THE ”STEIN AND SOUDS-NOT-FAT COMENT
IN THE MILK 0F JERSEY COWS

Thasis for tho Dogma cf M. S.
MICHIGAN STATE UNIVERSITY

Lawrenco W. Specht
1955

 

 

THE PROTEIN AND SOLIDS-NOT—FAT CONTENT
IN THE MILK OF JERSEY cows

By

LAWRENCE W. SPEggT

AN ABSTRACT

Submitted to the School of Graduate Studies of Michigan
State University of Agriculture and Applied Science
in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Dairy

Year 1955

J

 

1

AESTRACT LAURENCE W. SPECHT

The investigation was concerned with the protein and

solids-not-fat content in the milk of individual cows.

These components have assumed an increased importance in

the milk pricing picture. The object of the study was to
determine how these constituents vary and what factors cause
them to fluctuate.

Two herds of Jersey cows were sampled. An inbred
herd of twenty-three cows and a randomly-bred herd of
twelve cows provided 475 and 259 samples respectively
for analysis. The samples were drawn at bi-weekly intervals
over a fifteen month period.

The fat content was determined by the Babcock method,
the total solids by the Mojonnier, and the protein by the
Pyne modification of the formol titration. The solids—
not-fat value was obtained by difference.

Comparisons of the morning and evening samples from
the randomly-bred herd showed no significant differences
in the content of the fat, protein, total solids, and
solids-not-fat. The morning milking yielded a significantly
larger quantity of milk.

The inbred herd produced significantly more pounds of
milk and protein per cow per day. Significant differences
were found for the protein and solids—not-fat content in

favor of the inbred herd. It could not be determined

ABSTRACT LAWRENCE M. SPECHT

whether heredity or environment had the greater influence
on these values.

Analysis of the data by seasons indicated that the
highest values for fat, protein, total solids, and solids-
not-fat were reached during the winter months and that the
lowest values occurred in the summer months.

The second month of lactation yielded the highest
values for milk production and the lowest figures for the
milk solids studied. A gradual decline in milk production
and a concurrent rise in the milk solids occurred from the
second until the tenth month of lactation.

Individual lactation records of each cow's protein
production were kept. Twelve complete lactations and twelve
ISO-day records (extended to 305 days) were available for
study. The number of sires' daughters and dam-daughter
comparisons were too small to Justify any conclusions.

The effect of bi-weekly, monthly, and bi-monthly
testing programs on the accuracy of the pounds of protein
produced per cow in a single lactation was considered.

The values obtained from the bi-weekly testing intervals
were used as a basis for comparison. Values from the
monthly and bi—monthly testing intervals were in excellent
agreement with the figures obtained from the bi-weekly

testing interval.

ABSTRACT LAWRENCE W. SPECHT

Regression and correlation values were calculated
among the fat, protein, and solids-not-fat values for the
two herds and for the pooled data. Although of limited
value because of the small number of samples involved,
these values indicated the herds studied were representa-

tive of the Jersey breed.

LU

THE PROTEIN AND SOLIDS-NOT-FAT CONTENT

IN THE MILK OF JERSEY COWS

By

LAWRENCE N. SPECHT

A THESIS

Submitted to the School of Graduate Studies of Michigan
State University of Agriculture and Applied Science

in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Dairy

ACKNOWLEDGEMENTS

The author wishes to thank Dr. N. P. Ralston, under
whose direction the project was developed, and Dr. Earl
Weaver, who made it possible for the writer to undertake
graduate study. He is indebted to Dr. Dale Madden for
many helpful suggestions in the preparation of the manu-
script and for assistance with the statistical analysis.
Grateful acknowledgement is also due Dr. J. R. Brunner
for his words of encouragement and his assistance with

the analytical procedures.

ii

361562

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . .
REVIEW OF LITERATURE . . . . . . . .

Heredity . . . . . . . . . . .
Breed

Jersey Breed . . . . . . . .
Individuals within the breed

Age . . . . . . . . . . . . . . .
Stage of lactation

Nutrition . . . . . . . . . . .
Milking practices

Disease . . . . . . . . .

Season . . . . . . .

EXPERIMENTAL PROCEDURE

Analytical methods

Sampling procedure

RESULTS AND DISCUSSION . . . . .

Morning and evening variations
Differences between herds
Season

Stage of lactation

Individual lactations

Daughter comparisons by sires

111

+4 v4 H H H H
-4 ex \h 3' DJ r4

19

Testing intervals

Regression and correlation values

SUMMARY 0 O O O O O O C 0

BIBLIOGRAPHY . . . . .

iv

TABLE

II.

III.
IV.

VII.

VIII.

IX.

XI.

XII.

XIII.

LIST OF TABLES

Breed Values for Percentages of
Fat, Protein, and Solids-not—fat

Fat, Protein, and Solids-not-fat
Values Reported by Sherman . . . . . . .

Summary of Jersey Protein Values . . . . . .
Breakdown of Factors Compared . . . . . .
Morning and Evening Variations

Herd Differences by Season . . .

Pooled Seasonal Values . . . . . .

Average Daily Composition of Milk
by Stage of Lactation .

Number of Lactation for H rd Members . . . .

Milk and Protein Production
in Individual Lactations .

Dam-Daughter Comparisons

Pounds of Protein Produced in Single
Lactations with Three Testing Intervals

Regression and Correlation Values Among
Fat, Protein, and Solids-not-fat Percentages

Page

INTRODUCTION

Milk is an excellent source of high quality protein.
However, most milk has been purchased on the basis of its
butterfat content with little regard for the value of its
non-fat solids. The dairy industry is slowly changing
its pricing system to pay on the basis of the solids-not-
fat and fat content in milk. In order to give a more
equitable share of the pricing load to the solids-not-fat
portion, we must know more about the levels of protein
and solids-not-fat produced by the individual cow. It
should be determined how these levels vary and what factors
cause them to fluctuate.

Occasionally, herds produce milk which falls below
the legal limit for solids-not-fat. Until recently, this
problem has received more attention in England than in
this country. Ways of improving the solids-not-fat levels
of such herds are being sought.

To avoid being penalized for milk low in the non-
fat solids, the dairyman must either pay a premium for
cows that produce at the higher levels for the non-fat
solids and/or provide an environment that will encourage
the ultimate in solids-not-fat production. In this

respect, the information available from prior research

2
studies is not useful to the dairyman. Most research work
on the non-fat solids was collected before 1940 and was
based either on an inadequate number of samples or was
drawn from mixed herd milks.

For these reasons, this study is concerned with the
protein portion of milk and its relationship to the non-
fat solids. The assumption is made that protein is the
most important of the non-fat solids and that it is a
reliable indicator of the total solids-not-fat. Richardson
(1952) stated that the protein percentage bears a definite
relationship to the solids-not-fat percentage, and, that
if the former is known, the latter may be calculated with
a high degree of accuracy. Such a relationship makes it
possible to increase the solids-not-fat level by mani-
pulating the protein level.

The major purpose of this study is to obtain inform-
ation on protein and solids-not-fat levels in the lactations
of individual cows. Some of the points that will be
considered are variations between night and morning milkings,
herd differences within a breed, and variations due to
season and stage of lactation. In addition, comparisons
will be made of the accuracy of calculating the total
protein production of a lactation from two-week, monthly,
and bi-monthly testing periods. Interrelations among
the non-fat solids, fat, and protein percentages will be

calculated.

REVIEW OF LITERATURE

A large number of factors that influence the protein
and solids-not-fat content of cow's milk have been reported
in the literature. This review will discuss only those
major factors likely to affect a cow's normal lactation.
These may be divided into two groups:

(1) conditions due to the individual cow, arising
from either her genetic or physiological make-up. Breed
and strain within a breed are considered genetic factors,
governed by heredity, in contrast to the physiological
states of age and stage of lactation.

(2) conditions due to environmental factors. These
may or may not be readily controlled by the dairyman.

The plane of nutrition and the milking practices followed
are easily adjustable by the dairyman, while it is more
difficult to influence the effects of disease and season
on protein and solids-not-fat production.

The factors listed above will be discussed separately,
although they overlap in their effect on the protein and

solids-not-fat levels of individual cows.
Heredity

A major portion of the variation found in the protein

and solids-not-fat levels among individual cows is

kn

attributed to heredity. Bailey's review (1952) stated
that there is little reliable information on the variation
of the non-fat solids. He believed that the widely-held
assumption that the level of the solids-not-fat fraction
in milk tends to be similar for dams and their daughters
depends primarily on the fact that different breeds tend
to produce at different and characteristic levels.

However, Bonnier and Hansson (1946) statistically
analyzed the data from 2,245 samples from 29 monozygotic
twin pairs and reported the percentages of protein at
fixed values of fat percentage are equal for two identical
twins, not equal for two fraternal twins, and very unequal
for two unrelated animals. This was also the case with
lactose. It was concluded the relationships between the
fat, protein, and lactose are determined largely by inheri-
tance. Further work on this study reported by Hansson
and Bonnier (1949) substantiated their earlier findings.
These authors stated that the possibility of using this
fact in practical breeding depended upon the feasibility
of determining the protein and lactose components of the
milk in the field. Richardson, Young, Dalal, Pearce, and
Narula (1953) suggested the use of the formol titration
for protein as a possible solution to this problem.

Moore and Keener (1942) believed "proven sires could
be used to raise the percent of solids-not-fat in herds

low in this respect. They suggested that by selecting

bulls with the ability to transmit high solids-not-fat
and by eliminating mastitis it would be possible to im-
prove the compositional quality of herd milk. Rowland
(1938) and Bailey (1952) made similar statements.

The workers cited above outlined the effect of
heredity on the production of the non-fat solids, but
they did not give heritability estimates for these con-
stituents. The Swedish workers, with their monozygotic
twin studies, are probably the most advanced in this area
of study, and Hansson (1955) has suggested privately that
the heritability value for protein is close to that of the
fat percentage. The most used figure for the heritability
of fat percentage is 0.4. This is sufficiently high to
allow definite progress to be made in raising the protein
and, therefore, the solids-not-fat levels by breeding
procedures. These facts form a basis for much of the

work that must be done on the influence of heredity.
Breed

The obvious examples of hereditary are the breed
of cattle and the strains or individuals within a breed.
In order to discuss variations of protein and solids-not-
fat content, it is necessary to establish representative
levels for the various breeds. The data on breed variation
is voluminous and well-substantiated. A summary, Table I,

published by Overman, Garrett, Wright, and Sanmann,

TRBLE I

BREED VALUES FOR PERCENTAGES OF
FAT, PROTEIN, AND SOLIDS-NOT-FAT

 

 

 

 

 

 

Fat 5_ Prot. g S.N.F. g

N* Min. Max. Mean Min. max. Mean Min. Max. Mean
Ayrshire

208 2.92 5.66 4.14 2.92 4.58 3.58 7.20 10.38 8.94
Brown Swiss

428 2.92 6.48 4.02 2.60 5.74 3.61 7.99 11.44 9.40
Guernsey

321 3.65 7.66 5.19 2.65 5.45 4.02 8.19 11.10 9.68
Holstein

268 2.60 6.00 3.55 2.44 6.48 3.42 7.82 11.90 8.97
Jersey

199 3.28 8.37 5.18 2.93 5.83 3.86 7.68 11.07 9.51

* Number of samples

(1939) gives the minimum, maximum, and mean values for

the five major breeds for the percentages of fat, protein,
and solids-not-fat. These authors stated that the results,
especially in regard to the relationships between the
various components, are representative of the milk of

each breed of cows studied. A discrepancy does exist
in the values given for the Jersey and Guernsey fat

percentages, when compared to what is considered normal.
Jersey Breed

This study deals specifically with the Jersey breed,

hence a more detailed review of existing data is presented.

One of the earliest investigations was reported by
Collier (1891). The average fat percent on 238 samples
of Jersey milk was 5.61, while the solids-not-fat percent
averaged 9.80. Collier did not list protein as such, thus
its value cannot be compared with the figures given in
later studies. A figure of 3.91 percent was given for
casein content, although this makes the total protein
figure rather high. Collier's values are higher for the
average composition of fat and solids—not-fat than the
Illinois data cited.

Sherman (1906) reported on a five year study on
600 grade and registered Jerseys. The data involved the
analysis of 60 monthly samples. The samples were accurate
composites of the milk, produced by the herd at a single
evening's milking. The data were given as averages for
each month, with the lowest values for all components of
the milk occurring in July. The highest figures for the
various constituents were found to occur in the months
of December, January, and February. Table II gives the
minimum, maximum, and average values reported by Sherman
for fat, protein, and solids-not-fat.

Reder (1938) of Oklahoma gave protein percentages
for normal Jersey milk, with a range of from 3.33 t 0.04
(in the second month) to 3.85 e 0.07 (in the eleventh
month) for complete lactations. These data were compiled
from a single herd, utilizing 290 samples. The mean

protein percentage was 3.60 t 0.02.

TRBLE II

EAT, PROTEIN, AND SOLIDS-NOT-FAT VALUES
REPORTED BY SHERMAN

 

 

Minimum Maximum Average
Fat % 5.24 5.57 ' 5.42
Protein % 3.49 3.85 3.66
S.N.F. % 8.96 9.43 9.22

 

The Arizona study by Davis et a1 (1947) on two 30
cow Jersey herds gave a minimum value of 2.9 percent, a
maximum value of 3.9 percent and an average of 3.5 percent
for protein. These figures are lower than those shown
in most other references. The authors suggested that
the higher temperatures which exist in the Southwest could
have had a depressing effect on the protein and solids-
not-fat levels.

One of the most recent compilations is Richardson's
summary (1953) of data reported by the California,
Illinois, and Oregon Experiment Stations. For milk samples
that tested from 4.0 to 7.0 percent butterfat, the com-
parable range in protein values was from 3.56 to 4.15 percent
on 1,081 samples. Twenty-six samples of Jersey milk,
testing below 4.0 percent, were omitted from the tables.

The minimum, maximum, and average protein percentages
for Jersey milk reported for the work cited are summarized

in Table III. These data suggest the mean protein percentage

10
TABLE III
SUMMARY OF JERSEY PROTEIN VALUES

 

 

 

Reference Minimum Maximum Average
Sherman 3.49 3.85 3.66
Overman et a1 2.93 5.83 3.86
Reder 3.33 3.85 3.60
Davis et a1 2.90 3.90 3.50
Richardson 3.56 4.15 ----

 

for normal Jersey milk is in the range of 3.5 to 4.0

percent, but individual values may vary considerably.
Individuals within the Breed

Variations between individual cows of the same breed,
under the same environment,suggest that protein percentage
is inherited separately from either milk yield or fat
percentage.

There are reports in the literature of individuals
and strains within a breed producing low levels of solids—
not-fat. Davis et al (1947a) mentioned one "cow family"
of four individuals, that persistently secreted milk lower
in protein than the rest of the animals in a Jersey herd.
Most of the reports have been concerned with the solids-
not-fat portion of milk rather than specifically mentioning

protein in relation to this problem. Rowland (1938)

11
reported ”families" within breeds producing low solids—
not-fat milk. Richardson and Folger (1950) cited several
other researchers who reported abnormally low levels of
solids-not-fat that could not be traced to a pathological
condition in the udder.

The discussion cf abnormal levels of solids-not-
fat secretion is centered in two schools of thought. One
group believes the heredity factor is of major importance.
Others believe that the majority of cows producing milk

low in solids-not-fat have some type of udder infection.
Age

The question of changes in milk composition due to
age is still unsettled. Bailey (1952) reviewed the work
on this factor and reported that several workers found the
second lactation yielded the highest quality milk. In
this discussion, "quality" refers to the higher levels of
non-fat solids. However, Bailey's own work (1952a) showed
that the first lactation produced the highest non-fat solids
values, and that the quality of milk declines slightly
as the cow advances in age. This vieWpoint is supported
by the work of White and Drakely (1927) and Bartlett (1934).
One might conclude that younger animals produce higher
quality milk, and age does affect slightly the level of
solids~not~fat. Rowland (1938) suggested that some of the
decrease may be due to a higher incidence of udder infections

(mastitis) in the older cows. While many older cows show

12
no outward signs of infection, their milk is often lower
in quality than the average for the herd. Foot and Shat-
tock (1938) reported that often infection cannot be detected
by casual examination on the farm. Laboratory methods
are necessary to locate this "sub-clinical" infection.
These findings would suggest that many more of our dairy
cows are afflicted with some type of udder infection than
we normally think of as having "mastitis."

Bonnier and Hansson's work (1946) with monozygotic
twins is not in complete agreement with the work previously
cited. These workers found that the differences between
first and second lactations, in regard to percent of lactose
and percent of protein (which make up the bulk of the solids-
not-fat), were to a large extent due to random variations.
They concluded, "If any differences in protein percentage
between the first and second lactations did exist at all,
they were so small that they could be ignored." These
authors stated that by analogy the same conclusion could
be drawn concerning differences between all lactations,
and, therefore, all the lactations in their study were
pooled for statistical treatment. It was also stated that
most of the samples were drawn from first and second lacta-
tion cows. Therefore, their information on older cows
was limited, and the study was not complicated by a high

incidence of mastitis in these older individuals.

13
Stage of lactation

Protein and solids-not-fat content during a lactation
period vary in the same direction as the fat percentage,
but not necessarily in the same relationship. Azarme
(1938) and Bailey (1952b) both showed a decreasing level
of solids-not-fat for the first four to six weeks of the
lactation. A gradual rise occurred from the second month
until the late stages of lactation, when a sharp rise
occurred in the solids-not-fat curve. Bailey and Azarme
agreed that pregnancy is a major factor, affecting the time
of the sharp rise of the curve.

Davis et al (1947b) stated that the protein content
of the Jersey milk decreased 11.57 percent during the
first month of lactation and increased 8.99 percent from
the second month until the end of the lactation. Bailey
(1952b) showed that the levels of solids—not-fat appear
to be similar for both barren and pregnant cows for the

interval from the middle of the second month to the middle

of the sixth month of the lactation. However, he cautioned 1
against using either end of the lactation curve for making ' '
repeated observations. When measuring individual differences,

samples should be drawn from the first 180 days production

in order to avoid the error induced by stage of gestation.

14

Nutrition

Perkins, Krauss, and Hayden (1932) reported little
effect of the level of protein feeding on the composition
of milk produced. Davis et al (1947c) observed that cows
on marginal rations had essentially normal solids-not-
fat. However, recent reviews by Bailey (1952) and Herr-
mann (1954) reported a number of investigations which showed
a marked decline in the solids-not-fat level for cows on
low levels of nutrition. English workers Rowland (1944),
Bailey (1952c), and Baker and Cranfield (1933) all sug-
gested that low rainfall and high temperatures combine
to lower the amount and quality of roughage available for
grazing in the summer months. This lowered nutrient intake
is often serious enough to cause a decline in the solids-
not-fat percentage. Cranfield, Griffiths, and Ling (1927)
attributed this summer decline in solids—not-fat to lower
lactose values and believed the lower values for solids-
not-fat near the end of the winter feeding period were
due to lowered protein percentages. Bailey (1952c) and
Rowland (1944) also suggested that low levels of solids-
not—fat in late winter were due to "low quality" feed.

They believed wiser feeding practices and the practice of
having a higher percentage of young cows in the late stages
of lactation during this period would better enable the
farmer to produce herd milk above the legal limit for

non-fat solids.

15

Under normal barn feeding conditions or on good pas-
ture, it is unlikely that nutrient intake would cause any
significant change in the levels of protein and solids-
not-fat secreted. There is considerable evidence that
underfeeding, as, for example, when feed supplies are re-
duced by drought, reduces the solids-not-fat content of
milk. .

Milking practices

Variations in the protein and solids-not-fat content
of milk due to milking practices have not been extensively
investigated. Bartlett (1934) examined the first and last
drawn milk at a milking, and found small difference in
the protein and solids-not-fat. This is markedly different
from the manner in which fat is drawn from the udder.
Bailey (1952) cited Bartlett and Kay who investigated day
to day variations and found them to be small.

Davidson (1924) and Dicks, Eaton, and Carpenter
(1951) reported that interrupted milking for several days
did not appreciably affect the protein level of the milk.

Bailey (1952) and many other investigators stated
that the morning milk contains a lower level of solids-
not-fat than evening milk. It was suggested that this is
due to the slightly longer milking interval, resulting in
a higher milk yield of slightly lower quality.

In the literature previously cited, the changes due

f

10
to various milking procedures were small. Many of these
practices are not considered good management. It is not
likely that proper milking practices would produce any
long-term effect on the levels of protein and solids-not-

fat secreted by individual cows.
Disease

The only reported work on the effect of disease on
the solids-not-fat of milk is related to mastitis infections.
It is well-known that acute mastitis drastically alters
the composition of milk and makes it unfit for human
consumption. This condition occurs less frequently in the
well-managed herd and is not the major problem with regard
to the long-term levels of protein and non—fat solids pro-
duction. Rowland (1938) and Foot and Shattock (1938)
stated that chronic low grade infections are the most
serious dairy problem. These mild infections escape the
notice of the dairyman and are detected only by micro—
scopic examination. Rowland (1938) reported the levels of
casein and lactose are abnormally low in mastitic milk.
This would account for low solids-not-fat values. Rowland
found that in seven samples out of nine, low solids-not-fat
percentages were due to "subclinical" mastitis. The other
two samples Rowland believed to be low in solids-not-fat
for physiological rather than pathological reasons. Foot
and Shattock (1938) studied 29 herds with over 900 cows

in the winter of 1947-48 in England. Those cows secreting

1?
milk of less then 8.5 percent non—fat solids were examined
closely for signs of mastitis. If the barn inSpection did
not show obvious signs of mastitis, the sample was sent to
the laboratory for examination. Over 60 percent of the
cows giving milk low in solids-not-fat had mastitis. There
were twice as many cases of "subclinical" infection as
there were of the readily apparent type. These authors
stated that 60 percent was too conservative a figure and
believed that the actual incidence of udder infections was
close to 70 percent. The authors reported this conclusion
agreed closely with the findings of Rowland and Zein-el-dine
(1938). Richardson and Folger (1950) cited Kay, who reported
that 80 percent of the English milks low in solids-not-
fat could be traced to mastitis. The authors remarked that
Kay's work conformed closely to their own unpublished

data 0
Season

Seasonal effects on protein and solids-not-fat refer
to the particular combination of climatic variables which
make up an animal's environment for any given period.

The following references show that when any combination
of climatic conditions exists that causes severe discomfort
to the cow or a scarcity of nutrients, the levels of all
major milk components will be affected. It is agreed that
fat percentage and temperature have an inverse relation-

ship. Protein and solids—not-fat vary in the same direction

18
as the fat percentage, although not as markedly. Regan
and Richardson (1938) reported that temperatures above
80-85 degrees Fahrenheit caused a decline in the casein
and solids-not-fat values. Sherman (1906) noted that
values for protein and solids-not-fat were lowest for the
month of July and highest in the months of December, Jan-
uary, and February.

Bakalor of South Africa (1947) and Baker and Cran—
field of England (1933) reported a positive correlation
between the amount of rainfall and the level of solids-
not-fat in milk.

The data suggests that the non-fat solids' values
are lowest during the summer months and highest during
December, January, and February. Rainfall, humidity,
temperature, and the resultant crop of roughage exert a

definite influence on milk quality.

EXPERIMENTAL PROCEDURE

The cows used in this study were from two purebred
Jersey herds owned by Michigan State University. Twelve
cows of a randomly-bred herd were used, primarily for com-
parison. All the milking animals of an inbred Jersey
herd were sampled. Samples were taken for a period of
fifteen months, at two-week intervals, alternating between
the two herds. These data provided information about the
changes which occurred in the composition of the milk with
advancing lactation. To avoid the effects of colostrum,
milk samples were not drawn during the first twelve days
of a cow's lactation.

It was desired to obtain 305 day lactations on all
cows. However, the individual calving dates were well

distributed throughout the year, and only twelve of the
cows had a complete ten month lactation record. The major-
ity of the cows had portions of two lactations represented.
'It was possible in twenty—four cases to obtain inforiation
on the period from calving until the end of the sixth
month. The first 180 days include the period Bailey (1953b)
suggested for use in making repeated observations.

Most of the individuals of the inbred herd were the

offspring of two sires. Ten of the cows in the randomly—

19

20
bred herd were daughters of a third sire. Thus a comparison
could be made of the protein production of the offspring
of three sires.

The randomly-bred herd was utilized to obtain inform-
ation on the variations between morning and evening milkings.
The morning and evening samples from the randomly-bred
herd were analyzed separately. The morning and evening
samples from the inbred Jersey herd were mixed to give a

one day "composite" before being analyzed.
Analytical Methods

Composition. Determinations were made of the per-

 

centages of butterfat, total solids, and protein. The
solids—not-fat content was calculated by subtracting the
fat content from the total solids content. The content of
butterfat was determined by the Babcock method; the total
solids content was determined by the Mojonnier method.

Formal Titration for Protein: Protein content was

 

 

found by the Pyne modification of the formal titration, as
outlined by Richardson et a1 (1953)."To 10 ml. of milk

at a temperature of 30-25 degrees Centigrade, add 10 ml.
distilled water, 0.4 ml. saturated aqueous potassium
‘oxalate, and 1 m1. of 1.0 percent phenolphthalein. Allow
to stand about two minutes. Neutralize to a faint but
definite end point to compare exactly with a standard
prepared by the addition of one to two drops of 0.01

percent aqueous rosaniline hydrochloride solution to

21

10 ml. of the milk plus 10 ml. water and 0.1 ml. potassium
oxalate. The amount of dye to use varies with the natural
color of the milk being used. Jersey and Guernsey milks
require slightly more than those of the less highly pig-
mented milks. The color of the standard should correspond
to a milk-phenolphthalein color at a pH of 8.3

To the neutralized milk, add 2 ml. of clear, well-
preserved 40 percent formaldehyde. Neutralize to the
same end point as before with 0.1 normal sodium hydroxide.
Subtract from this latter titration, the titration of a
blank containing 20 ml. water, 0.4 m1. oxalate, 1 ml.
phenolphthalein, and 2 ml. formaldehyde. An experimentally
determined factor of 1.83 is used to convert the remainder
in terms of m1. of 0.1 normal sodium hydroxide to percent

protein by weight."

In a number of samples from the randomly-bred herd
and in a few samples from the inbred herd, the addition
of the potassium oxalate was sufficient to cause these
samples to produce the characteristic pink color, when
phenolphthalein was added. Therefore, it must be assumed
the end point of 8.3 was reached before the initial titration
with 0.1 normal sodium hydroxide was begun. This was
remedied by adding one or two drops of 10 percent acetic
acid to the samples. This caused the pink color to dis-
appear. Richardson (1954) observed a similar phenomenon

for the milk of individual cows and stated that it was not

necessarily associated with mastitic milk samples. He
indicated in the same reference that the method of acid-
ifying back to a pH of 7 before titrating would give a fair
degree of accuracy in the formol titration.

For several of the cows, this reaction was consistent
throughout their lactations, while with others it would
appear and disappear from one test period to another.

This is in agreement with the observations of Richardson,
reported above.

Cows which gave the reaction described above did
not show any visible signs of mastitis. An attempt was
made to exclude the milk from any quarter that showed clin-
ical mastitis on sampling day or that had been treated
for same within the previous three day period.

§amplingdprocedureL A sample was taken from the

 

total production of each cow at the morning and evening
milkings. Immediately after weighing, the milk was thor-
oughly mixed and about 450 ml. were placed in a pint milk
bottle and capped. All samples were refrigerated at about
40 degrees Fahrenheit until analyzed. Morning and evening
samples from the randomly—bred herd were analyzed separ-
ately to obtain information on morning and evening varia-
tions. The butterfat and protein percentages were determined
within the next twelve hours.

All samples were brought to room temperature in a
water bath and thoroughly mixed before fifty ml. were

removed from each, for use in the formol and Babcock tests.

23
Duplicate tests were run for the first five months on all
formol titrations. The average of the duplicates was used
as the protein figure for that particular sample. Repeat-
ability between the two determinations seemed sufficiently
good to permit discontinuing the duplicate work. This
decreased by one-half the amount of titration work. The
remainder of the sample (about 400 ml.) was preserved by
adding one mercuric chloride tablet and refrigerating at
about 40 degrees Fahrenheit until the total solids percentage
was determined. The solids-not-fat percentage was calcu-
lated by subtracting the fat percentage from the total
solids value (obtained by averaging duplicate samples
from the MoJonnier.)

The samples from the inbred herd were handled in
the same way, excepting that 400 ml. from both the night
and morning milkings were mixed. Fifty m1. of the resulting
sample were drawn for the formol and Babcock tests. The
remainder was used for the total solids test.

The night and morning samples were mixed without
regard to the differences in milk production between the
morning and evening milkings. Therefore, the daily averages
for this herd are approximations of the true values for
the total solids, fat, protein, and solids-not-fat percent-
ages.

Table IV illustrates the comparisons made in this

study.

24
TABLE IV

BREAKDOWN OF FACTORS COMPARED

 

 

 

MM
m..- __ __ i... 1....

 

 

 

 

 

Source of Lb. % % % % Lb.
SUbJeCt Data* Milk Fat T.S. S.N.F.Prot. Prot.
AM vs. PM R x x x x x x
Herd dif. I,R x x x x x x
Season I,R,C x x x x x
Stage of
lactation I,R,C x x x x x
Individual
lactations C x x x
Sires'
daughters C x x x
Testing
intervals C x
Regressions
and
correlations I,R,C x x x
* R a Randomly-bred herd
I a Inbred herd
C = Combined herd data

ESULTS AND DISCUSSION

The sections concerned with the morning and evening
variations, herd differences, and the effects of season
and stage of lactation present the mean values calculated
for the milk constituents. The differences between the
various sample means were tested for significance using

the "t" test. Dixon and Massey (1951)

Morning and evening variations
The morning and evening samples from the randomly-
bred herd were analyzed separately in order to compare
their compositional quality.

Pounds of milk. Two hundred fifty-nine comparisons

 

were made over a fifteen month period with a mean of
12.52 s 0.32 pounds for the morning samples and a mean of
11.45 s 0.30 pounds for the evening samples. This dif-
ference between the sample means was significant.

Percent of fat. The morning milk samples averaged

 

5.28 t 0.07 percent, while the evening samples averaged
5.23 e 0.11 percent. This difference between sample
means for 259 comparisons was not significant.

Pergent of total solids. The 259 morning samples
averaged 14.76 t 0.09 percent and the comparable evening

values averaged 14.95 t 0.09 percent. The difference

was not statistically significant.

Percent of solids-not-fat. The same number of

 

morning values averaged 9.46 i 0.04 percent, while the
evening values averaged 9.52 i 0.04 percent. This dif-
ference was not significant.

Pergent_pf protein. Two hundred fifteen samples

 

over a twelve month period averaged 4.15 e 0.03 percent

for the morning milk and averaged 4.15 e 0.04 percent for
the evening samples. There was no significant difference
between the protein content of the evening and morning milks
for individual cows of this herd.

Pounds of protein;_ Two hundred fifteen samples

 

gave a morning mean value of 0.497 t 0.013 pounds of
protein produced per cow. The evening mean value was

0.455 a 0.011 pounds. The difference between these means
was significant. Since the protein percentage was the

same for the night and morning samples, this significant
difference in pounds of protein produced was due to the
difference in the amount of milk produced at these periods.

Table V summarizes the values reported for morning
and evening variations.

These data agree with the usual concept that morning's
milk is slightly greater in quantity but lower in compo-
sitional quality. However, the lower fat percentage
values for the evening samples are contrary to this accepted

idea. There were a number of abnormally low butterfat

27
TABLE V

MORNING AND EVENING VARIATIONS

.— -‘-_‘..-..‘ _-~—~—.n_-_F ..--——-——.———_—-_~_ .

 

 

 

-A-‘— _—-~->-———~—h—.—‘-M-“O- '—~—._.— ”»+~-_‘—~.~‘—_ _— ~-- —._...._.~.~—\—._—

 

 

 

 

No. Morning Evening- ”t"
Factor Comp. Samples Samples value
Lb. milk 259 12.52 i 0.32 11.45 i 0.30 2.43*
% fat 259 5.28 i 0.07 5.23 i 0.11 0.37
% tot. 801. 259 14.76 t 0.09 14.95 t 0.09 1.52
% s.n.f. 259 9.46 t 0.04 9.52 t 0.04 1.11
% protein 215 4.15 t 0.03 4.15 t 0.04 0.00
Lb. protein 215 0.497 t 0.013 0.455 t 0.011 2.35*
_.“;VSignificant ‘_I -_—‘~-

tests among the evening samples. The milker often was
unable to get a complete let down from several members

of the herd. A similar behavior occurred in the morning,
although not as frequently. Therefore, the lower evening
value for fat percentage could reflect the effect of these
low testing samples. This observation supports the con-
tention of Richardson and Folger (1950), who stated that
milk samples should not be drawn from physiologically
disturbed cows.

It can be seen that the protein and solids-not-fat
values vary less than do the values for fat and total
solids. Since the protein values were nearly alike for
this experiment, it must be assumed that lactose was

responsible for the difference in the solids-not—fat values.

28
Differences between herds

The fifteen month test period was divided into five
periods: April-June 1954, July-September 1954, October-
December 1954, January-March 1955, and April-June 1955.
Morning and evening values for the randomly-bred herd,
discussed above, were averaged to provide a figure that
could be used for comparison with the "composite" values
from the inbred herd. The comparisons between the two
herds were made by periods, which roughly approximated
the seasons of the year. Table VI summarizes the values
found for the milk constituents by these three-month in-
tervals for each of the herds.

ffg}9§_§$ The mean value for the percent of butter-
fat was significantly larger for the randomly-bred herd.
The mean percentages of total solids and solids-not-fat
were significantly higher in the randomly—bred herd.
The inbred herd averaged 2.67 pounds of milk per cow per
day more than the individuals of the randomly—bred herd.
The figures presented in Table VI for these milk consti—
tuents are in agreement with the usual concept that larger
quantities of milk contain slightly lower amounts of milk
solids. No protein values were available for this period.

Period_11: Highly significant values were obtained
for the pounds of milk and pounds of protein per cow per

day in favor of the inbred herd. The significance of the

pounds of protein figure is a result of the difference

TABLE VI

 

 

 

 

 

-— .—--

—~—...——- --.— ——--— w----

 

 

HERD DIFFERENCES BY SEASON

29

 

 

 

 

Fa t Randomly-bredwyfrgf Inbred Herd__ 'tm
c or No.a Mean No.a Mean value

Period I April-June 1954
Lb. milk 44 26.36 t 1.23 83 29.03 t 1.16 1.58
% fat 44 5.72 i 0.14 83 5.35 t 0.09 2.26*
5 t. sol. 44 15.57 i 0.17 79 14.98 i 0.11 2.94**
% s.n.f. 44 9.86 i 0.06 79 9.59 t 0.03 3.80**
Period II July-September 1954

Lb. milk 63 21.63 t 1.16 95 28.82 t 1.28 4.16**
% fat 63 5.09 i 0.09 95 5.25 i 0.08 1.33
% t. sol. 63 14.26 i 0.18 95 14.68 t 0.10 2.10*
% s.n.f. 63 9.37 f 0.07 95 9.43 t 0.04 .70
% prot. 63 3.89 i 0.00 95 3.83 t 0.03 .94
Lb. prot. 63 0.81 i 0.04 95 1.09 t 0.05 4.69**
Period III October-December 1954

Lb. milk 42 23.58 t 1.45 90 26.25 t 1.22 1.41
% fat 42 5.14 t 0.12 90 5.82 f 0.07 4.89**
% t. sol. 42 14.49 i 0.18 90 15.50 t 0.09 4.98**
% s.n.f. 42 9.36 i 0.09 90 9.68 i 0.06 3.08**
% prot. 42 4.34 i 0.09 90 4.47 t 0.05 1.97*
Lb. prot. 42 1.00 f 0.05 90 1.14 t 0.05 2.04*
Period IV January—March 1955

Lb. milk 55 27.16 t 1.19 90 24.38 t 0.82 1.93

% fat 55 5.53 i 0.10 90 5.93 t 0.07 3.17**
% t. 801. 55 15.05 t 0.14 90 15.38 t 0.08 2.04*
% s.n.f. 55 9.56 t 0.05 90 9.46 t 0.03 1.67
% prot. 55 4.22 i 0.05 90 4.49 i 0.04 3.91**
Lb. prot. 55 1.13 f 0.04 90 1.07 t 0.03 1.16
Period V April-June 1955

Lb. milk 55 22.03 1 1.37 117 25.84 s 0.82 2.38%
% fat 55 5.46 s 0.17 117 5.54 1 0.07 .44

% t. sol. 55 14.82 i 0.19 117 14.90 i 0.08 .38
% s.n.f. 55 9.38 f 0.08 117 9.37 i 0.03 .13

7 prot. 55 4.26 i 0.00 117 4.2 f 0.04 .14
Lb. prot. 55 0.91 t 0.05 117 1.08 i 0.03 3 06**
All Periods April 1954 - June 1955

Lb. milk 259 23.97 t 0.59 475 27.09 i 0.48 4.02**
% fat 259 5.20 i 0.00 475 5.57 i 0.04 2°96**
% t. 801. 25, 14.86 t 0.08 471 15.07 t 0.04 2.90**
% s.n.f. 259 9.49 f 0.03 471 9.51 i 0.02 .54
% prot. 215 4.15 f 0.03 392 4.26 f 0.03 2.36*
Lb. prot. 215 0.95 t 0.02 392 1.09 t 0.02 4.66**

“—-._-

 

* Significant

——.—-

** Highly significant

i
I

 

 

.afiNumber of samplegm

_--——-_ —.—.-.-

30
in pounds of milk produced between the two herds, since
the difference between the protein percentage values
favored the randomly-bred herd but did not approach sig-
nificance. The inbred herd had slightly higher butterfat,
total solids, and solids-not-fat means, but these were
significant only for the total solids figure.

The lower figures for both pounds of milk and the
percentages of most of the solids calculated for the ran-
domly-bred herd can be assumed to be due at least partially
to the number of low fat and low solids-not-fat samples
obtained from those cows which did not let down their milk
completely.

Period III. The figures for the inbred herd were

 

larger than the figures for the randomly-bred herd for all
factors considered, although the difference in pounds of
milk produced per cow per day was not significant. Highly
significant values were obtained for the percentages of fat,
total solids, and solids-not—fat. Significant differences
were obtained for the percentage of protein and pounds of
protein. It is suggested that lower temperatures contributed
to the increased values reported for all of the milk

solids, since the inbred herd was housed in a pen-type

barn, where the inside temperatures were only slightly
above the outdoor temperature. The randomly-bred herd was
housed in a conventional stanchion type barn, where the

temperature seldom fell below 45 degrees Fahrenheit.

31
Period IV. The fat, total solids, and protein

 

content again favored the inbred herd. The fat and protein
concentrations were highly significant, while the total
solids figure was significant. However, the percentage
value for solids-not-fat favored the randomly-bred herd,
although not significantly. A possible explanation is that
the randomly-bred herd samples contained a higher percentage
of lactose. This higher lactose content could have been
sufficient to overcome the differences between the protein
values and to cause the higher solids-not-fat values
observed for the randomly-bred herd. Such an explanation
is in agreement with the inverse relationship between lac-
tose and protein, shown by Overman, Sanmann, and Wright
(1929) and many other research workers. The randomly-bred
herd also had higher values for daily milk production

and for pounds of protein produced per cow per day, but
these values were not significantly different from those

of the inbred herd. The lower milk production level for the
inbreds is attributed to poorer quality hay and silage

fed during the latter part of the winter.

Period V. This period gave herd figures more nearly

 

alike than any of the other test periods. The milk pro-
duction figure was significantly higher for the inbreds
and this contributed to the highly significant value ob—
tained for the pounds of protein produced per cow per day.
All other mean values were nearly equal. The "t" values

for these constituents were very small. It should be

mentioned that both herds were under the same climatic
conditions during periods I, II, and V.

All periods. The overall comparisons on 259 samples

 

from the randomly-bred herd and 475 samples from the inbred
herd reveal the following points of interest.

The mean value for pounds of milk produced per cow
per day by the inbreds was 3.12 pounds greater than the
randomly-bred herd value. The figures of 23.97 t 0.59
pounds for the randomly-bred herd and 27.09 i 0.48 pounds
for the inbred herd refer only to cows in milk. These
mean values are not herd averages per cow per day, since
dry cows are not included. Any cow giving less than 4
pounds of milk per day was considered dry and not sampled.
The difference between the herds is statistically highly
significant. These data agree with prior milk production
records available on both herds. There is little doubt
that the inbred herd members were better producers than
the individuals of the randomly-bred herd. In addition,
they were more carefully milked and managed.

The inbred herd averaged 5.57 e 0.04 percent on all
butterfat samples. The randomly-bred herd samples averaged
5.26 t 0.06 percent. This is a highly significant dif-
ference and may be accounted for partly by the combination
of lower environmental temperatures to which the inbreds
were exposed during the winter feeding period and by the
large number of abnormally low fat samples produced by the

randomly-bred herd.

33

The mean percentage of total solids for the inbred
herd was 15.07 t 0.04, while the value for the randomly-bred
herd was 14.86 t 0.08. This difference can be accounted
for by the difference in the mean fat percentage values.
Therefore, for the reasons given above, most of the dif-
ference in the total solids values, although statistically
highly significant, is likely due to environment rather
than to heredity.

The solids-not-fat mean value was 9.51 t 0.02 for the
inbred herd. The mean value for the randomly-bred herd
was 9.49 t 0.03. The difference between the herd means
for solids-not-fat did not approach significance. It is
not possible to say from the data whether the herds are
genetically similar for levels of solids-not-fat or whether
the lactose and protein fractions balance one another to
give nearly identical solids-not-fat values.

Two hundred fifteen samples from the randomly-bred
herd gave a mean of 4.15 s 0.03 percent for protein,
while 392 inbred herd values gave a mean of 4.26 s 0.03
percent. Statistically this difference is significant.

As previously pointed out, differences in protein percent-
age were greatest during the two cold weather periods
(Periods III and IV). The protein values did not show

the wide variation found in the fat and total solids values
during any of the periods.

The inbred herd produced significantly more protein

34
per cow per day during Periods II, III, and V. The ran-
domly-bred herd produced more during Period IV. A compar-
ison of the 215 samples gave a mean value of 0.95 i 0.02
for pounds of protein for the randomly-bred herd and a
figure of 1.09 t 0.02 pounds for the inbred herd. This is
highly significant and due principally to the difference
in milk production.

In this study, the non-genetic factors were not
adequately controlled. Therefore, it cannot be stated
definitely that the differences in milk composition between
the herds were due chiefly to environment. Both herds
would have to be housed and managed under the same condi-

tions in order to eliminate the influence of environment.

Season

Much of the data concerned with season have been
presented in the discussion on the differences between the
herds. Seasonal values by three months' periods are given
in Table VI for each herd. In this section of the dis-
cussion, the data were combined to give a larger number
of samples per season. Table VII presents the means and
standard errors for all factors considered by season.

The amount of milk produced per cow per day was
highest during the April-June 1954 season, with a mean
value of 28.10 t 0.88 pounds. All other seasonal values

were within the narrow range of 24.62 s 0.72 pounds to

35

 

 

00.0 s :0.H 00.0 s 00.0 30.0 a 00.0H 00.0 s 00.: 00.0 a 00.0 00.0 a 00.00 0002
s00 000 000 000 ems :00 nmflasmm
000a csnus00fl 000
00.0 s 00.H 00.0 « 00.0 00.0 « 00.0H 30.0 a 00.: no.0 s H0.0 00.0 a 00.00 0002
msa msa 00H msfi msfl 00H nmfiasmm
000a 0:0-saa
00.0 + 00.H 00.0 a 00.0 00.0 « 00.0H 00.0 s 00.: 00.0 + 0s.0 00.0 « ss.0m 0002
00H 00H 00H 00H 00H 00H nmfiassm
000a 002-000
$0.0 « 00.H 00.0 a 00.0 00.0 « 0H.0H 30.0 a 00.: s0.0 « 00.0 00.0 a 00.00 0002
00H 00H 00H 00H 00H 00H mmfiasmm
s00H 000-000
00.0 a 00.0 00.0 a H:.0 00.0 « H0.:H 00.0 a 00.0 00.0 « 0H.0 s0.0 « 00.00 0002
00H 00H 00H 00H 00H 00a meanssm
s00H 0000-H00
00.0 a 00.0 00.0 s 0H.0H A s0.0 « 03.0 00.0 « 0H.0m cams
--- 00H 00H --- ems 00H mmfiassm
s00H 000-000

Isamposm .00 .m.z.0 m mcflaom .e m camposme 000 0 mez .00

 

 

WMDAdS Adzomfimm quoom
HH> mug?

36

35.95 t 0.97 pounds. The differences between these latter
values are not considered to be significantly different.

The fat concentration mean value was 5.48 i 0.07 per—
cent for the April-June 1954 period. The mean value for fat
content dropped to 5.19 t 0.07 for the summer months of July,
August, and September. It rose to 5.60 i 0.07 in the Oct-
ober-December period and increased again during the January-
March 1955 period to 5.78 f 0.06. Most of these two in—
creases were due to the large increase in fat concentration
encountered in the inbred herd and discussed in the section
concerned with differences between the herds. The average
pooled value was 5.46 i 0.03.

The total solids mean value was highest during the
January-March period, averaging 15.25 t 0.08 and was lowest
for the July-September period, averaging 14.51 t 0.09. This
is good agreement with the literature. The pooled mean
value was 15.00 t 0.04 percent.

The solids-not-fat values followed the same overall
pattern as the fat and total solids, but did not exhibit
as much variation. These values ranged from 9.37 i 0.03
in the April-June 1955 period to 9.58 t 0.05 in the Octo—
her-December 1954 period. The pooled mean value was
“.50 i 0.03. This agrees closely with the value of
9.49 i 0.47 found by Combs (1954). The values reported
by Overman et a1 (1939), Davis et a1 (1947 ), and Richard—
son and Folger (1950) are also in good agreement.

The lowest protein values were found in the July-

37

September 1954 period. Their mean value was 3.85 t 0.03.
This figure agrees closely with the data presented in the
literature review. The highest protein values were obtained
in the October-December period with a mean of 4.43 i 0.04.
The January-March 1955 period gave a mean percentage of
4.39 i 0.04. The value for the April-June 1955 period

was 4.23 i 0.04. The pooled value was 4.22 i 0.02.

This value is somewhat higher than the figures cited in
the literature review. The range in values was from

2.87 percent to 5.89 percent. The lower figure resulted
from a sample drawn in July from a three year old inbred
Jersey cow, which was in the second month of lactation.
The high value was from a randomly-bred five year old

cow in late lactation. This range of values agrees very
closely with the one given by Overman et a1 (1939).

'Since these values were not obtained from true composites,
they are higher than would be expected for mixed herd milk
samples.

The pounds of protein produced per cow per day de-
pends mainly upon the cow's daily milk production, rather
than on the percent of protein in the milk. In this
instance, the daily average values for milk production were
nearly identicaliiefll periods, thus, the pounds of protein
produced followed closely the changes in the protein per-
centage. The pooled values averaged approximately a pound
of protein per cow per day. No other animal can approach

the dairy cow's ability to produce this valuable nutrient.

Stage of lactation

Samples drawn between the twelfth and 365th days
of the lactation were sorted into twelve intervals. The
first interval included only the samples from the twelfth
through the thirtieth day. Ten thirty-day intervals fol-
lowed. The last interval included samples from the 33lst
to the 365th day of lactation. A small number of samples
from cows in production more than 365 days were omitted
from the calculations. Table VIII gives the mean values
by stage of lactation for each of the herds and for the
pooled data.

The second month of lactation generally gave the
lowest values for the milk solids and the highest fig res
for milk production. A gradual rise occurred in all milk
solids percentages from the second month until the tenth
month of lactation, while milk production declined slowly.
The values for the last two intervals were somewhat irre-
gular because of the small numbers. In addition, some cows
were still several months from calving but producing well
during the eleventh and twelfth months of their lactation.
These cows' samples lowered the percentages for milk solids,
but increased the pounds of milk produced per cow per day

during the last two intervals.

331
to
335

301
to
330

d

f‘\

271
to
300
11-1‘

241

to

270
20-22

211

to

240
24-26

7

,3

22-0

181
to
210

151
to
180

TABLE VIII
Days in Lactation
21

to

150

1

91
to
120

31
to

J

90

to
60

AVERAGE DAILY COMPOSITION OF MILK BY STAGE OF LACTATION
31

30

Randomly-bred herd
14-16

 

V

 

 

No.*
Lb.

Olfl

HJF»N\OCD

C\

10.2

14.0

N

F.

"O\
in

I“

\

N
7
LL

mil
fat

4.24

prot.

~/

w
fl

in O‘\
H

14.99
9-75

5.62
10.03

14.85
9.41

s.n.f.

10-11

4

,J

1“
L.

17-

9-34
17.8

3
L

34-44 35—47 36-47 35-43 37-42

40-47 42—51

36.4

herd

lured

\

L

n
Th

T
.L

riOkilfl

U1U\m

F—U\d’U\

r;

16.3

30-7

38.8
I;

milk

39.9

P4
\Offlffl
[\qu
ind-Ln
r-i

O (in O

:-

O\

9.74

F!

Chd'WHO

wmein
H

O\

OJfiWF-fi

O\K\WHO

:9-
(\IO
my:

I‘J

9.36

O\

G\
W\
O\

9.59

9-5?

9.37

9.30

3..

40-46

60-69 57—64

58-77 58-74

39-46

Pooled data

Ho.*

0 7‘]
[\- -=!'

U\
O\EM#
ri

Olflib‘
O O

x

H 0

r4 +3

E-p 0
0 h
Q4Cu

15.39
9.58 9.72

15.27 15.38
les analyzed for protein percentage.

4

9.55

I"

15.13 15.

14.97

p

14.95
9.53

9.4;

14.76
* First figure refers to the number of sam

9-33

14.5,

14.46
Second figure refers to the total number of samples analyzed.

 

 

 

40

Individual lactations

The combined data for this study represents 51 partial
or complete lactations from 35 cows. There were 24 lact-
ations that included the first 180 days in milk. Twelve
of these were complete lactations, between 275 and 305
days in length. The remaining 12 lactations were cut off at
180 days and factored by 1.43 to allow more comparisons to
be made. This is the standard Dairy Herd Improvement Associ—
ation factor used to extend butterfat records. Eleven cows
did not calve so as to allow a sampling of their first six
months in milk. Table IX gives the breakdown on all lacta—

tions of at least two months' duration.

TABLE IX

NUMBER OF LACTATION FOR HERD MEMBERS

m_,__ ”
M -‘ — - I __ _ _ -—

Lactation Number

 

 

 

 

A wag—0..

 

 

 

 

 

 

1 2 3 4 5 6 7 Total
Complete
lactations O 4 2 l l 2 2 l
180 days
(extended to 305) 1 1 4 h 1 0 1 12
All-partial
or complete 8 1; 12 9 4 3 3 51

 

 

_
_..-‘ J»‘.—-.-'

The breakdown by lactation number gives an indication
0f the range in age for the herd members. The majority
of the cows were in their second, third, or fourth lactation.
Age at calving ranged from one year and nine months to

nine years and two months. There were not sufficient

41
numbers within the sub-classes to allow an analysis of the
data for the effect of age.

Table X gives a comparison of the 180 and 305 day
values for the pounds of milk and protein produced and
for the percentage of protein in a lactation. The cows are
grouped by sires.

In most cases where 180 and 305 day actual production
figures were available for comparison, the percentage of
protein for the sampling period increased approximately
0.2 percent. The extension factor of 1.43 does not allow
for this increase in protein percentage. Therefore, no
values are given for protein content on those lactations
extended from 180 to 305 days. For this reason, the values
given for pounds of protein on the twelve extended records

are low.

Daughter comparisons by sires

Sires I and II were inbred bulls. Sire III was the
randomly-bred sire. The individual daughter values are
listed in Table X. Sire I had two daughters that completed
a lactation and two with 180 day production figures that
were extended to 305 days. These four animals averaged
8,281 pounds milk, 338 pounds of protein and 4.08 percent
protein. Sire II had four daughters that completed a 305
day lactation and three others that had 180 days in milk.
These latter three records were extended to 305 days and

the resultant seven lactations averaged 8,747 pounds milk,

MILK AND PROTEIN PRODUCTION IN INDIVIDUAL LACTATIONS

180 days in milk

 

 

TABLE X

g.—

m”

A‘.

‘

_—

 

Complete lactations

 

 

’“LBT”“““""LBT ‘%“’ Lb. ‘"“““IBT‘””“’ZT“‘
Cow milk protein prot. Days milk protein prot.
Sire
3 7,585.4 299.0 3.94 305 9,531.4 395.6 4.15
5 6,328.7 258.7 4.09 305* 9,050.0 370.0 -
13 7,277.3 275.2 3.78 £75 8,728.1 346.5 3.97
15 4,064.7 168.0 4.13 305* 5,812.5 240.2 -
Sire II
8 6,651.8 262 5 3.95 305 9,965.4 411.1 4.13
10 4,396.9 201.8 4.59 305* 6,287.6 288.6 -
14 5,581.9 230.6 4.13 305* 7,982.1 329.8 —
16 4,708.6 197.1 4.19 305* 6,730.7 281.9 -
18 6,074.8 234.8 3.87 305 8,174.1 334.4 4.09
19 8,178.8 311.1 3.80 305 10,911.o 437.7 4.01
21 7,258.2 286.2 3.94 305 11,177.o 448.9 4.02
Sire III
101 5,154.3 224.1 4.35 305* 7,370.6 320.5 -
193 4,766.0 182.6 3.83, 305* 6,815.4 261.1 -
196 6,017.1 231.5 3.85 305* 8,604.5 331.1 -
1107 4,365.8 191.8 4.39 284 5,364.0 241.2 4 :0
1108 5,361.3 205.7 3.84 286 7,629.0 308.2 4.04
1109 4,942.3 194.9 3.94 305* 7,067.5 978.7 -
1111 5,181.8 214.8 4.14 305* 7,410.0 307.1 —
Various other sires
2 6,427.8 463.3 4.10 305 8,101.7 347.8 4.39
7 6,713.8 267.9 3.99 305 8,881.0 373.3 4.20
72 4,957.2 204.7 4.13 305* 7,088.8 292.7 -
‘3 4,974.8 218.2 4.30 305* 7,114.0 319.1 -
176 5,492.7 "11.3 3.85 305 ,886.1 315.7 4.00
184 5,670.3 100.3 3.53 305 8,349.1 301.7 3.41
* Extended from 180 days to 305 days with the DHIA iactil

 

 

43

332 pounds of protein, and 4.14 percent protein. The small
numbers involved prevent drawing any conclusions as to
which sire is the more prepotent for high protein production.
However, this method could be adopted for proving sires
on the basis of the protein production of their daughters.

Sire III had seven daughters in the randomly-bred herd
with at least 180 days in milk. Only two of these had a
completed production record. It was necessary to extend
the 180 day records for the other five daughters. The
seven lactations averaged 7,180 pounds milk, 293 pounds
of protein, and 4.07 percent of protein.

These data from eighteen daughters with single records
(ten of which are extended) are not sufficient to postulate
any real differences among the three sires.

The dam-daughter comparisons are given in Table II.

Testing intervals

The twelve completed lactations and the twelve 180
day lactations which were extended to 305 days were pooled
for use in this section. It is realized that the most
accurate estimate of a cow's protein production for a lact-
ation results from daily milk weights and daily determinations
of the protein percentage. This is not a practical testing
procedure for field use. An effort was made to compare
the results from a two-week testing interval with those
from monthly and bi~monthly intervals. The bi-weekly

interval was considered to be the standard, since it

 

-——- -—_—

TABLE XI

.0- ”fi'-

DAM-DAUGHTER COMPARISONS

 

44

w. "~-m~.. -.4-

 

 

a un ound rc nt
Dam Dau. Sire milkzd Pfiilis Pprot? Pgroi.
3 I 305 9.531 396 4.15
8 II 305 9.965 411 4.13
5 I 180* 9,050 370 4.09
13 I 275 8,728 347 3.97
21 II 305 11,177 449 4.02
15 I 180* 5,813 240 4.13
22 IV 180* 7,089 293 4.13
176 v 305 7,886 316 4.00
1109 III 180* 7,068 279 3.94
184 v 305 8,349 302 3.61
1108 III 286 7,629 308 4.04

.—

* Extended to 305 days

—————-.—

involved the largest number of samples. The monthly period
was considered because of its universal use by Dairy Herd

Improvement Associations. The bi-monthly period was used

in the hope of saving labor in a large scale sampling pro-
gram, taking advantage of the small variation shown by the
protein percentage. Standard Dairy Herd Improvement Associ-

ation rules were followed regarding centering dates, days

in milk, and the use of back credit. Table XII gives the

total Pounds of protein for the three testing intervals
and the percent of deviation from the two-week values for
the monthly and bi-monthly values.

The monthly testing interval gave excellent agreement
with the bi-weekly interval method.

The range of deviations

for the monthly figures from the bi-weekly values was

POUNDS OF PROTEIN PRODUCED IN SINGLE LACTATIONS
WITH THREE TESTING INTERVALS

 

—-—--§. -——————_.p.n_ .— w-

TABLE XII

_a.__

215.

 

 

 

a , "6
Cow xa Yb 20 4 S?" ’ 35v
Y on X Z on X
2 347.84 350.24' 342.90 +0.69 —1.42
3 395.58 406.96 407.51 +2.88 +3.02
7 373.32 374.49 386.11 +0.31 +3.43
8 411.08 406.53 405.32 -1.11 -1.40
13 346.53 339.50 337.15 -2.03 -2.71
18 334.37 339.22 345.45 +1.45 +3.31
19 437.67 445.87 438.97 +1.87 +0.30
21 448.91 457.97 456.80 +2.02 +1.76
176 315.73 322.07 331.90 +2.01 +5.12
184 301.73 295.09 313.05 .2.20 +3.75
1107 241.20 242.45 229.44 +0.52 -4.88
1108 308.20 307.83 308.51 -0.12 +0.10
5* 369.96 374.49 377.99 +1.22 +2.17
10* 288.57 285.93 299.54 -0.91 +3.80
14* 329.79 32 .47 330.86 -0.10 +0.32
15* 240.17 238.70 248.16 -0.61 +3.33
16* 281.90 287.47 286.40 +1.98 +1.60
22* 292.68 298.34 313.01 +1.93 +6.95
23* 312.07 309.24 313.84 -0.91 +0.57
101* 320.46 312.61 313.86 -2.45 —2.06
193* 251.13 259.66 267.45 -O.56 +2.42
196* 331.10 322.44 344.47 -2.62 +4.04
1109* 278.71 268.71 268.38 -3.59 -3.60
1111* 307.14 293.49 292.25 -4.44 —4.85
* Cows with records extended from 180 to 305 days
a Bi-weekly testing intervals
b Monthly testing intervals
0

Pi-monthly testing intervals

from -4.44 percent to a +2.88 percent. The mean percent
deviation for the monthly method was +(DJX) percent. The
correlation between the two sets of values was .995 .
Comparison of the individual values indicates there is no
real difference between these two methods. Therefore, the
monthly testing interval would be preferred, since it would
save labor and the cows could be sampled for protein in
conjunction with the butterfat sampling procedure.

Comparison of the bi-monthly interval with the bi-
weekly interval gave surprisingly good results. The range
in deviations was from ~4.88 percent to +6.95 percent.

The mean percent deviation was +1.04 percent. The corre-
lation for these two sets of values was .986 .

The bi-monthly method is not as convenient to calculate
as the monthly method, but it saves a considerable amount
of time and labor in sampling. The fewer the number of
samples available on any single lactation, the greater
the chance of distortion of the final value by abnormal
tests, sickness of the cow, errors of calculation, and
other chance factors. As long as the Dairy Herd Improvement
program tests on a monthly basis, it will be most convenient
to sample cows for protein at the same time the butterfat
sample is taken. This is especially true in cases where
the cows are milked with a pipeline milker and buckets

are used only on the testing day.

47
Regression and correlation values

The regression and correlation values were determined
for the solids-not—fat, fat, and protein percentages in
order to compare these figures with the results from other
studies. These values were calculated on each of the two
herds and also on the pooled data. Table XIII gives the
regression and correlation values.

The regression value calculated for the estimation
of the protein percentage from the fat percentage from
the randomly-bred herd data was Y = 3.27 + 0.890 X,for the
inbred herd Y = 2.08 + 0.388 X, and for the pooled data
Y = 2.67 + 0.281 X. The corresponding simple correlations
were +0.34, +0.61, and + 0.50. Overman et a1 (1939) gave
a regression value of Y = 2.40 + 0.282 X for protein on
fat from Jersey milk samples and a correlation of +0.53.

The regression value of the solids-not-fat percentage
on the fat percentage was 2 = 8.33 + 0.216 X for the
randomly-bred herd, Z = 6.91 + 0.465 X for the inbred herd,
and Z = 7.80 + 0.307 X for the pooled data. The corre—
lations were +0.40, +0.68, and +0.50 reapectively.

The regression most often mentioned in the liter—
ature is the one calculated by Jack, Roessler, Abbott, and
Irwin (1951), which is Z = 7.11 + 0.444 X or the percent of
Salids-not-fat equals 7.11 plus 0.444 times the pe°cent
(Bf Eat. This value was calculated from mixed herd milk

without regard to breed. The widely quoted equation of

TABLE XIII

REGRESSION AND CORRELATION VALUES
AMONG FAT, PROTEIN, AND SOLIDS-NOT—FAT PERCENTAGES

 

 

 

 

 

 

 

Regression Correlation
Randomly-bred herd Y = 3.27 + 0.890 X +0.34
Z = 8.33 + 0.216 X +0.40
2 a 6.92 + 0.602 Y +0.56
Inbred herd Y 2: 2 08 + 0.388 X +0.61
Z = 6.91 + 0.465 X +0.68
Z = 3.93 + 0.589 Y +0.67
Combined herds Y = a 67 + O -81 X +0.50
Z = 7.80 + 0.307 X +0.50
Z = 6 95 + 0.594 Y +0.62
X = percent of fat
Y a percent of protein
Z = percent of solids~not-fat

Jacobson (1936), which was also on mixed milks, is
Z = 7.07 + 0.40 X. Jack et al (1951) also gave a regression
equation for estimating the percent of solids-not-fat from

the fat percent for mixed Jersey milk samples. This was

0 nr" 3 A
z = f . + 0.253 x. evermau et a1 (1939) gave a Similar

8043t10n, Z =_8.33 + 0.747 x_

The regression values for the solids-not—fat percent

on the protein percentage were as follows: far the randomly-

/’ w ” "
bred herd, 7, = 0.9‘ + O.-10.;' Y; for the inbred herd,

49
Z = 6.93 + 0.589 Y; and for the pooled data, Z = 6.95 +
0.594 Y. The correlation coefficients were +0.56, +0.67,
and +0.62 respectively. Richardson et a1 (1953) gave a
regression equation for the solids-not-fat percentage on
the protein percentage of Z a 5.85 + 0.99 Y for Jersey
herd samples, while Overman et a1 (1939) gave a value of
Z = 6.61 + 0.749 Y. Overman's correlation value was +0.67.

The values obtained in this study compare favorably
with those Cited from other sources. No attempt was made
to obtain all the regression values available in the liter-
ature. Only those reported by recent workers in the field
were cited.

One might conclude that a similar equation would be
suitable for estimating the solids—not-fat content of milk
from the protein percentage. The protein percentage could
be obtained from the formol titration test. Whether the
formol titration test is sufficiently rapid and simple for
routine field use is still questionable. Such a question
must be resolved before this test could be seriously Con—
sidered as an addition to our Dairy Herd Improvement Pro—
gram. The determination of the pounds of protein produced
per cow for each lactation, and, subsequently, their con?
version into energy units might provide a better basis
than our present system supplies for the evaluation of

the individual cow's efficiency.

SUMMARY

Two herds of Jersey cattle provided 734 samples of
milk for the analysis of the fat, protein, total solids,
and solids-not-fat percentages. The study was most con-
cerned with the protein and non—fat solids values because
of the increased emphasis on the economic and nutritional
aspects of these two constituents.

A randomly-bred herd of 12 cows yielded a total of
259 samples. The samples were taken at two-week intervals
over a fifteen month period to ascertain the variations
in the percentages of the fat, protein, total solids, and
solids-not-fat components between night and morning milk-
ings. The morning milkings averaged 12.52 t 0.32 pounds
per cow, while the evening milkings averaged 11.45 + 0.30
pounds of milk per cow. This difference was statistically
significant. The morning values for the content of fat,
total solids, and solids-not-fat were 5.28 i 0.07 percent,
14.76 i 0.09 percent, and 9.46 i 0.04 percent respectively.
The comparable evening values were 5.23 t 0.11 percent,
14.95 t 0.09 percent, and 9.52 + 0.04 percent. None of
these differences were significant. Two hundred fifteen

morning samples analyzed for protein gave values of

4.15 i 0.03 percent and 0.497 pounds of protein produced

Q

51
per cow. The same number of evening samples yielded values
of 4.15 t 0.04 percent and 0.455 pounds of protein pro-
duced per cow. In summary, the morning milk was slightly
greater in quantity, but contained slightly lower amounts
of the total solids and solids-not-fat.

The data from the randomly-bred herd were also util-
ized for the purpose of comparison with the 475 samples
obtained from an inbred Jersey herd. The average daily

values for the two herds are given below:

 

 

Randomly-bred herd Inbred herd
Lb. milk 33.97 t 0-59 27.09 t 0.48
8 fat 5-35 t 0~06 5.57 + 0.04
¢ protein 4.15 t 0.03 4.26 + 0.03
a total solids 14.86 + 0.08 15.07 + 0.04
% solids-not-fat 9.49 f 0.03 9.51 t 0.02
Lb. protein .95 i 0.02 1.0; t 0.02

A highly significant difference existed between the
two herds for milk production with the inbred herd aver—
aging about three pounds more milk per cow per day. It
is believed that the inbred herd was genetically superior
for milk production, but the erratic response of several
members of the randomly—bred herd to the milking procedure
also contributed to this difference. The larger inbred
herd values were statistically significant for the protein
content and highly significant for the fat and total solids

Content. During the October-December and January-March

52
periods, the inbred herd was exposed to the lower temper-
atures of a loose housing barn. A marked increase was
observed in the solids content of the milk from this herd
during that time. Herd mean values for the percentages
of fat, protein, and solids-not-fat were more nearly alike
during the spring and summer periods. The protein and solids—
not-fat values showed less variability than the fat and
total solids' values.

The samples from both herds were pooled to observe
the differences due to season. The average amount of mi k
produced per cow per day ranged from 24.62 i 0.72 pounds
to 28.10 t 0.88 pounds. In general, the solids content of
the milk varied inversely with the temperature of the seasons.
The lowest values fur the fat content were observed in the
July-September period, when 158 samples averaged 5.19 t 0.07
percent. The highest values appeared in the January-March
period with 145 samples averaging 5.78 i 0.05 percent.
The low values for the protein and total solids content
also occurred in the July-September period. These values
were 3.95 t 0.03 percent and 14.51 f 0.09 percent. The
lowest average value for the solids-not-fat was 9.37 e 0.03
percent, which occurred in the April-June 1955 samp es.
The highest values for the protein content were found in
the October-December 1954 period and averaged 4.43 a 0.04
percent. The highest average total solids value was

15.75 i 0.09 in the January-March 1955 p riod. The highest

‘vLJ'J

 

solids~not-fat average value was 9.3) i 0.03, obtained
during the April-June 1954 period

The stage of lactation values exhibited character-
istic variations for milk production and the percentages
of milk solids. The mean value for the daily milk production
was highest during the second month of lactation, when it
averaged 38.2 pounds, and then it declined slowly until
the tenth month of lactation, when it averaged 15.8 pounds.
The lowest milk solids values were evident during the
second month, and a gradual rise occurred from the second
to the tenth month of lactation for both the fat and non-
fat solids. The second and tenth month pooled mean values
were as follows: fat percentage, 5.22 and 5.83 percent;
protein percentage, 3.74 and 4.50 percent: total solids
percentage, 14.46 and 15.38 percent; solids-not-fat, 9.28
and 9.72 percent.

Limited information was obtained in regard to the
protein production of individual cows in a single lactation.
Only twelve complete lactations were available for com-
parison. Twelve other lactations were extended from 180
to 305 days to increase the amount of information. Host
of the lactations were from young cows in their second,
third, or fourth lactation. Nastitis was not a complicating
factor in this study.

Grouping the cows by sires allowed a comparison of
the pratein production of the offspring of three bulls.

No difference in transmitting ability for protein production

 

54
could he postulated between the two inbred Sires. The
inbred sires' daughter averages for pounds of protein and
milk production were markedly higher than the averages for
the daughters of the randomly-bred bull. There was little
difference among the bulls for the percentage of protein
produced by their daughters.

A few dam-daughter comparisons are presented. No
conclusions can be drawn from these small numbers.

There is virtually no difference in the accuracy of
the bi-weekly, monthly, and bi-monthly testing periods
for protein production per cow lactation. It is suggested
that the monthly testing interval in connection with the
Standard Dairy Herd Improvement testing program is the most
convenient method of sampling to use in a long term study
to obtain information on the protein and solids-not-fat
values.

Regression values and correlation coefficients were
determined for the relationships among the fat, protein,
and solids-not-fat percentages. The regression equations
calculated from the pooled data were:

Percent of protein = 2.67 + 0.281 x fat percentage

Percent of solids-not-fat 7.80 + 0.307 x fat

percentage

Percent of solids-not-fat

6.95 + 0.594 x protein

percentage

Although their usefulness is limited due to the small

 

55

number of samples involved, the values are in good agreement
with the literature.

Any project involving the sampling of individual cows
for any of the milk solids must be a long time study in
order to test the milk of a single cow over several lact-
ations, before many of the causes of variation can be

fully evaluated.

Azarme, E

1938.

BIBLIOGRAPHY

Variations in the Protein Content of Milk during
Lactation. Jour. Dairy Res. 9:121—152.

Bailey, G. L.

1952.

Variations in the Solids-not-fat Fraction of Milk.
Dairy Sci. Abs. Review Article 12, pp.894-902.

 

1952a.

Studies on Variations in the Solids-not-fat
Content of Milk. I Variations in Lactation Yield,
Milk-fat Percentage, and Solids-not-fat Percentage
that are due to the Age of the Cow. Jour. Dairy
Res. 19:89-101.

 

1952b.

Bakalor,
1947.

Baker, A.
1933.

Bartlett,
193”

Bonnier,
1946.

Studies on Variations in the Solids-not-fat
Content of Milk. II Variation due to Stage in
Lactation. Jour. Dairy Res. 19:102-108.

Studies on Variations in the Solids-not-fat
Content of Milk. III Seasonal Variations. Jour.
Dairy Res. 19:109-114.

S.

Investigation into the Composition of South African
Milk. Seen in abstract only. Dairy Sci. Abs.
310:A67.

G. and H. T. Cranfield. ,
Variation in the Composition of Milk in certain
Midland districts of England during the years
1923-31. Jour. Diiry Res. 4.245-254,

Variations in the Solids-not—fat Content of Milk,
I and II. Jour. Dairy Res. 5:113-123, 179-18A.

G. and A. Hansson.

Studies on Monozygous Cattle Twins. VII 0n the
Genetical Determination of the Interdependency
between the percentages of Fat, Protein, and
Lactose in Milk. Acta Agriculturae Suecana II:
171-18u.

 

Collier, P.
1891. Report of the Director of the N. Y. Agr. EXpt.
Sta. N. Y. (Geneva) State Sta. Ann. Report,
No. 9-13.

Combs, T. N.

1954. A Study of the Fat, Solids-not-fat, and Total
Solids Content of the Milk of Individual Cows.
Unpublished M. S. Thesis, Virginia Polytechnical
Institute, Blacksburg, V3.

Cranfield, H.TP., D. G. Griffiths, and E. R. Ling
1927. The Composition of Milk. I Variation in the
Solids-not-fat, Fat, and Protein Content of Cows
Milk and their Relationship. Jour. Agr. Sci.
17:61-71.

Davidson, F. A.
1924. The Effect of an Incomplete Removal of Milk from
the Udder on the Quantity and Composition of Milk

Produced at Subsequent Milkings. Jour. Dairy
Sci. 7:267—292.

Davis, R. N., F. G. Harland, A. B. Caster, and R. H. Kellner.
1947a. Variation in the Constituents of Milk under Arizona
Conditions. I Variations of Individual Cows
Within Breeds by Calendar Months. Jour. Dairy
Sci. 30:415—#24.

, , , and .
1937b. Variation in the Constituents of Milk under Arizona
Conditions. II Influence of the Month of Lactation

in Cows of Different Breeds. Jour. inry Sci.
30:425-434.

 

 

 

 

, _’ ’ and .
1937c Variation in the Constituents of Milk under Arizona
Conditions. III Variation in Milk from Jersey,

Guernsey Holstein, and Mixed Herds. Jour. Dairy
Sci. 3o:fi35-442.

 

 

 

 

Dicks, M. w., H. D. Eaton, and C. H. Carpenter.
1951. The Effect of Interruption of Milking on the
Protein Fractions of Milk. Jour. Dairy Sci.
34:52-57.

Dixon, w. J. and F. J. Massey Jr.
1951. Introduction to Statistical Analysis, ist Edition,

MCGraw — Hill Book Company Inc., New York. 370 pp.

58

Foot, A. S. and P. M. F. Shattock.
1938. Incidence of Mastitis in Cows Yielding Milk low
in Solids-not-fat. Jour Dairy Res. 9:166-173.

Hansson, A.
1955. Oral Communication.

, and G. Bonnier.

1949. Further Studies on the Genetical Determination
of the Composition of Cows' Milk with regard to
Fat, Protein, and Lactose. Acta Agriculturae
Suecana III:2:179~188.

 

Herrmann, L. F.
1954. Indirect Estimates of the Solids-not-fat Content
of Milk. Unnumbered Report Agr. Marketing Service,
U. S. Dept. of Agr.

Jack, E. L., E. B. Roessler, F. A. Abbott, and A. w. Irwin.
1951. Relationship of Solids-not-fat to Fat in Califor—
nia Milk. Calif. Ag. EXpt. Sta. Bull. 726, 12 pp.

Jacobson, M. S.
1936. Butterfat and Total Solids in New England Farmers'
Milk as Delivered to Processing Plants. Jour.
Dairy Sci. 19:171-176.

Moore, H. C. and H. A. Keener.
1942. Improving the Solids-not-fat Content of Milk by
Selective Breeding. Annual Report of the Director
cg the Agr. EXpt. Sta. Univ. New Hampshire Bull. 354
1 pp.

Overman, O. R. 0. F. Garrett, K. E. Wright, and F. P. Sanmann.
1939. Composition of Milk of Brown Swiss Cows. Ill.
Agr. Expt. Sta. Bull. 457, pp. 575-323.

, F. P. Sanmann and K. E. Wright.
1929. Studies on the Composition of Milk. Ill. Agr.
EXpt. Sta. Bull. 95, pp. 51-174,

 

Perkins, A. E., W. E. Krauss and C. C. Hayden.
1932 Chemical Composition and Nutrient Preperties of
Milk as Affected by the Level of Protein Feeding.
Ohio Agr. Expt. Sta. Bull. 515, 43 pp.
Reder, R
1338. The Chemical Composition and Properties of Normal
and Rancid Jersey Milk. II Fat, Total 3>lijs, ‘
and Protein Content. Jour. Dliry Sci. 01:-“fb gin

Regan, W. M. and G. A. Richardson.
1138. Reaction of the Diiry Cow to Changes in Environmental
Temperature. Jour. Dairy Sai. 21:73-78.

Richardson, G. A.
1952. Pricing Milk at the Producer Level on the Basis
f both Butterfat and Skim Milk Solids Content.
Address before Dairy Conference of the American
Farm Bureau Federation.

w -

1954. wPitten Communication.

and A. H. Folger.
1950 Compositional Quality of Milk. I The Relationship
of the Solids-not-fat and Fat Percentages. Jour.
Dairy Sci. 33:135—146.

 

, J. 0. Young, S. H. Dalal, S.J. Pearce and

P. M. Narula.
1953 The Use of the Protein Test (Formol Titration)
to Estimate the Solids-not-fat Content of Milk.
Proceedings 34th Annual Meeting, Western Division
A. D. S. A. pp.70—79.

Rowland, S. J.
1938 The Protein Distribution in Normal and Abnormal
Milk. Jour. Dairy Res. 9:47-57.

 

1944. The Occurrance in Winter of Milk with a Low Content
of Solids—not-fat. Jour. Dgiry Res. 13:2ol—So3.

and M. Zein-el-Dine.

193%, The Investigation of Milks low in Protein and
Solids-not-fat Content. Jour. Dairy Res. 9:182-184.

 

Sherman, H. C.
1906 Seasonal Variations in the Composition of Cows'
Milk. Proceedings of Agricultural Chemical Society
S8zl719-1723,

White, M. K. and T. J. Drakely.
1097. The Influence of the Age of the Cow on the Yiel:
and Quality of the Milk. Jour. Agr. Sci. 17:420-427.

a“ ”'6 fin?!)

iLigtj .73

If}:

“35...

0??! V