Prediction of body composition in live animals.

Abstract

PREDICTION OF BODY COMPOSITION IN LIVE ANIMALS H. R. GHARAYBEH*, G. W. ARNOLD, W. R. McMANUS* and M. L. DUDZINSKI Summary Merino and Border Leicester x Merino ewe and wether weaners within the bodyweight range of 20 to 40 kg were slaughtered and body composition determined. Highly significant relations were established between full bodyweight or empty bodyweight and weights of skin, head, blood, internal organs, carcase weight and carcase composition. Total protein, fat and energy in the animals were also significantly related to full bodyweight and empty bodyweight. Significant differences existed in some of these relations between sexes and breeds. The data are compared with other published results. I. INTRODUCTION Most available techniques for predicting body composition in the live animal are either tedious or difficult to use in the field, e.g. tritium dilution technique (Panaretto 1963; Panaretto and Till 1963)) antipyrine and N-acetyl-4-aminoantipyrine (Wellington et al. 1956; Panaretto 1962; Bensadoun et al. 1962 , 1963), and with urinary creatinine as a predictor of body protein (Van Niekerk et al. 1963) . Other techniques, though promising, are still in the developmental stage, e.g. body fat from specific gravity determined by air displacement (Beeston 1964). Precise regression relations between parameters of the live animal and parameters of the composition of either the empty (ingesta-free) body or the carcase would have wide application. There has been little interest in such studies because the amount of ingesta in the intestinal tract is large and variable. It can range from 5 to 30% of the total weight of the body. Bensadoun et al. (1963), however, found good relations between the weight of the wool-free, empty body of sheep (EBW) and their 'shrunk ' ( 18-20 hour fast, water allowed) full bodyweight (FBW) when a comparatively wide variety of diets and body compositions were used (r = 0.994, S.E. of the estimate 3.8% of the mean shrunk weight of 41.62 kg) . After re-examining published work, Tulloh ( 1963) reported that in the sheep EBW and carcase weight are highly correlated with dissected carcase composition. He points out that carcase composition in sheep is more closely related to FBW than to age. Tallis, Turner and Brown ( 1964) report a 0.95 correlation between liveweight before slaughter and weight of edible meat in seven-month old Merino wethers. Gardner, Hogue and Bensadoun ( 1964) report good rela*School of Wool Technology, University of New South Wales. 'fDivision of Plant Industry, C.S.I.R.O., Canberra, ACT. Present address: W.A. Laboratory, Perth, W.A. $Division of Mathematical Statistics, C.S.I.R.O., Canberra, ACT. 284 Regional tions between wool-free empty body composition and the weight of the wool-free EBW of lambs. The present report details various regression relations obtained for pasturefed Merino and Border Leicester x Merino weaner ewes and wethers relating FBW and EBW to body composition. II. METHODS Twenty-four 9-month-old Merino and Border Leicester x Merino sheep were used. Six ewes and six wethers, paired for bodyweight within breeds, were selected from the flock to give as wide a range in bodyweight as possible (Merinos 20-3 1 kg, Border Leicester x Merinos 24-40 kg). The sheep were fasted for 36 hours, weighed and slaughtered. Weights of blood, head, feet, skin, internal organs and carcases wer e obtained. The skins were dried and shorn and the weight of wool recorded. The head and feet, internal organs and half of each carcase were minced separately. Minced samples were analysed for moisture, protein, fat and calorific value. The other half of each carcase was dissected into muscle, fat, bone and waste. Water content was determined on the separate parts. Edible meat was calculated as muscle plus fat plus waste. Linear regression of FBW and EBW on fresh weights, dry weights and chemical composition values of body and carcase components was done. The data showed no evidence of non-linearity. The mean linear regression and homogeneity of regression was tested. The four groups were compared using covariance. In this paper only the relations between FBW and EBW and body composition on a fresh basis are described. III. RESULTS Table 1 gives the estimated composition of animals with FBW' of 24 and s 31 kg for each breed and sex group. The components of the carcases given in Table 1 were each determined from an equation relating it to FBW. FBW' of s 24 and 31 kg cover the range common to all treatments. Analyses of variance, correlation coefficients of mean regressions and standard deviations of mean residuals are given in Table 2. FBW was significantly linearly related to all components measured at the 0.1% level of probability. Correlation coefficients varied from 0.745 for weight of head to 0.995 for EBW. There were no significant differences in the slopes of regression lines for sexes and breeds for any parameter except bone. But for skin, head, carcase and total body water weights there were significant differences in the adjusted means. Thus, at the same FBW all groups had the same EBW, weight of blood, internal organs, edible meat and total calories. Merinos had heavier skins and lower carcase weights than Border Leicester x Merinos; the latter was due to less bone. Wethers had heavier heads than ewes. The only difference between groups in total body water was that Merino ewes have significantly less than other groups. 285 TABLE 2 Analysis of varialnce and of covariance of fresh weight parts of the body, composition of carcase and total empty body (Y) on fasted liveweight (X) of 'r-Conservative - Least favourable in respect to the size of the (Xi - Xj ) component of LSD' S * P<O.O5 ** P<O.Ol *** P<O.OOl IV. DISCUSSION These data show that the composition of the live animal may be predicted with high precision from the FBW. Precision is slightly improved by using EBW - as seen from the correlation coefficients and residual standard deviations in Table 3. With the exception of internal organs, the differences between types of animals were the same when predicted by EBW or FBW. It is important that a fasted weight be taken because of the variability in weight of ingesta between animals or treatments (Large 1964). In this study, a 36-hour fast was used giving a correlation coefficient of 0.995 with a S.D. -t- 605 g for FBW on E'BW. This compares with 0.994 by Bensadoun et al. ( 1963) for an 18-20 hour fast. Large (1964) reported the relations between body parts and EBW for Suffolk x Halfbred ewe lambs aged l-l 6 weeks (4-40 kg EBW) . There is good agreement between his regression lines and ours only for the carcase weight o f Border Leicester x Merino sheep. Our Merino data has a parallel, but lower, regression line relating carcase weight to EBW. This suggests a true breed effect operates since the disagreement is difficult to explain on an age or nutrition basis (Tulloh 1963). Y (Carcase weight = 0.549 x = 0.602 X = 0.595 x where X = EBW in kg) -0.348 (Border Leicester x Merino) -2.580 (Merino) -0.719 Large (1964 ) in kg. Gardner, Hogue and Bensadoun (1964) reported relations between bod y chemical composition and EBW for Hampshire x Rambouillet-Columbia ewe and wether lambs on different planes of nutrition, slaughtered at 13 weeks ( 17-37 kg EBW). Their correlation coefficients for total protein, fat and calories were slightly lower than reported in the present study. The regression equations differed considerably as shown. TABLE 3 Comparison of relationships between body composition and full or empty bodyweight Body Component Blood Skin Head Internal organs Carcase Water Protein Fat Calories *Kilocalories 288 Full Bodyweight Empty Bodyweight Empty Bodyweight Residual Residual Residual S.D. (g) r S.D. (g) r (Large 1964) S-D* (8) 0.884 0.867 0.745 0.864 0.987 0.986 0.929 0.923 0.973 133 316 109 553 521 563 383 494 4183* 0.883 0.878 0.735 0.907 0.989 0.995 0.943 0.943 0.972 138 303 111 510 414 307 342 504 4210 61 148 59 416 Energy in whole animal (megacals) (Y) = 3 . 3 0 7 x -12.247. This paper (Y) = 4 . 7 3 3 x -47.47 Gardner, Hogue and Bensadoun (1964) Protein in whole animal (kg) 0 = 0.1816 X +0.16. This paper (y) = 0.1197 x t-O.92 Gardner, Hogue and Bensadoun (1964) where X = EBW in k g There was no difference between the breeds in whole animal calorific value. Merinos had more energy in skin and other organs which balanced the lower total energy from smaller carcases. Bensadoun et at. ( 1963) showed that FBW (18-20 hour fasting, but access) was highly correlated with empty body water up to 55 kg EBW animals not containing more than 30% fat. These workers proposed a whereby the energy value of the sheep body in vivo can be calculated from ledge of the empty body water and the FBW. water or for system know- For certain body components, a single regression equation describes the relations with FBW, but for others it does not. It was found in this and other studies (Gharaybeh, Arnold and McManus, unpublished data) that the fresh weight of edible meat differs very little at the same FBW or EBW between a wide range of breed, sex and grazing management treatments. It is concluded that the FBW (36-hour fasted) of sheep is a good predictor of body composition and the relations shown have special significance for field investigations. More evidence is needed to' show whether a general prediction equation covering a wide range of FBW, breeds, ages and nutritional histories of animals can be established. Preliminary information indicates that this is unlikely (Gharaybeh, Arnold and McManus, unpublished data). It is difficult to compare the precision of the regression relations reported here with the most accurate chemical technique (tritium dilution) because different workers present error terms in different ways. Use of residual standard deviation in all such studies would permit uniform comparisons to be made. V. ACKNOWLEDGMENTS The authors thank various members of the staff of C.S.I.R.O. Division of Plant Industry 's Grassland Agronomy Section and Mr. H. Lahoud and Mr. N. Watson of the School of Wool Technology for field and laboratory assistance. Particular thanks are due to F.A.O. and to the Andre Mayer Fellowship Programme for financial assistance to one of us (H. R. Gharaybeh) . A., V AN N IEKERK , B. D. H., PALADINES, 0. L., and R EID , J. T. ( 1962). Evaluation of the combined .use of antipyrine and N-acetyl-4-aminoantipyrine in predicting the body composition of sheep. Journal of Animal Science 21: 988. BENSADOUN, A., V AN N IEKERK , B. D. H., PALADINES, 0. L., and R EID , J. T. (1963). Evaluation of antipyrine, N-acetyl-4-aminoantipyrine and shrunk bodyweight in predicting the chemical composition and energy value of the sheep body. Journal of Animal Science 22: 604. BENSADOUN, VI. REFERENCES 289 BEESTON, G of the Australian Society of Animal Production 5: 353. J. W. U. (1964). Determination of the specific gravity of live sheep. Proceedings ARDNER , E. W., HOGUE, D. E., and BENSADOUN, A. (1964). Body composition and efficiency of growth of suckling lambs as affected by level of feed intake. Journal of Animal Science 23: 943. , R. V. (1964). The development of the lamb with particular reference to the alimentary tract. Animal Production 6: 169. PANARETTO, B. A. (1962). Body Composition in vivo. I. The estimation of total body water with antipyrine and the relationship of total body water to total body fat in rabbits. L ARGE l Australian Journal of Agricultural Research 14: 594. B. A. (1963). Body Composition in vivo. III. The composition of living ruminants and its relation to the tritiated water spaces. Australian Journal of Agricultural Research 14: 944. PANARETTO, B. A., and T ILL , A. R. (1963). Body Composition in vivo. II. The composi,tion of mature goats and its relation to the antipyrine, tritiated water, and the N-acetyl4-aminoantipyrine spaces. Australian Journal of Agricultural Research 14: 926. TALLIS, G, M., T URNER , H. N., and BROWN, G. H. (1964). The relationship between live measurements and edible meat in Merino wethers. Australian Journal of Agricultural Research 15: 446. T ULLO H, N. M. (1963). Relation between carcase composition and liveweight of sheep, Nature 197: 809. V AN N IEKERK , B. D. H., R EI D, J. T., BENSADOUN, A. and P ALADINES , 0. L. ( 1963 ) . Urinary creatinine as an index of body composition. Journal of Nutrition 79: 463. W ELLINGTON , G. H., R EI D, J. T., B RATZLER , L. J. and M ILLER , J. I. (1956). Use of anti,pyrine in nutritional and meat studies with cattle. Journal of Animal Science 15: 76. PANARETTO, 290