Further Studies on the Influence of Genotype and Dietary Protein on the Performance of Broilers 1

Further Studies on the Influence of Genotype and Dietary Protein on the Performance of Broilers 1 E. R. SMITH,* G. M. PESTI,*,2 R. I. BAKALLI,* G. O. WARE, and J.F.M. MENTEN3 *Department of Poultry Science and School of Forest Resources, The University of Georgia, Athens, Georgia 30602 ABSTRACT An experiment was conducted to quantify genetic differences in response to dietary protein level of male vs female broilers. Broilers (1 d old) from a high-yield strain cross () and a fastgrowing strain cross () were placed on fresh pine shavings in floor pens. From Day 0 to 18, all birds were fed a 23% CP starter diet. During Days 18 to 53 male birds were fed either a 16, 18, 20, 22, 24, or 26% CP diet (3,200 kcal ME/kg) and female birds were fed the 16, 20, or 24% CP diet. Significant differences (P < 0.05) were noted in the performance of the different strains. male birds had a higher body weight (3.37 vs 3.16 kg), higher feed intake (7.08 vs 6.78 kg), higher breast yield (31.76 vs 29.25%), higher carcass yield (73.90 vs 71.85%), and a lower adjusted feed conversion ratio (FCR; 2.10 vs 2.16 g:g) than males at 53 d of age. As compared to females, Ross Ross 208 female broilers also had a higher body weight (2.68 vs 2.55 kg), higher breast meat yield (33.61 vs 30.80%), higher carcass yield (75.31 vs 73.91%), and lower adjusted FCR (1.97 vs 2.04 g:g). Qualitative differences in the response of these strains were confirmed and better qualitative data is presented that can be used to predict the important output parameters from the import inputs in broiler production. (Key words: broiler, strain, sex, carcass yield, dietary protein, abdominal fat pad) 1998 Poultry Science 77:1678 1687 INTRODUCTION Improvements in the genetic composition of broilers is a compelling reason for producers to reevaluate their feeding regimens. Geneticists have selected for improved body weight on the assumption that it will increase the salable product (carcass weight) with the same or reduced inputs (mainly feed). Different growth potentials for different strains have been found (Proudfoot and Hulan, 1978; Malone et al., 1979; Boa- Amponsem et al., 1991; Holsheimer and Veerkamp, 1992). A genetic influence on feed conversion ratios (FCR; grams feed consumed per gram body weight change) was found by Hulan et al. (1980). Jackson et al. (1982), Leclercq (1983), Marks and Pesti (1984), and Barbato (1992) found genetics to play a role in abdominal fat pad weights. Orr et al. (1956) with turkeys and Renden et al. (1992) with broilers found a variance in carcass yield due to genetics. Relationships between Received for publication January 30, 1998. Accepted for publication June 1, 1998. 1Supported by state and Hatch funds allocated to the Georgia Agricultural Stations of The University of Georgia. 2To whom correspondence should be addressed: gpesti@uga. cc.uga.edu 3Present address: Dept. Zootecnia, Univ. Sao Paulo, 13418-900 Piracicaba SP, Brazil. Abbreviation Key: FCR = feed conversion ratio. 1678 genotype and dietary protein levels was reported by Leclercq (1983); Marks and Pesti (1984); Cahaner et al. (1987); and Smith and Pesti (1998). Much research has dealt with the relationship between dietary protein and broiler performance since Fraps (1943) showed that a manipulation of dietary protein can have various effects on broiler performance. He found differences in body weights, feed intake, and especially carcass composition of birds. Although this qualitative relationship has been known for over 50 yr, it is still not widely used in profit maximization models. Parsons and Baker (1982) showed a significant linear reduction in both rate and efficiency of gain when decreasing dietary protein level from 24 to 16%, but little effect in the 24 to 20% range. Summers et al. (1992) found no significant differences in weight gain, feed intake, or FCR with the feeding of various dietary protein levels. Recent studies by Leclercq and Guy (1991) demonstrated differences in the response to dietary protein and amino acids by genetic lines selected for high quantities of carcass lean vs high carcass fat. Differences in Leclercq s lines somewhat resemble differences in modern high yield vs fast-growing strains.

BROILER PERFORMANCE AS INFLUENCED BY PROTEIN AND SEX 1679 A difference in the growth response between males and females suggests that they have different nutritional requirements: Proudfoot and Hulan (1978) found a significant (P < 0.05) sex by diet interaction with broilers. Ajang et al. (1993) showed a significant (P < 0.05) diet by sex interaction, with males showing a greater response to protein increase than females. They also showed females to have a significantly (P < 0.001) lower carcass yield than males, probably due to a significantly (P < 0.05) higher skin yield. Smith and Pesti (1998) showed that a high yield broiler strain needed a higher dietary protein level to maximize body weight and feed efficiency than a fast-growing strain, very similar to what Leclercq and Guy (1991) found for their lean and fat lines. This experiment described here was conducted to confirm earlier results (Smith and Pesti, 1998) and quantify relationships and interrelationships between feeding various dietary protein levels to obtain better estimates of the performance of males of two commercial broiler stocks over a broader range of protein levels. Comparison of the responses of males vs females were also made and the results were extended to include breast meat yields. MATERIALS AND METHODS Eleven hundred and eighty-eight male (594 Ross Ross 208 and 594 ) and 216 female commercial broiler chicks (108 of each strain) were obtained from different commercial hatcheries. The strains were from different hatcheries. birds were feather-sexed and birds were vent-sexed at day of hatch. Average chick weights were 38.8, 38.0, 41.5, and 41.3 for Ross Ross males and females, and males and females, respectively. Chicks were randomly assigned to pens by strain cross and sex. There were 36 pens of males (33 birds per pen; 18 pens per genotype) and 12 pens of females (18 birds per pen; 6 pens per genotype). All birds were raised in floor pens (1.22 m 3.66 m) at 35 C for the first 3 d and then the temperature was decreased by 2 C every 3 d until 24 C was reached. Temperature was maintained at 24 C for the duration of the experiment. All chicks were fed a corn, soybean meal, and poultry oil based starter diet (23% protein, 3,200 kcal ME/kg) formulated to meet nutrient concentrations recommended by the NRC (1994) from Day 0 to 18 (Table 1). One bell-shaped waterer and one tube feeder (140 cm linear feeder space) were provided per pen. Water and feed were supplied for ad libitum consumption. Diets that contained CP levels of 16, 18, 20, 22, 24, and 26% were fed from Day 18 to 53 (Table 1). The calculated dietary energy level of all experimental diets was 3,200 kcal ME/kg. At Day 18, each of the diets was assigned TABLE 1. Composition of the experimental diets Experimental diets Level of protein UGA Ingredients Starter 16% 18% 20% 22% 24% 26% Ground yellow corn 57.34 72.10 66.03 59.95 53.88 47.81 41.74 Soybean meal 48% (dehulled) 33.48 20.32 25.48 30.65 35.81 40.97 46.14 Poultry fat (stabilized) 3.15 3,50 4.44 5.38 6.31 7.25 8.18 Poultry by-product meal 3.00.................. Iodized sodium chloride 0.21 0.40 0.40 0.40 0.40 0.40 0.40 DL-methionine 0.19 0.03 0.05 0.07 0.09 0.11 0.13 Vitamin premix 1 0.25 0.25 0.25 0.25 0.25 0.25 0.25 Trace mineral premix 2 0.05 0.08 0.08 0.08 0.08 0.08 0.08 Defluorinated phosphate 1.54 1.78 1.75 1.72 1.69 1.66 1.63 Limestone 0.79 1.45 1.43 1.41 1.40 1.38 1.36 Coban-60 3... 0.08 0.08 0.08 0.08 0.08 0.08 Bacitracin BMD-60 4... 0.01 0.01 0.01 0.01 0.01 0.01 Calculated composition 5 ME, kcal/kg 3.1 3.2 3.2 3.2 3.2 3.2 3.2 Analyzed composition Protein 22.66 15.6 18.1 19.9 21.7 24.1 25.9 1Vitamin premix provides the following per kilogram: vitamin A, 5,500 IU from all trans-retinyl acetate, cholecalciferol, 1,100 IU; vitamin E, 11 IU from all-rac-a-tocopherol acetate; riboflavin, 4.4 mg; Ca pantothenate, 12 mg; nicotinic acid, 44 mg; choline Cl, 220 mg; vitamin B 12, 6.6 mg; vitamin B 6, 2.2 mg; menadione, 1.1 mg; (as MSBC); folic acid, 0.55 mg; d-biotin, 0.11 mg; thiamine, 1.1 mg (as thiamine mononitrate); ethoxyquin, 125 mg. 2Trace mineral premix provides the following in milligrams per kilogram of diet: Mn, 60; Zn, 50; Fe, 30; Cu, 5; I, 1.5. 3Eli Lilly and Co., Indianapolis, IN 46285-0002. 4Alpharma, Fort Lee, NJ 07024. 5Based on NRC (1994) tables. (%)

1680 SMITH ET AL. TABLE 2. The influence of protein and age on the BW, feed intake (FI), and feed conversion ratio (FCR) of male broilers Age Protein BW 1 FI 1 FCR 1 BW 1 FI 1 FCR 1 (d) (%) (kg) (g:g) (kg) (g:g) 18 Starter 0.67 ± 0.03 0.76 ± 0.01 1.20 ± 0.01 0.66 ± 0.01 0.76 ± 0.01 1.21 ± 0.01 16 1.54 ± 0.02 2.56 ± 0.03 1.70 ± 0.01 1.57 ± 0.04 2.59 ± 0.05 1.68 ± 0.01 18 1.67 ± 0.05 2.62 ± 0.07 1.56 ± 0.04 1.65 ± 0.02 2.66 ± 0.04 1.62 ± 0.02 32 20 1.71 ± 0.03 2.63 ± 0.05 1.57 ± 0.02 1.66 ± 0.01 2.57 ± 0.02 1.59 ± 0.01 22 1.78 ± 0.02 2.57 ± 0.03 1.47 ± 0.01 1.65 ± 0.03 2.45 ± 0.03 1.51 ± 0.01 24 1.85 ± 0.05 2.63 ± 0.04 1.45 ± 0.02 1.69 ± 0.03 2.47 ± 0.02 1.46 ± 0.01 26 1.76 ± 0.03 2.46 ± 0.06 1.43 ± 0.01 1.66 ± 0.03 2.42 ± 0.05 1.48 ± 0.02 16 2.03 ± 0.01 3.72 ± 0.02 1.85 ± 0.01 2.06 ± 0.07 3.75 ± 0.08 1.84 ± 0.02 18 2.23 ± 0.04 3.86 ± 0.09 1.72 ± 0.03 2.17 ± 0.03 3.86 ± 0.04 1.78 ± 0.03 39 20 2.24 ± 0.04 3.86 ± 0.08 1.74 ± 0.03 2.21 ± 0.02 3.79 ± 0.04 1.74 ± 0.03 22 2.38 ± 0.03 3.83 ± 0.04 1.63 ± 0.01 2.14 ± 0.04 3.60 ± 0.03 1.70 ± 0.03 24 2.43 ± 0.05 3.92 ± 0.04 1.63 ± 0.02 2.20 ± 0.05 3.67 ± 0.02 1.70 ± 0.03 26 2.32 ± 0.04 3.68 ± 0.08 1.60 ± 0.02 2.19 ± 0.04 3.59 ± 0.09 1.64 ± 0.02 16 2.52 ± 0.04 4.81 ± 0.05 1.94 ± 0.02 2.56 ± 0.08 4.94 ± 0.13 1.95 ± 0.02 18 2.77 ± 0.06 5.12 ± 0.10 1.83 ± 0.02 2.65 ± 0.02 5.07 ± 0.04 1.90 ± 0.04 46 20 2.84 ± 0.11 5.20 ± 0.16 1.84 ± 0.01 2.71 ± 0.03 5.02 ± 0.05 1.88 ± 0.04 22 2.87 ± 0.05 5.11 ± 0.08 1.78 ± 0.01 2.61 ± 0.05 4.74 ± 0.05 1.82 ± 0.02 24 2.98 ± 0.03 5.24 ± 0.06 1.77 ± 0.02 2.66 ± 0.07 4.85 ± 0.02 1.81 ± 0.03 26 2.89 ± 0.04 4.95 ± 0.10 1.72 ± 0.02 2.68 ± 0.06 4.77 ± 0.13 1.77 ± 0.03 16 3.02 ± 0.04 6.64 ± 0.29 2.17 ± 0.10 2.99 ± 0.12 6.64 ± 0.15 2.23 ± 0.11 18 3.35 ± 0.07 7.37 ± 0.42 2.14 ± 0.10 3.13 ± 0.06 7.14 ± 0.37 2.21 ± 0.11 53 20 3.31 ± 0.06 7.39 ± 0.60 2.18 ± 0.13 3.17 ± 0.03 7.07 ± 0.40 2.19 ± 0.10 22 3.41 ± 0.10 6.93 ± 0.44 2.04 ± 0.09 3.02 ± 0.09 6.83 ± 0.34 2.19 ± 0.12 24 3.56 ± 0.06 7.56 ± 0.51 2.08 ± 0.09 3.05 ± 0.14 6.48 ± 0.22 2.06 ± 0.12 26 3.26 ± 0.08 6.56 ± 0.33 1.97 ± 0.07 3.11 ± 0.10 6.48 ± 0.45 2.06 ± 0.13 1Mean ± SE of three pens per treatment (33 chicks per pen). to six pens of male chicks (three pens per genotype). Only diets containing 16, 20, and 24% protein were assigned to females (two pens per genotype for each protein level). All diets were pelleted (66 C, 102 kg/ cm2)4 and crumbled before feeding. At 0, 18, 32, 39, 46 and 53 d of age, chicks and residual feed were weighed by pen. The FCR for each pen was calculated using the total weight gain for the period including weights of birds when they died (or were culled for lameness during the period). After weighing at Day 18, six male birds per genotype were removed at random for carcass analysis. At Day 32, 39, and 46, six male birds per treatment group were selected for carcass analysis. At Day 53, six males and eight females per treatment group were randomly selected for carcass analysis. All selected birds were wing-banded and held in crates. In the processing room, the crates were supplied in the sequence of the order in which they were stocked. Each bird was weighed immediately prior to slaughter. Birds were then killed by exsanguination, scalded, picked, and eviscerated. Abdominal fat pads were collected and weighed. Birds were rinsed, allowed to drip dry, weighed (hot carcass weight), and placed in ice water (0 C) overnight. The next day birds were allowed to drip dry ( 15 min) and reweighed (chilled carcass 4California Pellet Mill Co. Master Model, Modesto, CA 95358. weight) to obtain a percentage water retention. Whole breasts (skin and bone) were then removed and weighed to obtain breast yield. Breast yield was based on chilled carcass weight, carcass yield was calculated based on live body weight and percentage abdominal fat pad was based on the hot carcass weight. The study was conducted as a completely randomized design with a factorial arrangement of fixed treatment effects consisting of sex, strain, level of protein, and age. Statistical analyses were dependent on the type and frequency of response data measured for the males and females. The statistical models for body weight and feed intake included the effects of sex, strain, level of protein, age, and all possible interactions. The statistical model for the male carcass data at Days 18, 32, 39, 46, and 53 included the effects for strain, level of protein, age, and all possible interactions. The statistical model for female carcass data at Day 53 included the effects of strain, level of protein, and all possible interactions. Regression analyses were also employed for data analysis. Quadratic response models were fit to determine the behavior of the response variables to the set of independent factors. The models included all linear and quadratic terms and all pairwise cross products of linear terms. Reduced models were constructed for response variables by retaining independent variables that tested significantly at P < 0.10. Linear terms were always retained in the model when a quadratic or a cross

BROILER PERFORMANCE AS INFLUENCED BY PROTEIN AND SEX 1681 TABLE 3. The influence of protein and age on the BW, feed intake (FI), and feed conversion ratio (FCR) of female broilers Age Protein BW 1 FI 1 FCR 1 BW 1 FI 1 FCR 1 (d) (%) (kg) (g:g) (kg) (g:g) 18 Starter 0.63 ± 0.01 0.70 ± 0.01 1.18 ± 0.02 0.61 ± 0.05 0.65 ± 0.01 1.14 ± 0.03 16 1.36 ± 0.06 2.32 ± 0.06 1.75 ± 0.04 1.36 ± 0.05 2.32 ± 0.01 1.68 ± 0.03 32 20 1.42 ± 0.01 2.20 ± 0.04 1.58 ± 0.03 1.40 ± 0.01 2.41 ± 0.03 1.77 ± 0.03 24 1.61 ± 0.04 2.34 ± 0.09 1.49 ± 0.02 1.41 ± 0.01 2.04 ± 0.04 1.49 ± 0.03 16 1.79 ± 0.10 3.35 ± 0.17 1.90 ± 0.01 1.82 ± 0.01 3.37 ± 0.05 1.84 ± 0.01 39 20 1.84 ± 0.01 3.24 ± 0.10 1.78 ± 0.05 1.84 ± 0.03 3.48 ± 0.06 1.94 ± 0.01 24 2.08 ± 0.07 3.43 ± 0.17 1.68 ± 0.03 1.83 ± 0.01 3.05 ± 0.08 1.71 ± 0.03 16 2.22 ± 0.16 4.36 ± 0.29 1.99 ± 0.02 2.22 ± 0.01 4.36 ± 0.05 1.96 ± 0.02 46 20 2.29 ± 0.02 4.28 ± 0.13 1.89 ± 0.07 2.21 ± 0.04 4.44 ± 0.11 2.04 ± 0.01 24 2.38 ± 0.08 4.51 ± 0.19 1.88 ± 0.03 2.21 ± 0.01 4.01 ± 0.13 1.85 ± 0.05 16 2.57 ± 0.15 5.19 ± 0.04 2.05 ± 0.11 2.55 ± 0.01 5.16 ± 0.26 1.99 ± 0.11 53 20 2.63 ± 0.01 5.03 ± 0.25 1.92 ± 0.08 2.57 ± 0.05 5.58 ± 0.13 2.20 ± 0.01 24 2.85 ± 0.02 5.59 ± 0.01 1.95 ± 0.06 2.54 ± 0.02 4.78 ± 0.15 1.91 ± 0.07 1Mean ± SE of two pens per treatment (18 chicks per pen). product term involving that linear term met the above criterion. All analyses were conducted using the General Linear Model (GLM) procedure of SAS (SAS Institute, 1985). Performance RESULTS Analyzed protein content of the experimental diets were in acceptable agreement with formulation goals (Table 1). For the ages of birds studied, body weights and feed intake increased with age (Tables 2 to 4). As dietary protein level increased, the body weight response for males was nonlinear. Strain had highly significant effects on body weight and feed intake (Table 4). The Ross Ross 208 birds were heavier at each age (age by strain interaction; Table 4). Body weight and feed intake of both strains were very similar when the birds were fed 16% protein, but birds were heavier and ate more feed when protein levels were increased (significant protein by strain interaction; Tables 2 to 4). birds were heaviest when fed the 24% protein level, but birds were heaviest when 20% protein was fed (Tables 2 and 3). Sex had significant effects on body weight and feed intake (Table 4) and the sexes responded differently to protein over time as evidenced by significant sex by strain, sex by protein, and sex by age interactions (Table 4). Mortality averaged 3.6% to 18 d and 10.9% to 56 d; there were no significant effects of any treatment on mortality and the data are not presented. The sources of variation, degrees of freedom, and P values associated with the statistical tests for body weight and feed intake are provided in Table 4. Due to the significance of the interactions between strain and age, strain and protein, sex and age, sex and protein, and age and protein, separate quadratic response models for body weight and feed intake were fit to each combination of strain and sex (Table 5). Carcass Yield Carcass parameter yields for male (Table 6) and female (Table 7) broilers were affected by strain, age and dietary protein levels (Tables 8 and 9). Male birds had a higher carcass and breast yield at each age studied (strain by age interaction; Table 8). A larger difference was found in carcass yield between protein levels as the male birds aged (age by protein interaction; Table 8). The sources of variation, degrees of freedom, and P values associated with the statistical test for carcass TABLE 4. Analysis of variance summary of the effect of age, percentage protein, sex, and strain on chick performance Source 1 df Body weight P > F Feed intake Strain 1 0.0001 0.0013 Sex 1 0.0001 0.0001 Strain Sex 1 0.0701 0.4386 Age 4 0.0001 0.0001 Strain Age 4 0.0001 0.1876 Sex Age 4 0.0001 0.0001 Strain Sex Age 4 0.4359 0.8899 Protein 5 0.0001 0.0005 Strain Protein 5 0.0001 0.0015 Sex Protein 2 0.0002 0.1977 Strain Sex Protein 2 0.6556 0.1304 Age Protein 20 0.0239 0.4083 Strain Age Protein 20 0.2097 0.6737 Sex Age Protein 8 0.6037 0.9548 Strain Sex Age Protein 8 0.9467 0.9830 Error 150 R 2 0.9944 0.9892

1682 SMITH ET AL. TABLE 5. Coefficients for the best fit multiple regression lines to predict performance as function of age and protein level Body weight Feed intake Sex Source 1 Estimate P > t Estimate P > t Estimate P > t Estimate P > t Male Intercept 2.6770 0.0288 1.5369 0.0005 4.2256 0.0048 0.4143 0.1870 Age 0.1343 0.0009 0.0787 0.0001 0.0040 0.8203 0.0056 0.7192 Age 2 0.0010 0.0066 0.0001 0.0755 0.0026 0.0001 0.0023 0.0001 Protein 0.2647 0.0199 0.0738 0.0716 0.4171 0.0035 0.0228 0.0085 Protein 2 0.0048 0.0709 0.0016 0.0959 0.0100 0.0033 NS* NS Protein Age 0.0160 0.0001 NS NS NS NS NS NS Protein Age 2 0.0001 0.0054 NS NS NS NS NS NS Protein 2 Age 0.0003 0.0001 NS NS NS NS NS NS Female Intercept 0.3933 0.2462 0.5236 0.0001 2.3792 0.2008 4.7913 0.2838 Age 0.0448 0.0001 0.0644 0.0001 0.0952 0.0001 0.2210 0.0591 Age 2 NS NS 0.0001 0.0176 0.0005 0.0506 0.0004 0.0406 Protein 0.0017 0.9158 NS NS 0.3864 0.0443 0.6452 0.1611 Protein 2 NS NS NS NS 0.0101 0.0443 0.0165 0.1514 Protein Age 0.0007 0.0959 NS NS NS NS 0.0345 0.0056 Protein 2 Age NS NS NS NS NS NS 0.0009 0.0042 1Age in days; protein in percentage. *NS = The coefficient was not significantly different from zero at P < 0.10. parameter yields are provided in Tables 8 and 9. Due to the significance of the interactions between strain and age, and strain and protein for males, separate quadratic response models were fit to each strain (Table 10). Separate response models were also fit to each strain of females due to significant strain effect (Table 11). Carcass Weights Carcass parameter weights for male (Table 12) and female (Table 13) broilers were affected by strain, age, and dietary protein levels (Tables 8 and 9). Male Ross Ross 208 birds had heavier carcass and breast weights at each TABLE 6. The influence of protein and age on carcass parameters of male broilers by strain cross Age Protein Abdominal fat pad 1 Breast Carcass Abdominal fat pad 1 Breast Carcass (d) (%) (%) 18 23 1.12 ± 0.03 23.67 ± 0.20 62.96 ± 0.82 1.11 ± 0.02 22.80 ± 0.08 62.38 ± 0.69 16 2.08 ± 0.20 30.33 ± 0.52 67.77 ± 0.32 2.22 ± 0.16 25.11 ± 0.87 66.27 ± 0.32 18 2.04 ± 0.15 30.88 ± 0.64 69.12 ± 0.25 1.73 ± 0.17 27.33 ± 0.30 66.73 ± 0.47 20 1.51 ± 0.13 31.53 ± 0.50 68.03 ± 0.47 1.59 ± 0.15 27.71 ± 0.88 65.39 ± 0.74 32 22 1.16 ± 0.15 30.25 ± 0.46 66.72 ± 0.85 1.49 ± 0.07 27.26 ± 0.62 65.39 ± 0.33 24 1.19 ± 0.25 32.45 ± 0.09 68.25 ± 0.72 1.27 ± 0.18 28.94 ± 1.14 64.54 ± 0.85 26 1.19 ± 0.13 29.27 ± 0.96 67.34 ± 1.20 1.12 ± 0.21 28.68 ± 0.55 65.31 ± 0.70 16 2.01 ± 0.18 29.51 ± 0.81 68.54 ± 0.32 2.36 ± 0.21 27.34 ± 0.64 67.54 ± 0.96 18 2.07 ± 0.28 32.35 ± 0.97 69.32 ± 0.47 1.81 ± 0.24 27.58 ± 0.23 68.19 ± 0.73 20 1.65 ± 0.08 32.62 ± 0.97 68.85 ± 0.74 1.96 ± 0.23 29.11 ± 0.44 68.18 ± 0.50 39 22 1.42 ± 0.18 32.53 ± 0.77 69.65 ± 0.33 1.34 ± 0.12 28.97 ± 0.43 67.98 ± 0.94 24 1.14 ± 0.38 31.19 ± 0.89 68.77 ± 0.85 1.58 ± 0.18 27.70 ± 0.28 67.49 ± 0.87 26 1.51 ± 0.06 31.41 ± 0.31 69.74 ± 0.70 1.46 ± 0.29 28.29 ± 0.76 69.62 ± 0.58 16 2.55 ± 0.29 28.70 ± 0.90 73.42 ± 0.56 2.74 ± 0.11 28.15 ± 0.95 73.09 ± 0.79 18 1.93 ± 0.23 29.48 ± 1.19 70.66 ± 0.91 2.33 ± 0.25 27.98 ± 0.42 70.52 ± 1.08 20 1.81 ± 0.22 31.25 ± 0.69 72.60 ± 0.63 2.01 ± 0.19 28.43 ± 0.53 69.91 ± 0.30 46 22 1.81 ± 0.24 31.36 ± 0.56 72.99 ± 0.63 1.83 ± 0.15 27.36 ± 0.47 70.33 ± 0.84 24 1.71 ± 0.24 30.14 ± 1.15 71.29 ± 0.79 1.82 ± 0.27 29.22 ± 0.96 71.09 ± 1.45 26 1.36 ± 0.10 31.06 ± 1.12 72.91 ± 0.80 1.45 ± 0.13 28.00 ± 0.96 71.49 ± 1.20 16 2.92 ± 0.22 31.59 ± 0.74 74.13 ± 1.02 2.52 ± 0.29 28.82 ± 0.55 71.16 ± 0.51 18 2.18 ± 0.22 30.60 ± 0.57 74.09 ± 0.85 2.45 ± 0.35 29.18 ± 0.47 73.03 ± 0.91 20 1.87 ± 0.22 32.49 ± 0.89 74.52 ± 0.67 2.41 ± 0.25 30.26 ± 1.48 72.49 ± 0.67 53 22 1.96 ± 0.24 31.90 ± 0.66 73.58 ± 0.65 2.19 ± 0.23 28.90 ± 0.64 72.15 ± 0.86 24 1.80 ± 0.32 31.92 ± 0.89 73.92 ± 0.72 1.56 ± 0.12 28.83 ± 0.49 70.69 ± 0.53 26 1.42 ± 0.17 32.03 ± 0.73 73.14 ± 1.14 1.91 ± 0.20 29.48 ± 0.58 71.56 ± 0.37 1Means ± SE of three pens of two birds each per pen (except at 18 d, 6 birds per strain).

BROILER PERFORMANCE AS INFLUENCED BY PROTEIN AND SEX 1683 TABLE 7. The influence of protein and age on carcass parameters of female broilers by strain cross Age Protein Abdominal fat pad 1 Breast Carcass Abdominal fat pad 1 Breast Carcass (d) (%) (%) 16 3.13 ± 0.22 32.64 ± 0.48 75.13 ± 0.82 2.68 ± 0.30 31.03 ± 0.54 72.62 ± 0.85 53 20 2.56 ± 0.21 33.33 ± 0.45 75.49 ± 0.50 2.49 ± 0.26 30.24 ± 0.45 74.56 ± 0.77 24 2.04 ± 0.23 34.86 ± 0.72 75.32 ± 0.53 2.51 ± 0.26 31.14 ± 0.66 74.53 ± 0.58 1Means ± SE of two pens per treatment with eight chicks per pen. TABLE 8. Analysis of variance summary of the effect of age, percentage of protein, and strain on male chick performance Carcass Abdominal Breast Carcass Abdominal Breast Source 1 df weight fat pad weight yield fat pad yield P > F Strain 1 0.0001 0.8778 0.0001 0.0001 0.1375 0.0001 Age 4 0.0001 0.0001 0.0001 0.0001 0.3955 0.0001 Strain Age 4 0.0550 0.9551 0.0070 0.0537 0.7135 0.0283 Protein 5 0.0001 0.0001 0.0001 0.0319 0.0012 0.0007 Strain Protein 5 0.0288 0.4020 0.0209 0.2224 0.4606 0.3523 Age Protein 15 0.8478 0.3076 0.9553 0.0043 0.3149 0.4386 Strain Age Protein 15 0.0369 0.0178 0.0820 0.6825 0.7155 0.1539 Error (df) 244 R 2 0.9330 0.7315 0.8975 0.7995 0.2042 0.7045 TABLE 9. Analysis of variance summary of the effect of protein and strain on female chick performance Carcass Abdominal Breast Carcass Abdominal Breast Source 1 df weight fat pad weight yield fat pad yield P > F Strain 1 0.0017 0.6397 0.0001 0.0121 0.9091 0.0001 Protein 2 0.0004 0.0928 0.0001 0.1558 0.0331 0.0419 Strain Protein 2 0.2210 0.3385 0.0210 0.3469 0.1517 0.1295 Error 41 R 2 0.4489 0.1544 0.6414 0.2401 0.2172 0.5668 TABLE 10. Coefficient for the best fit multiple regression line to predict male broiler yields as functions of age and protein level Carcass yield Abdominal fat pad Breast yield Source 1 Estimate P > t Estimate P > t Estimate P > t Age 0.2900 0.0001 NS* NS 0.5250 0.0001 Age 2 NS NS NS NS 0.0053 0.0001 Protein 1.1591 0.0655 0.1626 0.0078 NS NS Protein 2 0.0260 0.0823 NS NS 0.0015 0.0898 Intercept 58.170 0.0001 2.9513 0.0001 120.17 0.0004 Age 0.3093 0.0001 0.0212 0.0001 5.2371 0.0010 Age 2 NS NS NS NS 0.0591 0.0015 Protein 0.0251 0.5568 0.0989 0.0001 3.5007 0.0238 Protein 2 NS NS NS NS 0.0503 0.0022 Protein Age NS NS NS NS 0.2644 0.0001 Protein Age 2 NS NS NS NS 0.0030 0.0003 1Age in days; protein in percentage. *NS = The coefficient was not significantly different from zero at P < 0.10.

1684 SMITH ET AL. TABLE 11. Coefficient for the best fit multiple regression line to predict female broiler yields as functions of age and protein level Carcass weight Abdominal fat pad Breast yield Source 1 Estimate P > t Estimate P > t Estimate P > t Protein NS* NS NS NS 106.76 0.0451 Protein 2 NS NS NS NS 2.759 0.0387 Intercept 133.932 0.0001 124.68 0.0001 333.01 0.0003 Protein 29.034 0.0018 2.8547 0.0073 17.152 0.0002 1Age in days; protein in percentage. *NS = The coefficient was not significantly different from zero at P < 0.10. age and protein level studied (strain by age and strain by protein interactions; Table 8). Male carcass and breast weights increased at a higher rate for each protein level as the birds aged (significant strain by age by protein interaction; Table 8). Female birds had a higher breast weight at each protein level (strain by protein interaction; Table 9). Due to the significance of interactions between strain and age, strain and protein, and strain and age and protein for males, separate quadratic response models were fit to each strain (Table 14). Separate response models were also fit to each strain of females due to the interaction of strain and protein (Table 15). DISCUSSION These results (Tables 2 to 14) with modern commercial broiler crosses are qualitatively consistent with earlier research results: Protein level, genotype, and sex have profound effects on body weight and the response to protein is dependent on genotype (significant interactions). Significant strain by protein level interactions for body weight were different from results found earlier by Proudfoot and Hulan (1978), but of course the strains were also different. As found by Nakhata and Anderson (1982), Parsons and Baker (1982), and Pesti and Fletcher (1984), the feed intake of males TABLE 12. The influence of protein and age on carcass parameters of male broilers by strain cross Abdominal Breast Carcass Abdominal Breast Carcass Age Protein fat pad 1 weight 1 weight 1 fat pad 1 weight 1 weight 1 (d) (%) (g) (kg) (g) (kg) 18 23 7.9 ± 0.17 107 ± 2.87 0.44 ± 0.02 6.9 ± 0.17 90.1 ± 4.27 0.38 ± 0.02 16 34.4 ± 3.5 350 ± 18.1 1.09 ± 0.05 33.8 ± 3.1 261 ± 7.6 0.97 ± 0.03 18 34.9 ± 3.1 370 ± 9.7 1.14 ± 0.26 27.8 ± 3.4 302 ± 15.9 1.04 ± 0.05 20 26.6 ± 3.1 388 ± 17.0 1.16 ± 0.49 26.0 ± 2.4 311 ± 18.5 1.05 ± 0.05 32 22 19.5 ± 2.3 356 ± 20.0 1.11 ± 0.45 25.7 ± 1.8 318 ± 13.3 1.09 ± 0.03 24 22.0 ± 4.5 406 ± 12.3 1.25 ± 0.30 22.0 ± 3.6 316 ± 23.1 1.08 ± 0.05 26 21.2 ± 2.9 358 ± 27.1 1.17 ± 0.74 18.6 ± 3.7 312 ± 16.5 1.05 ± 0.05 16 42.3 ± 3.8 430 ± 10.3 1.40 ± 0.03 45.2 ± 4.6 358 ± 22.9 1.25 ± 0.07 18 48.5 ± 7.1 523 ± 25.1 1.57 ± 0.05 32.8 ± 5.4 399 ± 16.1 1.39 ± 0.04 20 36.4 ± 2.9 495 ± 42.2 1.48 ± 0.09 42.8 ± 6.3 444 ± 16.7 1.43 ± 0.07 39 22 33.8 ± 4.3 551 ± 24.3 1.63 ± 0.05 28.2 ± 3.7 423 ± 32.4 1.40 ± 0.08 24 26.5 ± 7.3 528 ± 34.0 1.62 ± 0.07 35.3 ± 4.1 428 ± 14.1 1.48 ± 0.04 26 33.5 ± 1.3 501 ± 24.2 1.52 ± 0.06 31.4 ± 6.5 422 ± 20.3 1.46 ± 0.04 16 61.7 ± 8.7 508 ± 25.6 1.70 ± 0.07 70.0 ± 3.4 534 ± 24.8 1.80 ± 0.06 18 52.1 ± 8.1 563 ± 45.8 1.84 ± 0.01 61.2 ± 7.5 519 ± 16.4 1.78 ± 0.06 20 51.2 ± 6.4 654 ± 33.6 2.01 ± 0.11 51.3 ± 6.1 514 ± 19.7 1.73 ± 0.06 46 22 51.1 ± 6.0 661 ± 32.0 2.03 ± 0.09 47.4 ± 4.5 504 ± 23.0 1.77 ± 0.06 24 48.3 ± 6.6 621 ± 49.8 1.98 ± 0.09 44.5 ± 5.8 521 ± 32.85 1.71 ± 0.07 26 39.1 ± 3.0 667 ± 51.4 2.06 ± 0.12 36.4 ± 3.3 517 ± 22.76 1.76 ± 0.06 16 92.7 ± 5.4 740 ± 33.4 2.28 ± 0.06 77.6 ± 9.3 629 ± 23.73 2.11 ± 0.06 18 70.2 ± 7.4 738 ± 28.5 2.32 ± 0.06 81.4 ± 14.5 690 ± 44.13 2.29 ± 0.13 20 65.2 ± 7.8 852 ± 49.2 2.54 ± 0.09 81.1 ± 7.4 749 ± 68.61 2.38 ± 0.11 53 22 63.9 ± 8.8 830 ± 36.5 2.53 ± 0.10 65.0 ± 5.4 628 ± 25.97 2.10 ± 0.08 24 64.4 ± 12.0 850 ± 47.9 2.58 ± 0.08 46.2 ± 2.7 650 ± 24.23 2.17 ± 0.06 26 47.1 ± 6.4 777 ± 55.2 2.34 ± 0.12 60.2 ± 7.3 667 ± 28.12 2.18 ± 0.06 1Means ± SE of three pens per treatment with six chicks per pen.

BROILER PERFORMANCE AS INFLUENCED BY PROTEIN AND SEX 1685 TABLE 13. The influence of protein and age on carcass parameters of female broilers by strain cross Abdominal Breast Carcass Abdominal Breast Carcass Age Protein fat pad 1 weight 1 weight 1 fat pad 1 weight 1 weight 1 (d) (%) (g) (kg) (g) (kg) 16 79.8 ± 6.1 620 ± 18.8 1.83 ± 0.05 69.0 ± 8.7 582 ± 19.0 1.79 ± 0.05 53 20 65.5 ± 5.8 647 ± 20.4 1.86 ± 0.04 61.0 ± 7.3 552 ± 14.8 1.75 ± 0.04 24 56.9 ± 5.9 758 ± 26.3 2.06 ± 0.05 64.9 ± 6.7 611 ± 18.3 1.87 ± 0.04 1Means ± SE of two pens per treatment with eight chicks per pen. decreased as the protein level of the diet increased. Increased body weight of one genotype was indeed the result of an increase in feed intake, but feed intake was not proportional to body weight gain and resulted in lower FCR (Tables 2 to 5). males had a biphasic (increase followed by decrease) feed intake response as dietary protein increased. Birds receiving the 26% protein diet had the best FCR and lowest levels of abdominal fat. Birds (males and females) on the lower protein diets had the larger abdominal fat pads, qualitatively similar to results from Marks and Pesti (1984). As protein levels increased, abdominal fat pad weights decreased. The highly significant effect of genotype on body weight was particularly obvious at the 22 and 24% protein levels. At the 24% protein level a difference was seen in feed intake; however, larger differences were found in the FCR at the 22 and 26% protein levels. These results match results found earlier by Leclercq (1983) and Marks and Pesti (1984). The percentage carcass yield and breast yield were affected by genotype; however, only breast yield was affected by protein level. These results for carcass yield are very similar to results found by Renden et al. (1992). The maximum body weight response for Ross Ross 208 males was at the 24% protein level; whereas the nutritional requirement for maximum body weight of the males are within the 18 to 20% protein level range. males appear to have maximum carcass and breast yields at the 20% dietary protein level and males reach maximum yield in the 18 to 20% protein level range. These differences are particularly important during times of high feed prices. A severe reduction in dietary protein levels would cause a larger loss of outputs from birds and have a minimum effect on broilers. In times of high feed prices, if relatively low protein diets are fed to male broilers, then considerable potential will be lost by these birds. Sex also played a significant role in responses to dietary protein. females had significantly higher body weights and lower feed intakes at the 24% protein level than the ; TABLE 14. Coefficients for the best fit multiple regression line to predict male broiler performance as functions of age and protein level Carcass weight Abdominal fat pad Breast weight Source 1 Estimate P > t Estimate P > t Estimate P > t Intercept 1,191.7 0.0147 49.485 0.2946 558.90 0.0060 Age 56.789 0.0002 2.4517 0.0935 16.491 0.0001 Age 2 0.1516 0.1935 0.0219 0.0554 NS* NS Protein 35.277 0.0541 2.1436 0.2277 30.444 0.1130 Protein 2 NS NS NS NS 0.6805 0.1353 Protein Age 0.7242 0.0870 0.1059 0.0106 NS NS Intercept 262.03 0.2304 23.985 0.6129 429.52 0.0001 Age 29.178 0.0107 1.6273 0.2629 19.698 0.0001 Age 2 0.3240 0.0240 0.0240 0.0341 NS NS Protein NS NS 1.2792 0.4745 7.8485 0.0002 Protein Age NS NS 0.0797 0.0544 NS NS Protein 2 Age 0.0094 0.0001 NS NS NS NS 1Age in days; protein in percentage. *NS = The coefficient was not significantly different from zero at P < 0.10.

1686 SMITH ET AL. TABLE 15. Coefficients for the best fit multiple regression line to predict female broiler performance as functions of age and protein level Carcass weight Abdominal fat pad Breast weight Source 1 Estimate P > t Estimate P > t Estimate P > t Intercept 69.128 0.0001 2.5576 0.0001 30.801 0.0001 Protein 0.2389 0.0656 NS* NS NS NS Intercept 75.306 0.0001 5.3127 0.0001 28.050 0.0001 Protein NS NS 0.1366 0.0008 0.2787 0.0065 1Age in days; protein in percentage. *NS = the coefficient was not significantly different from zero at P < 0.10. therefore having significantly better FCR. Due to a lack of response by broilers to low protein levels, the greatest differences in carcass yield were found at the 16% protein level. Maximum differences in breast yield were found at the 24% protein level, due to a greater genotype response by the Ross Ross 208 birds. The best-fit equations (Tables 5, 11, and 14) can be used to predict the important output parameters from the important inputs in broiler production. Thus, if birds of a certain weight are desired, the days to produce that weight at a certain protein level could be calculated, or the protein level necessary to produce that weight in a certain number of days could be calculated. These equations will be used to construct maximum profit models based on the cost of each diet and desired slaughter weight and resulting carcass weights. Deficiencies in amino acids can cause reduced yields and performance. Amino acids need to be supplemented so that minimum levels can be reached that will maximize returns. Bilgili et al. (1992) showed dietary lysine to have a significant effect on the total deboned breast yield of broilers. Renden et al. (1994) showed lysine to have a significant effect on the yield of carcass parts. Moran and Bilgili (1990) and Acar et al. (1991) concluded that the lysine requirement may be higher than recommended by the NRC (1994). Proper essential amino acid levels must be provided for maximum economic returns. The studies cited above all helped define the best amino acid balance. We maintained amino acid minimums as a constant proportion of protein in formulating diets for this study so that there were no amino acid deficiencies (Table 1). By keeping the amino acids in a good balance, dietary protein level is the appropriate independent variable for studies of this type. Because the 16% protein diet was well balanced with respect to amino acids, adding any single amino acid would not be expected to result in a significant improvement in performance (Harper et al., 1970). To produce the improvements seen with the 16% protein vs the other diets, a balanced mixture of amino acids (protein) must be added. Because different genotypes respond differently to changes in protein levels, producers need to find the point of maximum economic efficiency for the strain of broilers they are rearing. That point could be maximum breast yield or carcass yield depending on the markets they are serving and current prices. Another important factor is feed intake: when ingredient costs increase, broiler companies have a tendency to feed less costly lower protein diets. With reduced efficiency, the cost of productivity may actually be increased. The point of maximum economic efficiency will naturally change with changes in feed prices and genetic strains. Therefore, feeding programs need to be re-evaluated whenever feed ingredient prices change, or new genotypes are utilized. REFERENCES Acar, N., E. T. Moran, Jr., and S. F. Bilgili, 1991. Live performance and carcass yield of male broilers from two commercial strain crosses receiving ratios containing lysine below and above the established requirement between six and eight weeks of age. Poultry Sci. 70: 2315 2321. Ajang, O. A., S. Prijono, and W. K. Smith, 1993. Effect of dietary protein content on growth and body composition of fast and slow feathering broiler chickens. Br. Poult. Sci. 34:73 91. Barbato, G. F., 1992. Genetic architecture of carcass composition in chickens. Poultry Sci. 71:789 798. Bilgili, S. F., E. T. Moran, and N. Acar, 1992. Strain-cross response of heavy male broilers to dietary lysine in the finisher feed. Live performance and further-processing yields. Poultry Sci. 71:850 858. Boa-Amponsem, K., E. A. Dunnington, and P. B. Siegel, 1991. Genotype, feeding regimen, and diet interactions in meat chickens. 1. Growth, organ size, and feed utilization. Poultry Sci. 70:680 688. Cahaner, A., E. A. Dunnington, D. E. Jones, J. A. Cherry, and P. B. Siegel, 1987. Evaluation of two commercial broiler male lines differing in efficiency of feed utilization. Poultry Sci. 66:1101 1110. Fraps, G. S., 1943. Relation of the protein, fat, and energy of the ration to the composition of chickens. Poultry Sci. 22: 421 424.

BROILER PERFORMANCE AS INFLUENCED BY PROTEIN AND SEX 1687 Harper, A. E., N. J. Benevenga, and R. M. Wohlhueter, 1970. Effects of ingestion of disproportionate amounts of amino acids. Physiol. Rev. 50:428 558. Holsheimer, J. P., and C. H. Veerkamp, 1992. Effects of dietary energy, protein, and lysine content on performance and yields of two strains of male broiler chicks. Poultry Sci. 71: 872 879. Hulan, H. W., F. G. Proudfoot, D. Ramey, and K. B. McRae, 1980. Influence of genotype and diet on general performance and incidence of leg abnormalities of commercial broilers reared to roaster weight. Poultry Sci. 59:748 757. Jackson, S., J. D. Summers, and S. Leeson, 1982. Effect of dietary protein and energy on broiler carcass composition and efficiency of nutrient utilization. Poultry Sci. 61: 2224 2231. Leclercq, B., 1983. The influence of dietary protein content on the performance of genetically lean or fat growing chickens. Br. Poult. Sci. 24:581 587. Leclercq, B., and G. Guy, 1991. Further investigations on protein requirement of genetically lean and fat chickens. Br. Poult. Sci. 32:789 798. Marks, H. L., and G. M. Pesti, 1984. The roles of protein level and diet form in water consumption and abdominal fat pad deposition of broilers. Poultry Sci. 63:1617 1625. Malone, G. W., G. W. Chaloupka, J. W. Merkley, and L. H. Littlefield, 1979. Evaluation of five commercial broiler crosses. 1. Grow-out performance. Poultry Sci. 58:509 515. Moran, E. T., Jr., and S. F. Bilgili, 1990. Processing losses, carcass quality and meat yield of broiler chickens receiving diets marginally deficient to adequate in lysine prior to marketing. Poultry Sci. 69:702 710. Nakhata, N., and J. O. Anderson, 1982. Describing the relation between dietary protein and energy levels and chick performance by mathematical equations. Poultry Sci. 61: 891 897. National Research Council, 1994. Nutrient Requirements of Poultry. 9th rev. ed. National Academy Press, Washington, DC. Orr, H. L., E. C. Hunt, and E. S. Snyder, 1956. A comparison of dressing percentages and yield of edible cooked meat in five strains of turkeys as broilers and mature stock. Poultry Sci. 35:333 338. Parsons, C. M., and D. H. Baker, 1982. Effect of dietary protein level and monensin on performance of chicks. Poultry Sci. 61:2083 2088. Pesti, G. M., and D. L. Fletcher, 1984. The response of male broiler chickens to diets with various protein contents during the growing and finisher phases. Br. Poult. Sci. 25: 415 423. Proudfoot, F. G., and H. W. Hulan, 1978. The interrelated effects of feeding diet combinations with different protein and energy levels to males and females of commercial genotypes. Can. J. Anim. Sci. 58:391 398. Renden, J. A., E. T. Moran, Jr., and S. A. Kincaid, 1994. Lack of interactions between dietary lysine or strain cross and photoschedule for male broiler performance and carcass yield. Poultry Sci. 73:1651 1662. Renden, J. A., S. F. Bilgili, and S. A. Kincaid, 1992. Effects of photoschedule and strain cross on broiler performance and carcass yield. Poultry Sci. 71:1417 1426. SAS Institute, 1985. SAS User s Guide: Statistics. Version 5 Edition. SAS Institute Inc., Cary, NC. Smith, E. R., and G. M. Pesti, 1998. Influence of broiler strain cross and dietary protein on the performance of broilers. Poultry Sci. 77:276 281. Summers, J. D., D. Spratt, and J. L. Atkinson, 1992. Broiler weight gain and carcass composition when fed diets varying in amino acid balance, dietary energy, and protein level. Poultry Sci. 71:263 273.