Sedimentary Geology, 21 (1978) 85--127 85

© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands

THE SILURO-DEVONIAN CLASTIC WEDGE OF SOMERSET ISLAND,

ARCTIC CANADA, AND SOME REGIONAL PALEOGEOGRAPHIC
IMPLICATIONS

ANDREW D. MIALL and MARTIN R. GIBLING

Geological Survey of Canada, Calgary, Alta. T2L 2.4 7 (Canada)
Geology Department, University o f Ottawa, Ottawa, Ont. K I N 6N5 (Canada)
(Received June 22, 1977; revised and accepted November 11, 1977)

ABSTRACT

Miall, A.D. and Gibling, M.R., 1978. The Siluro-Devonian clastic wedge of Somerset
Island, Arctic Canada, and some regional paleogeographic implications. Sediment.
Geol., 21: 85--127.

Upper Silurian and Lower Devonian rocks on Somerset Island were derived from
Boothia Uplift, a linear tectonic belt that underwent a pronounced period of positive
movement commencing in the Pridolian. The lower part of the clastic wedge is a succes-
sion of predominantly intertidal to supratidal dolomite and siltstone 150--400 m thick
(Somerset Island Formation). Markov analysis documents the presence of tidal cycles
in these rocks. The succession changes eastward, away from the uplift, into subtidal
marine limestone indistinguishable from that of the underlying Read Bay Formation. The
Somerset Island Formation grades vertically and laterally westward into alluvial deposits
o f the Peel Sound Formation, which consists o f red sandstone o f braided river and possi-
bly eolian origin, and two fanglomerate members. A variety of fluvial fining-upward and
thinning-upward cycle types has been documented by Markov analysis of a lower sand
member of the formation. The Peel Sound reaches a maximum thickness of 600 m in
northern Somerset Island.
Paleocurrent analysis o f crossbedding indicates eastward transport directions in the
fluvial rocks, except for cosets of very large scale crossbeds (up to 6 m thick) in the
northwestern part of the island, which indicate northwesterly flow. The large sets are
interpreted as the deposits of eolian dunes, or o f sand waves in a large trunk river which
carried clastic detritus northward, parallel to Boothia Uplift.
The Peel Sound Formation and its lateral facies equivalents on Prince of Wales Island,
west o f Boothia Uplift, contain coarser conglomerates and a higher sandstone/carbonate
ratio, indicating deposition under higher energy conditions than are thought to have pre-
vailed in Somerset Island. Relief may have been greater and depositional slope steeper in
the west, an asymmetry in Boothia Uplift that is reflected in the present-day structural
style of a narrow zone of tilting and reverse faulting in the west, and a broad zone of
gentle folding and normal faulting in the east. The similarity in structural and strati-
graphic asymmetry indicates a genetic link, which is further suggested by the presence of
syndepositional folds and unconformities at a few localities within the clastic wedge.
An estimate of the volume of sediment removed from Boothia Uplift indicates that
approximately one third can be accounted for in the present clastic wedge on either side of
86

t h e Uplift. Either t h e L o w e r Paleozoic f o r m a t i o n s were a t t e n u a t e d over t h e Uplift, or

rivers such as t h e h y p o t h e t i c a l t r u n k river were effective in t h e dispersal o f material
b e y o n d t h e region o f t h e clastic wedge.

INTRODUCTION

On Somerset and Prince of Wales Islands (Fig. 1) a thick succession of pre-

dominantly continental clastic sediments of Late Silurian and Early Devonian
ages forms the youngest part of a structurally conformable Cambrian to
Devonian sequence. The rocks are similar in many respects to the Old Red
Sandstone of eastern North America and northwestern Europe, much of
which was deposited in intermontane and marginal basins following closure
of the proto-Atlantic Ocean (Allen et al., 1967; Friend, 1969; Dineley,
1975). In the present case the sediments represent deposition on an alluvial
plain and adjacent tidal flats which developed on either side of a rising
mountain range within the craton. The present expression of the mountain
range is a structural belt named Boothia Uplift {Fig. 1), which contains a
core of Aphebian igneous and metamorphic rocks and is flanked by
upturned, folded and faulted younger Proterozoic and lower Paleozoic sedi-
ments. The genesis of Boothia Uplift is not yet fully understood. It has been
shown that movement t o o k place along it at several times during the Protero-
zoic and early Paleozoic (Kerr and Christie, 1965; Brown et al., 1969; Kerr
and de Vries, 1976; Miall and Kerr, 1977; Kerr, 1977), but the most pro-
nounced period of movement probably was that which occurred during the
latest Silurian and earliest Devonian, during which the rocks described in this
paper were deposited.
Thorsteinsson and Tozer (1963) applied the name Peel Sound Forma-
tion to the conglomerate, sandstone and shale which comprise the clastic
wedge. Miall et al. {1977) recognized an additional map-unit, the Somerset
Island Formation, which forms a transition sequence of interbedded carbon-
ate and fine clastic rocks between the entirely continental Peel Sound For-
mation and the underlying marine limestone and dolomite of the Read Bay
Formation.
The sedimentology of the clastic wedge of Somerset Island was discussed
briefly by Dineley (1966) and Brown et al. {1969) and that on Prince of
Wales Island was described by Miall (1969, 1970a, b, 1973). Detailed work
on the outcrops on Somerset Island was carried out b y M.R. Gibling (Univer-
sity of Ottawa) in 1973, 1974, and 1976 (Miall et al., 1977 and unpublished
data), and by parties from the Geological Survey of Canada in 1975 and
1976 {Reinson et al., 1976; Miall and Kerr, 1977), as part of a regional map-
ping project.
The purpose of the present paper is to describe the broad regional frame-
work of the clastic wedge in Somerset Island, based on mapping data, and to
analyze facies sequences and associations using the technique of Markov
 87

55
Cape Z~nne

74

Aston Boy

> -ZZ _;

"~Peel Sound Fm.

}-E-_--- ~ _

Somerset I. Fm.

I I
Cambrian -Silurian

NN.
Precombrion

•J'..•.
C r eBs w e l l 8 -
ay

4 "".. ? I I I
5 0 km
I

3 t~

Devon
~nlson ~.- "~_1. }
N-12 o

72

8oothia

Fig. 1. L o c a t i o n o f p r o j e c t area, s h o w i n g d i s t r i b u t i o n o f S o m e r s e t Island a n d Peel S o u n d

F o r m a t i o n s a n d p r i n c i p a l o u t c r o p sections.
88

chain analysis. A new technique, the contact matrix, is introduced in order

to supplement the Markov method. The techniques facilitate an interpreta-
tion of the depositional environments and lead to some comparisons and
contrasts with other parts of the clastic wedge, in Prince of Wales Island and
areas to the north. Work in progress by M.R. Gibling will cover other aspects
of the sedimentology in greater detail, including petrography and geochem-
istry, the depositional environments of the conglomerate beds and the
ecology of the abundant fossil fish.

STRATIGRAPHY

Description of formations

A discussion of the clastic wedge in Somerset Island requires consideration

of three formations, the Read Bay, Somerset Island and Peel Sound, in
ascending order. Contacts between these units are gradational. The Read
Bay Formation, as recently redefined on Somerset Island (Miall and Kerr,
1977), consists predominantly of nodular, mottled limestone, with subordi-
nate dolomite. The Read Bay Formation contains a rich fauna of brachio-
pods, trilobites, corals and stromatoporoids, and is marine in origin. It ranges
in thickness from 170 to 240 m.
The Somerset Island Formation (Miall et al., 1977) consists of inter-
bedded dolomite, limestone, dolomitic siltstone, shale and subordinate sand-
stone. Two members are distinguished: a lower, predominantly carbonate
member, and an upper member containing a more variable succession,
including red siltstone and red dolosiltite (Fig. 2). The contact between the
two members is drawn at the first appearance of abundant red siltstone. The
position of this contact varies from west to east across Somerset Island
because of lateral facies changes (siltstone units die out eastward), as dis-
cussed below. The Somerset Island Formation ranges in thickness from 150
to more than 400 m, the thickest successions occurring near the north--
south centre line of the present area of outcrop (Fig. 1). The formation is
the youngest present in this area, (apart from minor Cretaceous--Tertiary
outliers) and an unknown (possibly considerable) thickness has been
removed by erosion.
The Peel Sound Formation has been redefined (Miall and Kerr, 1977) so
that the first prominent red, crossbedded sandstone unit constitutes the base
of the formation. The Peel Sound consists mainly of sandstone and con-
glomerate with lesser amounts of siltstone and mudstone. It reaches a maxi-
m u m thickness of 600 m in the Cape Anne area of northern Somerset
Island. In that area the succession has been divided into four members, based
on the work of Dineley (1966) and Brown et al. (1969):
(4) Polymict conglomerate 120 m
(3) Pebbly sandstone, conglomerate 240 m
(2) Dolomitic conglomerate 280 m
 89

150-
If :l../ -1
m section continued below
[i/i/i
300- J • gray
m
L : ;: iF laminated dolomite

SDsi2
F~ _ ~ gray
limestone

~) l: .i ;/:1 /.../ / =
~ red
silty mudstone
/ / ./
I.I o
~red
siltstone, doiosiitite
I00-
iiii I .t
~ red
-Z--/-@-L o ~, sandstone
250- / .... /:./ ..,+,

'--1 - - / -
/ / /
o ~ / ~ groy/red
/ /
thin bedded
l { dolomite/siftstone

C,$ / / /
ripple marks

N:
parting lineation

~'eeding trails

B
200-
/ / /
0 v
.._../_._i.?i__ B brochiopods
v i # # C corals
f { F fish
L__'__/.%. G gastropods
/ / 0 ostracodee
S stromatoporoids
bioclastic

~
, . sandy
0

8,C

Fig. 2. Stratigraphic section at the t y p e locality o f the S o m e r s e t Island F o r m a t i o n (Sta-

t i o n 6, Fig. 1). SDsi2 = t o p o f upper m e m b e r o f S o m e r s e t Island F o r m a t i o n , S D s i l = t o p
o f lower m e m b e r , Sr = t o p o f R e a d Bay F o r m a t i o n . ( F r o m Miall et al., 1977.)
90

(1) Sandstone, siltstone, mudstone 6 0 - - 4 0 0 m

These members are in part lateral equivalents of one another, as discussed
below.

Age

A variety of vertebrate and invertebrate material has been used to date the
three formations (Broad, 1968, 1973; Broad and Dineley, 1973; Jones, 1974;
T.T. Uyeno, personal communication, 1976). The results were summarized
by Miall et al. (1977) from which Fig. 3 is taken. This diagram also shows
correlations with the comparable successions in other parts of the Boothia
Uplift region. Although marked lateral facies changes can be mapped in these
rocks, in most cases biostratigraphic data are not yet precise enough to pro-
vide independent confirmation of the diachronism; and formation contacts
are, therefore, shown as approximate time lines. Jones and Dixon (1977)
showed that the Read Bay--Somerset Island contact becomes younger east-
ward away from Boothia Uplift, and probably is late Pridolian in north-
eastern Somerset Island.

RUSSELL I. I PRINCEWALESOF
W E SI TSOMERSET
ERNI. I. CORNWALLIS I.

Siegenian

Snowblind Bay
Upper Fm
Gedinnian Member
D Mbr
Upper Mbr I Peel
I Sound Fm
Pridolian Lower ; Lower Mbr Somerset I- ~ C Mbr
Member Fm

Ludlovian Read Bay Fm I ~eod Boy Fm t I zx B Mbr

Wenlockian " -I - - Cape Storm

/ Fm - -

Fig. 3. Age and correlation of the Somerset Island and Peel Sound Formations. (From
Miall et al., 1977.)
 91

Regional stratigraphic variations

Fig. 4 shows two reconstructed stratigraphic cross-sections through the

clastic wedge in Somerset Island. Facies variations within the Peel Sound
Formation are shown in the section through the Creswell Bay area (Fig. 4B)
but are omitted from the other section. They are given in Fig. 5 which illus-
trates the distribution of the four members of the Peel Sound Formation in
northern Somerset Island.

iiiiii

0 km 5
~.',-.~,.~.°~..~.: ~ . * I O0
2 C.¢b+4~.PEEL4 +. I
~:~ a:~T+:SOUN[ o m
; +" o~,~v~*
+o:,.o,I;..... +6 ~.~..:::::
+'..~.~:,~'qrO~i~.-.~ :,~. .:..'..
• • •
• I
~' : . .. .:. .e~ 0.... U. ,Q' + :.. . . ~• ..:' .'. '.. ... :". : . :.. :.- :.. .: -.: ..: ... .' . . . : . : . : .• : ..: . : . : . : . . . . L
0
- O 0 ~ . O. ! ] ~ . : . :
..~,.,y..:.:
' PEEL SOUND FM ' . : . : ~ -- _-~-_-_---_ :_ -- --- ---:- --- :_-- -_- --[ _ - + -
.:.:.......+...............:...+ _
~ i i ::: +o',,as.~,+e':~ ~ - - - _ - : _ - - ~ - - _ r--_~__

! ~ I , I I I I
.............
,
.~.:~ ,-.,.,., , , ,
1 , 1 , 1 , 1
, ,
. .
..........
. . . . . . . . . .
,,,,
.

Fig. 4. Generalized cross-sections through the clastic w e d g e o f S o m e r s e t Island, s h o w i n g

lateral facies changes. L o c a t i o n s o f cross-sections are given in Fig. 1. ( F r o m Miall and
Kerr, 1 9 ' 7 7 . )
 92

// L! SOs,
-]
l,<:.~:~:.:,..~..-, ~ ~"'~ . i ~.
/
oo \ J :./~,',',*,',*° . • . . . . .
°BpZo f ~,,.:::::::::: : ",,::::,~:: .:
. . . . .
" :
~::.,..=========================== : .: - - ~ ,
.:°:.:.:..~.: - : -....
Pre : : : : : : : : : : : : : : : : : : :~oO%o
: : : : :.-: :~1~:.:.:.:.:.:-...:° :: :
Poil ° °°°% ' D. . P i ~::::::'.
...:~:"
.':':::::::::::~.
,-',:,:,:,:,:,2.: :
" :
• :
: ": "

°~" ° v "." :: ii :: ::: ::. :: :

} ~ ~ ~ : ~ ; : : : : : "*fie " . .... . ::. . : . . .: . :...

. /,
/ ~ ~ ~ X ~ ~° °o ~~' ~ 7 ~ 1 . : ~ : . • . . • z 11

o°o°o°o Oo°2,~,~-:~7:. : :: " • . . . . : • / ,

.::oO°O°o°O~.'=-.'~': :: ::: : ., : . : . )~-~ <i

Aston Boy X 0 I0 km
I I I i i

Dp4 - 400 m

--sea lev~

SDsi

- 40On-

Fig. 5. D i s t r i b u t i o n o f t h e f o u r m e m b e r s o f t h e P e e l S o u n d F o r m a t i o n ( n u m b e r e d Dpl to
Dp4) in n o r t h w e s t e r n S o m e r s e t Island. ( F r o m Miall a n d Kerr, 1 9 7 7 . )

It can be seen in both illustrations that the clastic wedge comprises a

broad, coarsening-upward succession, and that each facies association (as
represented by formation members) spreads gradually eastward, away from
the Boothia Uplift. The lower member of the Somerset Island Formation
(mainly dolomite) generally is thin in the west {Stations 5, 9), reaches its
 93

thickest and most typical development on central Somerset Island (Stations

7, 12) and, towards the east, contains interbedded units of mottled, nodular
limestone of Read Bay type (Stations 2, 8). East of the areas represented by
Figs. 4A and B limestone predominates in this member. The upper m e m b e r
of the Somerset Island Formation and the lowermost m e m b e r of the Peel
Sound Formation maintain similar lithofacies characteristics across the
island, b u t vary in thickness.
Mapping indicates that in the Cape Anne area there are rapid lateral facies
changes within the Peel Sound Formation (Fig. 5). The lower conglomerate
m e m b e r (Dp2) grades eastward, in part, into the sandstone of member 1 and
the pebbly and conglomeratic beds of m e m b e r 3. At any given level within
the succession the lateral transition takes place over a distance of less than
10 km and, in some instances, possibly as little as 1 to 2 km. The youngest
m e m b e r of the Peel Sound Formation (Dp4: polymict conglomerate) is in
part younger than, and in part a lateral facies variant of m e m b e r 3. However,
westward, towards the Boothia Uplift, it appears to cut down across older
beds to rest with a local internal unconformity on the dolomitic conglome-
rate of m e m b e r 2. This relationship cannot be demonstrated elsewhere in
Somerset Island. Exposures in the Creswell Bay region (Station 4) show
polymict conglomerate (Dp4) resting directly and probably conformably on
sandstone (Dpl).

FACIES ANALYSIS

Facies spectrum

The sediment of the clastic wedge can be divided into twelve principal
lithofacies. They are described below, and each is assigned a two-part code
for convenience. The code consists of a prefix for rock type and a suffix for
dominant internal structure. Prefixes are as follows:
G (= gravel) pebble, cobble and boulder conglomerate
S sand
F (= fine-grained) silt and mud, c o m m o n l y dolomitic
C (= carbonate) limestone, dolomite
The facies can be arranged in a spectrum (Fig. 6) which corresponds to the
order (upward) of increasing non-marine influence and increasing fluvial
energy, and also corresponds closely to the order of appearance of the facies
within the clastic wedge. This serves to emphasize the gradational nature of
stratigraphic variation within the wedge, and the absence of major erosional
breaks or marine transgressions within it.

Facies descriptions

Agr~llaceous limestone (Ca). This lithofacies consists of relatively hard

lumps of micritic limestone embedded in a matrix of argillaceous, in part
94

Gm conglomerate (polymict)
member 4
Sr ss with ripple marks
Sh laminated sandstone
PEEL
member 2
Gm conglomerate (oligomict)
SOUND Sr ss with ripple marks
FORMATION L Sh laminated sandstone
member 5
1 Sg ss with giant crossbeds
J member I lest ss with trough crossbeds
Sp ss with planar crossbeds
Sr ss with ripple marks
upper mbr Ii Sh laminated sandstone
. . . . 4 i FI siltstone
SOMERSET
ISLAND Li Fm mudstone
FORMATION .... Iowe_r mbr ~! Cs stromatolitic dolomite
i CI laminated d o l o m i t e / I s
READ BAY !-I Cb bioclastic Is/dol
ii L
FORMATION LL Ca argillaceous limestone

Fig. 6. The facies spectrum in the clastic wedge. Facies codes (Gin, Sr, etc.) are described
in greater detail in the text.

dolomitic limestone. Oscillation ripple marks and bioturbation are common.

A rich brachiopod fauna (mainly Atrypella), plus corals, stromatoporoids
and trilobites c o m m o n l y is present. A nodular or rubbly weathering appear-
ance is characteristic. This is the dominant lithology of the Read Bay For-
mation. Data provided by Jones and Dixon (1976, table 1) show that at one
locality in northeastern Somerset Island it constitutes 82% of the total
exposed thickness. Jones and Oldershaw (1977) state t h a t the nodular
texture is due to repeated events of early lithification and minor erosion
prior to burial.

Bioclastic limestone and dolomite (Cb). Fragments of brachiopods, corals,

trilobites, crinoids and gastropods are set in a micritic matrix. Small-scale
ripple marks are common, as are laminations caused by varying proportions
of shell debris and matrix. Facies Cb almost invariably rests on a smooth or
scoured erosion surface (Fig. 7).

Laminated dolomite and limestone (Cl). A finely crystalline texture, fine

lamination, platy weathering habit, and b u f f to pale yellowish brown
weathering colours distinguish this lithofacies. Fossils are scarce, but include
feeding trails, gastropods, ostracodes, fish and rare nautiloids and brachio-
pods (mainly Protathyris). Fossil occurrences generally consist of individuals
of one species scattered along a bedding plane. Oscillation ripple marks and
desiccation features are abundant locally. The latter consist of shrinkage
cracks, c o m m o n l y with clast displacement or complete brecciation (Fig. 7).
Scattered sand and silt grains and sandy laminae are common, particularly in
 95

Fig. 7. Laminated dolomite with desiccation cracks overlain by thin lenses of bioclastic
limestone showing scoured bases, Upper member of Somerset Island Formation, 17 km
northwest of Station 8.
96

the lower member of the Somerset Island Formation (Fig. 2). Scattered
blade-shaped vugs representing dissolved evaporite crystals rarely are present.
Petrographic evidence indicates that some of the dolomite is detrital in
origin.

Stromatolitic dolomite (Cs). Laterally linked stromatolite hemispheroids

(LLH type of Logan et al., 1964) occur in units averaging 10 cm in thick-
ness. An intraclast conglomerate or breccia c o m m o n l y occurs at the base.
Flat-lying stromatolitic lamination also is present within facies C1. The algal-
mat types described by Logan et al. (1974), including colloform-, smooth-,
pustular-, and film-mats, are present. Some stromatolite units are very persis-
tent laterally; one unit near the base of the upper member of the Somerset
Island Formation was traced for 10 km, near Station 8. Oncolites are present
in some beds within the lower member of the formation.

Mudstone (Fm). Distinction between lithofacies Fm and F1 c o m m o n l y is

difficult in the field because there is a complete grain-size gradation between
the two. Both are coloured bright green or red-brown or are mottled, and
may be massive or laminated. Both contain an abundance of desiccation fea-
tures, which c o m m o n l y are so pervasive that the rock consists of a fine-
grained intraclast breccia, generally emphasized by colour mottling. The
mudstone units are more recessive and some are coloured purple rather than
red-brown. The geochemical significance of these colour variations has not
been examined.

Siltstone (Fl). Silt-sized rhombic dolomite grains comprise a subordinate

to d o m i n a n t component of this lithology, in addition to quartz silt, and
some beds are sandy and/or argillaceous. Lamination is common. Scattered
blades or small nodules of gypsum are c o m m o n locally, and cubic halite casts
were observed at one outcrop near Station 8. Other features of the litho-
facies are noted above.

Laminated sandstone (Sh). The sandstone is quartzose, fine to very fine

grained, red, green or grey in colour, and characterized by horizontal
bedding (mnemonic h = horizontal in facies code) or lamination. Vugs, prob-
ably representing evaporite nodule solution cavities, are c o m m o n locally.
Parting lineation is abundant.

Sandstone with ripple marks (Sr). Grain size is typically fine- to very fine-
grained. Thin argillaceous partings may be present. Ripple marks are small to
very small in scale, with amplitudes c o m m o n l y less than 1 cm. All are
current-formed (they contain foresets with unimodal orientations) and most
occur as mutually interfering and erosive cosets. Lithofacies colour is invari-
ably red-brown, as is that of most of the remaining facies to be described.
 97

Sandstone with planar crossbedding (Sp). This is a relatively rare litho-

facies. Grain size is fine to medium. Crossbeds are of alpha type (Allen,
1963); they occur as solitary, tabular sets ranging in set thickness from 6 to
35 cm.

Sandstone with trough crossbedding (St). Facies St is very abundant

t h r o u g h o u t the Peel Sound Formation. Grain size ranges from fine to coarse
and, particularly in member 3 of the formation, scattered pebbles c o m m o n l y
are present. Crossbed t y p e consists of shallow scoops or troughs. Some are
solitary sets (theta type, of Allen 1963) but grouped sets are more c o m m o n
(pi type, of Allen 1963). Set thickness ranges from 8 to 40 cm, although
larger sets are present in Peel Sound member 3, grading up in size to those
defined (below) as giant crossbed sets.

Sandstone with giant crossbedding (Sg). Crossbeds ranging in thickness up

to 6 m are present in Peel Sound Formation member 3 (Fig. 8). They are
composed of medium to very coarse sand (rare angular grains up to 2 m m in
diameter) and m a n y show complex internal structures. Foresets are lami-
nated, and are arranged as overlapping, planar or broad trough-like structures.
Cosets of two or more sets reaching at least 8 m in total thickness have been
observed, but exposures large enough to reveal completely structures of this
magnitude are not common. Beds of this facies are invariably red in colour,
but the enclosing strata c o m m o n l y are drab.

Conglomerate (Gm). Pebble and cobble conglomerate constitute the bulk of

Peel Sound members 2 and 4. Most beds are massive, but crude planar- and
cross-stratification is visible in some outcrops. Thin lenses of facies St, Sr or
Sh c o m m o n l y are present (Fig. 9). Conglomerate clast types and matrix were
described by Dineley (1966, p. 273--274) and Brown et al. (1969, p. 536--
537). In member 2 of the Peel Sound Formation, clast size reaches 23 cm
and the conglomerate consists predominantly of rounded dolostone and
quartzite clasts derived from lower Paleozoic and Proterozoic rocks, set in a
dolomitic or quartz sand matrix, with subordinate chert and quartz pebbles.
In member 4, there are at least ten distinct clast types, of which quartzite,
gneiss and dolostone predominate. Clast diameter reaches 30 cm, with a
modal diameter of approximately 7 cm. Clast sources include Proterozoic
arenites and Aphebian igneous and metamorphic rocks of Boothia Uplift.

Facies associations 1: methods

The twelve lithofacies can be ordered into six main associations and these
form the basis for the stratigraphic subdivision of the clastic wedge (Fig. 6).
One of the purposes of this section is to demonstrate t h a t some of these
associations are internally cyclic, a fact which greatly facilitates their envi-
ronmental interpretation.
 98

Fig. 8. G i a n t c r o s s b e d s in M e m b e r 3 o f t h e Peel S o u n d F o r m a t i o n , 10 k m s o u t h w e s t o f
Cape A n n e .
A. Aerial view. B. Detailed view o f l e f t - h a n d e n d of o u t c r o p .
 99

Fig. 9. Conglomerate (Facies Gm) with minor lenses of planar and ripple laminated sand-
stone (Facies Sh, St), some filling minor channels. Peel Sound Formation, member 4,
6 km east of Pressure Point. Scale is 1.5 m long.

The technique used in this paper to analyze for cyclicity is Markov chain
analysis, a m e t h o d which was fully described b y Miall (1973). The embed-
ded-chain type of analysis (Krumbein and Dacey, 1969) is used throughout.
As pointed o u t b y Cant and Walker (1976, p. 112) an important objection
to the use of Markov analysis is that it does n o t show the nature of facies
transitions (gradational or erosional) and, to rectify this, a new technique
has been devised in order to supplement the Markov method. It is termed the
contact matrix, and is described herein for the first time.

Markov analysis method. Transition c o u n t matrices have been assembled

from pooled facies sequences for all well-exposed sections through the upper
m e m b e r of the Somerset Island Formation (Table I), and the lowermost
m e m b e r of the Peel Sound Formation (Table IV). Data are insufficient for
this to be done for the remainder of the clastic wedge, and the sequence anal-
ysis for these members is more subjective. Probability matrices have been
calculated from the data in Tables I and IV (computational details in Miall,
1973) and the difference matrices (Tables II and V) are used as a basis for
reconstructing facies sequences for the two members.
100

TABLE I
Transition count matrix, upper member of Somerset Island Formation

Fm F1 Sh Cl Cs Cb

Fm 0 4 1 10 0 0
F1 10 0 5 34 0 5
Sh 1 5 0 3 C, 1
C1 4 39 5 0 5 3
Cs 0 0 0 5 0 0
Cb 0 2 0 7 0 0

Note: facies codes used in Tables I--VI are explained in text.

C o n t a c t m a t r i x . T h e p u r p o s e o f this t e c h n i q u e is to assess the t y p e s o f con-

tact t h a t o c c u r b e t w e e n all possible facies pairs. Transitions are c o d e d in the
sequence d a t a sets as erosional, gradational or u n k n o w n ( p o o r e x p o s u r e ) .
F r o m these d a t a t w o p r e l i m i n a r y matrices are derived, in e x a c t l y the same
w a y t h a t t r a n s i t i o n c o u n t matrices are calculated in the M a r k o v m e t h o d . A
' g r a d a t i o n m a t r i x ' tabulates t h e n u m b e r o f c o n t a c t s b e t w e e n each lithofacies
pair t h a t are gradational. E l e m e n t s in this m a t r i x are t e r m e d Gij. A n 'erosion
m a t r i x ' , consisting o f elements Eiy, is derived in the same way. The c o n t a c t
m a t r i x consists o f elements C;j, and is calculated as follows:

Ciy - Gij - - Eli

G v + E/j
Values in t h e c o n t a c t m a t r i x range f r o m +1 t o --1. Negative scores indicate a
p r e p o n d e r a n c e o f scouring o r s e d i m e n t breaks, positive scores indicate a
d o m i n a n c e o f gradational c o n t a c t s . Zero scores indicate t h a t c o n t a c t t y p e s
are equally c o m m o n , o r t h a t n o d a t a were available f o r t h a t facies transition
(this m u s t be c h e c k e d in the c o u n t matrices). C o n t a c t matrices f o r the u p p e r
m e m b e r o f t h e S o m e r s e t Island F o r m a t i o n and the l o w e r m o s t m e m b e r o f
the Peel S o u n d F o r m a t i o n are given in Tables I I I and VI, respectively.

TABLE II
Difference matrix, upper member of Somerset Island Formation

Fm F1 Sh C1 Cs Cb

Fm 0.00 --0.14 --0.01 0.25 --0.04 --0.07

Fl 0.03 0.00 --0.01 0.04 --0.05 --0.00
Sh --0.01 0.11 0.00 --0.10 --0.04 0.04
Cl --0.09 0.12 --0.02 0.00 0.04 --0.04
Cs --0.10 -0.38 -0.07 0.61 0.00 -0.06
Cb --0.11 -0.16 -0.07 0.38 --0.04 0.00
 101

TABLE III

Contact matrix, upper member of Somerset Island Formation

Fm Fl Sh C1 Cs Cb

Fm -- --1.00 --1.00 --1.00 -- --

FI 1.00 -- --0.20 --0.11 -- --1.00
Sh -- 0.50 . . . . 1.00
C1 1.00 0.86 1.00 -- --1.00 --0.33
Cs . . . . 1.00 -- --
Cb -- 1.00 -- 1.00 -- --

Facies associations 2: results and interpretation

Somerset Island Formation, lower member. A typical section through part of

this m e m b e r is shown in Fig. 10. Two complete cycles are present and can be
described using the following abbreviation:
/ Cb -~ C1
where / represents a scoured base and ~ represents a gmdational contact. A
more generalized section through the same stratigraphic interval is given in
Fig. 2, b u t in this example no clearly defined cycle types are present.
The Read Bay Formation and the units of Read Bay-type limestone
within the Somerset Island Formation (Fig. 2, 4B) are of shallow subtidal
origin, as indicated primarily by the abundance and diversity of the inverte-
brate fauna (Jones and Dixon, 1976). The laminated dolomite and limestone
(C1) are interpreted as predominantly intertidal to supratidal in origin, on the
basis of criteria given by Laporte (1969, 1971). These include the even
lamination, the presence of stromatolites, the scarcity and low diversity of
the invertebrate fauna, the absence of open-marine forms such as corals and
trilobites, and the presence of desiccation features. Rare nautiloids and
detached coral colonies may have been emplaced by wave action (G. Nar-
bonne, personal communication, 1976). Sodium distribution data in the car-
bonate rocks indicate a hypersaline depositional environment for the Somer-
set Island Formation and a hyposaline (open marine) environment for the
Read Bay Formation (Veizer et al., 1977).
The cycles shown in Fig. 10 probably represent tidal mudflats prograding
over beach or tidal-channel lag deposits. The scoured surface at the base of
the cycles probably was caused by wave activity during a storm, or by tidal-
channel erosion in response to tidal flushing energy. The increased silt or
sand c o n t e n t at the t o p of the cycles may represent an alluvial influx or a
zone of sand concentration in a high wave energy zone, probably near mean
high tide level.
102

25- 9h
rn

7
., ., + ,

20-
©

(JE '

+!iiii!iiii!
15-

/' /' j, j

I©-

:b , 9,G
• • ,+

5-
: :i .... ~
3~

i.:~', •.+ -
~'+ : '+ il

Fig. 10. Typical section through the Lower member of the Somerset Island Formation,
4 km northeast o f Station 7. Arrows indicate cyclic units. Facies codes (at left) are
explained in the text; other symbols as in Fig. 2.

Somerset Island Formation, upper member. The transition count matrix and
the difference matrix for the single dependency Markov analysis of this
member are given in Tables I and II. They are based on pooled data from ten
partial sections through the unit, totalling 149 bed transitions.
Significance tests were performed on the data using ×2 methods, with the
following results: eq. 4 of Miall (1973): 29.2 with 24 degrees of freedom;
eq. 5 : 1 4 4 . 8 with 19 degrees of freedom. The first value does not exceed the
value of X2 with 24 degrees of freedom at a 95% confidence level. The
second value does, but experience (by ADM) with this equation indicates
that it is not a sensitive significance test. Therefore, significance test results
 103

are ambiguous. However, there is no doubting the substantive significance of

the probability matrices and their utility in interpreting the facies assem-
blage, as discussed below.
Table III is the contact matrix derived from the same outcrops, using 66
contact observations (the remainder being obscure or not exposed). Fig. l l B
shows the transition path diagram derived for these rocks by following
through all the positive (greater than random) transition probabilities in
Table II. Using the contact matrix this may be broken down into several
cycle types, most of which commence with an erosional base and are built
up by largely gradational facies changes. The various types, and their fre-

FI

Fm F

FI

CS ~

0
o G
CI ct
B
///// / / /
Cb I ,-, I ~, I / /
o

A.
Cs Sh
I

Cb .CI_ "FI .... Fm

B.
L

Fig. 11. Cyclic sedimentation in the Upper m e m b e r of the Somerset Island Formation.
A. T w o typical cyclic units (generalized from field examples), symbols as in Fig. 2. B.
Transition path diagram derived from the Markov difference matrix, Table If. Solid
arrows indicate principal paths, dashed arrows indicate less numerous transitions.
104

quency of occurrence in the ten stratigraphic sections, are listed below

(upward transitions into facies Fm at the top of each cycle may or may n o t
occur; they are not shown here). Typical examples are illustrated in Fig.
llA.

Type Sequence Frequency Mean thickness

1 Cl --> F1 29 5.2 m
2 /C1 ~ F1 6 4.7 m
3 / C b -+ C1 -~ F1 3 7.9 m
4 /Cl ~ Cs ~ C1 ~ Fl 3 8.1 m
5 / S h ~ Fl 1 7.4 m

These cycles account for 63% of the total number of beds studied. Varia-
tions on these cycles (of low statistical abundance) and random events com-
prise the remainder. Many type 1 cycles, in fact, may be of t y p e 2 or 4
because the contacts between the units c o m m o n l y could n o t be seen clearly.
The main difference between these cycles and those in the lower member
is the presence of facies F1 and Fm. Cycle types 1 to 4 are interpreted as
tidal in origin, the upper part of the cycle including high intertidal to supra~
tidal rocks that are n o t present lower in the formation. In Fig. l l B it can be
seen that the appearance of bioclastic units (facies Cb) is a random event
(upward passage into Cb is not shown because it is of low probability),
which would be expected if the facies represents a storm deposit, the lag
deposit of a laterally migrating tidal channel, or a thin sheet of debris result-
ing from marine transgression across the flats. Progradation of the carbonate
mud flat continued, however, after this depositional event.
The invariable position of stromatolite units (facies Cs} is within the lami-
nated dolomites (facies C1), which tends to support the interpretation of the
stromatolites as intertidal to shallow subtidal in origin. Logan et al., (1964}
state that LLH-type stromatolites commonly occur in high intertidal posi-
tions where wave activity is slight and desiccation is frequent. Recent work
has shown that growth forms and microstructures related to stromatolitic
algae show a distinctive sequence across most shallow subtidal to supratidal
flats (Kendall and Skipwith, 1968; Logan et al., 1974; Schwartz et al., 1975).
A comparison between the m o d e m algal mats of Hamelin Pool and those of
the upper member of the Somerset Island Formation is shown in Fig. 12,
and suggests a probable supratidal to shallow subtidal origin for these struc-
tures. It should be noted, however, that the zonation described by these
authors is n o t applicable everywhere. Grazing organisms m a y control mat
distribution (Garrett, 1970}, and algal mats ar~e restricted to upper intertidal
and supratidal areas on Andros Island (Shinn et al., 1969).
The clastic facies (F1 and Fm) represent terrigenous material (including
detrital dolomite) deposited predominantly in supratidal environments. The
local abundance of evaporite casts and nodules suggests high evaporation
 105

ZONATION OF KEY
ALGAL MAT AT ]~ Desiccation cracks
HAMELIN POOL
-- -- -- ~ . . . . y Syneresis cracks m
Vugs ~ Algal domes 4
FILM MAT ~ ~ Oscillation ripples (~) Oncolites

Scours 0 Ostracods
,~ P U S T U L A R ~ ~ Gastropodcoquina
n,- MAT " ~ ,coo Intraclasts

,,, ¥ 0 2
SMDO,H MA,

N
COLLOFORM MAT
ool

"~--- H ....

Fig. 12. Algal mat types in Hamelin Pool, Australia (from Logan et al., 1974) at left, and
the distribution of mat types in the lower parts of cycles (mainly facies C1) from the
Somerset Island Formation, upper member. The incoming of smooth mat is correlated
with the onset of intertidal conditions.

rates, and their formation may be similar to that of evaporites forming in

m o d e m carbonate sabkha environments in arid regions such as the Persian
Gulf (Shearman, 1966; Kinsman, 1969, 1971). The mottled (brecciated)
siltstone and mudstone, although containing less evaporite, are somewhat
similar to the 'chaotic m u d ' facies of Thompson (1968, pp. 26--29, pl. 9),
which is the dominant sediment type on the high intertidal mud fiat zone of
the Colorado River delta area. The chaotic texture is the result of desiccation
and brecciation, plus the disruptive effect of evaporite crystal growth. These
features imply a climate which included dry periods, but not necessarily ari-
dity (strongly seasonal rainfall), as discussed below.
Table I shows that laminated sandstone units (facies Sh) are interbedded
with most o f the other five facies states, and their appearance seems to be
essentially random. The facies is interpreted as the product of alluvial sheet
floods, the proximal portions of which 'deposited the contemporaneous
fluvial sandstones of the Peel Sound Formation.
The complete cyclic model for the upper m e m b e r of the Somerset Island
Formation consists of an eastward progradation of intertidal carbonate
m u d flats, behind which advanced the distal fringe of an alluvial plain. The
cyclic types listed above are merely variations on this theme. The position of
106

the various facies zones shifted rapidly with time, as a result of variations in
the local sediment supply or the relative rates of subsidence versus accretion.
Individual cycles, however, may be persistent laterally; one prominent dolo-
mite unit has been traced for more than 20 km northward from Station 7.
Holocene tidal cyclicity of similar type in sediments at Shark Bay, W.
Australia has been related to transgressive-regressive events resulting from
relative changes in sealevel (Hagan and Logan, 1974).

Peel S o u n d Formation, m e m b e r 1. Markov analysis and contact matrix data

are given in Tables IV to VI. They are based on pooled data from eight
partial sections, totalling 209 bed transition observations and 114 contact
observations. ×: tests gave the following results: eq. 4 of Miall (1973): 65.9
with 24 degrees of freedom; eq. 5 : 1 7 0 . 9 with 19 degrees of freedom. Both
these values are significant at the 95% confidence level, disproving the null
hypothesis of random facies transitions. A Markov path diagram is given in
Fig. 13B and principal cycle types derived from these data are listed below.
Some examples are illustrated in Figs. 13A, 14 and 15.

Type Sequence Frequency Mean thickness

1 / S t --> F1 24 2.1 m
2 /St -~ Sr -> Fl, F m 14 1.5 m
3 / S p -+ F1 3 2.9 m
4 / S p ~ Sh 1 0.3 m
5 / S p -+ Sr -> F m , FI 2 2.6 m
6 /Sh -~ F m , F1 15 1.2 m
7 /Sr -~ Fro, F1 10 2.1 m

These cycles account for 73% of the total number of beds studied. Deriva-
tion of the complex facies relationships would be difficult without the use
of the Markov and contact matrix methods. One important item of data is
brought o u t particularly clearly by the latter: the facies have been arranged

T A B L E IV
T r a n s i t i o n c o u n t m a t r i x , l o w e r m o s t m e m b e r o f Peel S o u n d F o r m a t i o n

Fm F1 Sr Sh Sp St

Fm 0 2 8 10 2 7
Fl 7 0 9 8 6 29
Sr 13 19 0 1 0 4
Sh 7 11 4 0 0 6
Sp 0 4 2 2 0 1
St 1 25 14 6 1 0
 107

TABLE V
D i f f e r e n c e m a t r i x , l o w e r m o s t m e m b e r o f Peel S o u n d F o r m a t i o n

Fm F1 Sr Sh Sp St

Fm 0.00 --0.26 0.07 0.19 0.02 -0.02

Fl --0.07 0.00 --0.09 -0.05 0.04 0.18
Sr 0.18 0.17 0.00 -0.14 -0.05 --0.17
Sh 0.09 0.07 --0.06 0.00 --0.05 -0.05
Sp --0.14 0.15 0.04 0.08 0.00 --0.12
St --0.16 0.17 0.07 --0.05 -0.03 0.00

in the various matrices in order of increasing grain size and sedimentary

structure scale, from left to right and top to bottom. Cells below and to the
left of the principal diagonal therefore record upward-fining transitions, and
in the contact matrix (Table VI) these all have positive scores. Coarsening-
upward transitions, above and to the right of the principal diagonal, all show
negative scores. This clearly confirms that gradational cyclic sequences are
virtually all of upward-fining character, and that each has an erosive base.
The types of lithofacies, their contained sedimentary structures, and the
nature of the cyclic sequences, all are consistent with a fluvial mode of depo-
sition. Some may be point-bar sequences formed on river meanders, such as
those described by Allen (1970). A characteristic feature of the meandering
river cycle is the presence of accretion deposits resulting from lateral migra-
tion of point-bars. In some cases they consist of large-scale, low-angle cross-
beds, the epsilon type of Allen (1963), which have been described from
ancient sediments by Moody-Stuart (1966) and others. However, such accre-
tion surfaces are commonly obscured by down-channel migration of smaller-
scale structures, principally dunes and ripples. Leeder (1973) pointed out
that of 231 Devonian meandering river cycles described by Allen (1970),
only 11 contain epsilon cross-stratification. Another factor working against
the recognition of epsilon cross-beds in ancient deposits is that in modern
rivers their lateral slope is small. Leeder (1973, fig. 4) has shown that for

TABLE VI
C o n t a c t m a t r i x , l o w e r m o s t m e m b e r o f Peel S o u n d F o r m a t i o n

Fm F1 Sr Sh Sp St

Fm -- -- --1.00 --0.33 -- --1.00

Fl 1.00 -- --1.00 --0.33 --1.00 --1.00
Sr 0.40 0.40 -- --1.00 -- --I.00
Sh 1.00 0.56 1.00 -- -- --i.00
Sp -- -- i .00 -- -- --
St 1.00 0.78 1.00 0.33 -- --
108

Fm

sp Sr

Fm FI
Sr
Sh
I

Fm

FI
Sr

Sp
Ot St

A.
Sr
,-#1%
,'~I Fm
Sp:.
B.

Fig. 13. Cyclic sedimentation in the Peel Sound Formation, Member 1.

A. Idealized stratigraphic sections, based on field examples, showing six of the common
cycle types. B. Transition path diagram derived from the Markov difference matrix, Table
V. Solid arrows indicate principal paths, dashed arrows indicate less numerous transi-
tions.

cycles of the scale present in the lowermost Peel Sound Formation the dip
of the accretion surface probably would be between 2 and 25 ° , with a value
of around 10 ° being most typical. Very few such crossbeds have been
observed in the Peel Sound, although a possible example is illustrated in
Fig. 14. A second feature of deposits formed in meandering river environ-
ments is the abundance of overbank material (silt and mud) in the succession
(Allen, 1970; Cant and Walker, 1976, fig. 16). Facies F1 and Fm range from
minor constituents to more than 25% of sections in the Peel Sound. For
example a relatively thick, recessive siltstone lens is present above the epsilon
crossbed in Fig. 14. However, the type of sandstone-dominated sequence
shown in Fig. 15 predominates in member 1 of the Peel Sound, facies F1 and
Fm constituting a few cm, or less in each cycle.
Most of the sediments comprising this member are interpreted as the prod-
uct of deposition in a braided river, though some may represent a meander-
ing river environment. Vertical and lateral variation between the two types
 109

Fig. 14. Peel Sound Formation, lower member, 11 km south-southeast of Cape Anne,
showing fluvial fining-upward cycles. Lithologies are indicated by facies codes: E =
possible epsilon crossbed. Scale is indicated by the hammer, which is 33 cm in length.

Fig. 15. Peel Sound Formation, lower member, 11 km south-southeast of Cape Anne.
Trough crossbedded sandstone units (Facies St), with thin siltstone partings (at top of
hammer handle). Scale is indicated by the hammer, which is 33 cm in length.
110

of river probably took place in response to fluctuations in discharge and sedi-

ment load. Fluvial cycles of the type and scale listed above have been
recorded in modem ephemeral braided streams (Williams, 1971) and some
ancient rock sequences (Moody-Stuart, 1966; McGowen and Garner, 1970).
Miall (1977), in a review of the braided river depositional environment,
termed fining-upward braided river cycles the 'Donjek type', using the gravel-
dominated sequences of the Donjek River, Yukon (Williams and Rust, 1969;
Rust, 1972) as the type example. The Peel Sound cycles are thought to
represent a distal, gravel-free equivalent of the Donjek cycle type. The origin
of the cyclicity is unclear, but may be the product of lateral accretion within
a topographically differentiated braided channel (energy level decreasing
from channel floor to higher and less active areas) or vertical aggradation in
a decaying channel. It is not known whether the variations in mean cycle
thickness listed above are of any significance. The majority of the cycles are
less than 4 m thick, and the thickness of the sand member within them may
be approximately equal to channel depth. By comparison with modern rivers
these are relatively small channels. The climate of the Somerset Island area
probably included extended dry periods, as indicated by the Occurrence of
evaporites and desiccation features in the Somerset Island Formation, and
the Peel Sound streams may, therefore, have been ephemeral rather than
perennial. Some evidence of desiccation and brecciation is present in facies
F1 and Fm in the Peel Sound Formation, but it is not as abundant as in the
underlying unit.
The alluvial plain on which these fluvial rocks were deposited advanced
steadily eastward during Late Silurian and Early Devonian time (Fig. 4).
The ephemeral nature of the streams is suggested by climatic considerations,
as noted above, and by the fact that sandstone bodies representing major
channels or prograding deltas, virtually are absent from the laterally equiv-
alent tidal deposits of the Somerset Island Fornlation.
Peel Sound Formation, conglomerate members. At least 90% of most sec-
tions through members 2 and 4 of the formation comprise conglomerate
(facies Gm), the characteristics of which were described in an earlier section
of this paper. The association of conglomerate with subordinate lenses and
wedges of sand (Fig. 9) is characteristic of the proximal braided river or allu-
vial fan environment. Miall {1977) named the facies assemblage the 'Scott
type' after the Scott Glacier (Alaska) outwash deposits (Boothroyd and Ash-
ley, 1975). The conglomerate beds are multistorey units representing super-
imposed longitudinal and rare transverse bar deposits, and possibly torrential
sheet-type deposits. Sand beds represent low-water channel-fill deposits or
bar-edge sand wedges. Similar alluvial fan deposits in the Peel Sound Forma-
tion of Prince of Wales Island were described in greater detail by Miall (1969,
1970b).
Peel Sound Formation, member 3. This member is composed of an alterna-
tion of sandstone units with giant crossbeds (facies Sg, Fig. 8), and pebbly
 111

sandstone with thin conglomerate units and abundant trough crossbedding

(facies St). Most troughs are less than 50 cm in height, b u t there is a size gra-
dation between these and the giant sets. No cyclic relationships are apparent.
Two modes of origin are possible for facies Sg; they may represent eolian
dunes, or fluvial sand waves. The main arguments for an eolian origin are the
size of the crossbedding, the well-sorted nature of the sand, and the anoma-
lous orientation of the foresets (see paleocurrent analysis). Glennie {1970,
p. 11) stated that pebbles up to 1 cm in diameter may be present in eolian
dunes, so that the presence of laminae containing grains 2 mm in diameter is
n o t inconsistent with an eolian origin. However, there is little to distinguish
the internal style of eolian cross-stratification from that shown b y large-scale
fluvial sand waves, such as those described b y Coleman (1969, pp. 208--211),
although trough-shaped crossbeds are u n c o m m o n in many eolian dune fields
(Reineck and Singh, 1973, p. 199). Some minor differences were described
by Hunter (1976) and include the tendency for eolian cross-strata to wedge
o u t upward, and to be lensoid in the direction of foreset strike, in contrast
to the greater consistency in foreset thickness shown b y subaqueous cross-
beds. Other minor features of eolian crossbeds c o m m o n l y include ripple
marks on foreset planes with crests parallel or nearly parallel to the direction
of foreset dip (representing transverse eddies), raindrop impressions, lag
deposits, avalanche structures and contorted beds (Steidtmann, 1976; Walker
and Harms, 1976). None o f these features was observed in the giant crossbeds
of the Peel Sound Formation.
The alternative interpretation, that facies Sg represents fluvial sand waves,
also raises problems, principally that o f the contrast between the very well-
sorted, pebble-free sand of the giant crossbeds and the pebbly sand of the
interbedded units of facies St. However, this could relate to the difference in
slope between a large trunk river and that of steeper, proximal tributary
streams carrying a coarser bedload. Secondly, the large sand waves described
b y Coleman (1969) contain clay lenses indicating periods of quiet-water sedi-
mentation, b u t these were n o t observed in the Peel Sound examples.
No firm conclusions regarding the origin of the giant crossbeds are drawn
in this paper. This is, in fact, an example of a general problem with regard to
the interpretation of large-scale crossbedding and the recognition of eolian
depositional environments in the ancient record. There are no u n d o u b t e d
diagnostic criteria for the identification of eolian deposits; crossbed style,
thickness and foreset dip are similar in each environment, and negative evi-
dence concerning the absence of clay lenses, contorted beds, avalanche struc-
tures or transverse ripples, is not conclusive.
In the present case the anomalous paleocurrent directions could be
explained readily from the standpoint o f wind directions or regional fluvial
paleogeography. The presence of a large river during deposition of Peel
Sound m e m b e r 3 would be in marked contrast to the ephemeral drainage
system interpreted as existing during the formation of m e m b e r 1. This, and
other paleogeographic considerations, will be discussed in later sections.
112

PALEOCURRENT ANALYSIS

A total of 165 sedimentary structures were measured in order to obtain

paleocurrent directions. Data have been grouped into geographically distinct
sets and the statistics of their distribution calculated using the method of
Curray (1956). Results are given in Table VII and Fig. 16. For the purpose
of this analysis, giant crossbeds are defined as those with a set thickness of
1 m or greater. Weighting of crossbed data has been carried out using the
method of Miall (1974). Most of the data indicate easterly transport direc-
tions, away from the Boothia Uplift, with some deviation to the northeast or
southeast, possibly indicating that dispersion took place on large, sandy,
alluvial fans.
Giant crossbeds have a mean orientation of northwest. The most obvious
explanation of this anomalous orientation is that the giant crossbeds had a
different mode of origin to the other cross-stratification features in the Peel

I C (G) C 4©kin 1

3 I~C

Fig. 16. P a l e o c u r r e n t data, s h o w i n g v e c t o r m e a n a z i m u t h s and c u r r e n t rose diagrams.

(G) = giant c r o s s b e d s ; o t h e r figures are s t a t i o n n u m b e r s . Statistical d a t a for these dia-
grams are in Table VII. Current rose diagrams are equal area p l o t s ( s e g m e n t area p r o p o r -
tional to n u m b e r o f readings) and s h o w u n w e i g h t e d data, in o r d e r to p e r m i t inclusion o f
t h e a b u n d a n t parting lineation i n f o r m a t i o n .
 113

"" 0

0~ 00 ~ 0 0
woo o v v v v

d ~ l d d d d ~
,~ ~ . ~

? I ~T?1??T
v V o o V V V V V V

o~

g
0
II 0 ~.~

] °
 114

Sound. A possible eolian origin has been discussed above. Paleogeographic

reconstructions by Woodrow et al. (1973, fig. 1) showed that during the
Middle and Late Devonian the Boothia Uplift was located approximately at
latitude 20°N, and that latitude lines were oriented northeast--southwest
with reference to present day coordinates. Assuming a similar trade wind
pattern to the present day the d o m i n a n t wind directions would have been
from the Devonian northeast -- the present day north -- and, therefore, quite
different from the current directions indicated by the large crossbed struc-
tures.
However, it cannot be assumed that the paleogeographic reconstructions
are precisely correct (a difference of 5--10 ° of latitude could be significant)
or that wind circulation patterns in the Devonian were the same as at the
present. In addition, the presence of a mountain belt in the position of
Boothia Uplift may have strongly influenced local weather patterns. The evi-
dence of paleocurrent directions is therefore inconclusive with regard to the
mode of origin of the giant crossbeds.

REGIONAL EVOLUTION OF THE CLASTIC WEDGE

Asymmetry of Boothia Uplift region

The Somerset Island--Peel Sound succession on Prince of Wales Island was

derived from the west side of Boothia Uplift {Miall, 1970a, b), but the facies
variations within it are not a mirror image of those on the east side of the
uplift (Fig. 17). Contemporaneous rocks of eastern Prince of Wales Island
include a significant thickness of sandstone and conglomerate different in
character from those on Somerset Island. Following Read Bay time the
clastic wedge developed as follows: on the west side of the Boothia Uplift a
zone of mixed marine and alluvial sedimentation perhaps less than 10 km
wide developed; at the same time a broad area of intertidal and supratidal
flats prograded out from the east side of the uplift, reaching a width of at least
70 km during deposition of the upper member of the Somerset Island For-
mation (Miall et al., 1977).
On Prince of Wales Island there is a zone of interbedded marine carbonate
and shale and alluvial sandstone at the distal fringe of the Peel Sound alluvial
plain (sandstone--carbonate facies of Miall, 1969, 1970a), whereas no com-
parable succession is seen on Somerset Island. The marginal marine zone in
Prince of Wales Island probably consisted of small deltas separated by inter-
deltaic bays and barrier beaches, whereas in Somerset Island tidal flat condi-
tions probably persisted through earliest Devonian time, and there is little
or no evidence that fluvial sand bodies built out far enough to the east to
enter the sea and form deltas (some eolian or fluvially derived mature quartz
sand appears to have been incorporated in the Somerset Island Formation as
relatively rare sandy dolomite and dolomitic sandstone).
The conglomerates in the Peel Sound Formation are much coarser to the
 115

west _~ east

Prince of Wales I. = o Somerset I.

r ~-=-- :.::::::~: ~ SF-3

L ~-- .:.:.:.:.:-:.:-: : ~ ....
L--- --~ J'":':': :':':-:'::~'X~ . . . . . . .

~ : . : . : - : S S . F..:.:.:.:.:-:
:::::::::::::::::::::::::::::::::::::::
~ "-
•:.:: .'. , . . . . . : . . - ---------:~1 I I I
i ~---~--- ~ -:~ ~-- _-_o~ : - - : I l I I I I l I l J I I I I ;
I I I I I ] iji I
' K ) - , ~ = - L_ ~ -- - , , , , I ' I ' RBF I , [ . I . .I . . I I
IIiJiI,I,l, I , RBF , , I, II J , , , , I , ,
~I' F L I I I I rllIl I i i i i i i
r ] £_L I J I I I I I L 1 I II _J_ J _ ~ _ £ _IL[ I_ L _ L I I. . .I . I J - I J - - JI - - I - - ~ L_L . . .I . . . . . I I

. . . . /0
sandstone 20 km 0

Fig. 17. C o m p a r i s o n o f the clastic w e d g e s o n either side o f B o o t h i a Uplift. R B F = Read

B a y F o r m a t i o n , PSF = Peel S o u n d F o r m a t i o n , SIF = S o m e r s e t Island F o r m a t i o n , F =
Facies, as in S a n d s t o n e - C a r b o n a t e , S a n d s t o n e and C o n g l o m e r a t e Facies o n Prince o f
Wales Island. T h e S a n d s t o n e Carbonate Facies o f western Prince o f Wales Island has n o w
b e e n assigned to the upper m e m b e r o f t h e Drake B a y F o r m a t i o n (Mayr, in press).

west o f Boothia Uplift than to the east. Clasts up to 150 cm in diameter

were observed by Miall (1970b, p. 569) in Prince of Wales Island, as com-
pared with a m a x i m u m size of 31 cm in Somerset Island (Brown et al., 1969,
p. 540).
The present<lay structural geology o f Boothia Uplift also exhibits asym-
metry: a narrow folded and faulted zone west of the Uplift and a broad area
of block faulting and relatively gently folding to the east (Kerr and Christie,
1965; Kerr, 1977). This style of deformation probably originated during
deposition of the clastic wedge, because the sedimentological differences
between the two sides of the arch can most readily be explained in terms of
differences in depositional slope: steep to the west, resulting in transporta-
tion of coarser debris, and the building of deltas into the sea; and gentle to
the east, resulting in the development o f a broad, low-energy alluvial plain
and tidal-flat complex. On Prince of Wales Island the clastic wedge has a
m a x i m u m width of approximately 40 km, on Somerset Island it is 70 km. It
is unlikely that climatic (for example, rain shadow) effects contributed to
this difference. As pointed out earlier, paleogeographic reconstructions by
116

Woodrow et al. (1973) tentatively suggest a predominant northerly wind

direction, parallel to Boothia Uplift.
Syndepositional tectonics
Miall (1969; 1970b, fig. 5) mapped a syndepositional fold in southern
Prince of Wales Island, which he considered to be the result of uplift and
folding taking place along the margin of the depositional basin while alluvial
fans continued to prograde out from the mountain front. Earlier conglom-
erate beds are tilted, whereas later deposits are virtually undeformed. A simi-
lar type of structure has been mapped in Somerset Island (Fig. 5) where an
earlier conglomerate wedge (Peel Sound member 2) was uplifted and partly
eroded before a later conglomerate (member 4) prograded out from the
source area. The two units (or their lateral facies equivalents) probably are
conformable farther to the east. Nearer the source area it would be expected
that the younger conglomerate wedge would truncate older beds completely,
and rest directly on the Precambrian rocks from which it was derived. Such a
relationship nowhere is preserved on Somerset Island, but on northern Prince
of Wales Island there is one outcrop showing a basal Peel Sound breccia
resting unconformably on upturned Read Bay Formation, with an angular
discordance of approximately 20 ° (Kerr and Christie, 1965; Miall, 1969).
Local unconformities have been recorded also in similar rocks in Cornwallis
Island (Thorsteinsson and Kerr, 1968).
Syndepositional structures of this type appear to be c o m m o n in areas of
coarse clastic sedimentation adjacent to rapidly rising uplifts. They have
been recorded in the Devonian of Norway (Bryhni and Skjerlie, 1975) and
the Tertiary of the Spanish Pyrenees (Riba, 1976). It is tempting to attribute
all the observed structures in the Peel Sound Formation to the same tectonic
episode, but Riba (1976) showed that syndepositional unconformities may
be present at several different levels within a clastic sequence, each persisting
along strike for 5 km or less. They are local phenomena that commonly
occur at times of rapid uplift and rapid sedimentation. In the case of the Peel
Sound Formation the structures developed both in southern Prince of Wales
Island and northern Somerset Island at a time when the local Paleozoic cover
was removed. Clast types underwent a marked change from oligomict to
polymict, with the introduction of a variety of clast types derived from
Proterozoic sources. In Prince of Wales Island, the clast change and the
folding occur near the base of the formation whereas in Somerset Island they
occur between members 2 and 4. As noted above, the Boothia Uplift is an
asymmetric feature. Upward movement may have been greater on the west
side than on the east, and this would explain the earlier stripping away of the
Paleozoic cover rocks on the Prince of Wales side.
Paleogeography
The interpretations set forth in preceding sections of this paper are sum-
marized in Figs. 18--20, which show tentative paleogeographic reconstruc-
 117

F-
-q

II
I 0 5 0 km
/

Fig. 18. Paleogeography of the Boothia Uplift region during deposition o f the lower
member of the Somerset Island Formation (early to mid Pridolian?).

tions for the Somerset-Prince of Wales Island region. These diagrams are
based on the assumption that the Somerset Island Formation of Somerset
Island and the Lower Peel Sound Formation of Prince of Wales Island are
correlative (Figs. 3, 17). As discussed b y Miall et al. (1977) there is limited
biostratigraphic evidence to support this assumption, and it would seem
probable that the lowest units o f the clastic wedge on either side of Boothia
Uplift both reflect the same tectonic episode.
In Ludlovian time the sea probably occupied the entire area shown in
Figs. 18--20. The Read Bay Formation shows no evidence of shoaling or of
detritus derived from Boothia Uplift. Miall and Kerr (1977) reported a
thickening o f the formation from south to north along the present east flank
of the uplift, and this is probably related to the tilting of the craton towards
the Franklinian Geosyncline.
118

~:-:- ::-:~%, ~ _ _ U~

j"

C •7

!i~iiiiii!l ::: : i ?
I-- :::::::::::: :::\ - : ~:::: : ,a •
I ::::::::::::::::::::: /
/.:.:,:.:.:.y:.:.:l /
iiiiii/iii!ii!t /
~'<~'a ! lu via I fan I~;;;;;;;;):i::l

I
'!/ zone (bojadcl)~----"z'{:::i:i:::: 1 / /

/,J J 50 km
iI ( " .~ . . . . . .

Fig. 19. Paleogeography o f the Boothia Uplift region during deposition of Peel Sound
m e m b e r 1 (late Pridolian to early Gedinnian?).

Emergence of Boothia Uplift in Pridolian time was followed by the

growth and gradual outward progradation of alluvial and tidal-flat zones
(Fig. 18), the deposits accumulating thereon now constituting the Somerset
Island Formation of Somerset Island and the Lower Peel Sound Formation
o f Prince of Wales Island. Bathurst Island was part of the geosyncline at this
time. On Cornwallis Island the subtidal marine limestones of Read Bay
member C were deposited (Thorsteinsson, 1958; Gibling and Narbonne,
1977).
The relief on the uplift increased steadily and without apparent break
during the late Pridolian and early Gedinnian. The elevated region became a
mountain belt from which large quantities o f gravel were derived, building
up an alluvial fan system (bajada) on either side. More coarse clastic material
is preserved now on the west side of Boothia Uplift than on the east, which
 119

• :•1
................ ::::j

::::::!:i:i:!:l C)
~a / :::::::::::[.~

o:\:):):):):):):?:?~
~:~iii!!!iiiiiiii!l
~o ~:':':':':':':'.
~33 ..:-:-:-:.:.:4.
O i:i:i:!:i:i:i~
:i:i:i:i:i:i:k

~~_~~
land area ?
:ii:!:~!:i:i~ s
,Dt.L~:i:i::l
~.:-:~i_ //

)iiiiiiii!ii]
li

I 0 . . . . 50kin
/
/
/
Fig. 20. Paleogeography of the Boothia Uplift region during deposition of member 2 and
3 of the Peel Sound Formation (early to mid Gedinnian?), assuming a fluvial interpreta-
tion for the giant erossbeds of member 3.

may indicate steeper depositional slopes in the west (but could be in part an
accident of exposure). Fig. 19 shows a tentative paleogeographic reconstruc-
tion for this period of time, corresponding to the beginning of Peel Sound
sedimentation on Somerset Island.
Biostratigraphy and sedimentation patterns on Somerset and Cornwallis
Islands, discussed by U y e n o (1977) and Gibling and Narbonne {1977), show
that the incoming of clastic sedimentation occurred substantially later on
Cornwallis Island than on Somerset Island (Fig. 3). The Boothia Uplift
plunges northward in northwestern Somerset Island and northeastern Prince
o f Wales Island, as indicated b y outcrop patterns o f the Proterozoic and
Paleozoic rocks. Orientations of Devonian alluvial fans in Prince of Wales
Island (Miall, 1970b) and paleocurrent directions shown by trough and
120

planar crossbedding in the Cape Anne--Pressure Point area {Fig. 16), all
have a northerly component, suggesting derivation from a source area that
ended west of Aston Bay. It is probable that no major land mass existed
north of Somerset Island until the mid-Gedinnian.
Member D of the Read Bay Formation is similar to the Somerset Island
Formation, and probably was deposited under intertidal to supratidal condi-
tions. The Snowblind Bay Formation is a conglomerate and sandstone unit
generally similar to member 2 of the Peel Sound, with a lower contact
ranging locally from gradational to unconformable (Thorsteinsson and
Kerr, 1968). Paleocurrent data from the base of the Snowblind Bay Forma-
tion (Gibling and Narbonne, 1977) indicate a source to the west to south-
west, showing that the Boothia Uplift extended to the north of Somerset
Island during the Early Devonian. As a result of continued uplift during
Snowblind Bay sedimentation, the entire area shown in Figs. 18--20 may
have become land in the mid-Gedinnian, with proximal alluvial environments
shifting northward closer to the margin of the Franklinian geosyncline.
A fluvial interpretation for the giant crossbeds of Peel Sound member 3
raises several paleogeographic implications, as discussed below. The giant
crossbeds may represent a trunk river flowing northward parallel to Boothia
Uplift. Interbedded with facies Sg are units of trough crossbedded sand
showing northeasterly paleocurrent directions, and these could be the depos-
its of tributary streams which flowed directly from the uplift. There are
many modern analogues for this fluvial pattern, including the Ganges, Indus,
Mackenzie, and Tigris-Euphrates river systems, all of which flow parallel to
structural strike. A similar pattern is emerging for some of the Mesozoic
fluvial rocks deposited near the R o c k y Mountains of Alberta (Rahmani and
Lerbekmo, 1975; McLean, 1976). If this is a correct interpretation, it implies
that the flow was channelled into a trough parallel to Boothia Uplift, caused
by subsidence close to the uplift or b y elevation of eastern Somerset Island.
A generalized reconstruction b y Dineley (1975, figs. 3, 4) shows the Somer-
set Island area as a depositional e m b a y m e n t between Boothia Uplift and the
craton of Baffln Island to the east, (Fig. 19). In early or middle Gedinnian
time this e m b a y m e n t may have disappeared, as indicated in Fig. 20. The large
runoff implied by the development of the major river may reflect an
increased rainfall. This could have been caused b y local climatic changes
induced b y the appearance of an area of strong relief.
A trunk river flowing north from Somerset Island would have built a large
delta where it entered the sea at the margin of the geosyncline, but the depos-
its of such a delta have not been identified. They could have formed east o f
Boothia Uplift, in the Devon Island area, in which case they have been lost
to erosion. Alternatively the river could have continued flowing northwest,
towards Bathurst Island.
The Bathurst Island Formation of southeastern Bathurst Island (Kerr,
1974) is similar in age to the Peel Sound Formation and, in part, may repre-
sent a delta deposited b y rivers flowing north from the Boothia Uplift region.
 121

U. Mayr (personal communication, 1976) stated that the Bathurst Island

Formation is unusually thick in the Allison River N-12 well in southeastern
Bathurst Island (Fig. 1), and that it consists of fine-grained clastic sediments
that could have been deposited in a deltaic environment.
Northward extension o f the Boothia Uplift in the Early Devonian would
have terminated such a pattern of sediment distribution, and provided a new
source area for the Bathurst Island Formation.
Alluvial sedimentation may have continued into Middle or Late Devonian
time, as indicated b y derived Upper Devonian fish in the Tertiary--Creta-
ceous rocks of Somerset Island (Dineley and Rust, 1968).
A paleogeographic reconstruction b y Woodrow et al. (1973) indicates that
the Canadian Arctic Islands were located a b o u t 20°N latitude during the
Middle--Upper Devonian. The area was probably in the trade-wind belt and
was characterized b y a semi-arid climate locally with moderate to high rain-
fall, strongly seasonal in distribution.

A m o u n t s and rates o f denudation and sedimentation

In Somerset Island there is a sequence o f mainly carbonate lower Paleo-

zoic sediments approximately 1 km thick underlying the Somerset Island
Formation (Miail and Kerr, 1977). Some of the units may have been depos-
ited uniformly across Boothia Uplift, although there is evidence that some
units thinned over the uplift, as a result of mild positive movements, partic-
ularly during the Ordovician (Miall and Kerr, unpubl.). Much of this
thickness was removed b y subaerial erosion during the formation of the
Somerset Island--Peel Sound clastic wedge, as shown by clasts o f lower
Paleozoic rocks low in the Peel Sound Formation, and the upward change in
clast character, which indicates deep erosion of the underlying Proterozoic
(documented in detail on Prince of Wales Island b y Miail, 1970b). The
volume o f sediment removed can be estimated b y assuming an average depth
of erosion of 1 km over the area of Boothia Uplift from latitude 71°N to the
northern limit of exposure of the Precambrian basement at Aston Bay. In
places the depth o f erosion was greater, as indicated b y the abundance o f
basement (Aphebian) clasts in the Peel Sound; in places erosion m a y have
been less, as suggested b y the fact that in places Paleozoic carbonate clasts
are abundant in the youngest preserved conglomerate of m e m b e r 4. The
estimated volume removed is 28,000 km 3. Only a part of this is accounted
for in the Somerset Island and Peel Sound Formations. In Prince of Wales
Island the clastic rocks (including detrital carbonate units) in these forma-
tions can be assumed to form a triangular prism, with a base 500 m thick
near the Uplift, tapering to a feather edge approximately 40 km to the west,
at the edge of the sandstone--carbonate facies (Miall, 1970a). Assuming that
this prism extends for 300 km along strike, the volume of sediment pre-
served is approximately 3000 km 3. In Somerset Island the width of the
clastic wedge varies from 40 to 70 km and the preserved thickness ranges up
122

to approximately 750 m. The volume of sediment is estimated at 5100 km 3.

The total volume of the clastic wedge on either side o f Boothia Uplift, there-
fore, is of the order o f 8100 km 3, which is less than one third the volume of
sediment estimated to have been removed from the uplift (no estimate can
be made of the depth of erosion of the Precambrian rocks). It cannot be
argued that post-Devonian erosion of the Peel Sound Formation has removed
part or all of the missing sediment, because clast changes within the forma-
tion demonstrate that the Paleozoic rocks were largely removed during depo-
sition of that part of the Peel Sound succession that is now preserved. Rivers
flowing northward from Somerset Island and west of the Boothia Uplift may
have been a major conduit of sediment dispersal, out of the present project
area and into the region of the Franklinian Geosyncline, and much carbonate
material may have been carried away in solution.
Modern sedimentation rates in alluvial environments range from 0.08 m
to 12 m/1000 years depending on tectonic and climatic setting, and size of
dispersal system (Gole and Chitale, 1966; Beaty, 1970; Weimer, 1970;
Ryder, 1971; Smith, 1973; Allen, 1974). At an average rate of 1 m/1000 yrs
the 600 m of braided river and `alluvial deposits near Pressure Point would
have accumulated in less than 1 m.y. which is a small fraction of the time
available for Peel Sound sedimentation (the Ludlovian to Gedinnian time
period was 26 m.y. long, according to Braziunas, 1975). Schwab (1976) has
shown that sedimentation rates deduced from ancient rocks c o m m o n l y are
an order of magnitude less than those measured in m o d e m environments,
possibly because of periods of erosion or non<leposition that are not allowed
for in observations of the Recent. Schwab (1976) gave rates ranging from
0.04 m to 0.1 m/1000 yrs for molasse basins. According to these values the
Peel Sound clastic rocks near Pressure Point would have taken 6 to 15 m.y.
to accumulate.
For purposes of comparison, a calculation of denudation rate is given,
based on the highest denudation rate recorded at the present day. This is 800
tons/mi2/yr, in semi-arid areas with little vegetation cover to inhibit r u n o f f
(Langbein and Schumm, 1958). Assuming a mean rock density of 2.6 this
rate is equivalent to the removal of 0.003 km3/yr over the 28 000 km 2 of
Boothia Uplift, and it would take 9.3 m.y. to erode to a depth of 1 km over
the entire area.
Further attempts to investigate denudation and sedimentation rates are
hampered by a lack of knowledge of the complete age ranges of the Peel
Sound Formation and uncertainty over the a m o u n t o f sediment removed by
post-Devonian erosion. Both denudation and sedimentation rates may have
been much higher in the Devonian than at present because, even in humid cli-
mates, there was little vegetation to inhibit erosion and runoff. Although the
results given are very speculative, this section has been included in order to
demonstrate that, even in an area where the paleogeography is relatively
clear, and there are no structural complications, attempts to interpret
detailed relationships between a sediment wedge and its source area are beset
by difficulties.
 123

CONCLUDING REMARKS

This paper has attempted a regional paleogeographical analysis of an intra-

cratonic clastic succession deposited in various marine, tidal and fluvial
environments. Several problems o f interpretation have emerged, a summary
of which highlights certain areas of research that need further examination
b y clastic sedimentologists in modern environments and other ancient rock
sequences:
(1) Facies sequence analysis using Markov chain techniques is now estab-
lished as an invaluable aid to environmental interpretation. Data relating to
the nature of facies contacts (erosional or gradational) are critical to such
interpretations, b u t have not been analyzed rigorously in most previous
applications of the Markov method. A new technique, the contact matrix,
has been introduced herein to meet this problem, b u t it needs further testing
before its utility can be evaluated fully.
(2) In spite of the abundant literature and the numerous facies models
n o w available to aid in the analysis of fluvial rocks, the interpretations pre-
sented herein are still n o t fully satisfactory. Rivers can be braided,
meandering, anastomosing or straight with single or multiple channels, b u t
the range o f lithofacies and cycle types associated with this morphological
spectrum has y e t to be documented.
(3) Criteria for the interpretation of large-scale crossbedding and the
recognition of eolian depositional environments in ancient rocks are ambigu-
ous. The giant crossbeds of the Peel Sound Formation are an example of a
general problem, which has received considerable prominence recently over
the reinterpretation o f the Navajo Sandstone (Jurassic) of the southwestern
United States, long regarded as a classic example of an eolian deposit (Free-
man and Visher, 1975 and discussion t h e r e o f in J. Sediment. Petrol., 1977,
v. 47, pp. 475--497).
(4) Attempts to interpret rates o f denudation and sedimentation, and to
relate the volume of a clastic deposit to the a m o u n t o f uplift and erosion in
the source area, are beset b y many difficulties. The problems are particularly
well illustrated b y the Siluro-Devonian rocks of the Boothia Uplift region
because it is an area with a relatively well d o c u m e n t e d paleogeographic
history and lacking in structural complication. In short, an area that would
appear to provide an ideal subject for such an analysis.

ACKNOWLEDGMENTS

The authors are grateful to many of their colleagues for discussions of the
subject mattei~ contained in this report, particularly R. Thorsteinsson, J.W.
Kerr, B.R. Rust, G. Narbonne and D. Elliott. H.P. Trettin and J.R. McLean
critically read an early draft o f the manuscript.
Field work b y the senior author formed part of Operation Boothia, a
regional mapping project o f the Geological Survey of Canada, under the
 124

general leadership of J.W. Kerr; that by the junior author was part of a PhD
thesis at the University of Ottawa, supported by the National Research
Council of Canada, and the Department of Indian and Northern Affairs.
Both authors wish to acknowledge logistical support from the Polar Conti-
nental Shelf Project.
Field assistance was provided to Miall by M. Oliver, J. Cosgrove and C.E.
Miall, and to Gibling by J.M. Sempels and D. Elliott.

REFERENCES

Allen, J.R.L., 1963. The classification of cross-stratified units, with notes on their origin.
Sedimentology, 2: 93--114.
Allen J.R.L., 1970. Studies in fluviatile sedimentation: a comparison of fining-upward
eyclothems, with special reference to coarse-member composition and interpretation.
J. Sediment. Petrol., 40: 298--323.
Allen J.R.L., 1974. Studies in fluviatile sedimentation: implications of pedogenic car-
bonate units, Lower Old Red Sandstone, Anglo-Welsh outcrop. Geol. J., 9: 181--208.
Allen, J.R.L., Dineley, D.L. and Friend, P.F., 1967. Old Red Sandstone basins of North
America and northwest Europe. In: D.H. Oswald (Editor), International Symposium
on the Devonian System, Alberta Soc. Pet. Geol., 1 : 69--98.
Beaty, C.B., 1970. Age and estimated rate of accumulation of an alluvial fan, White
Mountains, California, Am. J. Sci., 268: 50--77.
Boothroyd, J.C. and Ashley, G.M., 1975. Process, bar morphology, and sedimentary
structures on braided outwash fans, northeastern Gulf of Alaska. In: A.V. Jopling and
B.C. McDonald (Editors), Glaciofluvial and glaciolacustrine sedimentation. Soc. Econ.
Paleontol. Mineral., Spec. Publ., 23: 193--222.
Braziunas, T.F., 1975. A geological duration chart. Geology, 3: 342--343.
Broad, D.S., 1968. Lower Devonian Heterostraci from the Peel Sound Formation, Prince
of Wales Island, Northwest Territories. Thesis, University of Ottawa (Unpublished).
Broad, D.S., 1973. Amphiaspidiformes (Heterostraci) from the Silurian of the Canadian
Arctic Archipelago. Geol. Surv. Can. Bull., 222, 35--51.
Broad, D.S. and Dineley, D.L., 1973. Torpedaspis, a new Upper Silurian and Lower
Devonian genus of Cyathaspididae (Ostracodermi) from Arctic Canada. Geol. Surv. Can.
Bull., 222: 53--91.
Brown, R.L., Dalziel, I.W.D. and Rust, B.R., 1969. The structure, metamorphism, and
development of the Boothia Arch, Arctic Canada. Can. J. Earth Sci., 6: 525--543.
Bryhni, I. and Skjerlie, F.J., 1975. Syndepositional tectonism in the Kvamshesten district
(Old Red Sandstone), western Norway. Geol. Mag., 112: 593--600.
Cant, D.J. and Walker, R.G., 1976. Development of a braided-fluvial facies model for the
Devonian Battery Point sandstone, Quebec. Can. J. Earth Sci., 13: 103--119.
Coleman, J.M., 1969. Brahmaputra River, channel processes and sedimentation. Sedi-
ment. Geol., 3: 129--239.
Curray, J.R., 1956. The analysis of two-dimensional orientation data. J. Geol., 64: 117--
131.
Dineley, D.L., 1966. Geological studies in Somerset Island, University of Ottawa Expedi-
tion, 1965. Arctic, 19: 270--277.
Dineley, D.L., 1975. North Atlantic Old Red S a n d s t o n e - some implications for Devo-
nian paleogeography. In: C.J. Yorath, E.R. Parker and D.J. Glass (Editors), Canada's
continental margins and offshore petroleum exploration. Can. Soc. Pet. Geol., Mere.
4, pp. 773--790.
Dineley, D.L. and Rust, B.R., 1968. Sedimentary and paleontological features of the
Tertiary--Cretaceous rocks of Somerset Island. Can. J. Earth Sci., 5: 791--799.
 125

Freeman, W.E. and Visher, G.S., 1975. Stratigraphic analysis o f the Navajo Sandstone. J.
Sediment. Petrol., 45: 651--668.
Friend, P.F., 1969. Tectonic features of Old Red Sandstone sedimentation in North
Atlantic borders. In: M. Kay (Editor), North Atlantic--geology and continental drift.
Am. Assoc. Pet. Geol. Mere. 12, pp. 703--710.
Garrett, P., 1970. Phanerozoic stromatolites: Noncompetitive ecologic restriction by
grazing and burrowing animals. Science, 169 : 171--173.
Gibling, M.R. and Narbonne, G.M., 1977. Siluro-Devonian sedimentation on Somerset
and Cornwallis Islands, Arctic Canada. Can. Soc. Petrol. Geol. Bull., 25: 1145--1156.
Glennie, K.W., 1970. Desert sedimentary environments, Developments Sedimentology.
Elsevier, no. 14, 222 pp.
Gole, C.V. and Chitale, S.V., 1966. Inland delta building activity of Kosi River, J.
Hydraul. Div. Am. Soc. Civ. Engrs. 92(HY2): 111--126.
Hagan, G.M. and Logan, B.W., 1974. History of Hutchison Embayment tidal flat, Shark
Bay, Western Australia. In: B.W. Logan et al., (Editors), Evolution and diagenesis of
Quaternary carbonate sequences, Shark Bay, Western Australia. Am. Assoc. Pet. Geol.
Mere. 22, pp. 283--315.
Hunter, R.E., 1976. Comparison of eolian and subaqueous sand-flow cross-strata. Am.
Assoc. Pet. Geol., 6 0 : 6 8 3 (abstract).
Jones, B., 1974. A biometrical analysis of Atrypella foxi n. sp. from the Canadian Arctic.
J. Paleont., 48: 963--977.
Jones, B. and Dixon, O.A., 1976. Storm deposits in the Read Bay Formation (Upper Silu-
rian), Somerset Island, Arctic Canada (an application of Markov chain analysis). J.
Sediment. Petrol., 46: 393--401.
Jones, B. and Dixon, O.A., 1977. Stratigraphy and sedimentology of Upper Silurian
rocks, northern Somerset Island, Arctic Canada. Can. J. Earth Sci., 14: 1427--1452.
Jones, B. and Oldershaw, A.E., 1977. Origin o f rubbly limestones in the Upper Silurian
Read Bay Formation, Somerset Island, Arctic Canada. Geol. Assoc. Can. Prog. with
Abs., 2: 27.
Kendall, C.G.St.C. and Skipwith, Bt., P.A.d'E., 1968. Recent algal mats of a Persian Gulf
lagoon. J. Sediment. Petrol., 38: 1040--1058.
Kerr, J.W., 1974. Geology of Bathurst Island Group and Byam Martin Island, Arctic
Canada (Operation Bathurst Island). Geol. Surv. Can. Mere., 378.
Kerr, J.W., 1977. Cornwallis Fold Belt and the mechanism o f basement uplift. Can. J.
Earth Sci., 14: 1374--1401.
Kerr, J.W. and Christie, R.L., 1965. Tectonic history of Boothia Uplift and Cornwallis
Fold Belt, Arctic Canada. Am. Assoc. Pet. Geol. Bull., 49: 905--926.
Kerr, J.W. and deVries, C.D.S., 1976. Structural geology of Somerset Island, District of
Franklin. Geol. Surv. Can., Paper 76--1A: 493--495.
Kinsman, D.J.J., 1969. Modes o f formation, sedimentary associations and diagnostic fea-
tures of shallow-water and supratidal evaporites. Am. Assoc. Pet. Geol., 53: 830---840.
Kinsman, D.J.J., 1971. Early diagenesis of carbonate sediments in a supratidal evaporitic
setting. Am. Assoc. Pet. Geol. Bull., 55: 167--168.
Krumbein, W.C. and Dacey, M.F., 1969. Markov chains and embedded Markov chains
in geology. J. Int. Assoc. Math. Geol., 1: 79--96.
Langbein, W.B. and Schumm, S.A., 1958. Yield of sediment in relation to mean annual
precipitation. Am. Geophys. Union, Trans., 39: 1076--1084.
Laporte, L.F., 1969. Recognition of a transgressive carbonate sequence within an epeiric
sea: Helderberg Group (Lower Devonian) of New York State. In: G.M. Friedman (Edi-
tor), Depositional environments in carbonate rocks. Soc. Econ. Paleontol. Mineral.,
Spec. Publ. 14: pp. 98--118.
Laporte, L.F., 1971. Paleozoic carbonate facies of the central Appalachian shelf. J. Sedi-
ment. Petrol., 41: 724--740.
 126

Leeder, M.R., 1973. Fluviatile fining-upward cycles and the magnitude of paleochannels.
Geol. Mag., 110: 265--276.
Logan, B.W., Hoffman, P. and Gebelein, C.D., 1974. Algal mats, cryptalgal fabrics, and
structures, Hamelin Pool, Western Australia. In: B.W. Logan et al. (Editors), Evolution
and diagenesis of Quaternary carbonate sequences, Shark Bay, Western Australia. Am.
Assoc. Pet. Geol. Mem. 22, pp. 140--194.
Logan, B.W., Rezak, R. and Ginsburg, R.M., 1964. Classification and environmental signi-
ficance o f algal stromatolites, J. Geol., 7 2: 68--83.
Mayr, U., in press. Stratigraphy and correlation of Lower Paleozoic subsurface sections,
Cornwallis, Devon, Somerset, and Russell Islands, Canadian Arctic Archipelago. Geol.
Sure. Can. Bull.
McGowen, J.H. and Garner, L.E., 1970. Physiographic features and stratification types of
coarse-grained point bars: modern and ancient examples. Sedimentology, 1 4 : 7 7 - - 1 1 2 .
McLean, J.R., 1976. Cadomin Formation: eastern limit and depositional environment.
Geol. Sure. Can., Pap. 76--1B, pp. 323--327.
Miall, A.D., 1969. The sedimentary history of the Peel Sound Formation, Prince of Wales
Island, Northwest Territories. Thesis, University of Ottawa (unpublished).
Miall, A.D., 1970a. Continental-marine transition in the Devonian o f Prince of Wales
Island, Northwest Territories. Thesis, University of Ottawa (unpublished).
Miall, A.D., 1970b. Devonian alluvial fans, Prince of Wales Island, Arctic Canada, J. Sedi-
ment. Petrol., 40: 556--571.
Miall, A,D., 1973. Markov chain analysis applied to an ancient alluvial plain succession.
Sedimentology, 20: 347--364.
Miall, A.D., 1974. Paleocurrent analysis of alluvial sediments: a discussion of directional
variance and vector magnitude. J. Sediment. Petrol., 44 : 1174--1185.
MiaU, A.D., 1977. A review of the braided river depositional environment. Earth-Sci. Rev.,
13: 1--62.
Miall, A.D. and Kerr, J.W., 1977. Phanerozoic stratigraphy and sedimentology of Somer-
set Island and northeastern Boothia Peninsula. Geol. Sure. Can., Pap. 77--1A, pp.
99--106.
Miall, A.D., Kerr, J.W. and Gibling, M.R., 1977. The Somerset Island Formation: an
Upper Silurian to Lower Devonian intertidal/supratidal succession, Boothia Uplift
region, Arctic Canada. Can. J. Earth Sci., 15: 181--189.
Moody-Stuart, M., 1966. High and low sinuosity stream deposits, with examples from the
Devonian of Spitzbergen. J. Sediment. Petrol., 36 : 1102--1117.
Rahmani, R.A. and Lerbekmo, J.F., 1975. Heavy mineral analysis of Upper Cretaceous
and Paleocene sandstone in Alberta and adjacent areas of Saskatchewan. In: W.G.E.
Caldwell (Editor), The Cretaceous System in the Western Interior of North America.
Geol. Assoc. Can. Spec. Pap. 13, pp. 607--632.
Reineck, H.E. and Singh, I.B., 1973. Depositional sedimentary environments. Springer-
Verlag, Berlin 438 pp.
Reinson, G.E., Kerr, J.W. and Stewart, W.D., 1976. Stratigraphic field studies, Somerset
Island, District of Franklin (58B to F). Geol. Sure. Can. Paper 76--1A, pp. 497--499.
Riba, O., 1976. Syntectonic unconformities of the Alto Cardener, Spanish Pyrenees: a
genetic interpretation. Sediment. Geol., 15: 213--233.
Ryder, J.M., 1971. The stratigraphy and morphology of para-glacial alluvial fans in south-
central British Columbia. Can. J. Earth Sci., 8: 279--298.
Rust, B.R., 1972. Structure and process in a braided river. Sedimentology, 18: 221--246.
Schwab, F.L., 1976. Modern and ancient sedimentary basins: comparative accumulation
rates. Geology, 4 : 7 23--7 27.
Schwartz, H.-U., Einsele, G. and Herm, D., 1975. Quartz-sandy, grazing-contoured stro-
matolites from coastal embayments of Mauritania, West Africa. Sedimentology, 22:
539--561.
 127

Shearman, D.J., 1966. Origin of marine evaporites by diagenesis. Trans. Inst. Min. Met.,
75: 207--215.
Shinn, E.A., Lloyd, R.M. and Ginsburg, R.N., 1969. A n a t o m y of a modern carbonate
tidal-flat, Andros Island, Bahamas. J. Sediment. Petrol., 39: 1202--1228.
Smith, D.G., 1973. Aggradation of the Alexandra-North Saskatchewan River, Banff Park,
Alberta. In: M. Morisawa (Editor), Fluvial geomorphology. Proc. 4th Am. Geomorph.
Symp., Publ in Geomorph., SUNY--Binghamton, New York, N.Y., pp. 201--219.
Steidtmann, J.R., 1976. Eolian origin of sandstone in the Casper Formation, southern-
most Laramie Basin, Wyoming. In: R.C. Epis and R.J. Weimer (Editors), Studies in
Colorado field geology. Prof. Contr. Colorado School of Mines no. 8, pp. 86--95.
Thompson, R.W., 1968. Tidal flat sedimentation on the Colorado River delta, north-
western Gulf of California. Geol. Soc. Am. Mere. 107.
Thorsteinsson, R., 1958. Cornwallis and Little Cornwallis Islands, District o f Franklin,
Northwest Territories. Geol. Surv. Can. Mem. 294.
Thorsteinsson, R. and Kerr, J.W., 1968. Cornwallis Island and adjacent smaller islands,
Canadian Arctic Archipelago. Geol. Surv. Can. Pap. 67--64.
Thorsteinsson, R. and Tozer, E.T., 1963. Geology of northern Prince of Wales Island and
northwestern Somerset Island. Geol. Surv. Can. Mem. 320, pp. 117--129.
Uyeno, T.T., 1977. Summary of conodont biostratigraphy o f the Read Bay Formation at
its type sections and adjacent areas, eastern Cornwallis Island, District of Franklin.
Geol. Surv. Can., Pap. 77--1B, pp. 211--216.
Veizer, J., Lemieux, J., Jones, B., Gibling, M.R. and Savelle, J., 1977. Sodium: Paleo-
salinity indicator in ancient carbonate rocks. Geology, 5 : 177--179.
Walker, T.R. and Harms, J.C., 1976. Eolian origin o f Flagstone Beds, Lyons Sandstone
(Permian) t y p e area, Boulder County, Colorado. In: R.C. Epis and R.J. Weimer (Edi-
tors), Studies in Colorado field geology. Prof. Contr. Colorado School of Mines no. 8,
pp. 110--122.
Weimer, R.J., 1970. Rates of deltaic sedimentation and intrabasin deformation, Upper
Cretaceous of Rocky Mountain region. In: J.P. Morgan (Editor), Deltaic sedimentation
modern and ancient. Soc. Econ. Paleontol. Mineral. Spec. Publ. 15, pp. 270--292.
Williams, G.E., 1971. Flood deposits o f the sand bed ephemeral streams of central Austra-
lia. Sedimentology, 17: 1--40.
Williams, P.F. and Rust, B.R., 1969. The sedimentology o f a braided river. J. Sediment.
Petrol., 39: 649--679.
Woodrow, D.L., Fletcher, F.W. and Ahrnsbrak, W.F., 1973. Paleogeography and paleo-
climate at the deposition sites of the Devonian Catskill and Old Red facies. Geol. Soc.
Am. Bull., 84" 3051--3064.