THE RELATiONSHlP OF CERTAIN BODY MEASUREMENTS TC \XI‘EEGHT N BEEF CATTLE Thesis for the. Degree of M. S. MICHIGAN STATE COLLEGE \X'Wlliam M. Barton 1938 THE RELATIONSHIP OF CERTADI BODY W333 T0 WEIGHT IN BEEF cm Theais for’Degrea of M. 30 Michigan State College Will ism Mo Barton 1938 THES'S THE RELATIONSHIP OF 0mm BODY kmamams TO WEIGHT IN Em CATTLE A THESIS SUBI‘EI‘ITED TO THE FACULTY OF MCHIQ'LN STATE COILEGE OF LCEICULTURE AND APPLIED SCIENCE BY WILLIME M. BARTON IN PARTIAL FUIFIIIIflVI' OF TIE Rzgmms FOR THE DEGREE OF MASTER OF SCIENCE IN mum. HUSBANDRY EAST LANSING JINE 1938 ‘1- 15900 .Acknowledgements 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 Instructor George J.‘Propp, 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. D. Baton, Associate in mathematics, tor his helpful advice in the statistical computations. I. II. III. IV. V. VII. VIII. IX. X. XII. Table of Contents Introduction Review of Literature Experimental Material Eethod of Calculation Measurements Used Plan of merment Experimental Results General Consideration and Discussion Sumnery Conclusion Bibliography Tables I. Feeder Cattle Measurements II. Fat Cattle Measuranents Grand Totals for Both Feeders and Fat Cattle III. Weights of Beef Cattle as Predicted tmm Heart Girth Measurements IV. Approximate Heights of Beef Cattle of Good Grade for a Given Heart- Girth Measurement, United States Department of Agriculture V. Estimating Weights of Dairy Cows from Heart-Girth Measurements, United States Department of Agriculture VI. Average Body Weights of Dairy Cattle for Given Chest Girths, C. Brody, Missouri Page 10 11 1.3 16 18 26 SO XIII. Figures 1. 2. 3. 4. XIV. Charts Measuring Instruments, Standard Equipment Obtained from the Bureau of Animal Industry, united States Department of Agriculture Measurements Taken for Width of Body Measurements of Length, Depth, Circumference, and Height Correlations between Measurements and Weights in Beef Cattle Page 14 l? 21 ms RELATIOINBHIP OF CERTAIN BODY W3 TO WEIGTI' IN BEEF ems Introduction It has long been thought that there is a definite relationship be- tween certain body measurements and the weight of any animal. Many of the old-time cattle buyers who traveled from place to place buying cattle, based the price they were willing to pay for an animal upon the amount it would weigh when driven ever the scales. The basis for this Judgnent on weight was usually the depth of body, width of body, height or distance from ground and various other body measurements as determined by the eye. Of course, very few, if any of them, would ever say that an animal was so many inches deep in body, or so many inches wide or around, or would weigh a given number of pounds. It is a known fact that wide, thick, deep-ribbed calves weigh more than calves of the same age and breed that are shallow-bodied, narrow- backed individuals. Without a doubt, any man acquainted in any way with livestock would be able to tell which of two such animals was the heavier. Comparatively few men have an opportunity to estimate weights often enough to make them proficient in the work. Could weights be estimated on a reasonably accurate basis from any body measurement, it would be of insati- mable help to many livestock producers. The question of how many centi- meters or inches in depth, width or height are required to indicate a definite weight is still unanswered. The further question of increase in these measurements as an animal grows needs a great deal of study. Another question arises with animals on full feed that are being fitted'for shows or even finished for market. A8 they become fatter, they also become heavier, but are they increasing in circumference of chest, -2- cr flank, height or width, in the same proportion to body weigit as they would increase were they being grown along, requiring a longer period of time to reach the same weight? Any method of predicting weight rather accurately by certain body measurements would benefit tremendously the commercial cattle raiser, especially the similar breeders through the Middle West and West who are better known as livestock farmers, and never have more than a few cattle to sell at any one time. The buyer has a much wider experience than the seller in most cases, and therefore, is a better judge of weigits and in many cases buys cattle from the farmer at an estimated weigit much below the real one. If animal husbandry workers could devise a tape for measuring and a table of weights based on measurements, it would be of incalcuable value to those who have livestock to sell. This study, to a limited degree, tends to deal with some of the above questions and problems, and discusses the correlation and relation.- ship betwaen certain body measurements and weight in beef cattle. It also brings together data in an effort to determine what and how much relationship there is battleen measurements and weight. .3- Review of Literature The study of body measurements and their relation to weight is not an extensive one. The literature is very meagre and of a varied nature. Most of the work has been done with dairy cattle although the United States Department of Agriculture, Bureau of Animal Industry, Division of Animal Husbandry, this past year (1) prepared a table predicting weights of beef cattle, from heart girth measurements. There is a wide divergence beheen this table and the one arrived at as a result of this study. The United States Department of Agriculture (2) , Bureau of Dairy Industry, has also prepared a table predicting the weights of dairy cattle from heart girth measurements. Their table is included in this paper and also one prepared by Brody of Missouri (3). In comparing these two tables on predicting weights of dairy cattle, even though both are based on large numbers, there are in places wide differences in predicted weights using the same heart girth measurements. There is also a rather wide difference betmen the table prepared as a result of this study and that prepared by the Bureau of Animal Industry, Division of Animal Husbandry. I. L. Lush has probably done more work with measurements with beef cattle than anyone else. In his study with Texas steers published in 1928 (4) in which masurements were taken on all calves as feeders and again as finished steers, he found that the greatest increase was in body widths, followed in order by lesser increases in body circumference, width of the pelvic region, body depth, body length, height at topline from ground, and head measurements. He states further that unavoidable errors in taking body measurements limit their usefulness in describing animal form, size, and fatness. Age plays a very important role in determining how a steer's shape changes during fattening. Relative to weight, nearly all changes were more extreme with older steers than with young ones. This was to be -4- expected as a logical result of the fact that the gins made by‘ the older steers consist more largely of fat than do the gains made by younger steers. The younger steers mde a definite increase in the absolute size of their skeletal and muscular tissues during the fattening period. £8 a result of their studies at Missouri, Wowbridge and Haigh (5) concluded fmm experiments using different aged cattle and different rations, that: (a) Height at hips is affected very little, if stall, by fatness. (b) Height at withers is affected practically not at all by fatness. (0) Length from shoulder to hips is slightly affected by fatness. (d) Length from shoulder to ischium increases distinctly with fatness. (9) Width of hips increases distinctly with fatness. (f) Heart girth increases 53?.“11 with increase in fatness. At the University of Wyoming (6) seven measurements were taken at the beginning and the end of the fattening period of four lots of calves, and three lots of ”0-year-old steers. These were analyzed chiefly with reference to their relation to type (1.6. to ranginess vs. low-setness) as determined visually by experienced Judges. Heart girth, width at hips, depth of chest, and paunch girth all showed large increases during the fattening of two-year-old steers. Height at withers showed a small increase and the distance from the floor of the chest to the ground showed a distinct decrease. All measurements of the calves increased -5— during fattening, but the greatest increases were in heart girth, width at hips and paunch girth. The increase in depth of chest was slightly less and the increases in cannon circumference, height at withers, and body length were much less than the first three measurements mentioned. The increase was least of all in the diatoms from chest floor to ground. In the steer feeding experiments reported by Evvard, Culbertson, Wallace and Hmond of the Iowa Agricultural Experiment Station ('7), heart girth, paunch girth, shoulder height and rump height were measured on steers in feeder condition and again after fattening. The average increase and percentage increase were calculated. For the most part these steers were two-year-olds with about four months batman the five sets of measurements. A few were calves and yearlings. All four measurements increased during fattening, but heart girth and paunch girth increased more than height. On a percentage basis, the increase in heart girth was slightly greater than the increase in paunch girth in all lots. In most lots shoulder height increased more than rump height, but there were some exceptions to this. These Iowa results give a clear picture of two height measurements, both increasing slightly and at very nearly the same rates during fattening, while the he girth measurements increase more than the height measurements and the heart girth increases slightly but significantly more than paunch girth. Luah sums his 1928 (4) report up in the following way. "In general, steers increase much more in 1133 during fattening than they do in length or depth of body and least of all in height and head measurements. In general, the soft parts of the body increase most rapidly, and the only bony measurements which even approximately keep up with the increase in live weights are the pelvic measurements. There is some breed difference beheen o6— Herefords and Brahmans in flank girth and paunch girth. The most generally useful single measurement among those which increase rapidly with increasing fatness seems to be chest girth, although chest width and loin width could be more useful if the errors in taking those measurements could be made as small as the errors in most of the other measurements. The most generally useful ratio studied seems to be the ratio of chest girth to wither height.‘ Lush states further: "Throughout the whole question of using body measurements to supplement studies of weight changes in cattle, there runs as an undertone the question of how accurate body measurements can be as an objective description of animal form. In general, great accuracy is attained only in the measurement of rigid, bony structures, such as head length, cannon circumferences, or width at eyes. However, these parts of the animal body are usually of very slight direct comnercial importance. The details of conformation which have a direct commercial importance are for the most part concerned with soft structures which have curved surfaces and are Joined together by movable Joints. They are therefore difficult or impossible to describe in a mathematical sense with anything like comlete- uses. It seems to us from our experience in this study that body measurements should be regarded as of minor importance compared with weight changes, and that in most cases a system of artificial grades, standardized either by the use of models or of pictures might more satisfactorily describe important details of conformation than simple linear measurements alone can. Linear measurements should be regarded as supplementary to other means of description rather than as a substitute for those other means. Ii'he great advantage of linear measurancnts is their high degree of objectivity, which advantage they share with body weight." In his work with dairy cattle (8) Lush states that the correlation between the size of a measurement on different animals and the usual error -7- in taking that measurement is slight. The error for the most part is dis- tinctly larger for larger animals. The accuracy of most measurements compares rather favorably with the accuracy of weights where the standard error of weighing was found in most cases to be between six and twelve pounds or not far from one per cent of the mean weight. Measurements are less affected by day to day changes in weather and other externalrcondi- tions than are weights. The principal objection to the extensive use of body measurements, at loast with dairy cows, seems to be not their inaccuracy but their inadequacy to describe the animals in a complete way. Twenty-five different measurements were each taken eleven times on each of nine cows and ten yearling heifers from the Jersey herd of the Texas Agricultural Experiment Station. In only a few measurements was the standard error of measuring much larger than two per cent of the measurement and in about one third of the measurements it was less than one per cent. The errors of measuring are of about the same magnitude as the errors of weighing when expressed as percentages of the mean. In another study on the accuracy of cattle weigts (9) Lush finds the experimental error in the accuracy of single weights of cattle may be expected, ordinarily under uniform conditions, to be between six and twelve pounds. The experimental error is somewhat smaller with younger cattle and is distinctly smaller under unchanging environmental conditions than when some sudden change is made in some important condition between tWo of the weighing“ The size of the experimental error may perhaps be rather closely related to the fat-free weight of the animal or to the size and capacity of the animal's digestive tract, due to "fills" or "shrinks". The size of the experimental error to be eXpected increases somewhat with the age of the animals. -8- Severson and Gerlaugh (10) at Pennsylvania State conducted a study in relation to changes in body weights and measurements of steers during the fattening period. They measured a total of 214 steers, taking circumference of chest, with an average measurement of seventy-three inches. Average weight for the 214 head was 900 pounds. In examining the table in this study on predicting weights from heart girth measurements, it is shown that the predicted wei ghts for steers with a heart circmnference of seventy-three inches is 913 pounds. This weight is much closer than most livestock producers are able to guess. Steers having a large increase in the circumference of the chest have been those that had made the greatest daily gains. (It is true that the more a steer gains, the greater is the amount of covering of flesh over the ribs and crops or around the heart circumference.) Increases in body measure- ments have a closer relationship with gains in live weight than the initial measurements. This would logically be expected, since gins are dependent upon the capacity of a steer to lay on fat and make growth in bone and body tissue. The correlation coefficients for the true body circumferences of chest and rear flank, the width of thurls (hip joint) and the distance of hip to buttock Show the closest relationship of all the initial measurements with gains in live weight. This suggests the possibilities of using these measurements in the selection of feeding steers, at least for experimental purposes, as a means of reducing the experimental error caused by individuality of animals. Circumference of paunch shows intermediate relationship with ability to gain. The 333: cumference of chest and rear flank are more important in ascertaining gains than feed capacity as indicated by the circumference of paunch. All experimental workers realize that there are errors made in weighing livestock. All scales do not balmce or break with the same -9- tension. There are also errors made in taking any kind of’measurement. These errors may be due to one or more of several different reasons, such as, the position the animal is in, the way the operator holds the tape, and the tension.applied to the tape when.reading the measurement. weather conditions tend to cause errors. Dirt and other foreign material on the animal are responsible to some degree for errors. Regardless of who is doing the work and the kind of'measurament being taken, under our present method of measuring and weighing, there will continue to be a certain.amount of error in all weighing and measuring of animals. From.the studies conducted by other experiment stations, it has been concluded that, with the errors included, the measurement which has the most relationship to body'weigmt in.cattle is that of heart girth.or chest circumference. -10- Eiirimental Material This study'was started in the fall of 1932 and was continued through a period of three consecutive years. The data studied consists of various measurements in regard to weight of thirty-five head of Hereford heifers and thirtybsix head of’Hereford steers. The data was collected incidental to the major objective of a fattening experiment being carried on by the Michign Agricultural Experiment Station and the Bureau of Animal Industry of the united States Department of Agriculture. Three experiments were conducted, one each in 1932.33, 1933-34, and 1954-35. In each experiment halve purebred Hereford heifers and twelve purebred Hereford steers from the United States Range Live Stock Experiment Station, Miles City, Montana, were used. Measurements were taken at the beginning and at the end of the feeding period. The calves weighed on an average of 3'75 pounds as feeders and an average of 762 pounds as fat cattle. All of the calves graded "choice' as feeders and ranged from ”choice" to 'xn‘ime" as finished cattle. Seventy. tvm head of Hereford cattle were used at the start of the experiments, but because the data on Heifer No. 7 of the 19:54.35 group was found to be incomplete, she was necessarily eliminated. Since measurements were taken at both the beginning and at the end of the experiment, this study really deals with measurements on 142 head of cattle, classified as feeders and as fat cattle. -1],- Method of Calculation Because these calves were selected for uniformity of type it seemed unnecessary to compute a multiple correlation batman the measure- ments used and the weight of the fattening animals, so for convenience the least square method of calculating was used as it is outlined by Arkin and Colton (11). The standard error of prediction is a measure of the variation or the scatter about the line of regression. One standard error will include sixtyoeight per cent of the cases measured off, plus and minus about the line of regression. -12- ' N14 ”is liars 1. Insuring mtrmsnts, Standard Equipent Obtained from the Bureau of Animal Industry, United states Department of Agriculture: (1) Measuring rule for height of aninl, (B) caliper for width measurements, (0) am used to replace short caliper arms for width measurement of shoulder and thurls, (D) centimeter tape. ~13- Measurements Us ed Some thirty measurements were taken and recorded in centimeters (Figure 1). Because of the insignificance of sonm of the minor measure- ments, in figuring relationship between measurements and weight, all but twelve were discarded. These twelve are: length of body, height at shoulder, height at hips, depth of brisket, depth rear flank, loin width, crop width, chest width, width chest floor, heart circumference, flank cir- cumference and belly circumference. The length of body was taken from a point just ahead of the top of the shoulder to the pin bone. Height at shoulder was taken as the animal was standing on a lave]. floor in as natural a position as possible. Height at hips was taken in the same way. Depth of brisket was taken, using the parallel bars on the meter stick, standing at an angle in front of the brisket, the top bar level over the shoulders or withers, and the bottom bar between the fore legs, level with the floor or point of the brisket. Depth of rear flank, was taken using the parallel bars, the top bar being level over the hips and the lower bar touching and passing under the rear flank. The loin width was taken, using the small bars midway between the last rib and the hip Joint. The crop width was taken, using the small bars, Just back of the shoulder. The chest width was taken using the longer parallel bars, just back of the shoulders, passing down over the body. Width of chest floor, was taken using the shorter bars, just back of the fore leg, or a measurement of the chest floor width. The circmnfer- ence measurements were taken with a steel tape, graduated in centimeters. Heart circumference was taken Just back of shoulder and fore legs, pass- ing around the body; belly circumference, around the middle of the paunch; and flank circumference, over and imediately in fmnt oi‘ the hip Joint Figure 8. Measurements Taken for Width of Body. (A) Loin width, (B) chest width, (0) amp width. -15- and through the rear flank. .All measurements on any one animal were taken by the same person, to avoid as much error as possible in drawing the tape too tight or allowing it to remain too loose. -16.. Plan of Experiment The objects of study were: (1) To determine what relationship existed between the mechanical measurement and the actual weight of an animal. (2) To set up a table, if possible, for predicting weights of beef cattle from different body measurements, or at least from measurements having the highest correlation or relationship to weight. (3) To find the difference in weight battleen beef cattle and dairy cattle, having the same body measuranent, namely heart girth, and to also show the differemes that might be calculated from different eXperimental data. Throughout the entire eXperiment single measurements were used in calculating the relationship to weight. It is probable that two or V more measurements, having a high correlation, when thrown together, could be used more accurately in predicting the weight of a beef animl, but at the same time, in order to make this study as practicable as possible, and understandable to all beef cattle producers, it was decided to use only the single measurement in figuring relationship and, if possible, in predicting weight. figure 3. measurements of Length, Depth, Circumference, and Height. (A) Depth of brisket, (13) height at shoulder, (a) width chest floor, (D) heart circumference, (3) belly circul- ference, (r) flank circumference, (G) depth of rear flank. (3) height at hips, (I) length of body. -18- marinental Results The first measurement used in figuring relationship of measure- ment to weight was that of length of body. The correlation coefficient betwaen length of body and weight in feeder calves was found to be .749 with a standard error of prediction of 21.75 pounds. This is a fairly high correlation and the standard error of prediction is small as com- pared with that of cone of the other measurements. This means that should a predicting table be set up, this table should be within 21.75 pounds plus or minus of the actual weigit of the animal in sixty-eight per cent of the cases. The correlation coefficient between langth of body and weight in fat cattle was .253 with a standard error of predic- tion of 106.58 pounds. It would, therefore, be useless to try to predict weights of fat cattle by length of “body measurements. In using the measurement ”height at shoulder" a correlation of .473 was obtained with a standard error of prediction of 28.93 pounds in feeder calves and a correlation of .808 with a standard error of prediction of 64.92 pounds in fat cattle. Here again the standard error of prediction is too great to be of much practical value. The correla- tion coefficient betIeen the "height at hips" measurement and body weight was found to be .521 with a standard error of prediction of 28.05 pounds and for fat cattle to be .784 with a standard error of 68.54 pounds. Here again, as in the above measurements, the standard error of prediction is lower in feeder cattle than in fat cattle, but is still too great in both cases to be of much practical value. In the depth measurements, a correlation coefficient of .742 was obtained between the measurement "depth of brisket" and body weight with feeder cattle. The standard error of prediction here was 22.03 pomds. This standard error of prediction is low enough to be of practical value -19- in predicting weights of feeder cattle from "depth of brisket" measurement but in fat cattle, with a correlation coefficient of .858 between the smne measurement and body weigit, we have a standard error of prediction of 56.54 pounds. This standard error of prediction in both cases is lower than that found in any other measurement with the exception of flank circumference. However, because of the way the measurement is taken and because of the way an animal my be standing when measured, this measurement is subject to a great deal of experimental error. Fifty-six and one half pounds is still too great an error to be of practical value to most beef cattle producers. The measuranent "depth of rear flank" had a correlation coefficient with body weight of .348 in feeder cattle with a standard error of pre- diction of 30.81 pounds. This correlation is too low to be of any practi- cal value. In fat cattle, using the same measurement, we have a correlation coefficient of .721 with a standard error of prediction of 76.05 pounds. Again we have a correlation that is fairly high, but the standard error of prediction is too great. The width measurements were all found to be of little or no practical value in predicting weights. The correlation coefficient between the measurement "loin width" and weight was found to be .638 with a standard error of prediction of 25.32 pounds in feeder cattle and .783 correlation with a standard error of predict ion of 69.12 pounds in fat cattle. This measurement could be used to some degree in predicting weights but is subject to a mat deal of experimental error. Using the measurement "crop widths, we have a correlation coeffi- cient of .282 with a standard error of prediction of 31.53 pounds in feeder cattle, and a correlation coefficient of .707 with a standard error of prediction of 77.93 pounds in fat cattle. The correlation coefficient is too low in feeder cattle to be significant and the standard error of prediction is too great in fat cattle to be of practical value. Chest width and width of chest floor both had too low a correla- -20- tion coefficient to be significant in feeder cattle and the standard error of prediction was too great in both cases to be of much practical value in fat cattle. The measurement"heart circumference”has a fairly high correlation with body weight in feeder cattle and has a standard error of prediction of 25.08 pounds. This standard error of prediction is low enough to be of practical value to most livestock producers. In fat cattle we have a still higher correlation coefficient of .853 but the standard error of prediction of 106.06 pounds is greater than the standard error of all but one other measurement. The correlation coefficient between the measuranent "flank circum- ference" and body weight is .603 with a standard error of prediction of 24.83 pounds. This correlation is rather low but the low standard error of prediction suggests the practicability of using this measurement. With fat cattle, using the same measurement, we have a correlation coefficient of .888, with a standard error of prediction of 51.32 pounds. This again looks like a good measurement to use in predicting weights. The measuranent "belly circumference" was found to have a low correlation with a rather high standard error of prediction for both feeder and fat cattle and is, therefore, of very little practical value. of all measurements taken, there seemed to be less chance of experimental error in heart circumference and depth of brisket than in any other measurements. The correlations between heart circumference and weight, and depth of brisket and weight were found to be higher for both feeder and fat cattle than the other measurements. The measurements for both feeder and fat cattle were combined and the correlation coeffi- cients between this particular group of figures and those of weight were O‘hmtad. -21.. Figure 4 Correlations beWeen Measurements and Weights in Beef Cattle Faeder Cattle Fat Cattle Standard Correlation standard Correlation EI‘ror of ryx Error of ryx Mediation Prediction Pounds Pounds Length of body 21.75 .749 106.58 .253 Height at shoulder 28.93 .473 64.92 .808 Height at hips 28.05 .521 68.34 .784 Depth of brisket 22.03 .742 56.54- .858 Depth rear flank 30.81 .348 76.05 .721 Loin width 25.32 .638 69.12 .783 Crop width 31.53 .282 77.93 .707 Chest width 28.51 .497 59.52 .841 Width chest floor 31.96 .074 98.10 .454 Heart circumference 25.02 .647 106.06 .953 Flank circumference 24.83 .603 51.32 .888 Belly circumference 32.27 .180 90.90 .584 Combination Feeders and Fat Cattle Heart circumference sy : 32.32 ryx : .988 Combination Feeders and Fat Cattle Depth of brisket S : 60.04 mm : .958 -22.. The correlation coefficient between the measurement "heart circum- ference” and body weight, using all animals, was found to be .988 with a standard error of prediction of 32.32 pounds. This is a very high cor- relation, and the standard error of prediction is small enough to be of practical value. The correlation coefficient between the measurement "depth of brisket” and weight, using all animals, was found to be .958, with a standard error of prediction of 60.04 pounds. Again we have a very high correlation, but in this case the standard error of prediction is too large for practical use. After studying the above results and the reports published by other experiment stations, it was decided that the best measurement to be used in predicting weight was that of heart circumference. This measure- ment was used because there is less chance for error in securing the measurement, and also as the animal grows and becomes fatter, whether he is in the feed-lot or on pasture, heart circumference increases in about the same degree as does that of weight. There is also a greater relationship between the measurement, "heart circmnference" and body weight, in all cases than there is with other measurements. Table 3 shows the weights of beef cattle as predicted from heart girth measurements, as prepared in this study. In comparing this table with Table 4, prepared by the United States Department of Agriculture, Bureau of Animal Industry, it can be seen that there is a difference in most predicted weights having the same measurement. Some of the differences are large enough to be significant; others are not. The weights in Table I are all predicted with a standard error of prediction of 32.32 pounds, plus or minus. In other Words, using these predicted weights for this particular type of cattle, one should be within thirty- two pounds of the exact weight of an animal sixty-might per cent of the time. -23- TWo other tables on predicting weights of dairy cows from baart girth measurements have been included as a means of comParing the differences betteen weights of beef cattle and dairy cattle having the same heart circumference and also to show the differences in the predicted weights of dairy cows, having the same heart circumference, as predicted by two entirely different experiments, but for the suns kind of cattle. Tables 7 and VI. In going back to Fiaire 4, the explanation for large standard errors of prediction in some cases may be due to experimental ermr in taking the measurement. It is a known fact that eXperimental errors are found in all measm'ements, but even then the standard errors of prediction, as shown in the tables, are nearly all too large to be of any practical value to the man who is accustomed to estimating the weight of beef cattle. The graphs included in this study show the scatter around the line of regression and are self-explanatory. -24— General Consideration and Discussion The data used in this study were collected incidental to a major objective of a "degree of finish experiment”. It is possible to suppose that had the erperiment been executed to fit the requirements of the measurement and weight study, greater correla- tions and smaller standard errors of prediction might have resulted. The fact that these animals were measured over a period of three years may have increased the experimental error to some degree and this may in some instances account for the large standard errors of prediction. Even though the same person took the measurements each year, he is likely to vary the position and manner in which the measurements were taken over a period of three years. The same type of cattle may also vary in measurements during a period of three years' time because of climatic conditions, whether or not they have had a good grazing season, or due to the thickness of hide or length of hair coat. The fact that the cattle studied in this experiment were all of the same type reduces the value somewhat of a study of this sort. The correla- tion coefficients and standard errors of prediction are based on "choice" to "fancy selected" feeders only and "choice" to "prime" fat cattle, and are, therefore, not of much value when used on common or poorer grades of cattle possibly having the same body measurements. The table predicting weights of beef cattle from heart girth measuranents, mde in this study, is based on measuranents, taken on cattle weighing from three hundred to eight hundred pounds. Had measurements been included from cattle in all classes, weighing from sixty pounds to two thousand pounds, this study would probably have been of more value. The predicted weights are, however, rather accurate for calves weighing -25- between two hundred and one thousand pounds, of choice beef type and should be of great value to the small livestock producer, who is not located near accurate scales and has to depend upon his "guess" or the trucker's esthmate to establish a value for his cattle. Sums}! The investigation reported in this paper is a study of the relation of certain body'msasurements to wehght in beef cattle. The data were collected from thirty-five head of Hereford heifers and thirty-six head of Hereford steers fed by the Michigan Agricultural Experiment Station during the years 1932-53, 1933-34, and 1934-35. The correlation coefficients between twelve body’measurements and total body weight were calculated. The standard error of prediction was calculated for both feeder and fat cattle. The measurements having the highest correlation are: heart circumference, with a correlation coefficient of .647 in feeder cattle, .953 in fat cattle, and .988 in a combination of both feeder and fat cattle; and depth of brisket, with a correlation.coefficient of .742 in feeder cattle, .858 in fat cattle, and .958 in combination feeder and fat cattle. The other measurements are as follows, in order of size of correlation: flank circumference, loin width, chest width, height at hips, height at shoulder, length of body, depth of rear flank, crop width, belly circumference, and width of chest floor. The standard error of prediction was found to be the anallest, using the measurement'length of body" in feeder cattle, but was larger for length of body in fat cattle than any other measurement. The standard errors of prediction are rather large in all cases and predicting weights with errors this large is hardly practicable. Many men dealing in live- stock can "guess" within twentybfive pounds of an animal's weight. The table for predicting weights of beef cattle from heart girth measurements, prepared in this study, is fairly accurate for calves weighing between.two hundred and one thousand pounds. It is not, however, of much practical use for cattle above or'below these weights. -27- Included in this study, for comparison, is the table predicting weights of beef cattle from heart girth measurements, as prepared by the United States Department of Agriculture, Bureau of Animal Industry. (1) . Also included, for comparison only, are the tables predicting weights of dairy cattle from heart girth measurements (2) , as prepared by the United States Department of Agriculture, Bureau of Dairy Industry, and by C. Brody (3) of Missouri. That there is a difference in weight of beef cattle having the some heart circtnnference is shown in these tables. Dairy cattle having the same heart circumference also vary in weight as shown by the two separate tables. There is also a difference in weight betWeem beef cattle and dairy cattle having the same heart circumference, as can be seen by comparing the two sets of tables. -28— Conclusion The standard errors of prediction obtained in these results indicate that there is a slight correlation between certain body measurements and the weight of beef cattle. The standard errors in feeder cattle show very little difference. As the cattle become larger and fatter, the differences in standard errors grow larger. Part of this increased difference no doubt is due to experimental error made in taking the measurement. The animals used in this study were uniformly alike in type. The fact that measurements from this particular type show a slight relationship to weight does not man that the same measurements taken on all beef cattle would show the same relationship. The measurement found to be the best to use in predicting weight was that of heart circumference followed by depth of brisket. The table predicting weights of beef cattle by heart girth measurements, as prepared in this study, has some practical value to the shell livestock producer. To the man who has had a great deal of experience in handling livestock this table is of minor importance. Men who are dealing with livestock from day to day can ”guess" the weight of an animal rather accurately. This study is based only on the best type of beef cattle and the predicted weights are necessarily for that kind of cattle. After all, a good set of livestock scales is the best method or weighing cattle, and with improved roads, better methods of travel, and numerous nnrketing centers, the livestock producers very seldom sell cattle on estimated weights, as was done in former years. There is too great a chance for experimental error to accurately predict weights of beef cattle from measurements. Variations also occur in cattle having the same body measurements, due to internal fat, size of bone, thickness and pliability of hide, climatic conditions and various other reasons. There is some correlation between certain body measurements and weights but because of the possibilities of experimental errors and the size of the standard error of prediction it is hardly possible to take a certain measurement on a beef animal and predict very accurately that particular animal's vei git. .30- Table I Feeder Cattle measurements 1932 - 33 Number We ight Circumfer— Circumfer- ence Heart ence Flank Centimeters Steers: 16 389 134 132 17 366 135 137 18 399 131 139 19 379 134 132 20 398 139 138 21 414 137 139 22 415 138 139 23 376 138 132 24 (375 (135 (133 26 (401 (143 (148 27 (392 (131 (134 23 ( 23 ( 23 ( Heifers: 1 (371 (130 (138 2 (370 (132 (133 3 389 137 140 4 353 133 126 5 372 133 137 6 404 142 140 7 373 132 129 8 400 142 142 9 360 130 126 10 345 138 146 11 369 132 140 13 380 135 132 -31.. Table I Feeder Cattle Measurements 1933 - 34 NUmber Wei t Circumfer- Circumfer- ence Heart once Flank Centimeters Heifers: 1 364 130 133 3 392 133 138 4 395 134 145 5 442 138 152.5 6 371 130 133 7 430 137 140 9 388 134 134 10 (419 _ (1:52 (140 11 (373 (1:50 (124 12 24 (399 24 (134 24 (137 13 (see (127 (134 15 (323 (127 (1:52 Steers: 16 362 131 144 17 394 132 130 18 370 131 132 19 358 134 130 22 442 138 141 23 405 134 140 24 377 133 134 25 415 140 136 26 441 137 135 27 419 133 142 28 463 139 142 29 37? 131 131 -32- Thble I Feeder Cattle measurements 1934 - 35 NUmber Wei t Circumfer- Circumfer- ence Heart ence Flank Centimeters Heifers: Nb Tag 323 127.5 140 1 338 130.5 2 311 133 131.5 3 323 128 131.5 4 364 134 139 5 346 136 (142 e 311 127.5 (135 7 (385 (139 (150 8 (361 (136.5 23 (142.5 9 (346 (133 (140 11 24 (326 24 (127 (130 12 (336 (130 133 ( ( Steers: 13 (377 (138 138 14 396 138 138 15 341 124 120 16 409 135 143 17 325 128 129 18 339 130 136 19 368 132 132 20 357 135 134 21 387 135 140 24 340 131 137 39 346 132 133 NO the 529 127 129 -33.. Table II Fat Cattle Measurements 1932 - 33 unmber Weight Circumfer- Circumfer- ence Heart ence Flank Centimeters Heifers: 3 751 169 173 5 710 168 170 7 812 174 169 Steers: 17 784 175 172 22 866 175 177 27 (821 171 166.5 ( Heifers: 1 (800 171 176 """"" 2 (731 (167 (168.5 9 24 (819 (178 (171 ( 24 ( 24 ( Steers: 19 (889 (180 (179 " ' ' 20 (927 (186 (179 24 828 176 168 Heifers: 6 702 168 168 4 550 149 155 13 600 155 160 Steers: 14 628 153 152 16 659 160 159 18 568 152 148 Heifers: 8 706 169 166 10 649 166 164 11 713 169 173 Steers: 21 747 168 169 23 756 166 163 26 773 170 175 -34- Table II Fat Cattle Measurements 1933 - 34 NUmber Weight Circumfer. Circumfer- ence Heart ence Flank Centimeters iHeifers: 3 784 180 173 4 805 177 183 15 735 175.5 176.5 Steers: 18 821 176 179.5 24 788 174.5 174 25 906 184.5 193 Heifers: 6 872 186 186 10 1037 187 189 13 822 177 180 Steers: 19 968 194 187 22 976 189 182 29 886 183 178.5 Heifers: 5 695 162.5 171 9 594 157 158 11 577 156 151 Steers: 16 631 157 160 23 588 156 154 28 688 165 165 Heifers: 1 701 167 167 7 773 171.5 182 12 683 166 167.5 Steers: 17 716 165 164 26 742 171 168 27 813 171 179 -35- Tabie II Fat Cattle'Measurements_ 1934 - 35 NUmber Weight Circumfer- Circumfer- ence‘Heart ence Flank Centimeters_ Heifers: 4 691 163 6 596 153 162 11 597 157 161 Steers: 16 713 161 166.5 17 666 161.5 165 39 (601 (151 (151 ( ( ( Heifers: 1 (711 24 (165 23 (170 5 24 (775 (168 (174 12 (631 (160 (169 ( Steers: 14 (818 174 178.5 15 726 163.5 173 24 780 168.5 173.5 Heifers: 3 804 172 180 9 834 177 182 No Tag 695 166 180 Steers: 13 831 173 185 18 793 169 175 20 843 180 181 Heifers: 2 876 177 184 7 867 181 192 8 782 170 176 Steers: 19 989 189 188 21 1009 191 191 Nb Tag 845 182 178 Grand Average -36- Grand Tbtals For Both Feeders and fat Catt}: 'Weight Circumderence Circumference Heart #- Flank 81,485 21,733 21,766.5 568.53 151.86 154. 26 -37- Table III ‘Weights of Beef Cattle as Predicted from.Heart GirchMeasurements IMade in a study of feeder and fat calves at Michigan State College Heart Girth Centimeters Inches 106 41.7 107 42.1 108 42.5 109 42.9 110 43.3 111 43.7 112 44.0 113 44.5 114 44.9 115 45.3 116 45.7 117 46.0 118 46.4 119 46.9 120 47.2 121 47.6 122 48.0 123 48.4 124 48.8 125 49.2 126 49.6 127 50.0 128 50.4 129 50.8 130 51.2 131 51.6 132 52.0 133 52.4 134 52.8 135 53.1 136 53.5 137 53.9 138 54.3 139 54.7 140 55.1 141 55.5 142 55.9 143 56.3 144 56.7 145 57.1 146 57.5 147 57.9 148 58.3 149 58.7 Weight 92 102 113 123 134 144 154 165 175 186 196 206 217 227 237 248 257 269 279 289 300 310 321 331 341 352 362 372 393 404 ' 414 435 445 455 466 476 487 497 518 528 539 Heart Girth Centimeters Inches 150 59.0 151 59.4 152 59.8 153 60.2 154 60.6 155 61.0 156 61.4 157 61.8 158 62.2 159 62.6 160 63.0 161 63.4 162 63.8 163 64.2 164 64.6 165 65.0 166 65.4 167 65.7 168 66.1 169 66.5 170 66.9 171 '67.3 172 67.7 173 68.1 174 68.5 175 68.9 176 69.3 177 69.6 178 70.0 179 70.5 180 70.9 181 71.3 182 71.7 183 72.0 184 72.4 185 72.8 186 73.2 187 73.6 188 74.0 189 74.4 190 74.8 191 75.2 192 75.6 193 76.0 Weight 549 559 570 580 591 601 611 622 632 643 653 663 674 684 695 705 715 726 736 746 757 767 778 788 798 809 819 830 840 850 861 871 882 892 902 913 923 933 944 954 965 975 985 996 -38- T8ble III Heart Girvth Heart Girth Cent imeters Inches Wei t Centimeters Inches We ight 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 76.4 76.8 77.1 77.6 78.0 78.3 78.7 79.1 79.5 79.9 80.3 80.7 81.1 81.5 81.9 82.3 82.7 83.1 85.5 83.9 84.3 84.6 85.0 85.4 85.8 86.2 86.6 87.0 87.4 87.8 88.2 88.6 89.0 89.4 89.8 90.2 90.6 90.9 91.3 91.7 92.1 92.5 92.9 93.3 93.7 94.1 94.4 94.9 95.3 95.7 96.1 96.5 1006 1017 1027 1037 1048 1058 1068 1079 1089 1100 1110 1120 1131 1141 1152 1162 1172 1183 1193 1204 1214 1224 1235 1245 1255 1266 1276 1286 1297 1307 1318 1328 1139 1349 1359 1370 1380 1391 1401 1411 1422 1432 1442 1453 1463 1474 1484 1494 1505 1515 1525 1536 246 247 248 249 250 251 252 253 254 255 256 257 238 259 260 261 262 263 264 265 266 267 268 269 270 96.9 97.2 97.6 98.0 98.4 98.8 99.2 99.6 100.0 100.4 100.8 101.2 101.6 102.0 » 102.4 102.8 103.1 103.5 103.9 104.3 104.7 105.1 105.5 105.9 106.3 1546 1557 1567 1577 1888 1598 1609 1619 1629 1640 1650 1660 1671 1681 1692 1702 1713 1723 1733 1744 1754 1764 1775 1785 1796 ~39- Table IV Apgoximate Weights of Beef Cattle of Good Grade for 6 Given Heart-Girth Measurement United States Department of Agriculture, Bureau of Animal Industry, Division of Animal Enabandry Heart Weight Heart Weight Heart Weight Girth in Girth in Girth , in Inches Pounds Inches Pounds Inches Pounds 50 76 50} 562 71 942 50% 62 51 595 71% 959 51 67 51} 404 72 977 51} 91 52 415 72; 994 52 96 52; 426 75 1011 52} 101 55 457 75% 1029 55 106 555 449 74 1047 555 112 54 461 74% 1065 54 116 54k 472 75 1065 54} 126 55 464 75} 1100 55 129 555 496 76 1117 555 155 56 506 762 1135 56 141 5% 520 7'? 1154 56} 147 57 555 775 1173 57 155 57k 545 78 1192 57} 159 56 556 76} 1211 56 166 565 571 79 1250 68% 175 59 565 795 1249 59 161 59% 596 60 1269 593 166 60 611 80% 1266 40 195 60% 624 81 1308 405 202 61 657 81% 1528 41 210 61‘} 651 88 134:8 41% 216 62 665 82% 1566 42 226 62% 679 65 1566 425 254 65 695 655 1409 45 242 65} 706 64 1450 a} 250 54 723 84% 1451 ‘4 259 54} 738 85 1472 44; 267 65 753 855 1493 45 276 65% 768 86 1514 45; 265 66 765 66} 1555 45 294 66} 796 67 1557 46'} 505 67 614 87% 1576 47 515 67}- 629 88 1600 47} 522 66 645 88% 1622 48 332 53% 861 89 1644 46} 542 69 677 89% 1667 49 552 695 695 90 1669 493’ 562 70 910 90; 1712 so 372 70% 926 91 1735 91‘} 17 57 -40- Teble V Etimating Weights of D6111 0058 from Heart-Girth Measurements United States Department of Agriculture Bureau of Dairy Industry Heart Weig‘t Heart Weight Heart Weight Girth Girth Girth Inches Pounds Inches Pounds Inches Pounds 25 so 48% 364 71 1027 25} 32 49 574 719 1046 2., 34 49% 564 72 1069 27.} 35 50 394 789 1090 28 39 50$ 404 73 1111 29% 9 2 51 414 739 1132 39 95 51‘} 424 74 1153 29.} 93 52 434 74% 1175 30 101 52% 445 75 1197 30% 104 55 456 75% 1219 31 103 53% 467 '76 1241 31.} 113 54 476 76%; 1265 32 113 59} 489 77 1285 32.} 135 55 501 77% 1506 :53 123 55} 513 78 1331 33} 133 56 526 785- 1554 34 133 56% 539 79 1377 34% 143 57 552 79% 1400 35 14a 57% 566 80 1423 35; 153 56 579 609 1446 35 153 689 593 81 1469 354} 155 59 607 81% 1492. 37 158 59% 622 62 1515 574} 174 60 657 82% 1556 33 130 609 652 65 1561 39} 135 61 666 659 1564 39 192 61% 684 84 1607 391 200 62 700 643 1650 40 903 629 716 65 1655 40* 215 63 732 85% 1676 41 334 659 749 66 1699 ‘14} 333 64 766 86% 1722 ‘2 240 64} 765 67 1745 431 343 65 600 674} 1766 4,3 257 653 617 66 1791 43.} 255 66 835 88% 1814 4‘ 275 66k 853 89 1857 - 442 234 67 671 699 1660 45 294 67‘} 669 90 1665 ‘5’ 30‘ 68 908 90} 1906 46 51‘ 68'} 927 91 1929 w} 324 69 947 91% 1952 ‘7 334 5% 9 57 9 2 1975 479 544 70 967 46 554 702 1007 -41- Table VI Average Body'weighta of Dairy Cattle for Given Chest Girths 16 prepared by c. Brody, Missouri 57.5 550 Chest Girth weight Cheat Girth Weight Inches __ Inches 22.5 39 58.5 578 23.5 44 59.5 606 24.5 50 60.5 635 25.5 56 61.5 665 26.5 62 62.5 696 27.5 69 63.5 728 28.5 76 64.5 761 29.5 84 65.5 794 30.5 92 7 66.5 829 31.5 101 67.5 865 32,5 110 68.5 901 33,5 120 69.5 939 34.5 130 70.5 977 35.5 141 71.5 1017 36.5 153 72.5 1057 37.5 165 73.5 1100 38.5 178 74.5 1148 39,5 191 75.5 1185 40.5 205 76.5 1231 41.5 219 77.5 1276 42.5 235 78.5 1324 43.5 250 79.5 1372 44.5 257 80.5 1421 45.5 284 81.5 1472 46.5 302 82.5 1523 47.5 331 83.5 1575 48.5 340 84.5 1630 49.5 351 85.5 1684 50.5 332 86.5 1740 51.5 403 87.5 1797 52.5 425 88.5 1856 53.5 449 89.5 1916 54.5 473 90.5 1976 55.5 498 91.5 2038 56.5 523 92.5 2102 F. “a -1. 74 1... 71* u?“ 2. 3. 4. 6. 7. 8. Bibliography Estimating the Weights of Beef and Dml-Pxn'pose Cattle from Heart- Girth Measurements. Bradford Knapp, Jr., Associate Animal Husbandman. United States Departnent of Agriculture, Bureau of Animal Industry. Mimeograph N0. 24. Estimating the Weights of Dairy Cows from Heart-Girth Measurements. Khndrick, I. F. and Parker, Jo BO. Bureau Of Dairy Industry. Mimeograph 695. Relationship between Live Weight and Chest Girth in Dairy Cattle of unknown Age. Brody, 0., Davis, H. P. and 116835410. 1. 0n University of Missouri, College of Agriculture, Agricultural Experiment Station. Research Bulletin 262. Changes in Body Measurements of Steers during Intensive Fattening. I. 1.. 11161:. Texas Agricultural Experiment Station. Bulletin Nb. 3850 Sept. 1928. Studies in Animal Nutrition. (1... changes in form and weight on different planes of nutrition). Moulton. Trowbridse md 381811. “1330911 Agricultural Experiment Station. Research Bulletin No. 43. Type in Beef Calves. university of Wyoming, Agricultural Experiment Station. Bulletin No. 155. Ste” Feeding Experiments. Evvard, Culbertson and Hammond. Iowa Agricultural Experiment station Leaflets N08. 16-22 and 25. y A Study of the Accuracy of Measurements of Dairy Cattle. J'. L. 141811 and 0. c. Copeland. Journal of Agricultural Research, VOL 41' -43- 9. The Accuracy of Cattle Weights. J’. L. Lush. Journal of Agricultural Research, Vol. 36. 10. A Statistical Study of Body Weights, Gains and Measurements of Steers during the Fattening Period. B. 0. Severson and Paul Gerlaugh. Pennsylvania State College Agricultural Experiment Station Annual Report, 1916-1917. 11. An Outline of Statistical Methods. Arkin, Herbert, and Colon, Raymond B. 1934. Wang h 7 Feed “'5 VI. Wtz)’ y:A-fbt¥ Ant-272.19 .bz’éoli X: 80.” Li. wfzr YaA “’7 A ‘3 *884-96 / b: [3036 Length -.. Body L73 0-? 80 A30 /25 130 Waist-.1- Fhodsr Coffin . #£fl> #oo , 3 - i __i 3 ' . g ‘ o .321 __ r . , :32” 350 s . ' _ . ____ ' -.-. W7‘ —’ . )’ A 76): . - 79-31/60! .300 . b: 1.1/5 ”=7- ”21:1 922: ~ ,1... floor . _///// ”EL- 42 AL 22 Al 44! . W29 hf . F91 691-He 75o 8’50 7.50 A r. 297. 72 b : 1.5. 63 X ; Wit! 7‘6 642” Fhur' 35 4o \Mflghf FeJAOP aaffle H50 ‘ 72A+bx .Az-IZbuiz b: 5.3 X=FIQNK cuffle £250 A1~ 24 .7 350 b:2b§§7 XDF/ON K 750 $50 Depth R20» Fluw/ :5 a g; SQO ,Iqo'b We‘gh‘r Farr ta'rfle 9.50 ye. W'f'. YtA 'f b)! A 7- -8820 2‘ b: 26.13 X'— brisket 850 6.50 Depth BH’s Kc'r WOIQhT Feed 2 ’- (La-HIG- .506 yzAbe A: 4224/ b: 9.58 400 X -. law's ref 35o DcPfh BH'sh'e'f Joe 45 .50 5.5 60 65 M / 4' / ngh‘f’. Fo'f Coffle Hear-1’ Cnrcu Info-cue: cu-caphucucc-g ’50. I70 1‘ s ‘Zoa Weog hf Panda :- Ca-r-grl a #00 3.5a .300 HT. A T. Hi P5 [050 We'fibf rev (,0 1‘er 950 U0 7.50 [.50 Hi. A 7? Hi Ps' .5250 y: "f: i: AM»! Ag-3a5.82 b: 6.5? X: H7‘. HI}: 9‘2 )’= 47%)! :-/73’4.i7 A; .7/. 75 )1: Hi flips loo [Jo A25 [Jo W639 MT Feeder H0045? . A “'9 Shoumfl- I650 Weigh 1‘ Fa ‘I' (191er $950 y: A #13} A: -“1657. Z50 ' b: 2LX3 g5 . x 1.. ”7'. At Sh' loo [.5 Ila ' 1’5 120 r25 Wag!“ y: w+. Feeéck =A+bx daft/e )’ -. - 3.0? .— 50 b 1'. 2- 77’ Flo N K bravura» o IVC C _ 11¢ ' 1“ EL! ffo 7.50 6:70 Flaw/C Cot-Cnuf’. .550. L50 H a lie Mo /i .7 0 OD Wezgfl Fee 4 0 |- y: Wf. . 1' we ’!&f y: W'f. FZade» :A+b (.af/le i} : [SJ-:2 b :7. 47 [/50 x: WI'J‘IA Chas /. Wméw y: Wf. Feede» Gaff/55 I: ‘Hby a A7- 2.5 ./ b : é.fzf xv; Chef WIJ‘IA £»OP Wld fl). / 5 If g o .21 / 19‘ ' [050 Wetgln‘ Fa'f buff/e. ?50 I50 750 650 {HIP Wl'c'fb. r550 ‘ AI R 150 ’% ”'Tfiifilfflfi‘mflfijfliufiflflflfflfliflfliflfififlfif’