till

1
I

HHII

HMH

ll

4:.

 

THE EELATICI‘VSHEP CF CERTAIN
BODY I‘AEASL’REMENTS OF FEEDER
CALVES TO THEIR PERFORMANCE IN

' THE FEED LCT

Thesis {car We Degree of M. 5.

George. j. Pref: p
1937-

V

TH ES], ‘ /
. Q JIM so

 

I This is to certify that the
thesis entitled
NON UNIFORMITIES IN THE SIZE DISTRIBUTION
OF THE NASCENT CHAINS OF GLOBIN
FROM RABBIT RETICULOCYTES
presented by
° Alberto Protzel

has been accepted towards fulfillment
of the requirements for

 

 

: . ,' r
. r _ , . .' x
L' .nL

Ir

Major professor

 

Date_1-,/(il- —’I’4”1'//7//
/

0-7639

fi~\

w...‘ - ‘

 

‘5:

5‘

 

 

 

TIE RELATI CZC HIP OF CERTAIN BODY LEAS’533LEZITS CF mm
CAL-YES TC THEIR PERFCPJAITCE IN TEE FEEL L‘T
Thesis for the Degree of Master of Science

George J. Pro-pp

1337

THE RELATICITSHIP CF GETAI‘T BOD DY LEAS’JRJ BETTS C? F“ “ ER

CALIF 3 TC TI!“ u-IR P3“ t: “CHEESE-ICE Il-I‘ TEE FEED LOT

A THESIS

SUBLZITIED TC THE FACULTY

CF

MI CHIGAlx STATE COLLEGE

CF

.AGRICULTURE KID APPLIED SCIEEICE

By

1'

Georrre J Propp

“fits-n."

‘7
A

PARTIAL FULFILIJsZElIT OF THE REQUIFET. 32? T5

FOR THE

DEC‘a-LT"l1 OF thSTER OF SCIENCE

IN

ANIEiAL HUSBAJTDEY

East Lansing

June 1337

ACKNOWLEDGEMENT

The author wishes to express his sincere appreciation
to Professor G. A, Brown, Head of the Department of Animal
Husbandry, for his advice and guidance and to Professor G..A.
Branaman, Assistant in Animal Husbandry, for his helpful
assistance and constructive criticism which were invaluable
in the preparation of this manuscript. Appreciation is also
due Dr. W. A" Baten,.Associate in Mathematics, for his helpful

advice in the statistical computations.

II.

III.

VIII.
IX.

XII.

TABLE OF COITEHTS

INTRODUCTION

REVIEW CF LITERATURE

EXPERIMENTAL MATERIAL

METHOD OF CALCULATION

MEASUREMENTS USED

PLAN'CF EXPERIMENT

EXPERIMENTAL RESULTS

1. Use of Measurements in Estimating Feeder Grade

2. Use of Measurements and Feeder Grade in Estimating.Average
Daily Gain and Total Digestible Eutrients per 100 Pounds
of Gain

GEKERAL CCKSIDERATIOH AID DISCUSSION

SUZMARY

CCECLUSION

BIBLIOGRAPHY

TABLES

l. Feeder Calf Measurements

2. Standard Errors of Estimate‘Using Measurements to Predict
Feeder Grade

3. Standard.Error of Estimate'Using Measurements and Feeder
Grade to Predict Awerage Daily Gain and Total Digestible

Nutrients per 100 Pounds Gain

XIII.

TABLE or COI-ZTEITTS (Con't)

FIGURES

1. Measuring Instruments, Standard Equipment Obtained From
the Bureau of Animal Industry, United States Department
of Agri cul ture

2. {easurements Taken For Width of Body

3. Beasurements Taken For Depth of Body, Length of Body
and Fore Leg Length

CHART

l. Feeder Cattle Grading Chart

-1-

THE RELATICESHIP CF CERTAIU'BODY’MEASURLJEXTS OF FEEDER CALVES TO

THEIR PERFORMAKCE IN THE FEED LOT

INTRODUCTION

The idea that body shape or conformation is intimately related
to the subsequent performance of the individual is as old as the breeds
. themselves. If such a relationship significantly exists it should.be
possible by carefully studying and observing the outward appearance of
the animals to successfully select those individuals which would most
nearly fulfill the desired purpose. This relation is embodied in the
one-word expression "type" which Vaughan (3) defines as being “an ideal
or a standard of perfection, combining all the characters which contribute
to the animal's value and efficiency for the purpose Specified." It is
upon this doctrine that stock shows and stock judging are based. Just
what constitutes the ideal type has been the result of agreements reached
through an exchange of Opinions and experiences of the breeders themselves.
Though there was no systematic study made of the performances_of this
correct type, the type was quite universally accepted and selected for.
However, from time to time such factors as ”fads“ and differences of
Opinion caused the Judging standard to change and with these changes
taking place one cannot often be sure whether the change was an actual
improvement or the reverse. .As particularly vivid illustrations of such
changes one may cite the radical type changes that have taken place in
swine during the period of 1910 to 1925, and to a lesser degree, the
changes that have taken place in the modern draft horse. Beef cattle,
however, have been bred and selected since the earliest work of’Bakewell
with the same objective in mind. Their type has gradually approached

perfection without the infusion of any radical “fads" or OpiniOns.

{‘0

Because this type has met with the approval of breeders and judges over
such a long period of time, it has certainly been subject to a most
severe test, yet there is no specific evidence which would give this
type any advantage.

This study brings together data in an effort to determine
more accurately the extent of the relation, if any, that the present

body shape has to the future performance of the animal in the feed lot

and also to determine the value of the mechanical measurements.

REVIEW OF LITERATURE

The study of body measurements and their relation to the
ultimate performance of the animal is not an extensive one. he literature
is quite lacking and of a varied nature. Most of the work that has been
done in their field has not been done with beef cattle. Rather extensive
work has, however, been done with dairy cattle and to a lesser degree
of draft horses and lambs. The greater part of the available data does
not attempt to relate the future outcome of an animal with its present
shape. The studies have followed the nature of a change in measurements
of growing or fattening animals. None of this work that has been done
has attempted in any way to determine the efficiency of the animal in
regard to the variety of type that may be present, nor has an extensive
survey been made of the average daily gain and the feed required to
produce 100 pounds of gain and their relation to body shape. .A considerable
amount of work has been done with dairy cattle but none of this material
has any relation or bears any facts that might assist one in determining
the value of shape so far as fattening animals are concerned. There has
also been a small amount of work done with draft horses, but here again

this work has no bearing on the fattening of cattle, consequently, this

-3-

field is quite wanting for information.

Hultz (1), working with range-bred Hereford calves, observed
the changes in the individuals during the fattening period. He did not
mechanically measure any of the animals. He selected them to fit into
a type series ranging from low set to verv rangy. The results of this
experiment indicate that the low set calves tend to become more rangy
and that the rangy calves tend to improve in type during the fattening
period.' This study did not classify the efficiency of the different
types during the fattening period and the results obtained from this
experiment seemed to be more or less contrary to the present belief
that the low set individuals are potentially the good doers.

Severson and Garlaugh (2) used linear measurements and obtained
rather low correlations, and since body shape depends upon a.pr0portion
of measurements, higher correlations might have been obtained by calculat-
ing the multiple correlation or by using a.pr0portion of measurements.
Their work dealt with the change in measurements as the animal fattened
and does not represent a study of efficiency or performance.

Lush (5) has undoubtedly done more work in the measurement of
beef cattle than any other investigator. He worked with a large number
of range—bred Hereford steers. He made a large number of body measure-
ments and calculated the multiple correlation of these measurements with
the rate of gain, dressing per cent, and the value of the dressed carcass.
He found that in Spite of the accuracy of measurements and weights that
the size and the shape of the feeder steers only slightly indicated the
extent of this desirability at the end of the feeding period. The data
also indicated that the long4bodied tall steers with large middles, small
flank girth, and thin loins made the faster gain. Here again the results
obtained'by measurements are contrary to the belief that we have of the

present type steer, Lush states "that no score card or standard based

- h -

on conformation could ever be so accurate that the future performance

of the individual steer could be predicted from it with but few mistakes."
His conclusions were that form and function were not closely enough
correlated, which simply means that we have been placing too much weight
or emphasis on the shape of our fattening cattle. No attempt was made

in this study to correlate type of body or'body shape with the efficiency

of the gain.

EXPERIMENTAL MATERIAL

This study was started in the fall of 1932 and 1933 and was
continued through a.period of three consecutive years. The data studied
consists of various measurements and performances of thirty-five head
of Hereford heifers. The data were collected incidental to the major
objective of a fattening experiment being carried on by the Kichigan
Agricultural Experiment Station and the Bureau of.Anima1 Industry of the
United States Department of Agriculture. Three experiments were conducted,
one each in 1932-33, 1933—3M, and 193H-35. In each experiment twelve
purebred Hereford heifers from the United States Range Live Stock Experi-
ment Station, Miles City, Montana, were used. Periodic killings were
made at intervals of about M9 days, with the first kill starting at about
121 days. The purpose of these samplings was for the major objective
and not necessarily planned to fit the measurement and.performance study.
There were four such samplings which resulted in nine individuals for
each kill. The last, or fourth, kill is represented by only eight animals.
The data on heifer No. 7 of l93h-35 was found to be incomplete and she
was necessarily eliminated.

Each kill was treated as one single unit in the study of

average daily gain and of feed required for gains in weight. The entire

-5-

group was treated as a single unit in the case of feeder grade since all

calves were graded at the beginning of the experiment.

METHOD OF CALCULATION

Because these calves were selected for uniformity of tzpe, it
seemed unnecessary to compute a multiple correlation between the measure-
ments used and the performances of the fattening animals, so for convenience
the least square method of calculating was used as it is outlined by Arkin
and Colton (3). The standard error of estimate is a measure of the variation
or the scatter about the line of regression. One standard error will in—
clude 68% of the cases measured off plus and minus about the line of

regression.

MEASUB"KEHTS USED

Some thirty measurements were taken and recorded in centimeters
(Figure 1). To study width of the animals a group of measurements consist-
, ing of width of shoulder, width of crep, width of last rib, width of loin,
width through the thurls and width at the rump were taken (Figure 2).
This group of six widths were summed and an average width figure obtained
with which to work. The depth measurements were made at brisket, fore
flank, belly, rear flank and round (Figure 3). These were also summed
and averaged to give an average depth figure. The length of body Was taken
from a point just ahead of the tOp of the shoulder to the pin bone and the
fore leg length was obtained.by taking the difference between the height
of body at the withers and depth of chest. Besides these measurements
there was for each animal an average daily gain figure, a feeder grade
and the total digestible nutrients per 100 pounds gain (Table I). The

animals were graded by a committee of three men, using the charts made up

 

Figure l. Aeasuring Instruments, Standard Equipment Obtained from the
Bureau of Animal Industry, United States Department of Agriculture
(A) Measuring rule for height of animal, (B) Caliper for width measure—

ments, (C) arm used to replace short caliper arms for width measurement

of shoulder and thurls, (D) centimeter taps.

 

Figure 2. Measurements Taken for Width of Body

(A) Width at shoulders (long arm caliper), (B) width at craps, (C) width
at last rib, (D) width at loin, (3) width at thurls. Width at rump
falls on same line as (E) but was taken with short arm caliper while
thurl width was taken with long arm caliper. (F) Width at pins (not

used in mak ng up average width).

 

'Figure 3. Measurements for Depth of Body, Length of Body and Fore Leg Length
(A) Depth at brisket, (3) depth of fore flank, (C) fore leg
length, (D) depth of belly, (E) depth of rear flank, (F) depth

of round, (G) body length.

TABLE I.

FEEDER CALF H3

H7. fins-{"111
JPJJJJB» .55

Cattle slaughtered in First Kill

Year

Animal No.

Average Width of Body.

 

Shoulders
Craps
Last Rib
Loin

Thur l s
Rump

I
.Average Body Depth

 

Brisket

Fore Flank

Belly

Rear Flank

Round
Body Length

Fore Leg Length

Feeder Grade (5)

Average Daily Gain (lbs) 1.56 2

T. D. 1:.
Gain (lbs

per 100 lbs.

 

 

 

 

 

 

 

t 6 13 5 9 11 h 6 11
2h.75 26.08 25.25 27.5 25.9 25.2 26.0 25.u2 21.83
22.0 22.50 21.0 2u.5 23.5 2u.0 21.0 22 5 19.0
19.0 20.0 19.5 23.0 21.0 21.0 22.0 20.0 17.0
28.5 25.5 28.0 30.0 26.5 27.0 30.0 30.5 21.0
22.5 25.0 2h.0 2u.5 23.5 22.5 23.5 22.0 20.0
3u.5 38.0 33.0 35.0 3h.o 31.5 3h.o 31.5 31.5
22.0 25.5 26.0 28.0 27.0 25.0 25.5 26.0 22.5
133,6 50.8 37.8 39:31 1.6.0 116.6 131.9 142.2 313:0
51.0 53.5 5u.5 5u.5 53.0 52.0 51.5 37.0 h9.0
u8.5 5u.0 50.5 52.5 50.0 50.5 h9.o h6.5 8.5
M6.5 5u.5 50.0 5u.5 h8.o h9.0 51.5 us.5 M8.5
39.0 140.5 311.5 h1.5 31w 37.5 110.0 36.0 38.5
M8.o 51.5 h9.5 h6.0 h5.o uu.o h7.5 h5.o u5.5
99.0 103.0 100.0 107.0 103.0 107.0 to7.o 95.0 103.0
149.0 116.0 146.5 115.0 1+7.0 16.5 $8.0 M60 8.5
78.33 85.0 78.33 91.66 85.0 85.0 91.66 85.0 81.66

.37 1.75 2.u1 1.96 1.9h 2.us 2.16 2.05

501.3 u9o.8 M96.6

{All measurements in centimeters.

 

 

h28.h h95.3 h5h.8 h20.1 hh3.8 h25.6

- lO -

TABLE I. (Con't) FEEDER CALF unnsunsnnrrs

Cattle Slaughtered in Second Kill

Year

Animal No .

Average Width of’BodyT 26.83

8

 

Shoulders
CrOps
Last Rib
Loin
Thurls
Hump

Average Body Depth*

 

Brisket
Fore Flank
Belly
Rear Flank
Round
Body Length

Fore Leg Length

Feeder Grade (fi)

Average Daily Gain (lbs)l.75

T. D. N. per 100 lbs.
Gain (lbs.)

I"All measurements in centimeters.

 

23.5
19.5
30.5
26.0
35.0
21.0
111.1
5u.o
51.0
h8.0
35.5
50.0

”7.0

85.0

2213:12.

10 11
2u.83 g3g33”
23.5 23.0
18.0 19.0
26.5 23.0
22.0 20.0
33.0 31.5
26.0 23.0
51.0 51.5
u9.5 u9.0
M9.0 h9.o
37.0 37.0
50.0 h7.o

10u.0 103.0 101.0
u8.0 u8.o

78.33 85.0

1.7M

1.97

5h7.0 h83.0 u81.9

 

 

 

 

nn-a
1 7 12
25,h0 25.10 26.0
23.0 21.5 23.5
21.0 19.5 22.5
28.0 28.0 27.5
22.5 2h.5 2h.o
33-0 33.5 33.5
25.0 23.5 25.0
g5;5 M8.2 h7.2
u9.5 53.5 51.5
M620 50.5 118.5
50.0 52.0 50.5
36-0 39.5 39.5
hh.0 h5.5 h6.0
101.0 108.0 109.0
15.0 1.8.5 148.0

88.33 81.66 85.0

2.19

2.23

1.8%]

 

1 5
21.5 21.5
21.0 19.5
29.5 31.0
23.0 2u.0
33.0 33-0
23.5 27.5
115211.829.
u9.0 52.0
#6.5 h9.o
u8.5 50.0
38.0 h1.0
un.5 M8.o

 

.7:
01
O

01

.r.‘
-l
0 U1 0

99.0 103.0 103.

51.5

2.05

50.0

2.36

u7s.u 501.9 533.5 510.6 hh6.9

O

53.

91.66 91.66 81.66

1.62

522.8

TABLE I. (Con't)

Cattle Slaughtered in Third Kill

Year
Animal No.

Average Width of Body?

 

Shoulders
CrOps
Last Rib
Loin
Thurls
Rump

Average Body Depthl

 

Brisket
Fore Flank
Belly
Rear Flank
Round
Body Length

Fore Leg Length

Feeder Grade (%)

Average Daily Gain (lbs)l.62 1.96

T, D, K. per 100 lbs.
Gain (lbs)

‘.All measurements in centimeters.

.—

n“ a 71 ‘pfi mqum.vms
InnunR CAL: amvP.L..JalnL

 

 

1223:21
3 5 7
g5;5_ 26.h2 26. 5
2u.0 23.0 23.5
21.0 20.0 21.0
26.5 30.0 29.0
23.0 2u.0 25.0
33.5 3h.o 36.0
25.0 27.5 26.0
g§;5_ §§;B_ u6.6
52.5 50.5 h9.5
50.5 h8.0 50.0
52.5 u8.5 h8.0
no.0 37.0 36.0
M7.0 h8.0 h9.5
10h.0 103.0 102.0
145.5 116.0 1+8.0
81.66 91.66 78.33
1.96

587.2 578.7 507.h

 

 

 

 

 

 

 

 

 

 

1| 1 ~ -7h 131E215
3 1+ 15 0 3 9
25.7 25.0 2h.1 2u.5 2u.25 26.61
22.0 23.0 22.0 20.0 21.0 21.5
20.0 20.0 20.5 18.5 20.0 22.0
28.0 27.5 26.5 29.5 28.5 32.5
2n.5 2u.5 22.5 22.5 20.0 2h.o
33.0 32.0 31.0 31.0 33.0 3‘4.o
26.5 23.0 22.0 25.5 23.0 26.0
_3§;3, 37,2 nu.6 h5.1 95.5 u7.3
51.0 52.0 h9.0 M9.5 50.0 u9.5
”9.0%.5 116.5 147.5 117.5 1+8.0
50.0 u9.5 h8.0 M8.o u7.0 50.0
737.5 38.0 37.0 36.0 39.0 111.0
uh.0 h8.0 h2.5 uh.5 hu.0 h8.0
102.0 109.0 99.0 96.0 103.0 107.0
us.o 50.0 u7.o u8.0 h9.0 u6.0
95.0 81.66 81.66 81.66 75.0 88.33
1.9M 2.02 2.03 1.51 1.9M 1.97
597.7 533.9 503 9 557 2 529.9 502 6

TABLE I.

(Con't)

FEEDER CALF MEASUREMENTS

Cattle Slaughtered in Fourth Kill

 

 

Year
.Animal No. 1
Average Width of Body, 26.0
Shoulders — 20.5
CrOps 17.5
Last Rib 30.5
Loin 23-5
Thurls 3u.0
Rump 26.0
Average Body Depth" 35.9
Brisket 50.0
Fore Flank “5.0
Belly 1+9.5
Rear Flank 3h.5
Round H6.0

Body Length

Fore Leg Length

Feeder Grade (%)

Average Daily Gain (lbs) 1.57

T. D. N. per 100 lbs.

Gain (lbs)

 

 

 

1 2-
2 9 6

26.0 2h.75 2h.3
211.0 23.5 20.0
22.5 19.0 19.0
27.5 27.5 26.5
22.5 22.0 2u.0
33.5 32.5 31.5
26.0 2u.0 25.0
h6.’+ 1m___._g_ 3g;
51.0 50.0 52.5
h7.5 M7.o 50.0
M9.5 96.0 h8.0
39.0 35.0 36.0
h5.ol u3.o nu.o

103.0 105.0 103.0 100.0
u8.o h9.0 nu.o

“9.5

81.66 81.66 81.66

1.32

1.68

615.6 686.7 591.1

tAll measurements in centimeters.

 

h8.o
50.5
h1.5
h7.o
110.0

h8.0

2.39

gggg
21.5
20.0
28.5
22.5
32.0
2h.5

r

02.1
50.0

h6.5
L+7.5
37.5
hu.0
10h.o
MS.O

88.33 81.66 81.66
1.93

1.76

506.1 522.0 560.0

 

 

2 8
25,h2 26.5
21.0 21.0
18.5 20.0
36.5 28.5
21.0 23.0
31.0 32.0
2t.5 20.5
n9.o 53.0
h8.0 h8.o
M9.0 u9.0
38.0 no.5
u5.0 n7.0
96.0 101.0
M7.5 h9.o
81.66 88.33

1.9M 1.h5
5h9.3 616.0

-13..

and used.by the Bureau of Animal Industry of the United States Depart-
ment of Agriculture. An average of these three decisions determined the
animal's feeder grade. The chart is so arranged as to divide each given
grade into three parts: namely, top, middle and bottom. In order to use
such an arrangement in these computations, it was necessary to assign a
per cent value to each of the grades. The same scheme of percentage was
used here that has been used.by the Bureau of Animal Industry (chart).
The figure of total digestible nutrients was rather easily

computed because these calves were all individually fed and the number

of pounds of feed that they required during the feeding period was at hand.
The therms of net energy per 100 pounds of gain were also computed but were

not used in these calculations because of their similarity to the figure

of total digestible nutrients.

PLAH CF EXPERIKEK

The first or major objective of this study was to determine what
relation existed between the mechanical measurement and the actual grade
value that was assigned to the animals by averaging the decisions of the
three judges and the performance of the animals in the feed lot. If no
relationship exists, then we have been kidding ourselves as to what type
is most efficient. The other objective was to determine, if possible,
which of the individual measurements was of greatest importance in the eyes
of the Judges in reaching their decision and whether or not various prOportions
between these measurements were not more important than the individual measure—
ments themselves. .As has been stated before, these animals were selected
for uniformity of type. It seemed, therefore, unnecessary to compute
multiple correlations, so for simplicity the standard error of estimate

has been computed in these results.

.m-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . do: . . 4P . . JP . . 4P . . if “I . .4.”
as; ewe. s, are; emf... we}, ,1?“ a;
o _ o o o o o o o o o o
MMMJmMO mmfimwmmfimfl nmfimfl mmfimummfim91
L1 m P b .P u - __ W‘ J- -
.8:on “cognac. 53o: " dooo .1 on «one bench odds
. . u . omnooo . dosgom _ “ dogmom pdoaosamom
Essa .833. 3235 338mm,... 85%.” a has 354%
3.39.. u use: ” J“ ":39. none . .. . use mo
bmosohfima " bpob" .335 u bapgwfinm bamponodoa 3039 noofioga
a 1. M a . n t. u u u n n .83an8
u u u n n u .. .. .. .. n m Aemflvlll
" use 0 n u "35 " "nose “ " ~32. smog so
_ has . . .53. .3235. 3382.0: 133.3 __ “foe. Swansea
.28 3:8 330:8 “ u 233qu 4. HE . .1 «
baofiontnm . baob . waoaowmofi .3333 bHopouodoz .. Hash venom
$3038 523an . " 23088 See . .. «
baoaonpxa m bnob . page «Mons. bavnmflnm baouonodoz .. fig 955
use. 7.33 m u u 3% Mafia. .+ .1
Smashes it? «.39 n .2333 3383.0: saga" £3
£8338 £8338 . " 233.25 :3. .1 u 1
bHoEouuHfi knob. pnowodmofi bapgmflm baouonodon . Hash nacho
tonnoz nonpoz " " 38.82 “ o3: . _ J. Econ
38.3% as». 38.34 _ 32min 3383§ 3.5 so 53..
Seems a. s13 3832 :83“ an 3:. o 29.... .13.: . as: as
1.8» 1. so... was 3338.. 33333. and €93 flea has seam
“133." " “ 23m ” pecan .1 .1 €95“
. knob ». " mson M 33:36.? baoponodon ppofim u. Tahoe.“ «won
g“ m " d3 “ poem—=00 “ « pagan. mnoqpodmmoo
“has n has: samba“ 338802 swaaaoom mass" 22ng
v p p p P b . 1J- 4 [17 4. 1‘ J}
enhanumm onuwmnmw amummaom adamapfi «Hague manna a 1..
H396 soaaopm pong dong
douche oaonh open nopadz pan

 

 

83 0.73958 2940 g

 

‘

~15-

The preportions used in this study were arrived at after consider-
ing the contributions of each single measurement toward the ideal type.
Some measurements contribute more to the ideal type because of their
greatness, while others are more valuable because of their smallness. It
would be a mistake to multiply such two figures together because the values
of the two would cancel each other. On the other hand, to multiply two of
these measurements of the same nature would tend to bring out the point that
is being emphasized. Because of this logic, the preportion body length
times fore leg length divided by average body width times average body

depth or L.l was used. The measurements multiplied by each other here are

W.D
of the same nature, that is, L.1 improve the animal's type as they‘become

smaller. The W and D measurements improve the animal's type as they get
larger. The preportions of body length times the fore leg length divided
by average width, and also fore leg length divided by average width were
used because of the results which were obtained in the standard errors of

the single measurements.

EXPERIMEETAL RESULTS
In computing the standard error for estimating feeder grade,

it will be noticed that the preportions of l_had the smallest error of all
the measurements and preportions used in thgs study. This error of h.l7%
is quite significant in view of the fact that each grade occupies the space
of 10%. Among the single measurements fore leg length was most important
with an error of n.525, while average body depth withastandard error of
n.60§ was possibly regarded as the least valuable measurement in the eyes
of the grading committee in arriving at their decision. Average body width,
which is often considered to be the most important dimension had a standard

error of N.5 % which is practically the same as the standard error of average

f
-10...

body depth. Body length has a standard error of n.59fi and is only .01%
smaller than the largest error of single measurement.

The preportion of L.1 was used to compute the standard error
W
after average body depth was found to have the largest error of the single

measurement. In comparing this preportion with that of glj it will be
TD
noticed that the average body depth had a different effect when used in

the prOportion than when used alone. The standard error of the preportion

L.l is significant and is only .15 of a.per cent larger than the smallest
W

standard error among the single measurements. Because the width of the
animal and low setness are often regarded as the greatest assets to ideal

type, the prOportion l_was also computed. This error of H.18fl is .01%
W
smaller than the proportion LL; (Table II). .A statistical test was made
W.D

to determine the significance of the differences between the standard
errors of the various measurements in estimating feeder grade. The
computations show that the standard error of g; and l is significantly
smaller than the standard error of average bod; widtg, average body depth,
body length and fore leg length. It is not significantly smaller than the
error of the prOportions Lil and %, The prOportion 9%}.i3 significantly

' 1

smaller than the standard error of average body width, average body depth,

and body length. It does not differ significantly from the standard error

of fore leg length, the prOportion L.1 and the preportion l. The single
W.D
measurements do not differ statistically from each other.

According to these data, the measurements given most consideration
by the members of this gradin committee in arriving at the feeder grade
of the animal would.be that of a.pr0portion Ll or 1, In other words,
the body balance or the symmetry of the bodngouldgbe the most valuable

index. A.proportion of body length and fore leg length over body width,

or L.1 would also be an accurate index for the prediction of feeder grades,
W

-17-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TABLE II. STASCDAED ERRCR CF ESTILLATE III PER CEITT CF FEEDER GRADE
Average Average Body Fore 14.2.1. I_.__._1_ 1_
Body Body Length Leg W.D W W
Width Depth Length
(I) (D) (L) 1
Feeder
Grade n.5u9 n.596 n.587 n.522 n.199 n.372 n.167
TABLE III. STANDARD ERRORS CF ESTIMATE IN POUNDS CF AVERAGE DAILY GAIN
AND TOTAL DIGESTIBLE NUTRIENTS
Kills Feeder Average Average Body Fore I_.__.___ .12}. 1
Grade Body Body Length Leg $7.13 I 7
Width Depth Length
(W) (D) (L) (1)
1 .1h .28 .27 .26 .26 .51 .29 .39
2 .21 .2h .28 .2h .25 .2h .23 .15
Average
Daily 3 .17 .17 .17 .15 .17 .17 .17 .11
Gain
11» .31 .28 .32 .31 .29 .32 .30 .30
Average .21 .2’4 .25 .21!» .234 .31 .25 .214
1 214.0 38.2 31.9 29.5 32.0 21L6 36.7 32.6
2 28.5 27.7 29.7 27.7 29.6 29.6 29.7 30.9
Total
Digestible 3 28.6 29.8 28.6 30.8 2h.6 28.1 29.2 29.h
Nutrients
per 100 h 56.3 u7.2 5h.8 5h.8 nu.s M9.1 55.0 5h.9
pounds
Gain Average3u.u 35.7 36.2 35.7 32.7 32.8 37.7 36.9
1L

 

 

 

 

,.13_

while in the single measurements the fore leg len .*th would be most
valuable. Body width is a more significant single index than is body
length, and body length in turn is more significant than body depth. The
poorest measurement to use in predicting feed er ‘ de would.be oody de;)th.

In computing st mdard error for the avera mge daily gain, a some-
What different arran'eraent was necessary. Due to the fact that these
animals were killed in four different periods, it was nec fly to co...put
standard errors for each individual kill. The number of days on feed f
the first kill for the three year average was 121 days. This lapse of
time was determined by the degree of finish that the animal attained.
The other kills followed at intervals of approximately fortyanine days.
Each single measurement; namely, average body width, average body depth,
body length and fore leg length and the preportion L1, L1, and l_were used

WD W W

in computing 3 andard errors for average daily gain. With this group of
measurements feeder grade was also used. It will be noticed in Table III
that there is a tendency for the standard error to become smaller toward
the third kill and larger for the fourth kill. In one case out of eight
measurements this is not true. The standard error of feeder grade is
smallest for the first kill instead of the third kill. The fourth kill
in this case is the largest, as is found to be true of the other measure-
ments used.

In computing the standard error of the feeder grade the standard
error for the first kill animals was approximately .lh of a pound. With
these calves gaining from one and one-half to two and one—half pounds

daily, a standard error of this size in predicting the gain is highly
sign ficant. The standard error for the second kill forty-. nin edays later
was .21 of a pound, while the standard error for the third kill another

fortyanine days later is .17 of a pound. The fourth kill has a standard

- 19 -

error of .31 of a pound. These figures are all small in view of the

fact that only nine individuals were used to compute the errors for each
of the first, second and third kills and only eight animals in the fourth
kill.

The average body width measurement was not as accurate an index
in predicting the average daily gain of these calves as was the feeder
grade. In this particular case the standard errors of the first and
second kills were both higher than the standard errors of the reSpective
kills when feeder grade was used. The first kill standard error of .28
of a.pound and the second kill standard error of .2N of a pound are not
necessarily large but are significantly larger than the standard errors
of the first and second kills of the feeder grade. The third kill standard
error of .17 of a.pound is identical to that of the feeder grade, while
the fourth kill standard error of .28 of a pound is approximately .03 of
a pound smaller than the fourth kill standard error of the feeder grade.

The computed standard errors of the four kills using average
body depth follow the same trend as is noticeable in all the measurements.
The first kill has a standard error of .27 of a pound and the second kill
.28 of a pound and the third kill .17 of a.pound. The standard error of
this third kill again is the smallest of this particular measurement and
is approximately the same as that of the feeder grade and the average
body width. The standard error of the fourth kill is .32 of a pound and
is larger than either of the reapective standard errors of the feeder
grade and average body width.

Body length has a standard error of .26 of a.pound for the first
kill, .2h of a.pound for the second kill and .15 of a,pound for the third
kill. The fourth kill again has the larger standard error. The third

kill of .15 of a pound is more significant than any of the standard errors

-20..

of the above mentioned kills and is .02 of a pound smaller than any error
of the individual measurement and feeder grade.

The fore leg length runs similar in standard errors of the four
kills as the previous results show. The first kill standard error is
.26 of a pound, the second .25 of a pound, and the third .17 of a pound.
The fourth kill standard error of .29 of a pound is smaller than all standard
errors of the single measurements except average body width.

V

Of the preportions that were used as measurements, L.l had tne

largest standard error in each of the four kills. This result 33 rather
strange in view of the fact that this prOportion had one of the smallest
standard errors when computed with the feeder grade. The first kill had
an error of .51 of a pound. This error is too larg , eSpecially for those
calves that had an average daily gain of about one and one—half pounds.
The standard error for the second kill is similar to the reSpective kills
of the single measurement. The third kill also has a standard error of
approximately .17 of a.pound, and the fourth kill a standard error of .32

of a pound which is also similar to the single measurement results. In

1 all kills but the first are similar to the results of

the preportion L.
W

the preportion 1. The first kill is approximately .2 of a.pound smaller

.D
prOportion Lpl. The preportion l has a standard error
WOD w
of .11 of a pound in the third kill, which is smaller than any standard

a

for the mention

(1)

error regardless of kill. It is smaller than the standard errors of the
respective kills of all measurements by .05 of a pound. The standard errors
of the other three kills do not differ greatly from the computed standard
errors with the other measurements.

The individual standard errors of each kill were averaged for

each of the eight measurements. The feeder grade average is .21 of a pound

and is the smallest standard error of the entire group. This means that

- 21 _

feeder grade is a better index to the animal's performance so far as
average daily gain is concerned than any of the remaining single measure~
ments or proportions. The other measurements and proportions have a
standard error ranging from .2M of a pound for body length to .31 of a
pound for the prOportion EL},

W.D

There is a certain amount of doubt cast upon the value of these
measurements when the results of the computations are tested statistically.
The measurements which were smallest in error when correlated with the
feeder grade should also be the smallest in error when correlated with
the average daily gain. This, however, does not prove to be the case in
these measurements. There are possibly two explanations for the smaller
standard errors for the third kill and the largest standard error for the
fourth kill. According to Morrison (8), as cattle become fatter, there
is a tendency for their daily gains to become smaller. This fact might
account for the larger errors in the fourth kill animals. It does not
account, however, for the third kill animals having the lowest standard
error in all cases. The other possibility for this general trend lies in
the effect of eXperimental error. The selection of the animals for each
kill was made chiefly on degree of finish. There perhaps was an unconscious
tendency to eliminate the less desirable animals in the first kill and then
by a process of elimination, the poorer animals that remained were again
left until the final or fourth kill.

In order to determine which of the eight measurements expresses
the efficiency of the animal most accurately, they were computed with the
total digestible nutrients required.per 100 pounds gain. The total
digestible nutrients per 100 pounds gain were computed from the pounds
of feed consumed.by the individual, using the Morrison feeding standard.

The total digestible nutrients were used because it was thought they would

-22..

be more accurate in determining the animal's efficiency than the actual
pounds of feed consumed.

When the feeder grade measurement was computed with the total
digestible nutrients per 100 pounds gain, the error for the first kill
was approximately twentyefour pounds. This error is not large. It would
be a great asset to a feeder to be able to predict within twenty-four
pounds the feed required to produce one hundred pounds of beef. The errors
of the second and third kills are approximately equal and are about four
and a half pounds larger than the error of the first kill. As far as
estimating feed for 100 pounds gain, these two errors are still significant
in the eyes of the livestock man. The fourth kill has an error of 55.31
pounds. This error is almost twice as large as the error of the second
and third kills. With the fourth kill animals consuming approximately six
hundred pounds of total digestible nutrients per 100 pounds gain, an error
of this size is not necessarily large. This is the largest error of all
kills and all measurements.

In the average body width measurement, the first kill has an
error of 38.17 pounds. mhis error does not compare in size with the
respective error of the feeder grade measurement, while the second and
third kills of the average body width measurement are in approximation
of the second and third kills of the feeder grade. The error of the fourth
kill is approximately nine pounds smaller than that of the same kill with
the feeder grade measurement. It will be noticed that here again there
is a tendency for the second and third kills to have the smaller errors,
while the first and fourth kills are generally larger. The fourth kill
is considerably higher in error than the first kill. In two cases out
of the sight, the errors of the first kill animals are smaller than the

errors of the second and third.kill animals. This is true of the feeder

.. 23 ..
grade and the preportion of L41.
W.D

The results of computing average body depth with the total
digestible nutrients per 100 pounds gain are similar to those obtained
when average body width was used. The error of the first kill animals
is smaller than the respective error of the average body width by about
six pounds. The errors of the second and third kill animals are approxi—
mately the same in size and the error of the fourth kill animals is about
seven pounds greater for the average body depth measurement than for
average body width.

The body length errors vary a little for the different kills
when compared to the average body width and the average body depth measure—
ments, but their results indicate that there is very little difference in
the choice of any of the three as to the value in estimating or predicting
the requirements of total digestible nutrients per 100 pounds of gain.

Fore leg length has a standard error of thirty-two pounds for
the first kill, twenty-nine and a half pounds for the second kill, 2U.5h
pounds for the third kill, and was pounds for the fourth kill. Although
this first kill standard error was a little higher than the respective
standard errors of average body depth and feeder grade, it is smaller by
six pounds than the first kill error for the average body width measurement.

Because the figures here were so variable, an average of the
standard errors of the four kills was made. 0f the single measurement
fore leg length has an average standard error of 32.71; pounds. This is
the smc lest error of the four single measurements in addition to the
feeder grade, and would be the most accurate index in selecting animals
for efficiency in utilization of feed. Feeder grade is the second most
accurate of this group of measurements while body length, average body

depth and average body width are nearly equal in this reSpect. Of the

- 23 -

three different preportions used, L.l is a more accurate index than either
W.D

L.l or l, The proportion is a little more accurate than is L.1. Fore

W W W

leg length with an error of 32.7 pounds is a more accurate index than any

ans:

of the prOportions, but is only .09 of a pound smaller than the error of
L.l. The proportion L.l ranks second in value with an error of 32.8 pounds.
W.D W.D
Feeder grade with an error of 3M.“ pounds is the third most accurate index.
The greatest error and the poorest measurement as an index to efficiency
in the utilization of feed is the prOportion L.l. It has an error of

W
thirty-seven and a half pounds. The larger errors of the four kills here
again possibly have two explanations. The animals were selected for slaughter
in groups of equal finish as nearly as possible. The animals of the fourth
kill were left after a process of elimination. The other explanation lies
in the experimental work cited by'Snapp (h) and Morrison (8), et a1. Ex~eri-
mental evidence shows that as fattening animals near a high degree of finish,

more feed is necessary to produce one hundred pounds of gain than in the

thin animals.

EKERAL CONSIDERATION AED DISCUSSION

The data used in this experiment were collected incidental to
a major objective of a “degree of finish experiment.” It is possible to
suppose, had the experiment been executed to fit the requirements of the
measurement and performance study, greater or lesser differences might
have resulted. The fact that these animals were killed at intervals of
four different periods has necessitated a separate calculation for each
standard error. It has reduced the number from thirty-five individuals
to nine animals in three of the four kills and eight in the other. The
reduction in numbers alone has tended to depreciate the value of these

standard errors. Another error introduced into the experiment was that

“5
-C:. -

of combining animals of a three-year period. IRegardless of ability,
there certainly is a difference in the way a judge will see an animal
from one time to the next. With these animals being graded in three
different years, it seems possible that an error could be made in the
grading of the feeder animals. These animals were selected for their
uniformity of type. This selection made these animals too much alike so
that little or no difference existed.between them. A study of the original
data (Table I) will sUpport this consideration. In many cases, especially
the results of the proportions used in this study, the differences were
as small as one millimeter. It is also possible to think that a greater
number of measurements calculated with the performance of the animals
might have been of more value and significance. This seems unlikely in
the light of previous work that has been done. It is possible to suppose
that there are external characters which are readily recognized by a
skilled judge which are not susceptible of measurements with either tape
or measuring rule. Such things as disposition, individuality, pliability
of skin and others are examples of these characters. In this study
mathematics assumes the position ranging from zero to one hundred, or
from animals of one extreme type that do not exist to those of the other
extreme which again do not exist. The animals used in this study were
uniformly of a type somewhere between these two extremes. The feeder
grades of these calves ranged from seventy-five per cent to ninety-five
per cent. The data and results here do not express the performance of

a highly varied type but of one general type. It seems, therefore, that
in order to have such an experiment work out successfully, animals of

the same breed should be selected but of a widely varied type. It is
also wrong to assume that there is a direct improvement in efficiency of

the animal as type improves. Winter (6) states that “there has been a

-25..
general assumption that as type improved, there was a similar improve—
ment in efficiency. We know this is not necessarily the case. To a
certain extent we have been kidding ourselves on this point all these
years." We also know that as fattening cattle become fatter, they require
more pounds of feed per 100 pounds gain. Therefore, each individual kill
in this experiment actually represents a different degree of finish and
with that a different feed requirement per 100 pounds of gain. The data
Show a tendency for the typier animals to be a little the more efficient
in their feed requirements per 100 pounds gain as compared to the somewhat
less desirable type individual. On the whole, the animals were too nearly
alike to specifically show a definite contrast in performance of the

different types.

‘KEARY

The investigation reported in this paper is a study of the
relation of certain body measurements to the performance of fattening
cattle in the feed lot. The data were collected from thirtyafive head
of Hereford heifers fed.by the Michigan Agricultural Experiment Station
during the years 1932-33, 1933-71; and 19314.35. The relation of certain
body measurements and feeder grade were used to compute standard errors
using feeder grades, average daily gains and total digestible nutrients
in the ration per 100 pounds of gain in body weight as the variables.
Because there is a change in feed requirements and average daily gains
as animals become fatter, each kill was necessarily computed separately
and later averaged. The data show:

The preportion is the most accurate index in estimating the

31““

feeder grades of the animal. The Judges, therefore, place more emphasis

,0

on the preportion of these measurements than on any one single measurement

in grading the animal.

-27..

The pr0portion.;;l_was second most accurate in estimating
the feeder grade. The grgdgng committee gave more consideration to this
pr0portion than any single measurement in arriving at a feeder grade.

Of the single measurement, fore leg length was given more
consideration than body length, average'body depth or average body width.
The low setness of the animal has been considered one of its greatest
assets toward ideal type.

Average body width is a more important measurement than either
average body depth or body length.

Ayerage body depth received the least consideration by the
grading committee in arriving at a feeder grade.

The standard errors for all measurements were small enough to
keep the animals within the assigned grade, although none of these errors
kept the animal within its own third of that grade.

,Average body width allowed the animal to shift 1.225 into the
third of a grade above or below the assigned third of the feeder grade.

,Average body depth allowed the animal to shift 1.26% above or

below the third of the assigned third of the feeder grade.

The prOportion l is the best measurement in estimating feeder

W
grade. It has a slight advantage over the prOportion L.l, but this
W.D
difference is not statistically significant. If the preportion L.1 is

W.D
equally just as valuable as an index in estimating feeder grade as is

the preportion 1, then it should be in close approximation to both feeder
grade and the pzoportion l in estimating average daily gains. This,
however, is not the case.w The results cast suspicion on the value of

the mechanical measurement as compared to the judgment of a skilled

grading committee.

The pr0portion l ranks second to feeder grade as an index to
W

-28..

average daily gains. This preportion function d as would be expected
in estimating the average daily gains.

Of the single measurement, body length is the most accurate
index of average daily gains, while fore leg length is th e second most
accurate index in estimating average daily gains. Here again the difference
between the two errors is not statistically si'nificant. Consequently,
these two measurements have functioned approximately as would be expected.

Average body depth is the poorest single meg asurement to use
as an index in estimating avera3e daily gains and the preportion L. l is

W. D
the least accurate of the t1 ree preportions in estimatir g average daily
gains. It is also tr e least accurate of all eight measurements used in
this respect.

The single measurement, fore leg length is the best index of
the group of measurements in estimatire the total digestible nutrients
per 100 pounds of gain.

The pr0portion 9;; has an error of only .09 of a.pound larger

W. D
than that of the fore leg length measurement and r; nks second in value

for estimating total digesfi ble nutrients per 100 pom ride of gain.

Feeder grade ranks third in accuracy in estimating the total
digestible nutrients per 100 pounds of gain.

The largest error and the least accurate measurement is that
of the preportion ELL. The other me easurements all rank about the same
in their value as an index to requirements of total digestible nutrients
per 100 pounds gain.

The best measurement for estimating average daily gains and
total digestible nutrients per 100 pounds of gain cannot be determined

from the results that have been ob tained. Each kill has its own standard

error and most of them are too nearly alike to show any significant

£3

differences in their sizes. .Although an average of the standard errors
of the four kills was made for each measurement to assist i. this discussion,
it is not mathematically correct. Hence, the significance of the differences

between them was not computed.

CONCLUSION

The standard errors obtained in these results indicate that
there is a slight correlatien'between the various body measurements and
the perform.nce of the animal in the feed lot. The standard errors show
very little variation between themselves, but their differences in many
cases are significant in Spite of their likeness. The animals used in
this study were uniformly alike in type. The fact that this particular
type shows a slight relation to performance does not mean that an improve—
ment in type would also mean a corresponding improvement in efficiency
or vice versa. The performance of the fattening animal does not follow
the trend of a straight line. There is a general decrease in average
daily gain during the fattening period and a general increase in the re—
quirements of total digestible nutrients per 100 pounds of gain as the
feed period progresses. The errors obtained are too large to predict the
animal's position within the one-third of the feeder grade but are small
enough to accurately predict the general feeder grade. The three prepor-
tions used in this study when statistically tested are significantly

better than the sinrle measurement in predicting feeder grade. Feeder

.L'C’a-

1

grade is a better index than the mechanical measurements in estimating
average daily gains by periods while the mechanical measurements are
practically the same in their value in estimating average daily gains.
All measur‘m nts are approximately equal in size of error in estimating

the necessary total digestible nutrients per 100 pounds gain. Fore leg

-30..

length and the preportion L.1 are a little the better index to total

W.D
digestible nutrients required per 100 pounds gain than are the other

measurements used. In the case of average daily gains and total di estible

nutrients, the animals were killed in periods. This division in two periods
has destroyed the effect of large numbers and has tended to depreciate

the value of the standard error. Whether these errors statistica 1y differ
liom each other is very doubtful, but they do show that there is an increase
in error for the fourth kill animals and that sometime during he fattening
period, there is a time when the performance of the animal is more closely
related to its body shape than at any other ine. The results of this

study are too meager to warrant any Specific decision. However, conformation

is often the only basis available for judgment and should be given this

consideration at all times.

h.

Hulze, Fred 8.

Severson, B. 0.

-31..

BIBLIOGRAPHY

Type in Beef Calves
Wyoming oTicultural Experiment Station Bulletin 153, 1927

and Gerlaugh, P.

.A Statistical Study of Body Weights, Gains and Measure—
ments of Steers During the Fattening.Period.
Pennsylvania Agricultural Experiment Station.Annual
Report 1915-17, pp. 275-295.

Arkin, Herbert and Colton, Raymond.R.

napp, Roscoe R.

Lush, Jay L.

W131? 81's , L. M.

An Outline of Statistical Methods, 193M

Beef Cattle
Second Edition 1930

The Relation of Body Shape of Feeder Stters to Rate of

Gain, to Dressing Per Cent and to Value of Dressed

Carcass.

Texas.Agricultural Experiment Station Bulletin No. ”71, 1932

Breeding'Market Beef Cattle
Shorthorn World, April 10, 1937

Vaughan, Henry W.

Morrison, F. B.

Breeds of Livestock in.America, 1931

Feeds and Feeding ,
Twentieth Edition, 1930

'A
P

G

1,.
.a ‘

 

 

LIBRARY 800K 5:?

r.

Rworns
ANSI,

MICHIGAN STATE UNIVERSITY LIB RAR REI

LIL L ILI LLIIL II III LILIII II

 

'i‘s '