Roy E. RA 13332277 WIA
26 L PIc Johnson, B. RA 15281721 WIA
27 L M/Sg' Matthews, E. C. RA 06842776 WIA
TABLE El <Continued)
1300 F/Mortar
1300 FI
1315 8/Small Anns
1330 F/Grenade
1330 F/Mortar
1330 FI
1400 F/Mortar
1400 F/Mortar
1400 F/Mortar
1400 F/Mortar
1400 B/Small Arms
1400 F/Mortf!!'
1400 F /Mortar
1500 F /Mortar
1500 F /Mortar
1500 F/Mortar
1500
B/Small Arms
F /Mortar
1500 Il/ltifle
1515 FIMo"ar
1600 F / Artillery
Order
No.
Co Rank Name Serial No. Type Cas Time
Type of missile/
causative weapnn
Source Cas datoS
Remarks
Cas, I IOther
Rept tag
X
X
X Time est. from account of this casualty by company
member.
X
X Cas data from clinical records.
X
X X
X X Type of fragment from Service Record.
X
X Subject interviewed at 1400. Wounded while in
smwl crater on NE slope of Knob 2 at around
1330.
X
X Attachment to Co L inferred from K number.
X Attachment to Co L inferred from K
X Data on type of missile from clinical records.
Also, subject interviewed. Contusion sustained
when subject knocked down by blast.
X X Causative agent from hospital record.
X
Subject wounded by small-arms fire. then killed by
mortar fire. Time est. from account in affir.lavits
made out after action by company members.
X
X
X
0
:ll
0
,
>-l
,
'"
....
'"
28
29
30
31
32
33
34
35
36
37
L PIc
L EM
L EM
L SFC
L PIc
L Cpl
L M/Sg'
L Pfc
L rIc
L Cpl
Spencer, R. R.
Atch ROKA soldier
Atch ROKA soldier
Clark, Il. S.
James, H. H.
Jones. J. E.
Mitchell, J. H.
Stallman, R. G.
McGary, n. M.
I30yd, R. A.
RA 24269094
K 1107171
K 1107185
RA 33663492
RA 14329111
RA 18294576
RA 20212607
ItA 13333554
RA 16310534
RA 18338645
WIA
WIA
WIA
WIA
WIA
WIA
KIA
WIA
WIA
WIA
0 38 L EM Atch ROKA soldier K 1107172 WIA 1600
-I-
X Attachment to Co L inferred from K number.
~ -I
39 L EM Atch ROKA soldier K 1107186 WIA 1600 X
I
>'l 40 L Sgt Nonog, A. F. RA 39392116 WIA 1700 F/Mortar X Casualty data from clinical records.
I
41 L SFC Hohn, B. F. RA 6964759 WIA 2000 F/Mortar X
'"
....
'" 42 L pre Buraczewski, W. RA 13316047 KIA FI- X Body picked up 14 Se'p.
43 L Cpl Coleman, R. F. RA 11176894 KIA
FI- X Body picked up 14 Sep.
44 L Cpl Contarro, J. E. RA 18211610 KIA
-I-
X Body picked up 14 Sep. GSW rt. side, rt. arm.
45 L pre Craine, B. W. RA 18294363 KIA
-I-
X Tag made out 14 Sep. GSW forehead.
46 L pre Davis, D. H. RA 15251758 KIA
-1
47 L pre Gentry, T. L. RA 14322994 KIA
-1-
Listed as casualty on 13 Sep. Arbitrarily included
in 12 Sep.
48 L Pvt Hanshaw, W. H. RA 13339743 KIA
FI- X Body picked up 14 Sepo
49 ,L pre Hunter, W. C. RA 15278465 KIA
-I-
X Body picked up 13 Sep. GSW, chest.
50 L Cpl Johnson, D. RA 16303078 WIA B/- X Casualty data from Separation Form.
51 L pre Lindsay, R. L. RA 15277958 KIA
-1
52 L Ret Mabry, C, W. RA 13330119 WIA FI_ X
53 L Cpl New, G. E. RA 12107216 \VIA
-1
54 L pre Owsley, B. F. RA 15251725 KIA
-I-
X Tag made out 14 Sep. GSW chest and abdomen.
55 L Cpl Palmer, H. P. RA 13332886 KIA
-1
56 L pre Prescott, Sidney T. RA 24522825 KIA X Shrapnel, lower extremities.
-I
57 L Cpl Smith, J. R. RA 14248885 KIA
-1
58 L SFC Taylor, J. H. RA 38276411 KIA
-I-
X Secondary tag says 14 Sep. GSW chest, rt. ann.
59 L pre Thompson, J. R. RA 14339428 KIA
-I-
X
60 L pre Trueblood, H. P. RA 13331852 -KIA F /Mortar X Body picked up on 13 Sep and tagged at 1215.
c::
Casualty listed as 13 Sep.
Z
C?
r-
61 L pre Wade, T. H. RA 13319792 KIA
-I-
X Tag, made out 13 Sep'-says casualty on 12 Sep.
Casualty records give date as 13 Sep.
:x.
V>
62 L Pvt Wortman RA 15277345 KIA
-I
V>
c::
.." 63 I 2d Lt Weston, Marion E. o 945876 WIA 1130 B/Rifle X Casualty data from clinical records. Date ~ v e n as
Z
,..., 10 Sep but evidence indicates 12 Sep. Bullet
C?
0
'" from .25-calJap. rifle. r
'"
:x.
'"
V>
V>
:!!
,...,
0
TABLE El (continued)
Order
No.
Co Rank Name Serial No. Type Cas ..Time
Type of missile/
f8usstive weapon
Source Cas data
R
Remarks
Ca., IMed 10th.,
Rept tag
64 1st Lt Baynes, Marvin N. 02017542 KIA 1215 F/Mortar Casualty data from eyewitness accounts. Killed
near Sheridan (No. 10, above) hy next mortar burst.
65 PIc Jo)J.nson, J. C. RA 18334046 WIA 1230 B/Small Ann. X
66 PIc Engle, C. R. HA 15291797 IVIA 1300 F/Mortar X
67 2d Lt Engle, Marshall G. o 970018 WIA 1300 F /Grenade X
68 Cpl LaRue, H. F. RA 13293756 WIA 1300 F/Mortar X
69 SFC Lyon, R. A. RA 18328284 IVIA 1300
FI_ X Casualty data from clinical records.
70 SFC Moncrief, C. RA 6391018 WIA 1300
B/Small Ann.
F/_
Casualty data from interview with subjecL
71 Pvt Ryde" J. E. US 57300205 WIA 1300 F/MortftI' X
72 SFC Williams, J. D. RA 20455456 WIA 1300 -I
X Casualty data from original tag. GSW. H, arm,
charging enemy.
73 EM Atch ROKA soldier K 1105187 WIA 1300 F/Mortar X
74 I EM Atch ROKA soldier K 1105198 WIA 1300 F/Mortar X
75 II Pvt Heaton, B. T. RA 14342915 WIA 1330 F/Mortar X
76 Cpl Meadows, T. R. RA 13332785 WIA 1350 -/ X Tags state: GSW, grazing, left shoulder.
77 Pvt Brain, G. W. RA 12348599 WIA 1400 F /Mortar X Casualty data from clinical records.
78 PIc Busho, W. G. RA 17259764 IVIA 1400 F/Mo"", X
79 PIc Colbert, T. J. RA 13273216 IVIA 1400 F/MortllI' X X Cas. report said GSW; 1700.
80 SFC Lancaster, N. B. RA 6274279 WIA 1400 F/_ X
81 Pvt Mauldin, C. RA 14324594 WIA 1400 B/Small Anns X
82 PIc Mulinex, O. E. RA 15277341 KIA 1400 F /Mortar X Secondary tag, time corrected from
e
::tl
e
,
~
,
'"
'"
....
83 Pvt Scott, F. E. RA 18333808 DOW 1400 F/Mortar X X
0400 because of specified location
(vicinity of Hill 660).
Timing from eyewitness report of
place of this casualty. Morning and
casualty reports give date as 13
Sep.
0 84 Pvt Smith, J. L. RA 16310496 WIA 1400 F/Mortar X
::tl
85 CpI Stanfield, P. L. RA 14327772 MA 1400 F/Mortar X
0
I
o-'J
86 Pvt Walker, R. F. RA 13332732 WIA 1400 F/Mortar X X Type of fragment from entry in Service
I
.., Record .
'"
PIc Woodall, C. A. US 57300189 WIA F/Mortar X X ... 87 1400
88 Pvt Vance, D. RA 14321609 MA 1400 B/Rine X
89 EM Atch ROKA soldier K 1105185 WIA 1400
-1-
GSW, left arm.
90 Pvt Cooley, R. G. RA 34925513 WIA 1430 F/Mortar X
91 CpI McCleery, C. T. RA 17253580 WIA 1430 F/Mortar X X Clinical record states subject lost
consciousness with blast, came to
further down hill, had been standing.
92 CpI Muniz, I. M. RA 17256257 WIA 1430 B/SmalI Anns X
93 PIc Huitt, C. E. RA 14326885 'WIA
1435 F/Mortar X
94 PIc Conrad, G. S. RA 11185692 WIA 1500
FI_ X
;
95 CpI Griggs, W. C. RA 17259940 WIA 1500
FI_ X
;
96 Sgt Mannel, J. RA 18303856 WIA 1500 F/Mortar X Casualty data from clinical records.
97 Pvt Wilson, J. C. RA 16322760 WIA 1500 F/Mortar X X Type of fragment from clinical records.
i

98 CpI Marsh, C. W. RA 14314495 WIA 1515 F/Mortar X Timing from fact that eyewitness (No.
- 99. below) srud he WflB hit at "Hmll"!
time as subject. Clinical record
states subject lost memory for four
days from blast.
111 ... ", .. Tn ;n._... '; .. .., .".h;...... "'."' ... .1 h ....... 0 I.:.
99 SFC ................ "'.... 14218227 WIA 1515 X ..................... ......... ..... - .. .... ..
on NE slope of Knob 2 during air
strike. Was standing. and blast
knocked him 10-12 ft down the hill.
c:
Z
100 PIc Alva. J. RA 17252439 WIA 1530 FI- X Morning report lists casualty as 13
,.., c::
Sep . other reports 12 Sep.
2:
r
e;-,
::x:- 101 1st Lt Fields, J. A. o 1338329 MA 1600 F /Mortar X Cas report gives time as 1400. In
CI)
r-
interview subject said he was hit"late

in action and at same time as Perez CI)
-T1 CI)
(No. 102).
rn
,.,
0
'" 102
. I 2d Lt Perez-Chiesa, J. o 13288739 WIA 1600 F X
'" rr1
'"
I::'
TABLE E 1 (continued)
Order
No.
Co Rank Name Serial No. Type Cas Time
Type of missile/
causative weapon
Source Cas data
8
Remarks
Cas:! Med IOther
Rept tag
103 CpI Sexton, S. A. HA 6899:?J)2 IVIA 1600 -I
X X Tag states GSW, charging enemy.
104 Cpl Subilosky, E. F. RA 11078434 WIA 1600
FI_
X Time from original tag, missile from
secondary tag.
105 Pic Whitlock, H. C. RA 14329836 IVIA 1600 F/Mortar X X Clinical record slates subject knocked
20-30 ft down the hill and uncon
scious for a -few minutes. "
106 MlSgt McCollum, C. E. RA 34912827 WIA 1645 F/Mortar X
107 Pvt Rogers, R. L. RA 14323420 DOW 1800 F/Mortar X X DOW 19 Sep.
108 EM Atch ROKA soldier K 1105175 \VIA 2030 F/Mortar X
109 Pvt Bicker, E. R. RA 28772718 \VIA F/Mortar X Tag made out at 1830.
110 Brown, B. O. RA 18341857 \VIA
-1
Letter from hospital says GSW left
ann, chest.
III Pic Burton, L. L. RA 19341803 KIA
-1
Records give date as 13 Sep. Casualty
arbitrarily included in 12 Sep list.
112 I Pic Cox, J. A. RA 14294761 KIA
-1
Casually listed for 13 Sep, arbitrarily
assigned here to 12 Sep.
113 CpI lIeuss, P. F. RA 16300143 KIA
-I
X Body picked up 14 Sep. Casualty
listed for 13 Sep, arbitrarily re
assigned here to 12 Sep. Tag
states GSW, back.
114 Pic Neue, C. T. RA 19347717 KIA
-1
Casualty listed for 13 Sep, arbitrarily
reassigned to 12 Sep.
115 Pic Upton, J. C. RA 23484708 WIA -/
0
el
116 Pic Whitson, R. N. RA 14346097 KIA
-1
Casualty listed for 13 Sep, arbitrarily
'reassigned to 12 Sep.
I
o-l
117 EM Atch ROKA soldier K 1105182 \VIA
-I X GSW rt. leg.
I
'""
'"
...
118 EM Atch ROKA ooldier K 1105197 KIA
-I
X Casualty listed for 13 Sep, arbitrarily
reassigned to 12 Sep .
0 119 K PIc Brockhaus, W. O. RA 16301378 WIA 1100
FI_ X
el 120 K Sgt Mayo. P. S. RA 36345487 W1A 1200 F /Mortar X
,
o-l 121 K EM Atch ROKA soldier K 1105164 W1A 1300 F /i\:lortar X
,
""
122 K Sgt McBrayer, T. T. RA 14347248 WIA 1400 B/Rifle X
.
'"
123 K PIc Sei bert, R. F. RA 17240232 WIA 1400 FI- X Original tag says grenade fragments,
secondary tag shell fragments. At
stated time Co not in position to
receive grenades.
124 K PIc Sacker, C. R. RA 18334368 WIA 1400
FI_ X
125 K CpI Davis, J. A. RA 14328961 W1A 1430 F /Mort8r X
126 K Sgt Greene, B. H. RA 34305738 WIA 1500 F /Mortar X In interviews an EM stated the sub-
ject was hit on Knob 1 at time of
move out, the Co CO said he was
hit on SE slope of Knob 2.
127 K SFC Gri tta, P. B: RA 06251755 WIA 1500 F (Mortar X X In interview Co CO said subject hit
on SE slope of Knob 2.
128 K SFC Ridings, R. F. RA 20456924 DOW 1515 F (Mortar X Casualty data (rom Service Record
and hospital r e p o r t ~ which states
GSW wounds penetrating chest and
shell fragments penetrating face,
chest, arid anna. In interview, Co
CO said subject was hit on NW
slope of Knob 2.
129 K Pvt Stephens. D. L. RA 19358991 KIA 1515 F /Mortar X All available records give date as 13
Sep, but EM eyewitnesses inter
viewed said he was hit by mortar
fire on SE slope of Knob 2 in 12
Sep action. Time estimated from
their accounts.
c:
Z
130 K PIc Vanasco, F. RA 13313997 W1A 1515 F /Mortar X Cause of casualty from Service
C"")
Record. EM eyewitnesses said he
r- c:
was hit on SE" slope of Knob 2.
:l>
Z
(/) Time assigned accordingly.
C"")
(/)
131 K Pvt Dowdy, R. L. RA 34088212 IVlA 1800 F /Mortar X EM eyewitness said he was hit right
r
:>
on top of Knob 2. C/)
rn '"
-1 (/)
"
""
0
:::::!
/"T1
0
c
z
c-:>
r
:J:>o
(/)
(/) TABLE E 1 <Continued)
'"
ex>
-n '"
rT1
0
Source Cas dataO
Order
Co Rank Name Serial No. Type Cas
No.
Time
Type oE missile/
Remarks
causative weapon
CaSt I IOther
Rept tag
132 K EM Atch ROKA soldier K 1105167 WIA 2000 -I-
X GSW neck, chest.
133 K Cpl Cauley, J. E. RA 13315748 WIA FI_ X
134 K PCc R;(fie. J. F. RA 15272262 WIA
FI_ X
135 K Cpl Shrum, 8. J. RA 18344046 WIA F/Mortar X X C&sualty report said only GSW. EM
eyewitness interviewed said sub
ject hit somewhere between Knobs
1 and 2 by mortar. This location
does not specify atime.
I
F/Mortar X Report stated mortar and
136 K PIc Warner, L. J. RA 13311664 WIA
B/- bullet wound right baod and left
leg, 13 Scpo Other reports give
date 12 Sep.
137 K EM Atch ROKA soldier K 1105137 \VIA
-I-
X
138 K EM Arch ROKA soldier K 1105140 KIA -I-
X Casualty listed as 13 Sep. Arbitrarily
reassigned to 12 Scpo
139 K EM Atch ROKA soldier K 1105146 WIA F/Mortar X
140 K EM Atch ROKA soldier K 1105148 KIA X Casualty listed as 13 Sep. Arbitrarily
-I
reassigned to 12 Sep.
141 K EM Atch HOKA soldier K 1105163 KIA
-I-
X Casualty listed as 13 Sep. Arbitrarily
reassigned to 12 Sep.
142 M Cpl Zunrga, A. RA 19212303 \VIA 1430 F /l\1ortar X
143 M Sgt Hook, R. J. RA 13232791 WIA 1600 F /Mortar X
0 144 M PIc Monroe, H. F. RA 16313261 WIA 1600 F /Mortar X
:u
0 145 M PIc Di ckson, R. E. RA 13321163 \VIA
-I
I
>-l 146 H3 EM Atch ROKA soldier K 1105117 \VIA 1400 F /Mortar X
I
'"
147 H3 EM Atch ROKA soldier K 1105135 WIA 1400 F /Mortar X
....
'"
0
148 H3 Sgt Hayes, W. F. RA 31190651 WlA 1430 F /Mortar X In interview subject said he was hit
~
around 1600.
I
>'I
I
149 H3 EM Atch ROKA soldier K 1105125 KIA -I X Casualty listed as 13 Sep.
reassigned to 12 Sep.
Arbitrarily
""
'"
....
ISO H3 EM Atch ROKA soldier K 1107181 KIA -I
X COSllalty listed as 13 Sep.
reassigned to 12 Sep.
Arbitrarily
BThe first column under this heading refers to individual casualty and intennent reports made out in the field and to individual reports from hospitals, the
second column to origiMI or supplemental medical tags, and the third column ("Other") to miscellaneous material in the "201" files.
c
:z c
C'? Z
r-
C'?
:x>- r-
C/)
:x>
C/) '"
'"
C/)
CO
C/)
'"T'1
....,
,.."
0 ,.."
0
UNCLASSIFIED
SEER!1
Appendix F
ACTION NORTH OF TABU-DONG IN AUGUST 1950
361
ORO-T-261
UNCLASSIFIED
SUR"
THE ACTION NORTH OF TABU.OONQ IN AUGUST 19f10 AS VIEWED FROM BOTH SIDES
Sources: Friendly IIl.de: AfleracUon Report, 27th RCT,n paraphrased
Enemy side: Interrogation Report, S3, 19th NK Regiment (Item 1, Annex A), parapbrased
@
182000 Aug
o
I

I
CD
""
....
190230 Aug
20AIlg
202130 Aug
21074f1 Aug
Date Action of the 27th RCT Date Action of the 13th NK Dlv Commentll
After preparatory arllliery and tank fires, enemy attacks, tanks preceding
an estimated Infantry battalion In carriers or trucks. Fire of 8th FA Bn
halts advance. Enemy infantry continues attacking until about 2130,
attemptln& to penetrate positiOnS of 1st and 2d Bns, 27th RCT astride
road. Three enemy tanks supporting the Infantry reacb friendly forward
positions and are destroyed hy closeln artillery fire and bazooka fire of
Co F. Enemy withdraws.
An unknown number of enemy supported by two tanks probe positlono of
2d Bn, 27th Inf. Enemy force dispersed with artillery, mortar, and
small-arms fire.
The battalions of the 27th RCT Improve their poIIWons while friendly
air and artillery hit the enemy positions. These fires are thoogbt to
have prevented an enemy build-lip north of the 27th R,CT.
The enemy attempts to penetrate the friendly MLR with an estimated
battalion of Infantry and an Ilnknown nllmber of tanks IUld selfpropelied
gues. The enemy armOr Is used for Shock action, firing rapidly while
the Infantry deploys for the attack. Artillery and mortar fire Is placed
on the enemy. Friendly Infantry holds fire with small arms IlnUl the
enemy has approacbed to easy firing range. The tanks of Co C, 73d Tk
Bn attached to the 1st Bn, 21th 1m engage enemy at aoort range,
bu.t no hits are scored by either Side.
A limited objective attack Is launched by a platoon from Co B, a platoon
from Co G, and a section of tanks of Co C, 73d TIt Bn In order to deter
mine enemy losses and deceive the enemy as to friendly intenllons.
This force advances approximately fifteen hllDdred yardo to the north
and then wltbdraws on order when enemy SPguns open fire. In this
acllon the following materiel and personnel were destroyed or observed
already destroyed: one disabled tank with crew, five disabled tanks
{destroyed, apparently now, with thermlte grenades}, two SP guns, one
37-mm AT gun with crew, one 120mm mortar, and twenty troops.
The erterPY shells the entire 27th RCT area until
The enemy attacks positions of the 1st and 2d Bns, 27th RCT with an
estimated battallon and nine tanks. The RCT disrupts the attack with
coordinated fires from artillery, mortars, small arms, and tanks. Three
enemy tanks are destroyed JIlSt forward of the MLR, two by tank fire and
one by a 3.5ln. bazooka rd fired by Co F. At dayllgllt It Is Observed that
siX additlonal.tanks have been destroyed, bringing the total for the period
17-22 August \0 eleven. A tol\\l of six SP and two trucko have been
destroyed. [The reporting of these claims Is mllddled.]
Tactical air hits the enemy north and east of the 27th RCT. Observers
and POWs report that the enemy 10 withdrawing, evldcntly to redeploy.
The enemy probes the front with grOllps estimated 10 be one hllndred men
In apparent attempt to locate friendly pooltlons. The 27th lot holds fire.
Enemy artillery and mortar fire during the night Is Intcrmlttently
Silenced by the fires of the 8th and 37th FA Bns.
The enemy attacks with an estimated two companies of infantry and an
unknown number of tanks, alter artillery and SP-f1re Is concentrated on
Companies Band C 01 the 27th RCT. The attack is repllJsed by blocking
the enemy rear with mortar and artlllery fire and thcn drawing the enemy
troopo back Into these concentrations with smaUarDls fire. Enemy forces
Withdraw at 0200 w!thollt penetrating. Two were enemy tanks are deotroyed
by artillery. ToW'losses of enemy armor In the period 17-25 August are
13 tanks and 6 Sp.guns. Many tfllcks and smaller vehicles destroyed.
Total enemy casualties from all callses are estimated to nllmber 3000.
The lst and 2d Bno, 27th RCT are re1elved by lhe lS1 ROK Dl.v.
180flOO Aug
190001 Allg
20 Allg
212000 Aug
24 Aug
The 19th, 210t and 23d Regts of the 13th NKDlv engage the ROK Army
approXimately eight miles northweot of the road jllnetion north of
Tabv.dong and In heavy fighting lasting until 2100 advance to the vlclrilty
of the road jllnellon.
The 21st and 23<;1 NK Reg!s engage UN forces with the 19th NK Regt In
reserve. The sklrmlsb endS at 0400 with losses totaling approxlmately
one hllndred men. At this time, the 13th NK Dlv deployo 1 km south of
the road junction with the 23d Reg! on the blUo on the west olde of the ,
road, the 210t Reg! on the hUls on the eaot, and the 19th Reg! In reserve
at HaJangdong with the diviSional artillery.
The 23d Reg! remains in poslUon, wi the Ust Reg! lotns It on weot
side oj road because of Intense sbelling. The litll Reg! moves up to
location 1 km east of the road Junction.
From the positions of the previollS day, the 13th NK Dlv lallnches an
offenSive with Tabudong the objective. All Ilnlts except the 19th Regt
jump off at 2000. The attack Is SIlpported by five T34/85 tanks, of whlcb
four are lost. The 19th Regt, whose objective was to neutflllize artillery
positions 4 km north-northeast of Tabudong [thiS was the pooilion of
the friendly front line; US artillery was back In the vicinity of Tabll.dong],
jumped off 90 min late, at 2130. Its 2d Bn had becn disorganized by
friendly sbetling, and Its CO (the prisoner himself) had only assumed
command that day. The delay of the 19th Regt was responoible for the
disorganization 01 the dlvillion's attack. Enemy units were forced to fall
back to their original positions.
The 13th NK Div goes on defensive.
Friendly artlliery evidently
did Indeed dlstllrb and hold
off the enemy.
The enemy source need not
havc known abolll this
enemy attack, since his
regiment was In reserve.
Attack not confirmed by
reports of artillery and
tank battalions. Probably
a patrol, with SP guns
participating.
c:
z
('J
r

C/.)
C/.)
..."
I'T'l
212035 Aug
220137 Aug
23 ,,"ug
23_24 Aug
24_2f1 Allg
250100 Aug
280350 Aug
o
UNCLASSIFIED
mltf
Appendix G
ENEMY ARMOR
CoNTENTS
TABLES
Gl. DATA ON ENEMY ARMOR NORTH OF TABU-DONG CLAIMED
DESTROYED BY CO C. 73D TK BN
G2. KNOCKED-OUT ENEMY TANKS FOUND AFTER THE BATTLE
Page
367
368

365
UNCLASSIFIED
TABLE Gl
DATA ON ENEMY ARMOR NORTH OF TABU-DONG CLAIMED DESTROYED BY
COMPANY C, 730 TANK BATTALION
2
'
Dor Time Location Vehicle
Designation in
Table G2
and Fig. 9 Circumstances
Range of
friendly fire
Type of
ammunition Location of hits Result of hits
18 Aug 1730 bra Vicinity S o i ~ r i T34/85 1/7 Enemy tank advililcing and 500 yd HE & HYAP Bottom of gun shield; Po'rtion of gun shield
firing 011 troops right front side blown off, humed
19 Aug Chonpyong-dong 5U-76 NIH Gun had heen disabled by 600 yd HE Inside of vehicle Vehicle burned
US AT mille
1'1'1
""
T34/85 P/13 Tank parked in schoolyard, 40 yd HE&HVAP Front slope above Penetration. disabled
~
camoun,ed driver's hatch
""
T34/85 0/12 Tank parked in schoolyard, 30 yd HE & HVAP Right side below turret Penetration, disabled
21 Aug 1030 bra Chonpyong-dong
camouflaged I
5U-76 MIlO Gun in pass neaf blown-out 75 yd HVAP Front slope left of gun PeoctrntioD. burned
bridge
T34/85 HVAP&APC Front slope and driver's Penetration, burned
batch (picture of Jl6
H]
{sr
d
Tanka advancing toward us,
confinns such damage]
speed 15 mph
T34/B5 H/S 75 yd HVAP Front slope and driver's Penetration; crew
22 Aug 0330 hrs Vicinity $oi-ri
hatch abandoned
5U-76 K/a Gun advancing in column 100 yd APe & WP Left front side [picture Penetration, burned
c:::
:z
behind preceding two tanks of K/B confinns such
damage]
C"')
r
:x:- c:::
C/)
z
C/) .., C"')
0> r
"
...,
:x:
rrI C/)
0 C/)
"
rrI
0
c:
:z:
,
C'?
:t>
U>
U>
..,
."
0> f'T'1
co
TABLE G2
o
KNOCKED-OUT ENEMY TANKS FOUND AFTER THE BATTLE
Evaluated cause
Sunrey No.1, 27 Sep SO, Survey No.5, 11 Oct 50,
Remarks
designation and description designation and description Survey 1 Survey 5
A SU 76. Has 60 litera of gaa
[equivalent to 7 miles of travell;
1 rd left in gUD; hzeech gone; a
few small-arms penetrations,
Done vit'll; gun type 42Mi had
track knocked off by gun fire;
repaired by US; final drive
inoperative.
G SU-76. Direct bit by SOO-lb bomb;
completely demolished.
H T34/85. Burned; turret blown UPi
muzde split; combustion; credit
to air-ground is questioned.
~
o
J T34/85. Burned; holes in front
I annOfj lll1.illery (somewhat in
'"'I
question),
I
0>
""
.....
SP gun. NK. 8 miles N starting point in
Taegu. Vehicle all right side of road
near house. Damage: Done. Means:
none. Remarks: IlPparentiy abandoned
intact.
3 SP gun, NK. 1 mile N of No.1 vehicle in
draw on left side of road; gun covering road
e.t 5O-yd ~ e . n g e . Da.onae""e: hurned i=:l3ide,
ammo exploded; engine blown up. Means:
exact cause not detennined but artillery
fire probably caused damage.
4 SP gun, NK. 1.6 miles N of No: 3 vehicle on
left side of road in ditch. Damage: vebicle
completely demolished and strewn around
in several pieces at sides of road; gun and
cWTiage blown off; left side of vehicle
blown away. Means: artillery, direct bit of
HE round.
5 1'34/85. 2.1 miles N of No.4 vehicle on left
side of road. Damage: gun tube split at muz
zle; tank burned inside; ammo blown up;
gun shield damaged. Means: rocket, 3.5;
vehicle apparently abandoned and destroyed
by NK forces.
6 T34/85. 2.1 miles N of No.4 vehicle on
rigbt side of road. Damage: one peoetra
tion 00 gun shield and three in front slope;
tank burned; ammo blown up inside. Means:
90-mm AP.
Number -1324- stenciled on tUlTet.
Facing south.
"27tb--(no more visible) painted on
turret. Tnnk presumably knocked
out by 27th RCT on night 21-22
August. Tank lI.od gun facing
south.
Number "328- stenciled on turret.
Tank and gun facing south.
Tank considered knocked out on
night 21-22 August by Co C, 73d
1'11: On, but possible by Co I, 8th
CII.V 00 4 September.
Sure, Arty
Probable, bomb
Probable, abandoned
Unknown, Arty (?)
Sure. Ilhllndoncd
Probable, Arty
Sure, Arty
Sure, ahllndoned litotes
lilly bllzooka original
cause]
Sure, tank 9O-mm
T34/85. Demolished; probably hit
by air (in question)'
K SU76. Side penetration by flat
trajectory; an old sumple; credit
probably to tanks or artillery;
Borne fire.
L T34/85. Badly burned; probably
hit by air rocket.
M SU76. Side penetration; flat
trajectory; but by artillery or
tank; blew nut back; fire on ru.b
ber bogies only.
N SU-76. Probably gun fire; several
large-caliber boles.
o T34/85. Questionable; burned;
penetrated at all angles; tUITet
penetrated underneath; internal
explosion; credit plus grand
(sic). or air.
c
z
("")
P T34/85. Multiple penetrations;
,
hut on right aide; burned; could'
:l>
have been bazooka, artillery, or
en
air rocket, cauae in question.
en
""T'1
7 T34/85. 2.1 miles N of No.4 vehicle on left
side of road; 15 yd from No.6 lind 10 yd to
rear of No.5 [8]. Damage: tun-et off; pene
tration of left front slope; tank burned and
ammo blown up. Means: 9O-mm AP.
8 SP gun, NK. 2.1 miles N of No.4 vehicle on
left side of road 50 yd north of No.7.
Damage: penetrated hull at left front; tank
burned and anuno bltwn up; completely
destroyed. Means: 3.5 rocket.
9 T34/85. 500 yd N of No.8, 10 yd off left
side road at small. bridge. Damage: road
wheels burned; both tracks off; ammo ex
ploded inside; two cal-50 holes through gun
tube. Means: napalm and strafing.
Remarks: napalm tank found 10 -yd from
vehicle; exterior burning; indicates napalm
was cause of damage.
10 SP gun, NK. 50 yd N of No.9 on right side
road. Damage: penetration of left side of
fighting compartment; ammo exploded and
burned tank. Means: 9O-mm AP.
11 SP gun, NK. 50 yd N No. 10 lying upside
down on right side of road. Damage: pene
trations on left side of gun and through gun
shield; rounds passed cOiIlpletely through
vehicles. Means; 9O-mm AP.
12 T34/85. 75 yd N No. 11 on left side of road
in villsge. Damage: vehicle completely
burned inside and oul; tUlTet cracked at left
front, under gun shield; transmission had
been removed prior to time vehicle burned.
Means: napalm and strafing: Remarks:
napalm tank near vehicle; vehicle was prob
ably under repair at time it was destroyed.
13 T34/85. In courtyard on right side of road.
Damage: penetration of lelt rear comer at
hull, through engine comp.artment, into
fighting compartment; tank burned and ammo
blown up. Means: 90-mm AP.
Legend "C co 73 6.- painted upside
down on turrct. Tank preswnably
knocked out by Co C, 73d Tk Sn
on night 18-19 August and later
pushed or blown off road.
Tank facing south.
Facing south. Considered knocked
out on night 21-22 August.
Number '"308- stenciled on turret.
Tank and gun facing north. Thia
tank cannot be identified in reports
by friendly infantry or annor.
Number '"13--(partly obliterated)
stenciled on tWTet. Facing south.
Original cause of damage unknown.
Considered to have been furtber
damaged by short-range fire of
friendly tanks on 21 August.
Facing north. Considered to have
been immobilized by US mine and
destroyed by round ired by friendly
tank on 19 August.
Tank and gun facing aouth. Original
canae of damage unknown. Probably
further damaged by round fired at
short range by friendly tank on
21 August.
Number -320- stenciled on turret.
Original cause 01 damage unknown.
Further damaged by round fired at
short range by friendly tank on
21 August.
Unknown, air (?)
Probable, tank
Probable, rocket or
Arty
Probable, Arty or
tank
Probable, Arty or
tank
Unknown, multiple
Unknown, /Ilultiple
Sure, tank 9O-mm
Sure, bazooka, 3 . 5 - i ~
Sure, napalm
Sure, tank 9:O-mm
Sure. tank 9O-mm
Sure, tank 9O-mm
Sure, napalm [notes
say 90-mm APJ
rn
o
c:::
:z
..
C"'>
::t:
V)
V)
"'T1
rn
c
TABLE G2 (continued)
l!:valuated cause
Survey No. I, 27 Sep 50,
.
Survey No.5, 11 Oct 50,
Remarks
designation and description designation and description SUr'Yey 1 I Survey 5
Q SU-76. Right side hlown out;
Borne burn on blasted side;
cause unknown.
R SU_76. Hole in middle of right
side; no burning; gUD removed;
credit to artillery Of bazooka;
jagged bole.
S SU-76. ~ I b bomb holc; three
small strafing boles; probably
mechanical failure, or OVelTUDj
no other or fire.
T SU.76. Burned-out off road;
internal explosion; definitely
air, probably rocket.
u
V 1'34/85. Thue tanka up the valley
from Osangj all bit by napalm
w
@
o
I
>-l
I
en
'"
.....
14 SP gun, NK. Overturned on left Side of road,
75 yd N of vehicle No. 13. Damage: vehi
cle completely demolisbed; hlown into aev
eral pieces. Means: mortar or artillery.
Remarks: mortar or artillery; cratera near
vehicle indicate vehicle was destroyed by
one or the other.
15 SP gun, NK. In ditch on left side road.
Damage: gun removed, right track oU;
ammo removed. Remarks: vehicle
apparently abandoned.
16 SP gun, NK. On right edge of road.
Damage: near end used as ahutment ror
roadbed by US forces; no damage evident.
Meana: apparently nbandoned.
Tanle faCing north, gun facing south.
Possibly destroyed by friendly
iU1il1ery on evening of 21 August..
Facing road. Pouibly destroyed by
friendly artillery on evening of
21 August
Number "1326- stenciled on turreL
Unknown. multiple
Unknown, multiple
Probable, oven-un
Probable, rocket
Sure, napalm
Sure, napalm
Sure. napalm
Probable, Mort or Arty
Sure, abandoned
Sure, abandoned