UNIVERSITY OF CALIFORNIA

COLLEGE OF AGRICULTURE

AGRICULTURAL EXPERIMENT STATION

BERKELEY, CALIFORNIA

BREEDING
FOR EGG PRODUCTION

LEWIS W. TAYLOR and I. MICHAEL LERNER

BULLETIN 626

October, 1938

UNIVERSITY OF CALIFORNIA
BERKELEY, CALIFORNIA

CONTENTS

PAGE

Explanation of terms used in breeding 3

Historical development of methods of poultry breeding 5

Characters governing the annual egg record 8

Sexual maturity 10

Pauses 13

Persistency 16

Rate of production 17

General considerations of production characters 20

Characters affecting viability 21

Longevity 22

Disease resistance 22

Methods of measuring flock production 23

Characters governing the quality of eggs 24

Egg size 24

Egg shape 27

Egg color 27

Shell quality 28

Albumen quality 29

Hatchability 30

Breeding systems 30

Close inbreeding and line breeding 31

Outbreeding and crossbreeding 32

Grading 33

Value of breeding systems 33

Relation of genetics to poultry-breeding methods 33

Pedigree breeding 33

Pedigree records 34

Sib and progeny testing 35

Multiplication of superior stock 37

A breeding program for multipliers 39

Age of breeding stock 40

What breeding can accomplish 41

Egg production 41

Viability 43

Importance of other conditions 44

Literature cited 45

BREEDING FOR EGG PRODUCTION

LEWIS W. TAYLOR 2 and I. MICHAEL LEKNER 3

The purpose of this bulletin is twofold : (1) to review the present knowl-
edge of the genetics of egg-production and egg-quality characters and
(2) to demonstrate what may be accomplished by the application of the
principles and methods discussed in the text in bringing about rapid im-
provement in egg production. For this purpose, data from the California
Agricultural Experiment Station flock, as well as from literature on
poultry breeding, are used.

In the limited space of a bulletin, only a small part of the published
work on the subject can be reviewed. Only those papers which are di-
rectly drawn upon are cited. Similarly, the analysis of the Experiment
Station data has necessarily been confined to the more important general
considerations, with many details omitted. These data confirm in a strik-
ing manner the adequacy of the progeny-test method of breeding poultry,
when this method is applied to the individual characters affecting pro-
duction and viability. We believe that the application of similar princi-
ples should enable a poultry breeder to accomplish substantially the same
type of flock improvement. The presentation of dogmatic methods of pro-
cedure for pedigree breeders has purposely been avoided because con-
ditions vary widely from flock to flock. Emphasis is thus laid on broad
principles which may be adapted to specific circumstances found in dif-
ferent flocks. However, a more definite breeding program for multipliers'
flocks is advanced.

EXPLANATION OF TERMS USED IN BREEDING

Inheritance of characters is based on the transmission of genes from
parent to offspring. Genes are defined as the determiners of hereditary
characters. They occur in pairs (with a few exceptions), one member of
each pair being derived from the sire, and its mate from the dam.

One member of the pair may be different from its mate, as in the case
of a cross between two breeds. For example : A bird resulting from a cross
between a purebred rose-combed and a single-combed bird will receive a
gene for rose comb from one parent and for single comb from the other.
The crossbred bird will exhibit a rose comb, since the rose-comb gene is

1 Eeceived for publication April 19, 1938.

2 Associate Professor of Poultry Husbandry and Associate Poultry Husbandman
in the Experiment Station.

3 Instructor in Poultry Husbandry and Junior Poultry Husbandman in the Experi-
ment Station.

[3]

4 University of California — Experiment Station

dominant and prevents the expression of the single-comb gene, which is
known as a recessive. If two such crossbred birds are mated together, in
the next generation both rose- and single-combed individuals will appear
in the ratio of three rose-combed birds to one single-combed bird. Of the
rose-combed birds, one-third will be pure-breeding, and two-thirds will
not breed true (possessing like their parents, one rose-comb gene and one
single-comb gene) . The birds which possess two like genes and breed true
are homozygous for this gene. In the example given, the pure-breed-
ing rose-combed and the single-combed birds are homozygous for comb-
shape genes. Those rose-combed birds which possess two unlike genes will
not breed true and are heterozygous.

Various genes are usually designated by arbitrary letters. Capital let-
ters indicate the dominant condition, and small letters the recessive : the
gene for rose comb, for example, is represented as B and that for single
comb as r. The actual constitution of an individual with respect to these
genes is known as the genotype, which is designated by the genes of which
it is composed : the genotype of the single-combed bird, for example, is
represented by rr, that of the pure-breeding rose-combed bird by BB, and
that of the heterozygous rose-combed bird by Br.

The appearance of the bird with respect to these characters is called
the phenotype. In the example given, there are only two phenotypes —
single-comb, to which the bird with the rr genotype belongs, and the rose-
comb, to which the birds with both the BB and Br genotypes belong. The
two genotypes that show the rose-comb phenotype can be distinguished
only by breeding tests. If a rose-combed bird, bred to a single-combed
bird gives only rose-combed offspring, then it must belong to the BB
genotype, but if it gives some single-combed offspring, then it must be-
long to the Br genotype.

At times, when two unlike genes are present in the pair governing the
expression of a character, neither gene completely suppresses the action
of its mate. The case is then one of incomplete dominance. When this
occurs, the mating of two such heterozygous individuals produces three
phenotypes in the offspring, which correspond to the three genotypes
present.

In cases where characters are produced by the action of a large num-
ber of genes, such are known as multiple genes.

Sex-linked inheritance can probably be best explained by an example.
A simple sex-linked character in poultry is the barred plumage of the
Barred Plymouth Rock. When a male of this variety is crossed with a
black female, all the progeny is found to be barred. In a reciprocal cross,
however, when a black male is mated to a barred female, in the first gen-

Bul. 626] Breeding for Egg Production 5

eration, the females are found to be black, while the males are barred.
Sex-linked crosses for early sex identification are based on this principle.
The reason for the behavior of the sex-linked genes lies in the fact that
the male possesses a paired complement, as in the case of all other genes,
while the female has only one member of a sex-linked gene pair. Thus the
male genotype in the case of the pure Barred Plymouth Rock is BB,
while the female genotype is B-. In the case of the black birds, the male's
constitution is bb and the female's b-. In the second cross as described
above, bb x B- (black male x barred female), the male progeny will be
heterozygous barred (Bb) and the females will be black (5-) .

HISTORICAL DEVELOPMENT OF METHODS OF
POULTRY BREEDING

The history of the development of egg-production characters in strains
of chickens goes back centuries before there was any indication of a com-
mercial poultry industry. The types of fowl developed in the countries
bordering on the Mediterranean Sea were early recognized to have a
greater inherent capacity for producing eggs than birds from the Orient.
From these higher-producing fowl, the breeds of the Mediterranean class
of chickens, which includes Leghorns, Minorcas, and Anconas, were de-
veloped. During an intensive period of breed creation in the nineteenth
century, fowls of the Mediterranean class were extensively used to
cross with Asiatic breeds or with mongrel stocks to produce new breeds
and varieties, which are found in the American and English classes of
chickens.

With the development of improved systems of transportation and
marketing, accompanied by an increased demand for eggs for human
consumption at seasons other than those of normal high egg production,
an increasing importance was attached to the inherent ability of birds to
lay. Under some conditions of poultry keeping, the value of a breed no
longer rested entirely on its size and quality of meat, on its plumage
color or pattern, or on its peculiar appearance as evidenced by structural
differences in comb, feathers, or number of toes. Thus, by the middle of
the nineteenth century, writers on poultry subjects began to emphasize
the ability of certain breeds to lay a relatively large number of eggs.
Before the end of that century, the idea of trapnesting birds in order to
obtain exact records had been conceived and suitable devices to accom-
plish this work had been invented.

At the close of the nineteenth century, an experimental approach to
the solution of problems involved in improving egg production was made
by Gowell at the Maine Agricultural Experiment Station on a flock of

6 University of California — Experiment Station

Barred Plymouth Rocks (Pearl and Surface, 1909)/ His method of
breeding involved the selection and mating of his better-producing fe-
males in one line and of the poorer in another. No means of evaluating
the contributions of the breeding males was employed. After a period of
years, the egg production in the two lines was approximately the same.

Pearl succeeded Go well and was able to improve the production of the
same flock by the use of different methods of selection. He noticed that
there was a close correlation between the winter and the annual produc-
tion of any bird. The more eggs produced in winter, the higher was the
annual record. Using winter production as a criterion of selection for
mating, and analyzing his results genetically, Pearl (1912) developed his
theory of the inheritance of egg production, which proposed that two
pairs of genes were primarily responsible for high egg production.

A few years after Pearl's work was published, Goodale (1918) intro-
duced a new concept regarding the mode of inheritance of egg production
obtained from studies on the Rhode Island Red flock of the Massachusetts
Agricultural Experiment Station. Egg production was considered by
him to be the final expression of the action and interaction of a number
of diverse production characters. Of some ten characters first listed by
Goodale, five are important in contributing to an individual's annual
record. A sixth has been suggested recently. These characters may be
divided into two general groups : those affecting the length of time the
bird lays eggs, and rate or intensity of production when the bird is lay-
ing. The first group of characters include : the time when a pullet begins
to lay eggs, known as sexual maturity; the time when a hen ceases to lay
at the end of the pullet year of production, known as persistency ; and the
intervening nonlaying periods within the time limits of sexual maturity
and persistency produced by broodiness and winter pause. Spring or sum-
mer pause may be considered as another character (Lerner and Taylor,
1936) . These characters governing the length of time in production, com-
bined with rate of production, determine the annual record of the hen.

After the development of the idea of breaking egg production down
into a number of component parts, Goodale and later Hays carried on
studies to determine the method of inheritance of these characters. As a
result of these studies, Hays has advanced a series of hypotheses regard-
ing the number and nature of the genes that control the characters deter-
mining egg production (see Jull, 1932, for a bibliography of papers on
this subject) . Hays and Sanborn (1934) also have published data on the
improvement in the flock egg production and the various production

4 See "Literature Cited" at the end of this paper for complete data on citations,
which are referred to in the text by author and date of publication.

Bui,. 626] BREEDING FOR EGG PRODUCTION 7

characters in the Massachusetts Station flock over a period of twenty
years of breeding conducted by Goodale and Hays.

Several other investigators have published data on improvement in
egg production. Dryden (1921), by breeding from hens with high egg
records, was able to increase greatly the average production in the Barred
Plymouth Rock, White Leghorn, and crossbred flocks at the Oregon Agri-
cultural Experiment Station. He also commented that the use of the
progeny test in selection might be expected to hasten the progress to
higher production averages. Asmundson (1927a, 19275, 1928) has re-
ported results from six years of breeding Rhode Island Reds, Barred
Plymouth Rocks, and White Leghorns for improved egg production at
the University of British Columbia. The results of twenty years of mass
selection of high- and low-egg-production lines at the Cornell Agricul-
tural Experiment Station have been reported by Hall (1935) . Hall's con-
clusion that egg production has been improved by this method of selection
has been challenged by Petrov (1935) who has pointed out that, with the
exception of the first four years of selection, the increase in egg produc-
tion seemed to be due to improved environmental conditions, since both
high and low lines improved in average egg production.

During the early part of the present century, practical poultry breed-
ers selected birds for mating largely on the bases of physical appearance
and breed characters stressed by the Standard of Perfection. When trap-
nesting and pedigree systems came into common use, strong reliance was
placed on the ancestry of the individuals mated. This was soon followed
by a combined consideration of the individual's performance along with
its ancestry. During this period, the poultry industry began to strive for
high individual records. The production of 300-egg hens became the goal
of breeders.

Recently, there has been a tendency to add progeny testing to the bases
for selection of birds for breeding. This last step, introduced to the field
of poultry breeding by Pearl, combines all of the information available
from ancestry and individual records of performance with an actual test
of the bird's breeding ability. This type of selection has also tended to
change the emphasis from an individual basis of selection to a family
basis. Not only good individuals, but good individuals from good fam-
ilies, are wanted.

Prom a genetic standpoint, the progeny test is based on sound princi-
ples. Many good individuals with similar phenotypes are known to be
genotypically dissimilar, some being homozygous and some heterozygous
for different important genes controlling production characters. A good
bird from a uniformly good family is likely to be homozygous for and

8 University of California — Experiment Station

breed true for a greater number of desirable genes than a good individual
from a poor family. The final proof of the value of a breeding bird is,
then, found in the offspring, which show the various combinations of
hereditary factors that the parent can actually transmit to the next gen-
eration. This is especially so, since the actual number of genes involved
in the inheritance of egg production is not known. Estimates vary from
8 pairs postulated by Hays and Sanborn (1934) to 200-300 suggested by
Munro (1937) . Whatever the exact number may be, the phenotypes dis-
played by the progeny or by the sisters of the individuals considered give
a means of estimating the relative values of their genotypes.

CHARACTERS GOVERNING THE ANNUAL EGG RECORD

The annual record is a measure of the productivity of the individual and
reflects the contributions of all egg-production characters. The ideal con-
dition to strive for is the best combination of all of the desirable char-
acters involved. The number and importance of the independent factors
entering into the annual record can be adequately illustrated by actual
trapnest records of some individuals. A number of birds may have the
same annual record, but it may be conditioned by entirely different char-
acters in each case. Five representative individual trapnest sheets se-
lected to illustrate this point appear in figure 1. The five birds laid very
nearly the same number of eggs in the thirteen calendar months begin-
ning with September 1. Yet an examination of the course of egg produc-
tion in these birds will show five different reasons why the figure of
approximately 200 eggs in each case was not exceeded.

The bird whose record is shown in figure 1, A, was a late-maturing in-
dividual, not starting production until she was 305 days old. Although
she laid at a high rate, the late sexual maturity limited her record for
the year.

Similarly, the length of time in production of the bird in figure 1, B,
was limited by an early molt. She evidently was a nonpersistent bird and,
though fairly early-maturing and laying with reasonable intensity, did
not exceed the 200-egg mark because of early cessation of production.

Figure 1, C, on the other hand, illustrates the loss of time in production
during the laying year, rather than at its beginning or end as in the
previously mentioned cases. This bird went into a pullet molt or, in other
words, exhibited winter pause, during the months of December and Jan-
uary, which limited her possibilities of attaining a high record.

The case of the bird in figure 1, D, which became broody on four occa-
sions during the year, also shows a similar loss of time in production.

Finally, figure 1, E, shows the record of a bird that possesses most of

Bul. 626]

Breeding for Egg Production

2 3 4 5 6 7 9 10 li izuft )£l6liiei92(2IZ2l3242!2*Z7Ul93<J| Month To Date

MAR /// //

APR /////

S/SSS/'S//

JUNE • •/"/

yyyyyy

//'/

AaMaAa VA>V\ kPPv

• ••/' • •^

A, Leg band No. 896; hatching date Apr.10, 1934 ; age al first egg, 305 days

Month l2M j (la 9

1514 i5l6m8l9202l22232<2S2fc2722U95C3l Month To Dot

SEPT

_L9

/^4

lie

142
Ife7
166
193
193

///

^///

194
§ Leg band No. JI8; hatching date Mar. 20,1934; age at first egg, 173 days

2fi?9ioji Mont

///

////

/ / '//

C, Leg band No. Jl ; hatching date Mar.20,1934; age at first egg, 190 days

Z3tbi789io

19 2021 22 23 24 25 26 2.7 2S23 3031 Month To Date,

y ////

s-*yy

FEB • SS/S

S/S/S yys

MAR S S/S

/////

I4Q
147
162
176
199

/./

AUG /// /////

///// S

y s s/ */ yy s/s *s ss

D, Leg band No.J318; hatching date Mar. 20, 1934; age at first egg, 193 days

Month

Totals

2-4

Month

To Date

SEPT

OCT

•

NOV

•

DEC

JAN

FEB

1 1

MAR

1 1 1

APR

2<V

135

MAY

l&O

JUNE

168

JULY

ISO

AUG

192

SEPT

195

E, Leg band No. J 10; hatching date Mar.20,1934; age at first egg, 176 days

Fig. 1. — Conditions lowering the annual record:
A, late maturity; B, poor persistency; C, winter
pause; D, broodiness; E, low rate.

10 University of California — Experiment Station

the desirable characters governing time in production — her maturity
and persistency are satisfactory, she was nonbroody and lost very little
time by winter pauses. And yet her record was as low as those of the other
four. The explanation lies in her inability to lay at a high rate. While
the bird in figure 1, A exhibited monthly production figures of 24, 26,
28, 27, 27, and so on, this bird laid only 19, 24, 15, 18, and 12 eggs respec-
tively. Although the factors which need improvement in these five birds
are different, no indication that this is the case could be obtained from
the total annual egg records alone.

However, the conditions existing in reality are even more complicated.
The birds in figure 1 each exhibited one principal defect. Most birds
show various combinations of these defects, often two and sometimes
three or more at a time. Yet it is possible both for the trapnesting breeder
and the nontrapping multiplier to detect and estimate the severity of
these defects in the flock.

Methods of detecting and measuring each of these defects with and
without trapnesting are discussed in the following sections.

Sexual Maturity. — Sexual maturity is the character which expresses
the rate of sexual development of the bird and the onset of egg produc-
tion. This factor is the one that Pearl principally studied when he investi-
gated winter production. In his stock it was the early-maturing birds
which laid the most eggs before March 1 of their first laying year. When
maturity is considered on a time scale, a less arbitrary criterion of this
character is available. If all of the birds considered are hatched on the
same date, the date of the first egg laid provides such a measure. If, how-
ever, the date of hatch is variable, maturity may more logically be meas-
ured by the age at first egg.

There is, however, a complication involved here too. Climatic condi-
tions are known to affect the rate of growth of individuals ; early-hatched
chicks, as a rule, grow and develop faster than do late-hatched ones
(Kempster and Parker, 1936). The age at first egg is affected by a sea-
sonal retardation of growth, associated with high temperature. The Cali-
fornia Station flock is usually hatched in March and April, but even in
this short season, differences develop : each week's delay in hatching de-
lays the date of first egg by 10 to 12 days, or, in other words, the average
age at first egg is increased by 3 to 5 days. In a similar manner, nutrition
affects the age at first egg : for example, the addition of 15 per cent wheat
bran to a ration containing no bran used at the California Station low-
ered the average age at sexual maturity by approximately 15 days.

Artificial light is another environmental factor which may decrease
both the age and the date of sexual maturity. Thus Scott and Payne

Bul. 626] Breeding for Egg Production 11

(1937), by subjecting turkeys to artificial light, accelerated the onset of
production by nearly two months. Similarly, Bissonnette and Csech
(1937) and also Clark, Leonard, and Bump (1937) obtained earlier eggs
from pheasants by providing artificial light.

Age at first egg, however, seems to be a satisfactory criterion of sexual
maturity in chickens, and with the recognition of and provision for the
influence of hatching date, it can be safely used, when other management
factors are uniform. In turkeys, Asmundson (1938) has found date of
first egg to be superior to age at first egg as a measure of maturity.

That sexual maturity affects annual production is clearly seen from
figure 2, which illustrates the effect of the date of first egg on the length

AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT

414- DAYS

374 DAYS

336 DAYS

269 DAYS

2 24 DAYS

DATE OF FIRST EGG BEGINNING OF FALL MOLT

Fig. 2. — Length of laying year in relation to date of
first egg. Birds are grouped according to date of first egg,
in 50-day periods.

of laying year of a flock of pullets. The actual production is not shown
here largely because there are inherent differences in intensity of laying
between the individuals whose records were used for this chart. If equal
rates of production were found in all groups, the length of time in pro-
duction would be directly related to the annual record. Figure 2 indicates
that birds maturing in February have only slightly more than 50 per
cent as much time in production previous to the annual fall molt as birds
maturing in August. The fall molt occurs at about the same date regard-
less of date of maturity. In commercial practice, the factor of early ma-
turity is of added importance, for most of the birds are hatched in the
spring months, and hence those that are early-maturing usually lay more
eggs at the time of high egg prices.

That sexual maturity is inherited has been demonstrated definitely by
Pearl (1912), by Hays (1924), and by "Warren (1934). In determining
this character both sex-linked and non-sex-linked genes are involved.
Sex-linked inheritance in chickens operates in such a manner that the
sire exerts more influence than the dams on the character observed in the
daughters (see p. 4 in section, "Explanation of Terms Used in Breed-

12 University of California — Experiment Station

ing"). Consequently it is of paramount importance to use sires from
early-maturing families if one desires to improve his flock for this char-
acter.

Hays (1924) has suggested that two pairs of genes are involved in the
inheritance of early maturity : EE, which is sex-linked (E- in females)
and E'E', which is not. According to Hays, either gene in the dominant
state will produce early maturity, while late maturity will ensue when
both are recessive. Warren (1934) has produced evidence that both E
and E' are necessary for a satisfactory early maturity. The situation may
be even somewhat more complicated, but in any case, at least two pairs
of genes are involved, one of which is sex-linked.

There is a possibility of having birds which mature too early. Preco-
cious maturity is associated with small body size, which is one of the fac-
tors contributing to small egg size (see section entitled "Egg Size," p. 24) .
Precaution must therefore be taken to guard against developing too
early maturity as well as against too late maturity.

Hays (1936a) has produced evidence that the earliest phenotypic
group of Rhode Island Reds mature at less than 180 days of age. These
birds were assumed to carry genes E and E', since they produced off-
spring maturing on the average earlier than the offspring of dams ma-
turing at 180 days or more. The border line between early maturity and
precocity has not yet been determined. However, general breeding ex-
perience would indicate that birds maturing at less than 150 days of age
are apt to be small and to produce small eggs. Furthermore, at least in
one case on record (Knox, 1930), such birds did not produce as many
eggs as the birds maturing at between 160 and 210 days of age.

Identification of both early- and late-maturing birds is accomplished
very easily. The trapnest record, of course, is the best guide, but physical
selection of early-maturing birds is relatively simple. Any culling guide
gives directions for selecting early-maturing birds. 5 The late-maturing
birds should be marked in order to eliminate them from any future use
as breeders. They may, however, still prove to be fairly profitable egg
producers, if they are healthy and of average size.

5 Most general texts on poultry contain instructions for culling, for example :
Jull, M. A. Poultry husbandry. 2d ed. viii + 548 p. Illus. McGraw-Hill Book Com-
pany, Inc., New York, N. Y. 1938.

Lippincott, W. A., and L. E. Card. Poultry production. 723 p. Illus. Lea and Febi-
ger, Philadelphia. 1934.

There are also books dealing specifically with the subject of culling methods:
Eice, J. E., G. O. Hall, and D. E. Marble. Judging poultry for production, rru + 425 p.
Illus. John Wiley and Sons, New York, N. Y. 1930.

Payne, L. F., and H. M. Scott. International poultry guide for flock selection.
142 p. Illus. International Baby Chick Association, Kansas City, Mo. 1934.

Bul. 626] Breeding for Egg Production 13

Pauses. — Pauses occurring in the production record between the time
of the first egg laid and the normal fall molt may be broadly classified
into three groups : winter pause, pause due to broodiness, and spring and
summer pause.

Winter Pause : One very important problem in connection with pauses
in the winter as well as those in the spring and summer is the debatable
question as to what constitutes a pause. At various times, different work-
ers on the subject have used some given number of successive days of
nonlaying for a definition of pause, the number of such consecutive days
varying from 4 to 15. As yet no clear-cut definition, which would not be
arbitrary, has been suggested, the major difficulty being the fact that the
underlying reasons for such pauses are not well understood. In many
cases, pauses can be recognized in the individuals of a flock by head, neck,
or body molt. As has long been recognized, birds going out of production
usually start molting. Often the severity of the molt — which is judged by
knowledge of the order of the normal molt (head-neck; breast-abdomen-
back ; wings ; tail) — serves to indicate the length of the pause. The win-
ter-pause molt is usually confined to the head and neck. Some individ-
uals, however, lay through a molt, and others pause for a short period of
time without molting. In a trap-nested flock, all types of pausing birds
can be accurately recognized, but in a flock which is not being trapped,
some are improperly classified even by those well trained in the use of
known physical methods of culling.

Little is known about the factors involved in winter pausing. That
pauses are brought on by environmental factors in many cases is well
recognized, for sudden changes in management, extremely rough han-
dling, and sudden climatic changes often result in pause. Since earlier-
hatched groups of pullets of the same strain regularly show a greater
amount of pause than later-hatched lots, there is probably a direct rela-
tion between age of the bird and the internal conditions which cause her
to cease egg production. Yet at the same time that some birds in any
flock of a given age show a pause, others may go through such disturb-
ances and continue to lay. This gives evidence that there are differences
between birds in their thresholds of response to conditions producing a
pause. Evidence from the California Station flock indicates that such
differences are, at least in part, hereditary. If such is the case, the possi-
bility of selection and breeding for absence of pause immediately pre-
sents itself.

That such procedure is wise can be illustrated by the losses a poultry-
man sustains by having many pausing birds in his flock. Thus, some years
ago in the California Station flock, birds showing a pause of 7 or more

14 University of California — Experiment Station

days were found to lay on the average about 56 eggs during the four
winter months (November through February), while the nonpausing
birds during the same time laid on the average 77 eggs ; the winter pro-
duction of the nonpausing birds was over one-third greater.

Hays (1924) has postulated that the inheritance of winter pause de-
pends on a dominant gene M for its expression. Thus, birds of the con-
stitution mm would be free from pause. However, the mode of inheri-
tance of the factors responsible for the occurrence of winter pauses is
not so simple as this, for, if it were, the elimination of pause from a flock
would be easily obtained by breeding only from nonpausing {mm) fam-
ilies. At present, the best a breeder or multiplier can do is to select for
mating those families or individuals which show greatest freedom from
pause. For breeding selection against pause, pullets which are from 7 to
8 months of age in November will probably give a better differentiation
between genetic pausers and nonpausers than either older or younger
birds. The trapnest breeder can base his selection on periods of nonpro-
duction in the egg records. In the nontrapnested flock, reliance must be
placed on the appearance of molt in birds. Although some of the pausing
birds may prove profitable egg producers, they should be identified so as
to be eliminated from the breeding flock.

Broodiness : This type of pause is of greater importance in some breeds
than in others. The Asiatic breeds are very broody as a rule. The Ameri-
can breeds, such as Plymouth Rocks, Rhode Island Reds, and Wyan-
dottes, are subject to considerable broodiness, while most of the Mediter-
ranean breeds seem to be comparatively free from it ; but no strain of
any breed has ever been reported to be completely nonbroody. Selection
against broodiness should be very rigorous, since if unchecked it may
become an important problem, especially in small breeding flocks.

Of the different characters involved in egg production, broodiness is
one upon which considerable genetic and physiological data have been
obtained. Two pairs of dominant complementary genes, A and C, seem
to be involved in producing broodiness, according to Goodale, Sanborn,
and White (1920). The presence of either A or C alone will produce no
effect, but the two brought into combination will result in a broody bird.
Thus two nonbroody lines crossed together may produce broody progeny,
if one of the lines possesses the factor complementary to the one carried
by the other line. For instance, a cross between AAcc and aaCC will re-
sult in a broody bird, AaCc, heterozygous for both genes. In the Cornish
breed, Roberts and Card (1934) have obtained evidence of an additional
sex-linked gene for broodiness.

A situation involving complementary genes makes it difficult for any

Bul. 626] Breeding for Egg Production 15

breeder to eliminate broodiness completely from a strain. The incidence
of broodiness may be greatly reduced, however, by a method of testing
the genotype of breeding birds. This method involves the isolation of two
nonbroody strains : one carrying gene A without gene C, the other carry-
ing C without A. When crossed, these strains produce broody daughters.
Prospective breeding males and females can be mated to first one and
then the other of the test strains. If no broody daughters are produced
from the mating with either strain, then the prospective breeding bird
carries no genes for broodiness and is of the constitution aacc.

In a flock where such a program is not feasible, elimination of the
broody families or, in the case of nontrapnested birds, of broody individ-
uals from the breeding flock, should be made. Since broodiness is a very
conspicuous character, this plan can easily be carried out. In the simplest
form it involves placing a colored legband on a bird that at any time was
taken off the nest as being broody. Once identified, such birds can be dis-
posed of in any manner desired.

The importance of broodiness in the annual record of a bird depends
both on the number of times during the year that a bird becomes broody
and on the length of time before the bird resumes laying. Both of these
factors are extremely variable ; the latter is in part dependent on man-
agement, since the sooner a broody bird is detected and prevented from
sitting, the more quickly she will return to production.

The physiological basis of broodiness is better understood than that of
most of the other characters discussed. Broodiness is produced by pro-
lactin, a hormone secreted by the anterior pituitary body. This is the
same substance that controls the production of crop milk in pigeons as
well as of the milk of nursing mammals (Riddle, Bates and Lahr, 1935) .
Byerly and Burrows (1936) have found that broody females and males
from broody strains produce more of this hormone than nonbroody birds.

Spring and Summer Pauses : Besides the pauses occurring in the win-
ter months and pauses due to broodiness, there are other periods of in-
terruption of egg production, which may appear during the spring or
summer. Superficially they seem to be of the same nature as the winter
pauses, but actually they have been demonstrated (Lerner and Taylor,
1936) to be the result of other causes which have not as yet been deter-
mined. Whether hereditary factors are concerned in the production of
these pauses is not known at present, consequently neither is it known
whether or not selection against them is feasible. Superior birds, how-
ever, do not exhibit this condition, so that discriminating against spring
and summer pausers in a breeding flock is a safe policy because of the
possibility that this defect may be transmitted to their progeny.

16 University of California — Experiment Station

Persistency. — The character limiting or extending egg production at
the end of the laying period is known as "persistency" ; the cessation of
production is usually accompanied by the onset of the annual (fall) molt.
Its importance, recognized by Rice (1914) before the development of the
Goodale-Hays hypothesis, can be readily demonstrated in the same man-
ner as that of early maturity. Thus figure 3, based on the same stock as
used for figure 2, shows that the early -molting birds have a much shorter
period of lay than the late molters. Thus the birds molting in June were
in production about 60 per cent as long as those molting in December.
Figure 3, like figure 2, indicates that there is no relation between date
of first egg and the time of onset of fall molt,

OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

248

DAYS

302

DAYS

336

DAYS

390

DAYS

429 DAYS

DATE OF FIRST EGG BEGINNING OF FALL MOLT

Fig. 3. — Length of laying year in relation to beginning
of annual molt. Birds are grouped according to date of
molt, in 50-day periods.

A rather important question in connection with persistency is the
selection of an adequate method of measurement. There is a large number
of these available, the most prominent ones being the length of biological
laying year, suggested by Hays and Sanborn (1926) ; production in
August and September, used by Knox, Jull, and Quinn (1935) ; and age
at last egg or date at last egg, originally mentioned by Goodale (1918)
and recently brought back to usage by Lerner and Taylor (1937a). The
latter measures are deemed more adequate than that of Hays and San-
born on the grounds that length of biological laying year depends on
sexual maturity as much as on persistency and hence is not a true meas-
ure of the latter (Lerner and Taylor, 1937^) . In view of the above criti-
cism of the use of the biological laying year as a measure of persistency,
Hays's (1936&) claim that persistent production results from the action
of a single dominant gene P should be submitted to further experimental
test. The measurement used by Knox, Jull, and Quinn reflects the in-
herent rate of production as well as the ability of the bird to lay late in
the fall, and thus also fails to give an uncomplicated measurement of per-
sistency.

Bul. 626] Breeding for Egg Production 17

As far as the preference between age at or date of last egg is concerned,
there seems to be a greater effect of date of hatch on the age at last egg
than on date at last egg. Both can be demonstrated to be inherited to an
equal extent, when a population hatched during a limited season is con-
sidered. By analogy with sexual maturity, the age measure may be used
in preference to the date measure, especially if correction for hatching
date is made. Probably breeders hatching stock over a period of several
months will find it easier to use the date of last egg as a measure of per-
sistency. In turkeys, Asmundson (1938) has used this measure as an
adequate criterion of persistency.

Whichever method is used, selection by the breeder can be readily
practiced for this character. Such selection, if properly carried out,
should be effective in increasing the length of time the flock is in lay, but
caution must be exercised, just as in the case of early maturity, not to
overdo it.

In commercial practice, a flock of yearling hens which will lay through
the late fall at the time of the highest egg prices will be highly profitable.
A breeding flock, on the other hand, if in lay at a late date, may not return
to full production after the molt by the time hatching eggs are desired.
A practice of forced molting might be necessary in highly persistent
breeding flocks, but it would tend to nullify any attempt to select for
extreme persistency. In order to make possible adequate selection of
birds for persistency, forced molting probably should not be started
before the end of September.

Rate of Production. — The factor of intensity, or rate of production, is
also one for which a number of measurements have been proposed. The
most common way of representing rate is by the use of percentage pro-
duction. This may be designated as gross rate and has the distinct dis-
advantage of permitting confusion of nonpausing birds possessing low
intensity with birds which show pause but have a high rate of production
when in lay. Since some types of pause are inherited independently from
rate, they should be measured separately. The net rate of production
accomplishes this by eliminating pausing days from consideration. Thus,
instead of dividing the number of eggs laid in the period considered by
the total number of days involved, it is divided by the latter number less
the days broody and in pause. The difficulty arising here is again the one
of deciding just what constitutes a pause. Until more information on the
subject is available, only arbitrary standards can be used. In some of the
work carried out at this station, where 7 days has been used as a minimum
standard for pause, the comparative figures in table 1 were obtained.

As may be seen, the winter pauses are distinct from rate, since the net

University of California — Experiment Station

winter rate is almost the same irrespective of whether the birds paused
in the winter. The gross rate in the two cases, however, is different, the
winter-pausing birds falling about 20 per cent lower than the nonpausing
birds. Spring and summer pauses, however, seem to have a somewhat
depressing effect on rate, this being one of the points of evidence for a
distinction between them and the winter pause. The main point these
data illustrate, however, is that the gross rate may be a misleading cri-
terion of measurement of the inherent intensity of production.

Average clutch size has been used by Hays and Sanborn (1932) as a
measure of intensity. It is, however, somewhat more laborious to calcu-

TABLE 1
Gross and Net Rates of Production for Pausing and Nonpausing Hens*

Winter-pausing birds

Spring- and summer-pausing birds

Birds pausing both winter and spring or summer.
Nonpausing birds

Winter rate

percent
46
63
47
65

Net

percent
65
64
64
67

Spring rate

Gross

percent
74
58
56
74

Net

percent
77
70
71
75

* From Lerner and Taylor (1936).

late and, furthermore, does not reflect the number of nonlaying days
between clutches of eggs, which should enter intrinsically into the cal-
culation of rate. A hen laying an egg each third day would have the same
average clutch size as a bird laying an egg every other day, but the latter
would have the higher net rate. The suggestion of Hays and Sanborn
(1934) that two dominant genes I and V determine clutch size does not
necessarily solve the problem of inheritance of rate of production, par-
ticularly because of the arbitrary assignment of clutch sizes to specific
genotypes.

In using net rate, one should choose a short period, rather than the
whole year. This is largely because the average annual rate may also lead
one to an erroneous evaluation of the inherent intensity of a bird. Thus,
a bird pausing during the winter months will have undue weight laid on
her spring production, which is normally higher than the winter produc-
tion ; and a nonpausing bird will be penalized by the inclusion of her
winter record in the calculation of her rate. Since, however, rates at
different periods of the year bear a definite relation to each other, a
shorter period of evaluation can be used with a fair degree of precision.
In general, the rate during the spring months (March to June) is found
to give a higher correlation with the annual egg record than does the

Bul. 626]

Breeding for Egg Production

winter or the summer and fall rates ; but the correlation between winter
rate and annual production is sufficiently high to make it feasible to rely
upon winter rate. This has the additional advantage that, outside of the
birds which pause through most of the winter months, an evaluation of
the inherent rate can be thus made in pullets before the breeding season.

A, HIGH WINTER RATE

/ \ B, UNKNOWN
/ /—•—.- \s\ WINTER RATE

/ / -~X/ v J

/ / / K \y
/ if ^ \

7J-

1 \ M

/ / \ A

' v \\

\ V-

/ C, LOW WINTER RATE \ \

W, \\

\
\

V \

' % — .

/ \^

' \

\
\
\

—1 1 1 1 1 1 I I l

A/QV. A£C ^PIAL /Z& MA*. APR. MAW JUA/£ JVLY AUG. SEPT. OCT.

Fig. 4. — Relation of fall and winter rate to annual
rate: A, birds laying 21 or more eggs in each of any two
of the winter months (November, December, January) ;
C, birds laying less than 21 eggs in two of these months;
B, birds, information on which is lacking for these months.

Figure 4 illustrates the fair degree of accuracy obtained by such a
method of classification with nonpausing birds. An arbitrary standard
of 21 eggs a month in any two of the three winter months (November,
December, January) is used for high rate. Birds fulfilling such a require-
ment exhibit on the average a higher rate of lay throughout the rest of
the year than those falling below it. The third curve on the graph repre-
sents the production of birds which either paused during the winter
months or were extremely late-maturing. Birds with both low and high

20 University of California — Experiment Station

rates were present in this group, since the average production is found
to be intermediate between the other groups. The rate of such birds would
have to be judged by their performance in the spring. Since, however,
these birds would in all likelihood not be bred from, because they ex-
hibited a winter pause or exceedingly late maturity, no disadvantage in
delaying the collection of information on the rate of such birds is in-
curred. In all groups, the highest rate of production occurred from
March to May, and the differences in rate between high- and low-rate
groups were least in the spring and summer months.

The above methods of identifying birds possessing high or low rate are
obviously suitable only for trapnested flocks. Unfortunately, there are
no precise physical criteria by which rate can be judged, so that indi-
vidual selection is impossible in nontrapped flocks. A short trapnesting
period will achieve the desired ends; otherwise only pen selection is
possible. Where pause is not a significant factor, percentage production
of one pen as measured against another may serve a useful purpose.

General Considerations of Production Characters. — The general dis-
cussion of the genetic factors affecting egg production in relation to
practices of selection can now be summarized. Some of the characters
controlling time in production, such as maturity, persistency, and broodi-
ness, are inherited on a basis such as to make selection for the desirable
genes possible. The trapnester should have no difficulty in interpreting
his records on these characters and in setting up standards of selection.
Further, where trapnesting is not practiced, certain physical manifesta-
tions of maturity, persistency, and broodiness are easily observed ; and
most culling guides provide complete instructions for the recognition
both of superior individuals and of those to be eliminated from the breed-
ing flock.

Manifestations of pauses are not so readily interpreted. While neck
molt is a fair criterion of the expression of the winter-pause character,
there is considerable difficulty in setting up satisfactory and efficient
standards of selection in these cases. This applies less to the trapnesting
breeder than to the nontrapnesting multiplier.

The condition of summer pause is still more difficult to interpret, since
it shows some association with low rate, and rate is the hardest character
to measure unless some trapnesting is practiced. If at least part-time
trapnesting is practiced, selection for high rate is feasible. Otherwise,
little reliability can be placed on the use of physical standards of selec-
tion, and improvement of the inherent intensity of production in a flock
is beset with great difficulties.

The best birds are not the ones which are superior in only one par-

BUD. 626] BREEDING FOR EGG PRODUCTION 21

ticular character, but those which exhibit the best combination of all of
the characters considered. The annual record and the monthly and sea-
sonal distribution of egg production throughout the year are in the final
analysis the main criteria of the profitableness of a bird's production.
Their expression is conditioned by the combination of the characters
discussed ; hence, the latter should always be considered, not only singly,
but also in relation to each other and to their contribution to the annual
record.

When flock efficiency is considered, however, a whole new series of
characters, some of which are undoubtedly inherited, comes into the
picture. These are the characters affecting the viability of birds. Flock
efficiency will, of course, be lowered if mortality is high. Selection for re-
sistance to adverse environmental conditions or for the ability of the
bird to live must, therefore, now be considered.

CHARACTERS AFFECTING VIABILITY

The influence of the conditions affecting viability on the efficiency of pro-
duction of the flock begins with the egg and continues throughout the
productive life of the bird. The number of chicks hatched from a given
number of eggs, the proportion of these raised to maturity, and finally
the mortality of the laying birds, all have an effect on the size of the flock
and on its productivity. Just how much of the embryonic and post-em-
bryonic mortality is due to inherited factors cannot be estimated accu-
rately, but there is good evidence in many cases that at least part of the
mortality up to the completion of the first laying year is controlled by
genetic factors, and hence the condition is amenable to improvement by
proper breeding methods.

The mortality most closely related to egg production is that occurring
during the first laying year. This is evidenced by reports available on
commercial flocks where as high as 100 per cent replacement has been
necessary during the first laying year. The replacements are not entirely
due to mortality, since some of the birds removed, though nonproductive,
appear anatomically normal, and others are unprofitable because of non-
fatal pathological conditions. However, at least half of the replacement
in an average well-cared-for flock is likely to result from actual deaths
from nonspecific pathological conditions, that is, conditions not related
to definite infectious or dietary factors. This point was brought out by
Lubbehusen and Beach (1938) in an analysis of mortality in the Cali-
fornia Station flock.

The presence of birds that are doomed to die during the laying year
and also of birds which fail to lay because of specific diseases causes a

22 University of California — Experiment Station

flock to be less profitable. To illustrate this point, limited evidence on the
California Station flock indicates that birds which succumb to paralysis
and associated conditions in the spring months show a reduced produc-
tion during the previous fall months although they appear normal in
every other respect. In a similar manner Asmundson and Biely (1930),
among others, have reported that adult carriers of pullorum disease show
a reduction in egg production. On the other hand, we have evidence that
birds developing disturbances of the reproductive system seem to lay
normally up to the time of onset of the outward manifestations of the
condition.

In any case, if any of the various diseases to which poultry is subject
have a specific hereditary basis of susceptibility or are conditioned even
in part by a general constitutional weakness, breeding methods may be of
value in increasing the flock efficiency by the creation of lines more re-
sistant toward such conditions.

Longevity. — Partly on these grounds, Pearl (1923) suggested breed-
ing from aged birds only. He believed that the productivity of the breed-
ing birds would be improved along with longevity. Greenwood (1932),
however, has found that very old birds showed no marked superiority in
offspring, while their progeny was limited in numbers by low fertility
and hatchability. Whether or not such is the case, greater viability will
ensue in a flock as a consequence of selection of longer-lived strains. Dif-
ferences in mortality between families in the same flock are known to
persist for several generations, as they have in the California Station
flock, and hence are probably heritable. Despite the poultryman's inter-
est in obtaining long-lived birds, comparatively few reports are available
on the results of selection for longevity. This fact may indicate that
breeding for characters influencing longevity is a complex process.

Disease Resistance. — Some experimental evidence of the inheritance
of resistance to specific infections in chickens has been obtained. Roberts
and Card (1926 and 1936) have demonstrated inherited differences in
susceptibility to pullorum disease, and Lambert and Knox (1932) ob-
tained similar results with fowl typhoid. They failed, however, to obtain
strains in which all the birds were resistant. Furthermore, such diseases
can be more efficiently controlled by other means. On the other hand,
there is evidence on the hereditary nature of resistance to certain other
disease conditions for which methods of prevention and control are not
yet known. This applies particularly to neurolymphomatosis (paralysis)
and may also be true of other forms of lymphomatosis (big-liver disease
or leukemia), leucosis, and certain types of tumors.

Present conceptions of the nature of such pathological conditions are

Bul. 626] Breeding for Egg Production 23

conflicting, but there is considerable experimental and statistical data
from British Columbia (Asmundson and Biely, 1932), Louisiana (Upp
and Tower, 1936) , Idaho (Gildow, Williams, and Lampman, 1936), Iowa
(Wilcke, Lee, and Murray, 1938) and from the California Station flock,
to indicate the existence of definite differences in degree of occurrence of
this group of diseases between different families. These data point to the
possibility of partial control by selective breeding. Results obtained with
the California Station flock having a bearing on this problem are dis-
cussed in a later section (p. 43) .

A suitable method of selection against these diseases in an affected
flock involves breeding from families which are as free as possible from
paralysis and other neoplasms, rather than the selection of survivors
from susceptible families. Since the genetic nature of disease resistance
is usually fairly complicated, the latter method would result in perpetu-
ation of factors for susceptibility in the flock. The former method, on the
other hand, would tend to eliminate any family of birds which shows an
undue concentration of such undesirable factors.

All available evidence indicates that disease resistance is specific in
nature. Creating a strain resistant to one disease does not mean that the
strain will be resistant to any other disease. Nor does it mean that the
strain will be unusually susceptible to some other disease. Genetically,
every disease presents a separate problem of breeding for resistance.
Whether or not genetics can be effectively applied to the control of all
diseases is questionable. Those for which specific control measures have
been developed can undoubtedly be handled more easily by such means,
rather than by breeding. Extension Circular 8 a should be consulted for
other more direct methods of control of diseases.

METHODS OF MEASURING FLOCK PRODUCTION

Several ways of expressing the average egg production of a flock have
been commonly used. The least-accurate method is that of the average of
the records of surviving birds. A better method involves the computation
of hen-day averages. This method is well suited for cost-account work,
since the denominator represents the average number of birds in the flock
for a given period of time.

A measure of breeding-flock values known as the production index has
been developed at the Kimber Poultry Breeding Farm. This measure is
the quotient of the number of eggs laid by the group under consideration
(family, flock, etc.) over the original number of pullets in the group. Any

6 Beach, J. R., and S. B. Freeborn. Diseases and parasites of poultry in California.
California Agr. Ext. Cir. 8:1-110. 5th ed. 1936.

24 University of California — Experiment Station

birds that die at any time during the laying year are thus considered
equally with the survivors as members of the flock or family. In that
manner, a flock or family exhibiting high mortality will be penalized.
Just as the annual egg record reflects all the various characters influ-
encing the egg production of an individual bird, the production index
expresses them for a group of birds. While neither the productivity nor
viability of a family is reflected separately in the production index, this
index gives a true picture of the interaction of all of the factors involved.
To obtain measurements on the individual characters contributing to
production, one has to resort to the separate determination of all of the
variables previously discussed.

The production index is superior to both the average of the survivors
and the hen-day average as a measure of the productive efficiency of
breeding flocks. The average of the survivors completely ignores the im-
portant factor of mortality in the flock; the hen-day average fluctuates
greatly with the use of various culling practices, so that a heavily culled
flock may show a better average production than an unculled flock of the
same inherent productivity. Both mortality and culling decrease the
average egg production expressed as a production index, which thereby
shows the prevalence of nonproductive or nonviable birds in the flock.

CHARACTERS GOVERNING THE QUALITY OF EGGS

All of the discussion to this point has referred to the quantity of produc-
tion. Factors affecting quality of production may be discussed next.

Egg Size. — Probably the most important character concerned in the
quality of commercial eggs is that of size. The production of large eggs
within a reasonable period of time after sexual maturity is now recog-
nized as an essential in a good egg-production strain. Rigid selection in
breeding flocks has brought about during the last fifteen years a remark-
able improvement in the size of egg produced by many strains.

That there is a direct correlation between body size and egg size, within
a breed or within a strain, has long been known. Very small birds rarely
produce large eggs. Table 2 illustrates the relation between body weight
and egg weight. The 595 birds involved were weighed in December, and
the December and January egg weights were calculated for birds at dif-
ferent body weights. Sexually mature pullets which are still growing will
improve their egg size as they become larger.

Seasonal effects expressed through temperature also affect egg size :
Lorenz and Almquist (1936) have shown an average decrease of 1 ounce
per dozen in mature birds for each 30° F increase in maximum daily tem-
perature.

Bul, 626]

Breeding for Egg Production

Rate also may be related to egg size, producers of very small eggs often
having a phenomenal rate, while producers of exceptionally large eggs
seldom have a high rate. This relation is illustrated in table 3. However,
this does not mean that all birds laying small eggs have a high rate of pro-
duction. Since egg size is the result of the sum of the weights of the
various parts of the egg, any circumstance reducing the weight of a com-

TABLE 2

Relation between Body Weight and Egg Weight in 595

White Leghorn Pullets

Body weight at 8 months of age

Class
average

Number
of birds

Average
egg weight in

December
and January*

grams
1,100-1,399

pounds
2.76
3.42
4.08
4.74
5.40

236

272

ounces
1.621

1,400-1,699

1.781

1,700-1,999

1.900

2,000-2,299

1.975

2,300-2,599

2.000

* From 8 to 10 months of age.

TABLE 3

Relation between Winter Rate of Production and Egg
Weight in 145 White Leghorn Pullets

Number of birds

Net
winter rate
(November-
February
inclusive)

Winter
egg weight
(November-
February
inclusive)

per cent
56.0-63.9
64.0-71.9
72.0-79.9
80.0-87.9

ounces
1.852

1.855

1.841

1.810

ponent part reduces egg size. Thus, Buckner, Martin, and Insko (1930)
have shown that a deficiency of calcium in the diet reduces the amount of
shell per egg and also results in smaller egg size.

From a breeding standpoint, the most satisfactory short period in
which to determine the egg size of individuals is during the late winter
or early spring or, more specifically, in California during February and
March. By that time, pullets hatched in the previous breeding season are
approximately one year old, have reached adult body size, and produce
eggs under a moderate range of temperatures in most areas in California.
Egg size during this period is also reasonably close to the average size of
egg produced during the pullet year of production, as can be seen from

University of California — Experiment Station

table 4. Comparative inherent egg size can be determined at any season,
but the early spring period is least subject to adverse environmental in-
fluences. Where greater accuracy is desired, about 5 eggs a month may
be weighed. Weighing every egg scarcely adds sufficient accuracy to
warrant the labor involved.

Hays (1929, 1937) has advanced a hypothesis for the inheritance of
egg size involving a dominant gene A producing small eggs and dominant
genes B and C for large eggs. When A, B, and C are present, an inter-

TABLE 4

Monthly Variation in Egg Weight in a Flock of 116

White Leghorn Pullets

Month

Average
egg weight

Month

Average
egg weight

ounces
1.753
1.834
1.926
1.965
1.972
1.972
1.961

ounces
1.997

July

1.997

2.021

2.067

October

Annual average

2.081

May

1.962

mediate, but predominantly small size of egg is obtained. According to
this hypothesis, the genotypic constitution for largest egg size should be
aaBBCC, which would represent a pure-breeding large-egg type. Rigid
selection of both male and female breeders for egg size should act to
eliminate the undesirable A genes rather rapidly, and cause B and C
genes to become homozygous at the same time in an increasingly large
portion of the flock. While Hays' data lend considerable support to this
hypothesis, it is questionable whether or not the gene combinations sug-
gested by him have universal application. One practical difficulty in
flock breeding lies in the fact that the contribution of the male for all
egg characters can be determined accurately only by progeny tests. Prac-
ticing rigid selection of breeding males on the basis of egg size exhibited
by their female ancestors or their sisters is, however, fully as important
as selecting females with large egg size for mating.

The discrimination against small eggs on the market is sufficient to
encourage the production of large eggs. However, egg size may be se-
lected to the point where it becomes undesirably large. Minorca varieties
frequently produce eggs averaging 30 or more ounces per dozen. These
eggs are too large for the standard types of fillers and cases used for
chicken eggs. Occasional premiums for "jumbo" sizes may influence some
poultrymen to attempt commercial production of this type of egg. The

Bul, 626] Breeding for Egg Production 27

low hatchability associated with extremely large eggs (Dunn, 1922),
however, should be sufficient argument against breeding for this size.
Better hatchability results are usually obtained by selecting for breeders
hens with average egg size ranging from 24 to 28 ounces per dozen.

Egg Shape. — There are three common shapes of eggs : long, ovoid
(typically egg-shaped), and round. Long types are not commercially
desirable since the fillers of egg cases are not deep enough to take a large
long egg. Similarly, large round eggs are too wide for the ordinary filler.
From a hatchability standpoint, furthermore, ovoid eggs in common ex-
perience give better average results than either very long (Landauer,
1937) or round types.

The inheritance of egg shapes is not clearly understood. Benjamin
(1920) states that long-shaped types are dominant over normal types,
but Kopec (1924) has presented contrary evidence. Some casual observa-
tions made by the senior author would lead us to believe that round may
be dominant to ovoid. Axelsson (1936), however, has found no domi-
nance of any one shape over another in several breed crosses. Rigid selec-
tion for ovoid types tends to bring rapid improvement in the character
of the eggs produced by the flock. It is important that the breeding males
come from families producing ovoid-shaped eggs.

Many types of misshapen eggs seem to be produced by temporary or
chronic inflammation of the shell gland of the oviduct and are probably
not inherited. Unless the breeding hens displaying such conditions have
proved to be valuable breeders, such birds should probably be removed
from the breeding flock. Misshapen eggs show notoriously poor hatch-
ability.

Egg Color. — In the common breeds of chicken, egg color is usually
some shade of either brown or white. One peculiar breed, the Araucana,
lays a blue egg. Breeds of the Mediterranean class, such as Leghorns,
Minorcas, and Ancona, should lay a pure-white egg. Breeds of the Asiatic
and American classes should produce a brown egg. In these breeds a
uniform, medium-brown color is desired. Pale-brown or creamy colors
should be discriminated against.

The New York market has long been noted for its premium on pure-
white-shelled eggs, while the Boston market has similarly placed a pre-
mium on brown-shelled eggs. In recent years these premiums have been
lower than previously. However, in choosing a breed or strain for com-
mercial egg production, local market preferences for egg color must still
be considered.

Birds producing brown-shelled eggs and also individuals from white-
shelled varieties which produce tinted or creamy eggs characteristically

28 University of California — Experiment Station

show a decreasing intensity of color with advancing production (Benja-
min, 1920). Immediately after sexual maturity or after any period of
nonproduction, the color of the shell tends to be deeper in shade than
after long periods of production. Where tints and creaminess are to be
selected against, it is important to take records of color when the bird
first begins to lay and after pauses and the annual molt.

Egg color is apparently controlled by several pairs of genes (Hays,
1937) , some of them acting cumulatively to produce darker shades, others
acting as inhibitors of color (Punnett, 1923). Crosses of dark brown by
white strains usually produce eggs ranging in the medium-brown shades.
Crosses of medium brown by white strains may produce fowls with egg
color ranging from creamy to light brown. Blue color is dominant to
white ; crosses of blue and brown produce a greenish color, as Punnett
(1933) has shown.

In white-egg strains, several generations of selection of both male and
female breeders from white-egg families will reduce the incidence of
creamy or tinted eggs to a very low proportion. Frequently, crossing two
different white-egg strains produces a considerable degree of color in the
egg of the offspring. Similarly, mating creamy or tinted individuals to-
gether may actually produce some birds laying brown eggs. All of these
results are explained by the fact that there is probably some complemen-
tary action between some of the genes producing brown color in eggshells.

Shell Quality. — The most fundamental of the factors of shell quality is
that of thickness. Thick shells are desired both for market eggs and for
hatching eggs. In the incubation of eggs, characteristic differences in
hatchability between matings can sometimes be traced to the amount of
shell present.

Data from several generations of selection for shell thickness in White
Leghorns at the California Station are given in table 5. In this experi-
ment, birds from families selected for thick shell had a significantly
higher percentage of shell during four generations than corresponding
pullets bred from thin-shell lines. In these instances, the birds from both
lines received the same rations and management. This evidence confirms
the previous finding of Taylor and Martin (1928) that shell thickness is
in part dependent upon hereditary factors. Munro (1938) has also found
similar evidence for inheritance of the amount of ash in the shells of eggs
from a flock of Barred Plymouth Rocks. Probably the genetic basis for
inheritance of an ability to produce thick-shelled eggs is complex. The
importance of nutritional factors in producing shells of high quality
must also be emphasized, since deficiencies of calcium or vitamin D in the
diet quickly lead to production of defective shells.

Bul. 626]

Breeding for Egg Production

Other common types of poor shell quality are sometimes produced by
closely related individuals, and the inheritance of tendencies for their
production may be suspected (Hays, 1937). Such types include rough
or "sandy" shells, mottled shells, and "glassy," or very nonporous shells.
Hens producing these types of defective shells should be removed from
the breeding flock.

Another quality which seems to be inherited is the character of the
bloom or finish of the shell. While the Leghorn breed is said to produce a
chalk-white egg characteristically, many individual hens lay eggs having

TABLE 5

Proportion of Shell and Hatchability in Eggs from
Thick- and Thin-Shell Lines

Year

of
hatch

Line

Number

of
pullets

Average

proportion

of shell

Hatchability

of
fertile eggs

1934

/ Thick shell

per cent
9.84
9.44

9.94
9.24

per cent
82.1

\ Thin shell

58.5

1935

f Thick shell

90.3

\ Thin shell

76.2

a pronounced glossy finish. Data on the California Station Leghorn flock
would indicate that a single gene difference is usually found, the chalky
finish being dominant.

Albumen Quality. — Hoist and Almquist (1931) have produced evi-
dence that individual hens were surprisingly constant in the proportion
of total albumen found to be in the firm condition. Since then, other re-
ports (Lorenz, Taylor, and Almquist, 1934; Knox and Godfrey, 1934)
have given evidence for the inheritance of this character. From further
breeding results, the amount of firm white now seems to be the result of
the expression of a considerable number of genes ; crosses between high
and low lines give intermediate results. Strains producing eggs with
characteristically large quantities of firm white can be created by breed-
ing. All experiments designed to modify the proportion of the firm white
of freshly laid eggs by nutritional or other means have failed.

Van Wagenen and Hall (1936) use a different criterion of albumen
quality : they make a visual comparison of the appearance of the albumen
with a photograph of eggs classified according to quality. This measure,
known as the albumen-condition score, is intimately correlated with the
actual height of the firm albumen observed in egg contents removed from
the shell and membranes. On the other hand, there is considerable evi-

30 University of California — Experiment Station

dence to show that the percentage- firm-white and albumen-height meas-
ures are not necessarily closely correlated in eggs from certain hens. A
good albumen-condition score is probably dependent upon many genes
for its expression.

Good quality is required of a high-grade market egg. Lorenz and Alm-
quist (1935) found that eggs with an original high percentage of firm
white are not only of superior quality when fresh but also after storage
since the rate of liquefaction of the albumen in such eggs is lower than
in eggs with lower percentages of firm white. Hall and Van Wagenen
(1936) have also claimed higher hatchability for eggs with the best
albumen-condition scores. Some tests conducted at the California Sta-
tion have not confirmed this. There seems to be no association between
the percentage of firm white and hatchability.

Hatchability. — In our experience, the greatest differences found in
hatchability in commercial poultry flocks have been produced by nutri-
tional factors. Insufficient amounts of one or several vitamins and min-
eral substances in the rations of breeding birds frequently cause poor
hatchability under practical conditions of poultry breeding. In order to
obtain high hatchability as well as chicks of good quality, the breeding
ration must not be deficient in these essential nutrients. 7

Under good conditions of nutrition and management, variation in
hatchability between matings is common. Hays and Sanborn (1924) have
postulated a genetic explanation by which a single gene controls hatch-
ability : HH individuals hatch 85 per cent or more of fertile eggs, Hh 55
to 84 per cent, hh less than 55 per cent. Jull (1932) has produced evi-
dence which would indicate a more complex inheritance of hatchability.
His view is supported by the fact that some six or more lethal genes are
known to cause embryonic death in chickens. Practically, however, Hays'
method of selecting birds hatching 85 per cent or more is an effective
means of selecting inherently high-hatching stock ; at the same time, it
allows for a reasonable amount of embryonic deaths produced by non-
inherited conditions.

BREEDING SYSTEMS

The comparative value of various systems of breeding, by which improve-
ments of characters in poultry have been made, has long been a subject of
debate by poultry breeders. The principal systems which have been used
involve close inbreeding, line breeding, outbreeding, crossbreeding, and
grading.

7 Almquist, H. J., T. H. Jukes, and W. E. Newlon. Feeding chickens. California Agr.
Ext. Cir. 108:1-38. 1938.

Bul. 626] Breeding for Egg Production 31

Close Inbreeding and Line Breeding. — Some degree of inbreeding is
involved when any mated pair of birds have a common ancestor. The
closer the common ancestor is to the individuals mated and the greater
the number of common ancestors, the greater the degree of inbreeding.
Closest inbreeding is attained by successive generations of full brother
to sister matings. High degrees of inbreeding can also result from mating
a larger number of generations of less closely related individuals, such
as are involved in half brother by half sister, father by daughter, mother
by son, and cousin by cousin matings.

Numerous reports of the detrimental effects of close inbreeding are
available. Four or five generations of full brother by sister matings have
generally given reduced hatchability, decreased viability of chicks and
adult stock, decreased growth rates, and poorer average egg production
(Cole and Halpin, 1922 ; Dunn, 1923 ; Hays, 1924 ; Dunkerly, 1930; Jull,
1933a) . When less-intense inbreeding, involving the mating of more dis-
tantly related birds, has been practiced, high degrees of inbreeding have
been obtained with fewer detrimental effects in some instances (Jull,
1933a; Waters and Lambert, 1936). The latter procedure may permit
more rigorous selection against undesirable genes, which often seem to
be recessive and lethal or semilethal in action. Under intensive inbreed-
ing, such genes become homozygous very rapidly in a considerable pro-
portion of the progeny and lead to serious detrimental effects. A more
gradual approach to homozygosity permits a more effective elimination
of them as they are discovered. Since lethal genes are apparently very
widely distributed in chickens, utilization of very close inbreeding, while
satisfactory for quickly increasing homozygosity, is often accompanied
by unfavorable results which make the practice undesirable. Asmundson
(1935) and Marsden and Knox (1937) have obtained similar results in
inbreeding turkeys.

Line breeding involves the repetition of certain desirable individuals
and their offspring in successive generations for the production of inbred
strains. By this method a male may be mated to his daughters and then
later to the female offspring resulting from the mating with his daugh-
ters. Similarly a female may be mated to her son and later to a male pro-
duced from the mating with her son. Line breeding attempts to concen-
trate the favorable qualities of an individual in a flock by inbreeding in
the above manner.

Systems involving mating of less closely related individuals have been
widely used and have produced good results. It is common practice to
line-breed for several generations and also to mate individuals not more
closely related than first cousins. Even in such matings, detrimental ef-

32 University of California — Experiment Station

fects may be produced in some cases. But, if the stock is vigorous and lias
previously been selected for desirable characters associated with hatcha-
bility, growth, and viability, successful matings may be expected.

Outbreeding and Crossbreeding. — The opposite of inbreeding is in-
volved in outbreeding and crossbreeding. In outbreeding, or outcrossing,
birds of the same breed or variety but of nonrelated strains are mated.
In crossbreeding, the birds are of different breeds or varieties. One type
of outbreeding recently advocated (Waters, 1937) has involved topcross-
ing, in which two nonrelated strains are crossed and their offspring in
turn mated to another strain not related to the first two. Frequently the
strains so mated are themselves inbred to some degree, but after out-
breeding to nonrelated strains, the degree of inbreeding as calculated on
the basis of a reasonable number of generations is reduced to zero.

Outbreeding may be resorted to when some desirable characters are
lacking in the strains bred. Its proper use involves the introduction of
necessary genes into the strains to be improved. Care should be taken to
make test matings to measure the results of the outcross before new stock
is indiscriminately mated with known strains. The combination of char-
acters from the crossing of genetically different strains at times gives
rise to very favorable performance in the offspring (Warren, 1930), in
which case the two strains are said to "nick," and at other times they may
give an inferior type of offspring. Many strains of birds have been ruined
by the unwise introduction of new stock by matings involving the whole
flock, or the greater part of the best birds in the flock, without previous
test of the progeny of such matings. On the introduction of new breeding
strains into the flock, a number of test matings should be made and the
progeny given a chance to prove their qualities before extensive matings
between the two strains are attempted.

Successful crossbreeding requires the maintenance of two or more
breeds or varieties with such characteristics that, when they are crossed,
the first-generation offspring are superior to the parental stock (Warren,
1930). The product of the cross is said to show hybrid vigor and is of
superior commercial value. These crossbred birds are not satisfactory
breeders as a rule, and therefore the cross must be repeated when more
crossbred stock is desired. The average poultry breeder has sufficient
difficulties in breeding one strain of poultry without multiplying his
work with more breeds or varieties. Only in cases where a large produc-
tion of crossbred chicks is possible, does a sufficient emphasis on good
breeding of the parental stock become warranted economically. Some
breeds and strains are not suitable for crossing because they transmit un-
desirable characters to their offspring, and hybrid vigor does not result.

Bul, 626] Breeding for Egg Production 33

Grading. — When a flock is found to be deficient in many respects, a
poultryman may resort to a system of grading. In this system, males are
introduced each year from a superior strain or breed and are mated to
successive generations of selected breeding hens until the flock comes
to resemble more and more the quality of the superior birds purchased.
Lippincott (1920) has shown that flocks of mongrel fowls can be rapidly
improved by this system of breeding. It is also widely used by multipliers
of commercial egg-production stock, who return each year to the same
poultry breeder for breeding males.

A slight degree of inbreeding may be involved in the practice of grad-
ing, the extent depending on the relationship between the males pur-
chased in different years. Such relationship is usually not sufficiently
close to produce any detrimental effects and there is no reason to change
to another strain of breeding males if satisfactory results have been ob-
tained.

Value of Breeding Systems. — The purpose of inbreeding is to increase
the uniformity of stock by increasing the number of pairs of genes in
the homozygous condition. Outbreeding and crossbreeding, on the other
hand, definitely increase heterozygosity and lead to less uniformity in
breeding results in subsequent generations. If the poultry breeder desires
to obtain a flock which will breed reasonably true for desired characters
for a number of generations, then some degree of inbreeding must be in-
volved. If the breeder desires only one generation of stock possessing a
maximum of qualities for a specific purpose, then outbreeding or cross-
breeding may produce the type of stock desired.

The success of various systems of poultry breeding rests not so much
in the method chosen as in the use of records for the elimination of un-
desirable characters and selection of superior birds for mating. Blind
adherence to any method is apt to lead to failure. The value of adequate
records rests in the information made available to the breeder, who then
can intelligently choose the system most likely to produce good results.
The system, of course, may vary from year to year, with the quality of
the stock available.

RELATION OF GENETICS TO POULTRY-BREEDING

METHODS

Pedigree Breeding. — Knowledge of genetics can be applied to best ad-
vantage in the improvement of strains of poultry only with the aid of
pedigree breeding. Information on the characters exhibited by individ-
ual birds must be supplemented by knowledge regarding the same char-
acters in the ancestors of the mated birds in order to determine the

34 University of California — Experiment Station

degree of probable homozygosity for desirable genes. The best test of a
breeding bird's value is found in the performance of its progeny, the
determination of which demands complete pedigree records. Other meas-
ures of the probable value of a breeding bird rely on the performance of
the family of which the bird was a member. Here, too, pedigree records
are required to establish relationship.

With adequate records of pedigree and of individual characteristics
and with a knowledge of the principal genes concerned with egg produc-
tion in birds, results to be expected in the offspring from a given mating
may be predicted with a fair degree of accuracy. Once a mating is ade-
quately tested and known to give good progeny, it may then be repeated
as long as the original breeders give viable offspring with the knowl-
edge that the future offspring will perform reasonably like their older
brothers and sisters.

Pedigree breeding, coupled with a knowledge of genetics and progeny-
test data, eliminates many of the hit-or-miss results obtained by other
methods of mating in which the relationship of birds is not accurately
known. It also makes possible more rapid improvement in specific heredi-
tary characters which are controlled by several pairs of genes than can
be obtained by simply selecting individuals showing the desired char-
acters. Today, practically all of the leading breeders of egg-production
strains of poultry pedigree their stock.

Pedigree Records. — Trapnesting, pedigreeing, and record keeping are
futile if an analysis of records is not made in such a way as to reveal not
only the best individual performances but also the best means of repro-
ducing superior offspring. Record-keeping systems may be simple or
complex according to the number of characters for which selection is
practiced. Any system which demands so much work in its keeping that
adequate analysis of records is prevented is a detriment to, rather than an
aid in, breeding. The rule should be to simplify record taking to the point
where further reduction in scope would eliminate essential information
on characters of importance in the selections made.

We have often been asked to recommend record forms for pedigree
breeders of poultry. In our opinion no one system of records is best for
all breeding purposes. Pedigree breeders should be encouraged to de-
velop their own record forms to meet their own needs. In any case,
some form should be provided for each of the following : (1) daily egg-
production record of individual birds, (2) hatching and chick-banding
records, (3) mating records, and (4) family summaries. The latter
should summarize the performance of individual members of each family
with respect to the characters affecting production and viability. Wher-

Bul. 626] Breeding for Egg Production 35

ever possible, records of mortality and autopsy findings, as well as of
quality characteristics of the birds and of the eggs produced, should be
taken and summarized.

Sib and Progeny Testing. — Recent developments in breeding methods
which have come into popular use involve the analysis of records of fam-
ily performance. Males and females from families in which the sisters
have made known records of performance are called "sister-tested."
When records of growth, viability, feathering, and other characters are
also available for the brothers of such individuals, the term "sib-tested"
is used. The term "sib" here refers to the family of full brothers and
sisters. Similarly the sires and dams which produced the family are
called progeny-tested. The sib and progeny tests may reveal the family
to be superior, average, or poor. These tests enable the selection of breed-
ing stock from the superior individuals belonging to superior families.
The genetic composition of breeding birds is more easily revealed by a
study of family records than in cases where no sib or progeny data are
available.

The value of sib, sister, and progeny tests depends partly on the num-
ber of offspring tested for each mating and partly on taking unselected
groups of individuals for the tests. From matings which give a high de-
gree of uniformity in production and viability of the offspring, as few
as five pullet offspring may suffice for a reasonably accurate progeny
test, especially when the hatching season is limited. More offspring are
usually desired for matings giving a wider segregation of qualities ; in
these cases probably not less than ten should be tested. Any means of
selection which tends to eliminate certain types of offspring from the test
leads to a false measurement of the breeding value of the mated birds.
Thus an accurate progeny, sister, or sib test demands an unselected and
unculled group.

When data from accurate progeny tests are available, the breeder has
eliminated a considerable factor of chance in results to be expected from
repetition of matings. The egg production of the groups of offspring re-
sulting from the same mating may vary in noninherited characters from
year to year; also in cases where environmental conditions are known to
interact with genes controlling inherited characters, variation between
offspring produced in different years or seasons may occur. However,
the greater part of the genetic control of production characters will be
expressed alike in groups of pullets hatched in different years.

Table 6 gives some data on results obtained from the California Station
flock on repeated matings. Despite considerable differences in environ-
mental conditions during the years in which these data were taken, the

University of California — Experiment Station

relative order of superiority of the matings was maintained. Unfortun-
ately, data here presented are limited both in the number of repeated
matings and in the number of daughters per mating. Frequently, pro-
duction indexes of the daughters of a mated pair will vary in different
years by 20 to 30 eggs or more. Rarely, however, will a mating that has
proved to be superior in one year give inferior progeny in another year,
or vice versa. Jull (1933&) has reported even closer agreement between
groups of offspring produced by the same matings.

TABLE 6

Production Indexes of Progeny Produced in
from the Same Matings

Different Years

Class

by first year of

mating

Number

of
matings

Year of mating

Number

of
daughters

Production
index

100.1-125.0

f First

16
15

21
17

62
37

24
20

108.6
114.7

125.1-150.0

f First

149.6
170.9

150.1-175.0

175.1-200.0

( First

187.4
174.9

200.1-225.0

f First

207.9
185.6

The relative value of sister tests in breeding for egg production can be
illustrated by data on the same flock involving a greater number of birds
(table 7) . While the individual egg-production record of the dam showed
no correlation with the production index of the daughters, the produc-
tion index of the family of full sisters including the dam bore a direct
relation to the production index of the daughters. For this group of
birds, the production of the dam's family of full sisters is certainly a
better criterion of breeding value than the individual record of the dam.
These results indicate the desirability of selecting good families from
which good individuals may be chosen as breeders. However, Jull (1933fr)
has been able to find no correlation between the average production of
full sisters of the dam and that of the daughters in his work. An expla-
nation for the differences between Jull's results and ours in this respect
is not available.

Whether or not the individual record of the dam bears any important
relation to the production ability of the daughters has been strongly de-

Bul. 626]

Breeding for Egg Production

bated. Jull (19335) has found little or no correlation between dam's and
daughters' egg-production records. In this respect, his results are similar
to ours given in table 7. Goodale (1935), on the other hand, has found
some correlation between the production of the dam and that of the
daughters. These two conflicting reports and also the data given above
can perhaps be reconciled to a certain degree. In Jull's data and in ours,
the range of production of the dams was relatively narrow (between 200
and 300 eggs), while in Goodale's data a number of groups of dams had

TABLE 7

Belation of Production Indexes of 262 White Leghorn

Pullets to the Average Record of Dam and to

the Average Production Index of the

Dam and Her Full Sisters

(Each group represents approximately one-quarter of

the total number of pullets)

Classification by dam's record

Classification by family records

Average
record of

dams

Production
index of
daughters

Average

production

index of

dams and their

full sisters

Production
index of
daughters

eggs
287.6
270.5
250.2
226.6

183.4
214.4
191.0
185.3

230.3
201.0
191.6
108.5

215.1
213.0
205.0
139.2

produced less than 200 eggs and others had produced over 300 eggs. It
still is possible to conclude that when the production of the dams is be-
tween 200 and 300 eggs, close correlation with daughters' records is often
not obtained.

Another valuable use for progeny tests is found in connection with the
introduction of new and unrelated breeding stock into a strain, as pre-
viously discussed under "Outbreeding and Crossbreeding" (p. 32).

Progeny testing is also of value in experimental matings. The number
of such matings possible under commercial poultry-breeding conditions
usually must be small. However, in the breeding of any flock, conditions
will often arise which cannot be answered from previous experience. In
these cases, the poultry breeder is justified in making matings designed
to reveal new facts which will guide him in future breeding procedure.

Multiplication of Superior Stock. — The multiplier may buy superior
stock from the pedigree breeder and reproduce it by generations of non-
pedigree matings. His job is to increase the number of superior birds

University of California — Experiment Station

available for commercial poultry keeping with as little loss of quality in
the strain as possible. He may sell his products as pullets or adult stock
to commercial poultrymen or as eggs to the hatcheries, which in turn sell
baby chicks to the commercial poultrymen. In either case, the multiplier
serves a very useful purpose : he brings the improvements gained by
pedigree breeding within the means of the commercial poultryman, who
often cannot afford to buy pedigreed stock for the production of market
eggs, even if unlimited quantities were available. The multiplier who, by
lack of an effective breeding program, makes no effort to improve the
quality of his stock, is of doubtful service to the poultry industry.

TABLE 8

Production and Mortality with Respect to Quality of White

Leghorn Pullets as Determined at Five Months of Age

Grade

at 5 months

of age

Total

number of

birds

Per cent

flock

Production
index

Per cent
pathology

263
783
31

24.4
72.7
2.9

100.0

182.2
150.9

62.8

156.0

40.3

48.5
77.4

All grades

1,077

47.3

The males used in multipliers' flocks should be from pedigreed ances-
try. The most rapid dissemination of good qualities can only be obtained
by mating superior males with selected hens. The females can be culled
frequently or trapnested for short periods of time in order to select the
superior individuals. In flocks where the pullets will later be used for
breeders, a constant selection of stock should be practiced from hatching
time until the matings are made. Birds showing retarded development
should be eliminated throughout the growing period. At 5 to 6 months of
age, the pullets should be graded and segregated according to their de-
velopment. First-grade birds will, on the average, prove to be the best
stock. Second-grade birds often are profitable commercially but do not
average as well as the first-grade stock. Third-grade birds are under-
developed and should be culled, because they rarely prove profitable even
under favorable commercial conditions.

Table 8 gives production and mortality data on three grades of birds
from the California Station flock. This type of results has been repeatedly
obtained in grading birds over a period of years. Profitable levels of pro-
duction were reached only in the first and second grade. Pathological
conditions in the first and second grades were lower than in the third
grade.

Bul. 626] Breeding for Egg Production 39

Segregation into grades as here used was based on the comparative de-
velopment of the birds. In the first grade, only birds of good size and
weight, without coarseness, with good pigmentation of shanks and beaks
(in case of yellow-skinned breeds), and showing freedom from any appar-
ent disease, were included. The second grade consisted of birds not so well
developed as those in the first grade but also apparently free from dis-
ease. The third grade included all small birds, as well as those with poor
pigmentation and those showing signs of the onset of pathological condi-
tions. One of the most frequent of these was the beginning of blindness
produced by an iritis.

While many birds belonging to the second grade will make good
production records, the character of their development should cause hesi-
tation about their promiscuous use as breeders. A strain of birds char-
acterized by rapid and sturdy growth is desired in commercial poultry
production. Under the same flock conditions, the first-grade pullets de-
velop in a superior manner and, provided they meet all other standards,
are to be preferred as breeders. Third-grade birds should be marketed as
poultry meat or destroyed.

Where circumstances prevent the trapping of potential breeding birds
for a complete year of production, trapping for a period of two to three
months is a valuable aid in selection of stock. When combined with the
use of culling methods, part-time trapnesting can be used to determine
with good accuracy the productive ability of a bird. It is particularly
valuable for determining the ability of the bird to lay at a rapid rate,
which cannot be established by culling methods alone. However, culling
methods can be used to determine with reasonable accuracy the length of
time a bird lays if applied during usual periods in which the birds are
maturing, pausing, or molting.

A Breeding Program for Multipliers. — While it is not considered ad-
visable to present a rigid program of breeding procedure for pedigree
breeders, because of the variability of conditions encountered in different
flocks, a general breeding program for multipliers may be advanced.
Such a program can be presented in the form of a summary of the pre-
ceding discussion. The application of the following specific recommen-
dations should be limited to birds hatched in the spring season, from
February to April :

1. Remove all poorly developed birds at any stage of development.

2. Segregate the pullets at five to six months of age and discard all
birds falling into the third grade.

3. If part-time trapnesting can be used for birds aged six to eight
months, eliminate pullets laying at less than 65 per cent net rate. This

40 University of California — Experiment Station

procedure will also enable elimination of individuals producing unde-
sirable types of eggs.

4. Remove late-maturing or pullet-molting birds from the breeding
flock in November or December of the first laying year.

5. Remove all birds which become broody in the course of the year.

6. Cull the flock in the summer or early fall to eliminate early molters
and poor producers, as determined by the usual methods of judging past
production.

7. Remove all diseased birds and pullorum reactors.

8. Mate selected females to sib- or sister-tested pedigreed males from
high-production stock which also exhibits high viability under commer-
cial flock conditions.

9. Do not set small, overly large, misshapen, poor-shelled, or off-colored
eggs.

Age of Breeding Stock. — The comparative value of young and old
birds as breeders is frequently debated. If natural selection operates to
remove the weaker birds and if these birds were weak owing to genetic
causes, then older birds should unquestionably be superior to the young
birds of the same flock as breeders. However, the individuality of birds
frequently does not reveal their complete genetic constitution. Even in
families with low average viability, there are often individuals which
live to an advanced age. In numerous instances in the California Station
flock, these birds have given offspring with poor viability. Age, in itself,
is no guarantee of the ability of an individual to transmit high viability
to its offspring. On the other hand, a flock of old birds should have been
reduced by culling and natural selection to the point where its offspring
should excel offspring from the same flock before the poorer and weaker
individuals had been eliminated. So far as is known, the same bird pos-
sesses the same ability to transmit desirable inherited characters to its
offspring when it is a pullet as at any later time in its life (Hays, 1928) ,
although a change in the incidence of certain diseases may greatly affect
the amount of mortality in the offspring from different years of mating.

In commercial stock produced without the aid of pedigree records,
breeding from older hens and cocks is advisable unless younger stock
from strains characterized by superior viability are available. Where
pedigrees are known, family characteristics and proved breeding ability
rather than age of the breeding bird should be the decisive factors.

Bul. 626] Breeding for Egg Production 41

WHAT BREEDING CAN ACCOMPLISH

As previously stated, many characters of production and viability in
poultry appear to result from the interaction of genetic and environ-
mental factors. Improvement in any flock may represent actual increase
in number of desirable genes or it may be merely the reflection of im-
proved nutrition, disease-control methods, or management. Munro (1937)
has given strong evidence that much of the improvement in average egg
production attributed to breeding selection is not genetic in origin. Cer-
tainly no one can dispute the fact that improvements in knowledge of
poultry nutrition during the last two decades have made possible a bet-
ter average winter egg production than was previously attainable.

To present concrete evidence for an undiluted effect of breeding selec-
tion, all controllable nongenetic influences must be kept constant as far
as is possible ; or, if they do vary, their effect must be measured independ-
ently from the effect of the genes. Such conditions have prevailed for a
number of years in the Single Comb White Leghorn breeding flock of the
University, and it is possible to state that no environmental changes
associated with nutrition or management can account for the results ob-
tained, with the exception of an improvement in early growth and sexual
maturity obtained by nutritional means. The results presented in table 9
thus give a reasonably accurate measure of the effect of the breeding
selections practiced for all except the above-mentioned characters.

Egg Production. — As can be seen, a rather unusual improvement in
the average productive level of the flock was obtained. Associated with
the improvement in number of eggs produced were earlier sexual ma-
turity, higher rate, higher persistency as measured by age at last egg,
and a reduced pathology. No material improvement with respect to
broodiness was obtained except in the last two years of mating. While
the amount of winter pausing was reduced, improvement was not con-
sistently obtained. The amount of spring and summer pausing, however,
was greatly reduced. The improved production was not associated with
any sacrifice of desirable egg size or egg color.

During the years represented by the data in table 9, improvement was
obtained in family lines originally represented in the first year of the
study, as well as in new families originating from purchased stock
crossed with the foundation lines. Whether the stock originated from
families present in the original flock or from crosses with other strains,
the progeny produced was submitted to an analysis of production char-
acters and family performance before any further use as breeders was
made of the parents. Of five males from other strains which were mated

University of California — Experiment Station

with selected hens of the Station flock in the course of these years, two
were proved satisfactory by the progeny test for continuation in the
breeding flock. In the case of the three unsatisfactory males, only a few
progeny were saved for breeding. Since matings headed by males not

TABLE 9

Production and Pathology Characteristics of the California Station Flock

of Production-Bred Single Comb White Leghorns

1933

1934

1935

1936

Number of pullets banded

535

704

485

403

Average weight at banding time

2.59 lbs.

2.65 lbs.

2.98 lbs.

2.90 lbs.

Nov. 12

Oct. 13

Oct. 4

Sept. 16

Average age at first egg

211 days

195 days

187 days

174 days

Per cent of birds maturing under 200 days of

43 p.ct.

64 p.ct.

71 p.ct.

84 p.ct,

Per cent of birds with high rate*

52 p.ct.

76 p.ct.

73 p.ct.

Per cent of birds with winter pause, 7 days

67 p.ct.

65 p.ct.

52 p.ct.

Per cent of birds with spring or summer

37 p.ct.

34 p.ct,

15 p.ct.

18 p.ct.

13 p.ct.

9 p.ct.

Average date of last eggf

Oct. 13

Oct. 4

Oct. 8

Oct. 17

557 days

566 davs

Per cent of birds laying October 1 at end of

50 p.ct.

67 p.ct.

69 p.ct.

Per cent of birds attaining 2-oz. egg size

72 p.ct.

74 p.ct.

76 p.ct.

79 p.ct.

2 p.ct.

3 p.ct.

2 p.ct.

Per cent pathology to September 1J

38.3 p.ct.

32.7 p.ct,

27.8 p.ct.

33.8 p.ct,

Per cent pathology to October It

55.0 p.ct.

38.1 p.ct.

36.8 p.ct.

36.5 p.ct.

Per cent reproductive pathology for year§. . .

22.0 p.ct.

14.3 p.ct,

14.2 p.ct.

18.6 p.ct.

Per cent neoplastic pathology for year§

10.6 p.ct.

9.1 p.ct.

8.9 p.ct.

6.7 p.ct.

Per cent digestive pathology for year§

19.2 p.ct.

7.1 p.ct.

6.8 p.ct.

4.7 p.ct,

Per cent functional pathology for year§^

16.2 p.ct.

10.6 p.ct.

10.3 p.ct,

10.7 p.ct.

Average 365-day egg record of survivors

217 eggs

234 eggs

239 eggs

243 eggs

175.6 eggs

206.6 eggs

241.9 eggs

249.7 eggs

126.3

170.0

186.0

196.0

Production index to October 1 (average age

125.6

171.8

186.5

201.6

* See figure 4 for definition of high rate.

t Records closed on January 15.

% Records based on original number of pullets banded at five months of age. The difference between
pathology to September 1 and that to October 1 largely represents the proportion of the flock culled in
September which showed pathological lesions.

§ Includes mortality and birds disposed of at the end of the year which showed pathological lesions.
The figures for different types of pathology are nonadditive, since many duplications exist.

t Largely degeneration of the kidney and liver.

bred by us accounted for less than 20 per cent of the total matings in the
years cited, the whole improvement in the flock cannot be attributed to
outbreeding with birds of other strains. However, a valuable hereditary
contribution was made by the two introduced males which proved to be
satisfactory.

Bul. 626] Breeding for Egg Production 43

Viability. — During this period the flock has been free from all known
infectious diseases except coccidiosis which, however, caused the death of
a very few birds and was a minor pathological factor. The preceding
statement does not refer to neuro- and visceral lymphomatosis, the in-
fectiousness of which has not been demonstrated for this flock (Beach,
1938). Yet the proportion of the flock showing pathological conditions
has not been low.

An examination of the types of pathology present in this flock empha-
sizes the importance of loss from diseases or functional disturbances of
unknown origin and for which no means of control has been devised

TABLE 10

Occurrence of Neoplastic Diseases in the Offspring Produced

from matings of blrds from families showing

Frequent Occurrence of Neoplasms

Year

Number
of birds

Per cent
neoplasms

1933-34

52
129
152
134

13.5

1934-35

25.5

1935-36

27.6

1936-37

23.9

(Lubbehusen and Beach, 1938). Reproductive and functional types of
mortality have not been specifically selected against. Losses from pathol-
ogy associated with the digestive system in the 1933 flock were un-
usually high and were found predominantly in the families from one
male and his son. After elimination of these families, losses from dis-
orders of the alimentary system dropped sharply for the succeeding
years.

Neoplastic diseases (including neurolymphomatosis, lymphoid infil-
tration of organs, iritis, tumors, etc.) have declined slowly during the
four years reported. At the same time, deliberate efforts were made to
breed a strain of birds susceptible to these neoplasms. Data in table 10
indicate that deliberate breeding of birds from families showing a fre-
quent occurrence of neoplasms may result in increasing losses from
pathology of this kind. Birds produced from these matings received the
same rations and management as those from the production-bred line
in which a decrease in incidence of neoplasms has occurred (table 9) . In
fact, birds from high- and low-mortality selections were hatched in the
same incubators, reared in the same brooders, and kept in the same laying
houses for their first year of production. Yet while, in the production-
bred flock, a decrease occurred in the proportion of birds exhibiting neo-

44 University of California — Experiment Station

plasms, the families shown in table 10, after a marked first-year increase,
maintained approximately the same higher level of neoplastic pathology.
The evidence from these matings indicates a possible genetic basis for
susceptibility or resistance to certain neoplastic diseases. Similar evi-
dence has been obtained by Williams, Gildow, and Lampman (1938) and
Wilcke, Lee, and Murray (1938). Probably either the hereditary basis
for resistance is complex or it involves complicated interactions with
nongenetic influences.

Breeding for disease resistance is slow and costly. Where adequate
means are available to control diseases by means of agglutination tests,
vaccines, specific drugs, or sanitation, trying to breed for resistance is
uneconomic. Where no means of control are available and where there
are evidences of different responses by birds of different strains or fam-
ilies, genetic selection is advisable, even though improvement of the stock
comes slowly.

Some poultrymen are prone to suggest that, if birds are properly
selected and mated, strains will be created with such inherent vigor
that they can withstand adverse conditions commonly found on poultry
ranches. But this seems more than can logically be expected to result
from good breeding. Inadequate or deficient rations, insanitary condi-
tions, and generally poor management cannot be overcome by better
breeding of poultry.

Importance of Other Conditions. — In presenting the results obtained
with the California Station flock, we have attempted to show how iden-
tification of and selection for characters of production and viability have
resulted in improved average egg production in successive generations.
Poultry breeding is but one means for improvement of conditions in
poultry production. As such, its contribution must be added to the con-
tributions from other sciences as applied to poultry for the solution of
the problems which the poultryman faces in his business. Breeding can
create stock with improved potentialities for profitable production, but
such potentialities will not be realized unless the birds are well fed,
housed, and managed. Good breeding probably will not help the ineffi-
cient poultryman; it can expand the possibilities for success on the part
of the poultryman whose methods of feeding and flock management are
sound.

Bul. 626] Breeding for Egg Production 45

LITERATURE CITED

ASMUNDSON, V. S.

1927a. A study of first year egg production of S. C. Rhode Island Reds. Sci. Agr.

8:141-50.
1927&. A study of the changes in the distribution of first year eggs within a flock

of S. C. White Leghorns at the University of British Columbia. Proc. 3d

World's Poul. Cong. [Ottawa], p. 129-37.
1928. The first year egg production of Barred Plymouth Rocks. Sci. Agr. 9:90-102.

1935. Some results of turkey breeding work at the California Agricultural Station.
California Turkey News. August ; p. 3-4.

1938. Influence of various factors on egg production in turkeys. Jour. Agr. Res.
56:387-93.
Asmundson, V. S., and J. Biely.

1930. Effect of pullorum disease on distribution of first year egg production. Sci.

Agr. 10:497-507.
1932. Inheritance of resistance to fowl paralysis (neurolymphomatosis galli-
narum) I. Differences in susceptibility. Canad. Jour. Res. 6:171-76.
Axelsson, J.

1936. Untersuchungen iiber die Form der Huhnereier. Proc. 6th World's Poul.
Cong. [Leipzig]. 2:1-4.

Beach, J. R.

1938. Experiments in the transmission of fowl lymphomatosis by inoculation.
Poultry Sci. 17:67-71.
Benjamin, E. W.

1920. A study of selections for the size, shape and color of hens' eggs. N. Y. Agr.
Exp. Sta. (Cornell) Mem. 31:195-312.

Bissonnette, T. H., and A. G. Csech.

1937. Hatching pheasant chicks on Christmas day. Amer. Nat. 71:525-28.
Buckner, G. D., J. H. Martin, and W. M. Insko, Jr.

1930. Some chemical factors governing egg formation in the hen. Proc. 4th World's
Poul. Cong. [London]. Sec. B; p. 330-33.
Byerly, T. C, and W. H. Burrows.

1936. Studies of prolactin in the fowl pituitary. II. Effects of genetic constitution
with respect to broodiness on prolactin content. Proc. Soc. Exp. Biol, and
Med. 34:844-46.

Clark, L. B., S. L. Leonard, and G. Bump.

1937. Light and the sexual cycle of game birds. Science 85:339-40.
Cole, L. J., and J. G. Halpin.

1922. Results of eight years of inbreeding Rhode Island Reds. Anat. Rec. 23:97.
Dryden, J.

1921. Egg-laying characteristics of the hen. Oregon Agr. Exp. Sta. Bul. 180:1-96.

DUNKERLY, J. S.

1930. The effect of inbreeding. Proc. 4th World's Poul. Cong. [London]. Sec. A;
p. 48-71.
Dunn, L. C.

1922. The relationship between weight and the hatching quality of eggs. Connec-
ticut (Storrs) Agr. Exp. Sta. Bul. 109:92-114.

1923. Experiments on close inbreeding in fowls. Connecticut (Storrs) Agr. Exp.
Sta. Bul. 111:139-72.

46 University of California — Experiment Station

Gildow, E. M., J. K. Williams, and C. E. Lampman.

1936. The transmission of fowl paralysis (lymphomatosis) Poultry Sci. 15:244-48.
Goodale, H. D.

1918. Internal factors influencing egg production in the Ehode Island Bed breed

of domestic fowl. Amer. Nat. 52:65-94, 209-32, 301-21.
1935. Eelationships between egg production of ancestors and offspring. Poultry
Sci. 14:295.
Goodale, H. D., E. Sanborn, and D. White.

1920. Broodiness in domestic fowl. Data concerning its inheritance in the Ehode
Island Eed breed. Massachusetts Agr. Exp. Sta. Bui. 199:93-116.
Greenwood, A. W.

1932. The value of progeny in relation to age of dam. Harper-Adams Utility Poul-
try Jour. 17:478-80.
Hall, G. O.

1935. The value of the pedigree in breeding for egg production. Poultry Sci. 14:
323-29.

Hall, G. O., and A. Van Wagenen.

1936. The association of certain measures of interior egg quality with hatchability.
Poultry Sci. 15:501-6.

Hays, F. A.

1924. Inbreeding the Ehode Island Eed fowl with special reference to winter egg
production. Amer. Nat. 58:43-59.

1928. Eelation of age of parents to hatchability, livability and fecundity in the
domestic fowl. Poultry Sci. 7:106-15.

1929. The inheritance of egg weight in the domestic fowl. Jour. Agr. Ees. 38:
511-19.

1936a. Inheritance of sexual maturity in Ehode Island Eeds. Proc. 6th World's Poul.

Cong. [Leipzig]. 2:34-38.
1936b. Studies on the inheritance of persistency. Genetics 21:519-24.

1937. Inheritance of egg size and egg character. Massachusetts Agr. Exp Sta. Bui.
344:1-28.

Hays, F. A., and E. Sanborn.

1924. The inheritance of fertility and hatchability in poultry. Massachusetts Agr.

Exp. Sta. Tech. Bui. 6:20-42.
1926. Annual persistency in relation to winter and annual egg production. Massa-
chusetts Agr. Exp. Sta. Tech. Bui. 9:190-203.
1932. Types of intensity in Ehode Island Eeds. Massachusetts Agr. Exp. Sta. Bui.

286:1-11.
1934. Breeding for egg production. Massachusetts Agr. Exp. Sta. Bui. 307:1-27.
Holst, W. F., and II. J. Almquist.

1931. Measurement of deterioration in the stored hen's egg. Hilgardia 6(3) : 49-60.
Jull, M. A.

1932. Poultry breeding. 376 p. John Wiley and Sons, Inc., New York, N. Y.
1933a. Inbreeding and intercrossing in poultry. Jour. Heredity 24:93-101.
1933&. Progeny testing in poultry breeding as a means of evaluating the breeding

potentiality of an individual. Amer. Nat. 67:500-14.
Kempster, H. L., and J. E. Parker.

1936. The normal growth of chickens under normal conditions. Missouri Agr. Exp.
Sta. Ees. Bui. 247:1-47.

Bul. 626] Breeding for Egg Production 47

Knox, C. W.

1930. Factors influencing egg production. I. The influence of maturity upon egg

production in S. C. White Leghorns. Iowa Agr. Exp. Sta. Res. Bul. 119:

311-32.
Knox, C. W., and A. B. Godfrey.

1934. Variability of thick albumen in fresh -laid eggs. Poultry Sci. 13:18-22.
Knox, C. W., M. A. Jull, and J. P. Quinn.

1935. Correlation studies of egg production and possible genetic interpretations.
Jour. Agr. Res. 50:573-89.

Kopec, S.

1924. Some data referring to size, shape and weight of eggs in the domestic fowl.
Pam. Parist. Inst. Nauk. Gosp. Wiejsk. Pulawach. (Mem. Inst. Natl. Polon.
Econ. Rurale Pulawy) 5:294-327.
Lambert, W. V., and C. W. Knox.

1932. Selection for resistance to fowl typhoid in the chicken with reference to its
inheritance. Iowa Agr. Exp. Sta. Res. Bul. 153:262-95.
Landauer, W.

1937. The hatchability of chicken eggs as influenced by environment and heredity.
Connecticut (Storrs) Agr. Exp. Sta. Bul. 216:1-84.

Lerner, I. M., and L. W. Taylor.

1936. The relation of pauses to rate of egg production. Jour. Agr. Res. 52:39-47.
1937a. The measurement of sexual maturity and persistency. Poultry Sci. 16:

419-21.
193 7&. The spurious nature of the linkage between length of laying year and sexual
maturity in the fowl. Amer. Nat. 71:617-22.
Lippincott, W. A.

1920. Improving mongrel farm flocks through selected standard-bred cockerels.
Kansas Agr. Exp. Sta. Bul. 223:1-48.
Lorenz, F. W., and H. J. Almquist.

1935. Firm white of fresh and storage eggs. Poultry Sci. 14:340-41.

1936. Seasonal variations in egg quality. Poultry Sci. 15:14-18.
Lorenz, F. W., L. W. Taylor, and H. J. Almquist.

1934. Firmness of albumen as an inherited characteristic. Poultry Sci. 13:14—17.
Lubbehusen, R. E., and J. R. Beach.

1938. Poultry mortality of non-infectious origin. Jour. Amer. Vet. Med. Assoc.
(In press.)

Marsden, S. J., and C. W. Knox.

1937. The breeding of turkeys. Yearbook of Agriculture 1937:1350-66.
Munro, S. S.

1937. Inheritance of egg production in the domestic fowl. II. Increases in produc-
tion, their extent and characteristics with a discussion of causal factors. Sci.
Agr. 17:376-85.

1938. Effect of heredity on interior egg quality and shell composition. Poultry Sci.
17:17-27.

Pearl, R.

1912. The mode of inheritance of fecundity in the domestic fowl. Jour. Exp. Zool.

13:153-268.
1923. Duration of life as an index of constitutional fitness. Poultry Sci. 3:1-10.

48 University of California — Experiment Station

Pearl, R., and F. M. Surface.

1909. Data on the inheritance of fecundity obtained from the records of egg pro-
duction of the daughters of "200-egg" hens. Maine Agr. Exp. Sta. Bui.
166:49-84.
Petrov, S. G.

1935. Analysis of the development of high and low lines of Leghorns at Cornell.
Poultry Sci. 14:330-39.

PUNNETT, R. C.

1923. Heredity in poultry. 204 p. Macmillan and Co., London.

1933. Genetic studies in poultry. IX. The blue egg. Jour. Genet. 27:465-70.
Rice, J. E.

1914. Some practical points in the management of poultry for egg production.
Massachusetts State Board of Agr. Bui. 1:106-37.
Riddle, O., R. W. Bates, and E. L. Lahr.

1935. Prolactin induces broodiness in fowl. Amer. Jour. Physiol. 111:352-60.
Roberts, E., and L. E. Card.

1926. The inheritance of resistance to bacillary white diarrhea. Poultry Sci. 6:
18-23.

1934. The inheritance of broodiness in the domestic fowl. Proc. 5th World's Poul.
Cong. [Rome]. 2:353-58.

1936. Inheritance of resistance to bacterial infection in animals. Illinois Agr. Exp.
Sta. Bui. 419:467-93.

Scott, H. M., and L. F. Payne.

1937. Light in relation to the experimental modification of the breeding season of
turkeys. Poultry Sci. 16:90-96.

Taylor, L. W., and J. H. Martin.

1928. Factors influencing thickness of egg shell. Poultry Sci. 8:39-44.
Upp, C. W., and B. A. Tower.

1936. The incidence of blindness and paralysis according to family. Poultry Sci.
14:421.
Van Wagenen, A., and G. O. Hall.

1936. The inheritance of certain characters affecting egg quality. Poultry Sci.
15:405-10.

Warren, D. C.

1930. Crossbred poultry. Kansas Agr. Exp. Sta. Bui. 252:1-54.

1934. Inheritance of age at sexual maturity in the domestic fowl. Genetics 19:
600-617.
Waters, N. F.

1937. The use of inbred fowls for top-crossing. Poultry Sci. 16:348.
Waters, N. F., and W. V. Lambert.

1936. Inbreeding in the White Leghorn fowl. Iowa Agr. Exp. Sta. Res. Bui. 202:
1-55.
Wilcke, H. L., C. D. Lee, and C. Murray.

1938. Susceptibility and resistance of some strains of chickens to fowl leucosis.
Poultry Sci. 17:58-66.

Williams, J. K., E. M. Gildow, and C. E. Lampman.

1938. Some factors influencing the transmission of fowl paralysis. U. S. Egg and
Poultry Mag. 44:24-27.

18w-10, "38(4013)