H. E. (1966). J . appl. Bact. 28 (3), 403-411.

SCHEFFERLE,

The Microbiology of Built Up Poultry Litter

HENRIETTA E. SCHEFFERLEI
Department of Bacteriology, School of Agriculture, Edinburgh, Scotland

(Received 10 May, 1965)

SUMWY. The numbers of viable bacteria in built up poultry litter were found t o be
101o-lO1l/g fresh weight and appeared to be little affected by factors such as age,
temperature, moisture content and pH. Counta for unused litter and poultry droppings
were lower. I n built up litter of high alkalinity coryneform bacteria were predominant;
micrococci occurred sporadically and smaIl numbers of nocardias, streptomycetes,
aerobic spore formers and streptococci were encountered. A variety of Gram negative
bacteria also occurred, the numbers of which appeared to be controlled by alkalinity;
they were less abundant in litters where the pH and buffering capacity were high.
Strongly alkaline conditions also tended to lower the fungal counts but had no effect
on the count of enterococci.

TEE BUILT UP or deep litter system of poultry keeping is one of the intensive
methods of management whereby the birds are confined indoors during their
productive life. The method consists in starting with about 4-6 in. of peat moss,
chopped straw, wood shavings or some other form of litter and adding to this
periodically until a depth of 10-12 in. is attained. I n this way excess moisture from
the droppings of the birds is absorbed and the litter is kept in a moderately dry
and friable state. Under these conditions decomposition of the droppings takes
place without the production of offensive odours; ammonia is liberated and the
material becomes markedly alkaline,
The only recorded investigation into the microbiology of built up poultry litter
is that of Halbrook, Winter & Sutton (1951), who examined litter made from ground
corn cobs, shavings or bark and found the same trend in their microfloral content.
I n fresh litter under birds they observed during the first 8 weeks a general increase
in the total count at 37" and in the counts of lactobacilli, enterococci, coliform
bacteria, moulds and yeasts. I n litter in use for over a year, however, the numbers
of lactobacilli, coliform bacteria, moulds and yeasts had decreased appreciably, but
the enterococci had decreased only slightly. The aerobic pIate count remained
virtually unchanged with increasing age of the litter and always a t a higher level
than could be accounted for by the groups determined; therefore Halbrook et al.
(1951) postulated the existence of other bacteria not included in the differential
analysis. The general reduction in the counts of lactobacilli, coliform bacteria, moulds
and yeasts was thought to be due to the increase in pH of the litter because the
effect could be accentuated by the addition of lime. It might also be significant
that the total aerobic count was determined a t 37", a temperature a t which many
organisms originating from soil would fail to grow.
* The present address of the author is: Scottish Home and Health Department, 1India Buildings,
Victoria Street, Edinburgh 1, Scotland.
404 Henrietta E. Schefferle
The work described below is concerned with the numbers and types of micro-
organisms found in poultry litter of varying nature; one sample of droppings was
also examined. Although lactobacilli are numerous in the intestine of poultry, many
grow poorly or not at all under aerobic conditions (Shapiro, Rhodes & Sarles, 1949;
Harrison & Hansen, 1950a,b) and would not be expected to multiply under the
conditions of aeration and pH existing in built up litter. The occurrence of this
group of bacteria has therefore not been considered in the following investigation.

Materials and Methods

Nature of samples examined
A description of the samples and details of temperature, moisture content and pH
are lieted in Table 1.

T~LBLE 1
Nature of the samples of poultry titter and droppings examined
Mean
Sample TJrpe Age Month temp. Dry PH
number of litter sampled when matter
sampled ( %)
("C)
18 Chopped straw Before chicks June 4.3
and peat moss introduced
lb 1, 3 months Sept. 8.4
lo ,, 9 months March 6.6
2 - 2 months June 8.6
3 - 4 months May 8.9
4& Chopped straw 1 week May 16.0 884 7.2
and peat moss
4b ,* 6 weeks June 19.6 79.0 7 -0
40 I, 9 weeks JdY 18.9 69.6 8.0
4d , 3 months Aug. 20.0 72.8 7 43
40 ,, 44 months Sept. 12.9 76.8 7.6
4f ,, 9 months Feb. 8.0 54.2 6.4
5 Chopped straw 10 weeks Oct. 18.6 79.4 7.4
6 Chaff, peat 7 months June 21.0 87.6 8.1
moss and straw
7 Sawdust and 9 months June 26.0 68.2 8.5
oow manure
8 Droppings from
birds on 23.6
litter 4a
Samples having the same number are from the same litter at different ages.
- ,Data incomplete, but probably chopped straw and peat moss.

Collection of samples
The surface layer containing fiesh droppings was first removed. A sample was then
taken from the remaining depth of litter, the process being repeated a t several
points in the poultry house. The samples were thoroughly mixed and a portion
taken for bacteriological examination and for pH and moisture determinations.
 Microbiology of poultry litter 405
Handling of the samples was done as aseptically as possible. At the time of sampling
the temperature was taken a t a point midway down in the litter and at random
positions throughout the house, the results being averaged.
The sample of droppings was collected from the surface of one litter, care being
taken to avoid contamination from the litter material.

Moisture content and p H of samples

The moisture content was found by drying 60 g of the material a t 100" overnight.
pH was measured electrometrically (Cambridge Instrument Co., Ltd., Grosvenor
Place, London, S.W.l) on a portion of the suspension, 1 in 10 in water, prepared for
bacteriological examination.

Preparation of samples for bacteriological examination

Long lengths of straw were f i s t cut into small pieces. A 30 g sample was transferred
to a previously sterilized macerator jar, 270 ml of sterile tap water were added and
the whole was treated in a top-drive macerator for 1 min. Decimal dilutions were
then made in tap water. For most of the work described below only the higher
dilutions containing relatively low numbers of bacteria were suitable.
One ml quantities of appropriate dilutions were used for making plate counts.
In later work 0.1 ml of the inoculurn was spread over the surface of a dried plate
using a bent glass rod. This latter method had the advantage of showing the features
of surface colonies. Platings of each dilution were made in triplicate and carried
out as rapidly as possible after the dilutions were prepared.

Estimations of numbers of micro-organisms

All media were sterilized a t 128" (22.5 lb steam pressurelina) momentarily.
Total viable counts and the bacterial types present
Total viable counts were found to be best determined on nutrient agar containing ;
peptone (Difco-Bacto),6 g; peptone (Evans), 6 g; meat extract (Oxoid Lab-Lemco),
10 g; NaCl, 6 g; agar (Davis), 16 g; tap water, 1000 ml; pH 7. Plates were incubated
a t 22" for 10 days.
From the incubated. plates approximately 30 colonies were picked a t random so
that the relative abundance of the different bacterial types might be estimated.
Each colony was transferred to a nutrient agar slope and incubated at 22"; as soon
as growth appeared a Gram stained preparation was made and the young culture
was examined for motility.
Gram negative bacteria
From the examination of the colonies on the total count plates there was some
indication that the relative abundance of Gram negative bacteria might be influenced
by the pH of the sample. To investigate this further, plate counts for Gram negative
bacteria were made using nutrient agar containing 1 :600,000 of crystal violet
incubated a t 22' for 10 days. When the plates were crowded, the dye in this medium
became concentrated in many of the colonies and Gram positive bacteria then
406 Henrietta E. Schefferle

developed. To overcome this difficulty plates with (100 colonies were always
selected for counting. Streptococci occasionally appeared on these plates but their
colonies were tiny and easily recognized.
Enterococci
Enterococci are present in considerable numbers in the intestinal contents of
poultry and since they tolerate a high p H they probably withstand the alkaline
conditions in built up litter.
The medium used for their detection contained; peptone (Evans), 10 g ; meat
extract (Oxoid Lab-Lemco), 10 g ; glucose, 5 g; yeast autolysate, 50ml; agar (Davis),
15 g ; tap water, 1000 ml; p H 6. The autolysed yeast was prepared by holding 1
kg of brewer's yeast in 1 1 of water a t 50" for 24 h ; after centrifuging, the
supernatant was sterilized in bottles. Plates were incubated a t 45" for 2 days. The
p H of this medium was somewhat critical because a t p H 7 coli-aerogenes bacteria
(Escherichia coli type I) tended to outgrow the streptococci.
Moulds and yeasts
The colony counts of moulds and yeasts were estimated by surface plating on:
glucose, 10 g ; peptone (Evans), 5 g ; K,HPO,, 0-5 g; agar (Davis), 15 g ; tap
water, lo00 ml. The p H was adjusted to 3 4 4 . 0 by adding 1 ml of 0.5 N HaSO,
to 100 ml of the melted medium immediately before use. After 5 days a t 22" mould
and yeast colonies could be recognized.

Results
Total bmtericll counts and the types of bacteria predominantly present
The counts obtained on the various samples are summarized in Table 2. The counts
for the built up litter were all of the same order of magnitude, being in the range
10,800~ 1O6-153,0O0x 106/g of fresh material whereas those for unused litter (sample
l a ) and droppings (sample 8) were somewhat lower.
Examination of the colonies on high dilution plates showed that the majority
of organisms could be placed in one of three groups, coryneform bacteria, micrococci
or Gram negative types. Other bacteria such as aerobic spore formers, nocardias,
streptomycetes and streptococci were found only occasionally. From the results it
can be seen that coryneform bacteria became the predominant bacteria in built up
litter, particularly if the pH was high; the occurrence of micrococci was more erratic
and there seemed t o be no relationship between their numbers and the nature of
the litter. The percentages of the Gram negative types varied considerably but
there seemed to be fewer present in those litters of high pH.
Although the sample of droppings showed a lower total count than did built u p
litter samples, the proportions of the various types of organisms were very similar,
and on more detailed examination the strains isolated appeared to be identical with
those normally found in built up litter. This suggests that either the sample of
droppings had become contaminated or the organisms ingested from the litter
retained their viability when passing through the intestine of the bird. That the
 TABLE2
~ Total b a c t e d counts and the predominant types of bacteria in
poultry litter and droppings
Percentage of E
A -
Plate count r - c
Sample ( x 10"g Coryneform Micro- Gram Aerobic strepto- Strepto- 0
number* freah wt) btMteria cocci negative spore Nocardias mycetes cocci w
C.
types formers
0,
0
18 188 16 8 68 0 0 8 0
lb 153,000 83 8 0 3 0 3 3 3
IC 14,700 66 2 40 2 0 0 0 0
2 49,000 72 3 12 3 5 0 5 r
3 21,000 66 31 0 0 0 0 3 cd
0
pa 86,000 62 0 35 0 0 0 3
4b 35,000 41 3 48 0 0 3 3 5
4c 30,000 74 0 19 0 0 3 3
4d 36,000 65 0 35 0 0 0 0 .c"
48 47,000 76 10 10 3 0 0 0
rlf 10,800 60 0 31 0 6 3 0 6
5 72,000 76 7 14 0 0 3 0 (D
1
6 35,000 74 17 9 0 0 0 0
7 36,000 80 8 4 0 8 0 0
8 88 72 9 19 0 0 0 0
* For description of samples, sea Table 1.
408 Henrietta E. Schefferle
latter is possible was shown in later work when the caecal contents of a battery kept
hen were examined; coryneform bacteria and aerobic Gram negative bacteria were
among the isolates from platings on nutrient agar. Unfortunately no similar
examination was made on a hen kept under the built up litter system.
Gram negative bacteria
Counts of Gram negative bacteria determined on the crystal violet medium were
compared with those estimated from their percentage representation on the total
count plates; this showed that the selective medium detected 8-40% of the estimated
number. As each calculation was based on the examination of only about 30 colonies
it was felt that, although the selective medium was not giving a n absolute count
of the Gram negative bacteria, it was giving a roughly constant proportion so that
the results could be used for following fluctuations in the number of Gram negative
organisms.
No direct relationship appeared t o exist between the numbers of Gram negative
bacteria (counted either on the selective medium or on the total count plates) and
p H of samples but there did seem to be a correlation between numbers and percentages
of the total count and the amounts of acid required t o adjust the pH of 300 ml of a
1 : l O litter suspension to a value of 5 (Table 3). It would appear, therefore, that
the relative incidence of Gram negative bacteria in built up litter is diminished by
alkaline conditions and more effectively so if the material is well buffered.
TABLE3
The incidence of Gram negative bacteria in poultry litter and the
volumes of acid required to adjust samples to p H 5
count
( x loe/@;
fresh wt) % of No. of ml of
Sample of Gram negative Gram negative PH N H,S04 required
number* bacteria on & y s t a l bacteria on of sample to adjust 30 g of
violet medium total count plates litter to pH 6
4f 1,020 31 6.4 2.6
4d 2,900 36 7.6 8.8
4b 2,400 48 7.9 8.9
4a 8,900 35 7.2 9.1
4c 1,000 19 8.0 9.4
4e 870 10 7.6 11.8
6 820 14 7.4 13.0
6 560 9 8.1 13.6
7 700 4 8.5 2043
~

For description of samples, see Table 1.

Enterococci
Only enterococci grew on high dilution plates prepared from built up litter or
droppings and incubated at 45". They were identified by morphology, ability t o
grow at 45" and production of ammonia from arginine; all produced acid and reduced
the dye in litmus milk, in most cases this was followed by curdling. Species
differentiation within the group was not made. Their occurrence in built up litter
was consistent (Table 4)and variations in their numbers did not seem t o be related
to any characteristics of the litter samples.
 Microbiology of poultry litter 409
TABLE4
The types and numbers of bacteria f r m poatltry litter
and droppings capable of growth at 46"
Sample Plate count Types of bacteria
number* ( x 10a/gfresh wt) present
Unused litter
la 15 Bacillua licltengormis;
Escherichia coli type I;
unidentified
streptomycetes
Built up litter
lb 163,000
2 423,000
4a 82,000
4b 13,000 Enterococcoci
4c 14,000
4d 7,000
40 11,600
4f 2,700
5 60,000
7 56,000
Poultry droppings
8 1,100 Enterococci
* For description of samples, see Table 1.

The count a t 45" on the unused litter (sample l a ) was much lower than those for
built up litter and did not represent a count of enterococci. They were Bacillus
licheniformis (identified by the key proposed by Gibson & Topping, 1938), E . coli
type I end unidentified streptomycetes. I n the absence of strongly acid-forming
bacteria such as the enterococci, faecal coli-aerogenes bacteria were evidently able
to grow on the medium a t p H 6.

TABLE5

Sample
number*,

Built up litter
4a
p H of
sample

7.2
-
Plate counts of moulds and yeasts in poultry litter and droppings
Plate count
( x 108/gfresh wt) of

Moulds

11,000
Yeasts

88,000
4b 7.9 20,000 3,000
4c 8.0 5,400 1,000
4d 7.6 3,100 200
40 7.6 920 90
4f 6.4 1,300 1,000
5 7.4 4,200 <100
6 8.1 70 180
7 8.5 100 20
Poultry droppings
8 130 1,050
* For description of samples, see Table 1.
4'0 Henrietta E. Schefferle
Moulds and yeasts
Colony counts for moulds and yeasts are given in Table 5. From the figures for
litter sample 4 it may be aeen that there was a slight decrease in the numbers of
moulds and y w t s with increasing age of the litter. The reduction in counts observed
for this litter may have been due to the continuance of slightly alkaline conditions
in the older samples because the counts increased again in sample 4f in which the
pH had dropped to 6-4., The other samples of built up litter also show relatively
low counts with high pH values. Counts for poultry droppings are included for
comparison.

Discussion
When the droppings of hens are added to litter rapid multiplication of bacteria
appears to take place. The plate counts obtained in the present work from built up
litter were somewhat higher than those previously reported by Halbrook et al. (1951)
who, however, made counts at a temperature (37")which would inhibit some of the
bacteria likely to be present. Generally the total counts found during the present
investigation were all of the same degree of magnitude and seemed to be independent
of the age of the material. Slightly lower counts were found in two samples collected
during the winter; these were both cold and wet. Low temperatures in poultry litter
would be expected to restrict multiplication of the bacteria present ; in addition
the droppings tend to remain wet and the litter may become sodden, a condition
not conducive to the growth of the bacteria, the majority of which are strict aerobes.
I n normal built up litter bacterial decomposition of uric acid results in a rise in pH.
The relatively low pH of the samples collected in the winter possibly resulted from
the continual addition of droppings, normally slightly acidic, unaccompanied by
active decomposition of the uric acid.
The bacteria found on high dilution plates could be divided into three main
groups, coryneform bacteria, micrococci and Gram negative types. Although the
total counts showed little variation, the proportions of the different types varied
with the alkalinity of the samples. Generally, the greater the alkalinity the higher
was the proportion of coryneform bacteria and the lower the proportion of the Gram
negative types.
The number of enterococci found in the sample of poultry droppings was com-
parable with that reported for the faeces of hens (Johansson, Sarles & Shapiro, 1948)
and the colon contents of chickens (Shapiro & Sarles, 1949). The numbers found
in built up litter samples tended to be slightly higher and because the droppings
become diluted by the litter it seems certain that these organisms multiply after
excretion. Their numbers showed no correlation with the nature of the litter. The
results obtained in this investigation are in marked contrast to those of Halbrook
et al. (1951) whose counts for enterococci in poultry droppings and litter were never
>lOO/g. It is difficult to understand why they were so low eince the method of
enumeration was the same as that used by Johansson et al. (1948) and Shapiro t
Sarles (1949).
Fungi would not be expected to be favoured by alkalineconditions and, in agreement
with the results of Halbrook et al. (1951)) this was confirmed in the present work.
 Microbiology of poultry litter 41*
I t has been suggested that when using the built up litter system of poultry
management soil or old poultry litter should be used to inoculate fresh material.
Possibly some benefit from this practice would be derived in new houses, but in
old houses the bacteria responsible for the decomposition processes would be present
in adequate numbers and would certainly multiply provided the conditions of
temperature, aeration and moisture content were suitable.

The author wishes to express her thanks to Dr. T. Gibson for his guidance during
the course of this work and for helpful criticism given during the preparation of the
paper,

References
GIBSON, T. & TOPPING, L. E. (1938). Further studies of the aerobic spore-formingbacilli. Proc.
SOC.agric. Bact. p. 43.
HALBROOK, E. R., WINTER,A. R. & SUTTON, T. S. (1951). The microflora of poultry house
litter and droppings. Poult. Sci. 30, 381.
HARRISON, A. P. JR. & HANSEN, P. A. (1950a). The bacterial flora of the cecal feces of healthy
turkeys. J. Bact. 59, 197.
HARRISON, A. P. JR. & HANSEN,P. A. (1950b). Lactobacilli from turkeys. J. Bact. 60, 643.
JOHANSSON, K. R., SARLES,W. B. & SHAPIRO, S. K. (1948). The intestinal microflora of hens
as influenced by various carbohydrates in a biotin-deficient ration. J. Bact. 56, 619.
SHAPIRO, S. K., RHODES, R. A. & SARLES, W. B. (1949). Lactobacilli in the intestinal tract of
the chicken. J. Bact. 58, 689.
SEAPIRO, S. K. & SARLES,W. B. (1949). Micro-organisms in the intestinal tract of normal
chickens. J. Bact. 58, 631.