Inheritance of body weight and associated traits in young chickens

Modeling the growth of four commercial broiler genotypes reared in the tropics

Genetic improvement in commercial broilers worldwide is heavily focused on selection for higher final body weight at a given age. Although commercial broilers are mostly sold by their final body weight, it is important to carefully consider how this weight is attained and at what cost. The cost of feeding broilers, which constitutes about 70% of the total cost of broiler production, varies considerably at different stages of the bird. Careful consideration of the growth curve of broilers and the parameters of the growth curve is critical to optimize profitability of commercial broiler production. The objective of this study was to model the variations of the growth curves of 4 commercial broiler genotypes reared in Ghana using the Gompertz and polynomial growth functions. Data on body weights at 1, 7, 14, 21, 28, 35, and 42 days for 4 unsexed commercial broiler genotypes were used to model both the Gompertz and polynomial growth functions. The 4 genotypes ranked differently for Gompertz predicted early (1-28 days), late growth (28-42 days), and body weight at 42 days. Gompertz function predicted growth better for broiler chicken than the polynomial as the parameters of the Gompertz function are biologically meaningful and heritable. Selection of broiler genotypes for production based on their growth curve (slower early growth and faster late growth) could minimize cost of production and thereby increase the profitability of commercial broiler production in the tropics.

Effect of age and housing system on genetic parameters for broiler carcass traits

The effect of age and housing system on genetic parameters for BW and carcass traits was investigated. Traits were measured on broilers of different ages (48, 63, and 70 d). Birds in the 48 and 70 d groups were raised in group housing, whereas birds in the 63 d group were raised in the same housing up to 22 d and in individual cages between 22 and 63 d. Each group consisted of approximately 2,000 individuals from a single group of parents. Carcass, breast meat, abdominal fat, and back half were expressed as percentage of BW. The heritabilities of BW at 48, 63, and 70 d were 0.31, 0.26, and 0.19, respectively, and the heritabilities of back half percentage at 48, 63, and 70 d were 0.42, 0.38, and 0.21, respectively. For other carcass traits, heritabilities were in the same range in different age groups. A positive genetic correlation was found between BW and valuable parts of carcass (breast meat and back half) at 48 d; these relationships were negative at 70 d. The genetic correlation between BW and abdominal fat percentage at 70 d was higher than at 48 d. The increase in growth at 48 d was accompanied by increase in valuable parts; at 70 d it was accompanied by an increase in abdominal fat percentage. The genetic correlation of BW at 48 d between individual cage and group housing demonstrated a genotype by environment interaction for performance of birds, which has consequences for design of breeding schemes.

Correlated responses in body composition to divergent selection for exponential growth rate to 14 or 42 days of age in chickens

Chicks divergently selected for 14-d (14H and 14L) or 42-d (42H and 42L) exponential growth rate (EGR) over five generations were used to determine correlated responses between growth at different ages and body composition. Regression coefficient estimates across five generations of selection were not significant for any line at either age for percentage total body water or protein. Genetic correlations between EGR from hatching to 14 d of age (EGR14) and 42-d percentage carcass fat were -0.18, -0.57, 0.63, and -0.79 among the 14H, 14L, 42H, and 42L lines, respectively. Genetic correlations between EGR from hatching to 42 d of age (EGR42) and 42-d percentage carcass fat were 0.09, -0.67,0.50, and -0.75 among the 14H, 14L, 42H, and 42L lines, respectively. During the short-term selection experiment, selection for fast EGR14 or EGR42 increased fat at the age of selection. However, selection for fast EGR42 increased body weight and percentage fat at 42 d of age (DOA), whereas selection for fast EGR14 increased body weight but not fat at 42 DOA. Therefore, it is possible to simultaneously select for high body weight at, or near, the inflection point of the growth curve without increasing fat deposition or obesity by taking advantage of the lack of a genetic correlation between EGR14 and body fat percentage at later ages.

Symposium summary and challenges for the future

The poultry industry has grown and prospered over the past 50 yr by a repeated pattern of careful analysis of factors limiting production, followed by replacement of biological functions with management practices. Examples include assisted incubation, selection of sires, and survival via novel housing. Each resulted in a period of enhanced product output. Trends developing over the past decade raise the potential for consideration of another intervention, that of assisted reproduction. Examples illustrating the need to consider, and adopt at several levels, assisted reproduction are provided. Three critical aspects of poultry production should be monitored by careful documentation of: 1) genetic throughput from pedigree to product, best assessed by monitoring number of chicks produced per male; 2) product cost, best assessed by optimizing rate of lay and fertility of laid eggs for each hen; and 3) product quality, reflected in the homogeneity of progeny for desired traits. Each segment of the industry (turkey, egg or broiler; breeder or producer) will find unique solutions to these interacting factors. Presentations within the symposium are reviewed and integrated, and comments are provided relative to challenges facing the industry in the 21st century.

Mode of inheritance of unselected traits in lines of chickens selected for high or low antibody response to sheep red blood cells. 1. Resistance to marble spleen disease virus and juvenile body weight

Two lines of White Leghorns that had undergone long-term selection for high (HH) or low (LL) antibody response to sheep red blood cell antigen(s) formed the nuclear lines for this experiment. Matings were made in a full diallel cross to produce in a single hatch from age-contemporary breeders the parental lines, reciprocal F1 and F2 crosses, and backcrosses for 16 progeny types. For males and females, there were parental line differences in BW to 42 d of age, after which there was decline between lines for males. Differences in BW between reciprocal F1 crosses and maternal heterosis declined with age, primarily reflecting dissipation of effects of egg weight. Heterosis of BW was dependent on the particular F1 cross and recombination effects were not important. At 50 d of age chicks were inoculated with either a 1 or 10% suspension of spleen extract from chickens infected with marble spleen disease virus (MSDV). A third group served as uninjected controls. Response to MSDV was evaluated by spleen weight 6 d after inoculation. Spleen weights relative to BW of control chicks were heavier for the HH than LL line with evidence from the crosses of sexlinkage and negative heterosis. Line LL chicks were more resistant to MSDV than Line HH chicks was F1 crosses intermediate to and different from either parental line with no evidence of heterosis.

Inheritance of carcass variables when giant jungle fowl and broilers achieve a common physiological body weight

Ten chicks each from Giant Jungle Fowl (JF), broiler breeders (BB), and the respective F1 cross between them, were reared under typical broiler management practices until they reached a market weight of approximately 1,818 g. When individual birds reached market weight they were killed by CO2 asphyxiation and variables (length and weight) associated with organ, muscle, and skeletal growth were measured. Age to market weight demonstrated parental and heterotic effects. The JF had more deboned leg muscle and less Pectoralis major muscle than BB. Abdominal fat was higher in BB and showed positive heterosis. Organ weights were consistently lower for JF than BB. Positive heterosis was observed for spleen and pancreas, and negative heterosis was indicated for heart, liver, and lung. The digestive tract generally weighed less and was shorter in length for JF than for BB. Positive heterosis was observed for proventriculus and gizzard weight, and negative heterosis was noted for duodenum and jejunum lengths. Parental line differences were observed for most bone variables, including weight and length. Parental lines differed for femur breaking strength, but not for tibiotarsus breaking strength. A large negative heterotic effect was observed for tibiotarsus bone breaking strength.

Responses to long-term divergent selection for eight-week body weight in chickens

Thirty-six generations of divergent selection for body weight at 8 wk of age resulted in approximately an eightfold difference between the high (HWS) and low (LWS) lines for this trait. In both White Plymouth Rock lines, selection intensities declined over generations, with responses about five times greater during the first 18 than the last 18 generations of selection in Line HWS and about two times greater in Line LWS. Also, responses to selection were more irregular in the later than in earlier generations. Sexual dimorphism of body weight (male:female) was greater for Line LWS than HWS. A randombred control line (AC) maintained concurrently with the selected lines did not exhibit environmental trends across generations. Relaxed lines, established at different generations of selection, were used to evaluate effectiveness of selection and monitor environmental changes over short periods.

Crossbreeding effects for body weight and carcase traits in Pekin duck

1. Three sire lines of Pekin duck, three dam lines and 9 crosses between these two groups were tested in replicated tests from 1987 to 1991. Body weight at the age of 7 weeks (n = 5189), carcase percentage, percentage of abdominal fat and thickness of breast muscles were recorded (n = 1432 for carcase traits). 2. The data were analysed by linear models. All models contained the factors test and sex. When analysing only crosses, general and specific combining abilities (GCA and SCA respectively) were included as well as their interactions with the test. Using the complete data set (lines and crosses), direct genetic and heterotic effects were considered. 3. The factors test and sex were significant in both data sets for all traits. 4. For body weight, all genetic factors were significant. The average heterosis was 65 g +/- 9 g (2.2%); the highest observed heterosis was 7.7%. 5. For carcase percentage, the heterosis was always negative (up to -2.1%), whereas in abdominal fat, a desirable negative heterosis was observed only in one cross. No significant heterotic or combining ability effects were observed in the thickness of breast muscles. 6. Significant interactions between at least one genetic factor and test were found in all traits. 7. The relative importance of GCA and SCA, possible sources of confounding estimates of genetic effects, the importance of heterotic effects and genotype x environment interactions are discussed.

Growth, feed conversion, and carcass composition in females of four broiler crosses fed starter diets with different energy levels and energy to protein ratios

Body weight, feed conversion, and carcass composition were studied in females of four commercial broiler crosses that differed in weight at 7 wk of age. The birds were fed starter diets for the first 3 wk containing different energy levels but the same protein level (Experiment 1) or varying energy or protein to equilibrate energy:protein ratios (Experiment 2). Experiment 1 was conducted in floor pens; Experiment 2 in batteries. All birds received a common grower diet from 4 to 7 wk of age. No significant cross by starter diet interactions were found for body weight, feed conversion, carcass weight, or protein percentages. In the floor experiment, 3- and 7-wk body weights were lower in birds receiving the highest energy starter than in those receiving lower energy starters; in the battery experiment, body weights were unaffected by energy of the starter. Feed conversion was lower in the birds receiving the high-energy starter in both experiments. Carcass weights essentially paralleled live weights in both experiments; however, percentage of protein in carcasses of 7-wk-old broilers was unaffected by starter diets. Feeding a high-energy starter for the first 3 wk posthatch resulted in higher carcass fat percentages at 7 wk of age but lower absolute and relative abdominal fat pad weights. The significant cross by starter diet interactions indicated that the fastest growing broilers responded to the high-energy starter diet by a significant reduction in abdominal fat without an overall reduction in carcass fat.

Divergent selection for growth in Japanese quail under split and complete nutritional environments. 7. Heterosis and combining ability among diallel crosses following twenty-seven generations of selection

Growth patterns of quail lines divergently selected for 4-wk BW under split and complete nutrition environments were investigated utilizing a diallel mating scheme. The design, involving 16 mating combinations, allowed investigation of heterotic effects, reciprocal cross effects, and combining ability. Progeny from Generation 27 breeders were evaluated in two hatches under both selection diets. Heterosis for hatch weight was essentially zero; however, percentage heterosis from High x High crosses and Low x Low crosses ranged from 5 to 18% after 1 wk of age. Quail progeny from Low x High crosses were consistently larger than quail from reciprocal High x Low crosses under both selection diets. Reciprocal differences were greatest immediately posthatch and declined with age. However, in crosses involving males from High and Low split diet lines mated to females from High and Low complete diet lines, large reciprocal differences in BW remained at 8 wk. Mean heterosis values for BW across ages from crossing High and Low lines both within and across selection environments were negative in six of eight comparisons. Mean values ranged from +3 to -11% and indicated that greater selection responses may have been made in divergent selection for low 4-wk BW than for high BW. General combining ability of lines was greater when transmitted via females as opposed to males, and was similar under the two diets (split and complete). There was evidence that high-BW lines exhibited greater general combining ability under both dietary environments than did low-BW lines.

Divergent selection for exponential growth rate at fourteen or forty-two days of age. 1. Early responses

Athens-Canadian randombred chickens were divergently selected for exponential growth rate (EGR) to 14 or 42 days of age, resulting in four unique populations: 14L, 14H, 42L, and 42H. Exponential growth rate was calculated as: EGR = [ln b-ln a] + t; where a = hatching weight; b = weight at age of selection; and t = age of selection. In the S4 generation, 14H chicks had significantly greater EGR than 42H chicks at 14 but not 42 days. Chicks from the 42H line had significantly greater EGR than 14H chicks from 14 to 42 days. This meant that 14H chicks were significantly heavier than 42H chicks at 14 but not 42 days. Indeed, 14H chicks were consistently 5 to 10% heavier than the 42H chicks in each generation. Selection for EGR at 14 days was at least as effective in altering 42-day body weight as selection for growth rate at that age during the early generations of selection. Evidence of asymmetry existed for selection for EGR at 14 but not 42 days. Realized heritability of fast EGR14 was .29 +/- .08 (SE) whereas that of fast EGR42 was .15 +/- .09. Genetic correlation between fast EGR14 and fast EGR42 was 1.45, whereas the converse genetic correlation was .67. This would suggest that all the genes that influence 14-day growth influence 42-day growth, but not vice versa. Growth curve analysis supported the general hypothesis of growth, in that instantaneous growth rate was fastest in the 14H line, as was the decay of growth. Indeed, 14H birds were smaller at and beyond sexual maturity than 42H chickens.

Genetic architecture of carcass composition in chickens

A cross-sectional diallel experiment was used to estimate the contribution of various types of genetic variation for the chemical components of the chicken carcass. Populations tested included a commercially developed sire line (CM), a fertility-selected broiler line (FS), and the pureline Jersey Giant (JG). Average line effects were significant for body weight and total carcass water, fat, and protein. Line effects were significant only for percentage fat and water. Maternal effects and additive sex linkage were not significant for any carcass component on an absolute or percentage basis. Significant average heterosis and sex by heterosis interactions were observed for abdominal fat pad weight and total carcass fat. Male hybrids exhibited a 29% decrease in abdominal fat pad size, but female hybrids exhibited a 7% increase. Significant heterosis was observed among females, but not males, for percentage carcass fat. Changes in carcass fat via hybridization produced the largest percentage change among males due to heterosis of the sex chromosomes. In the present experiment, negative heterosis for abdominal fat among male progeny accounted for approximately 65% of the overall reduction in fat content of the carcass.

Correlated responses in lines of chickens divergently selected for fifty-six-day body weight. 3. Digestive enzymes

Levels of amylase, trypsin, chymotrypsin, and lipase in the pancreas and small intestinal chyme were measured in females from four lines of chickens. Two of the lines had undergone 32 generations of divergent selection for 56-day body weight, and in the other two lines selection for high or low weight had been relaxed for 5 generations. The diet used in the present experiment was that under which selection had been practiced (20% crude protein and 2,685 kcal of ME/kg). Comparisons between divergently selected lines at common ages revealed higher enzyme levels for high- than low-weight lines. When comparisons were made at a common body weight (80 +/- 5 g) there were no differences between lines. These results suggested that correlated responses in feed intake were mediating the regulation of digestive enzyme levels in the pancreas and in intestinal chyme of growth-selected lines of chickens. Chicks from high-weight lines had elevated enzyme levels after a mild feed restriction compared with those provided ad libitum access to feed. It was hypothesized that hyperphagia associated with the high-weight lines in combination with a mild feed restriction and the associated meal feeding stimulated synthesis and secretion of digestive enzymes.

Correlated responses in lines of chickens divergently selected for fifty-six-day body weight. 1. Growth, feed intake, and feed utilization

Correlated responses for growth, feed intake, and feed utilization were measured in female chickens from lines after 32 generations of divergent selection for 56-day body weight and in sublines in which selection had been relaxed for five generations. The diet used in the present experiment was that under which selection had been practiced (20% crude protein and 2,685 kcal of ME/kg). Feed intake between each selected line and its relaxed subline was equalized by paired-feeding. Ad libitum-fed controls were also maintained. At 21 days of age, differences between the selected lines for ad libitum-fed chickens were 404% for body weight, 279% for feed intake, and 138% for feed conversion ratio (FCR). Respective percentages for the high and low relaxed lines under ad libitum feeding were 267, 223, and 121%. When pair-fed, the growth of the line selected for high weight was unimpeded and FCR improved. When chicks of the low-weight relaxed line were pair-fed, FCR also improved, but growth was reduced and appetite development inhibited. Hyperphagic behavior was observed in the line selected for high weight. In the low-weight selected line, the chicks exhibited hypophagia. Although residual feed consumption was unchanged by selection for 56-day body weight, variation was present for this trait, with lower values under restricted than ad libitum feeding.

Inheritance of antibody response to sheep erythrocytes in lines of chickens divergently selected for fifty-six-day body weight and their crosses

Experiments were conducted to determine the principal mode of inheritance for antibody response to SRBC in selected populations of chickens and their crosses. Additive genetic variation was important in the inheritance of both primary and secondary responses to this antigen. Reciprocal differences and heterosis also influenced the secondary response. Kinetics of primary and secondary antibody responses were evaluated in pairs of lines divergently selected for high (HW) or low (LW) juvenile body weight and high (HA) or low (LA) response to SRBC antigen. Different groups of chickens were sampled on Days 3 to 7, 10, 14, 24, 26, 28, and 31 postprimary inoculation (PPI). Half of the chickens sampled after Day 14 received an additional injection of SRBC antigen. Plasma from each individual was examined for total, mercaptoethanol-resistant, and mercaptoethanol-sensitive titers. Antibody titers PPI to SRBC were consistently highest in Line HA and lowest in Line LA. Lines HW and LW reacted similarly to Line LA early in response, but showed higher levels later. In both primary and secondary responses, the weight lines peaked at similar levels. Thereafter, Line LW maintained a higher antibody titer level to SRBC antigen than Line HW.

Genetic architecture of growth curve parameters in chickens

Genetic improvement in growth of poultry has traditionally proceeded via selection for body weight at a fixed age. Due to increased maintenance costs and reproductive problems of adult broiler breeders, the potential for genetic manipulation of the growth curve has been receiving increased interest. Research of both male and female progeny of a three-way diallel cross was used to investigate the inheritance of growth curve parameters. The Laird form of the Gompertz equation was used to determine growth curve parameters, and was suited to the juvenile growth data frequently collected from meat-type chickens. Growth rate exhibited significant heterosis due to both autosomes and the sex chromosomes. Age at inflection point also exhibited significant average heterosis, though only among females. Growth rate was also influenced by average line effects, as was age at inflection point. Maternal effects had no influence on growth curve parameters, while additive sex linkage was observed for growth rate. Phenotypic and genetic correlations were calculated among the growth curve parameters and suggest that specific breeding programs could alter the growth trajectory of the contemporary broiler chicken. Moderate heritabilities were observed for the growth curve parameters and support the hypothesis that the growth curve could be altered via genetic manipulation of early postnatal growth, especially during the first 14 days post-hatch.

Genotype, feeding regimen, and diet interactions in meat chickens. 1. Growth, organ size, and feed utilization

Allomorphic growth of organs and feed utilization were measured in cockerels from two meat lines known to differ in growth potential under alternate-day and daily feeding of diets differing in nutrient density. The faster growing line consumed more feed and utilized feed more efficiently than the slower growing line. Relative weights of heart and all segments of the gastrointestinal tract except the esophagus were smaller in the faster than in the slower growing line whereas breast, thigh, and drumstick were greater for the faster than the slower growing line. These patterns existed regardless of diet. In contrast, line by feeding regimen interactions were noted for most traits; the faster growing line was influenced more by restricted feeding than the slower growing line. Thus, the practice of selecting individuals in a restricted-feeding environment when commercial stocks have feed available to them at all times may be faulty. Although the alternate-day feeding regimen improved feed efficiency, results indicated that full benefit of feed restriction would be realized by feeding a diet higher in nutrient density.

Sex-linked and maternal effects on growth in chickens

A study was designed to estimate the contribution of various types of genetic variation in growth of chickens using populations that widely differed in growth rate. Populations tested included a commercially developed sire line (CM), a fertility-selected, broiler-type line (FS), and the pureline Jersey Giant (JG). A full set of F1 crosses and purelines of both sexes was used to estimate the following genetic effects: heterosis due to the autosomes, heterosis due to the sex chromosomes, average line effects, maternal effects, and additive sex linkage. Average line effects were highly significant beyond hatching for both body weight and shank length, with rank-order CM greater than FS, which was greater than JG. Significant differences between the sexes within a line were observed for CM and JG line effects, confirming the existence of sex-linked factors. There was no average heterosis or sex-by-heterosis interaction for growth-related traits, but specific heterosis was frequently significant. Maternal effects were sporadic and generally of little importance beyond hatching. Additive sex linkage, however, was highly significant and influenced body weight beyond hatching. These data indicate that the primary advantages in increasing growth via cross-breeding in chickens comes from average line effects and additive sex linkage. Heterosis and maternal effects can influence performance, especially in specific crosses, but these effects appear to be sporadic and of less importance than sex linkage.

Effects of monosodium glutamate on lines of chickens having different juvenile exponential growth rates

MSG (4 mg/g BW) or the equivalent molar ratio of NaCl were injected subcutaneously into chicks from four populations of chickens derived from a double selection experiment [(1) low exponential growth rate (EGR) to 14 days of age (14L), (2) high EGR to 14 days of age (14H), (3) low EGR to 42 days of age (42L) and high EGR to 42 days of age (42H)]. MSG significantly influenced growth in chickens, but the absolute effect was genotype and sex dependent. Weights of abdominal fat pads were ranked among lines according to the sequence 42H, 14H, 42L and 14L. MSG significantly increased fat pad size by 51%, independent of genotype or sex. MSG significantly decreased breast weight in females and males by 7%. The higher percent fat pads and lower breast weights associated with MSG are indicative of growth due to fat accretion rather than protein deposition.

Growth-associated traits in parental and F1 populations of chickens under different feeding programs. 3. Eating behavior and body temperatures

In two experiments, parental populations of chickens and their F1 crosses were fed ad libitum or on alternate days. Eating and drinking behaviors, ability to compensate for fasting, and surface and cloacal temperatures were measured. Genetic and environmental (due to fasting) differences in feeding and drinking behaviors were found. Chicks from lines selected for larger body weight were better able to compensate for a 24-hr fast than those selected for smaller weight. Crosses were similar to the parental line that was the better compensator. Correlations between surface temperatures and body weights were consistently positive.

Growth-associated traits in parental and F1 populations of chickens under different feeding programs. 1. Ad libitum feeding

Two experiments were conducted to compare body composition, plasma concentrations of glucose, lipid and protein, gastrointestinal tract (GIT) morphology, and digestive enzyme activities among populations of chickens exhibiting wide differences in growth. In the first experiment, males from a high weight (HW) line, a low weight (LW) line, and crosses between the lines (HL) were examined. In the second experiment, the same high weight (HW') line, a White Leghorn (LEG) line and their crosses (HLEG) were compared. In both experiments, comparisons were made at both 25 and 61 days of age. The HW chicks were heavier and had larger breasts and longer shanks than the LW or LEG chicks. Although the crosses (HL, HLEG) exhibited values for these traits intermediate to those of the parental lines, HLEG chicks tended to resemble one of the parental lines more closely than did the HL chicks. Differences among all mating combinations were significant for body fat, but the crosses (HLEG, HL) tended to more closely resemble their HW parental line. The HW chicks had significantly higher concentrations of glucose, lipid, and protein in plasma at 25 but not 61 days of age. No differences in glucose, lipid, or protein concentrations in plasma were observed in the second experiment. When expressed as a percentage of body weight, LW and LEG chicks tended to have larger GIT than HW chicks, whereas crosses more closely resembled the HW line. Although significant differences in digestive enzyme activities were obtained, these differences were generally associated with differences in body weight. It was concluded that genetic variation in the traits examined was dependent on both population and age.

Thermoregulation in newly hatched chicks

Chicks from seven early-feathering populations were used in a series of experiments to detail the course of thermoregulation in newly hatched chicks. Lines HN and LN were composed of White Plymouth Rock chickens that had been selected 24 generations for high and low 8-week body weight, respectively. Reciprocal crosses of those lines (HL and LH) and an F2 produced from HL as well as lines originating from HN and LN with the sex-linked dwarf allele (dw), HD and LD, respectively, were also used. Surface (foot pad) temperature, core (rectal) temperature, and body weight were obtained on female and male chicks at 2, 4, 6, 8, 11, 14, 17, 20, and 23 days in all populations and at 105 days of age in HN and LN lines. Although LN and LD chicks generally had lower surface temperatures than those from the other lines from 11 to 23 days of age, surface temperatures of HN and LN birds were not different at 105 days. No significant differences due to the dwarf allele were found in surface temperatures, but there was a consistent tendency for dwarfs to be cooler than normals in the high lines and warmer than normals in the low lines. The LN chicks had lower core temperatures through 23 days of age than the other populations measured, but the narrow range of core temperatures in all populations precluded a definitive analysis to reveal mode(s) of inheritance. Surface-to-core temperature ratios were also lower in LD and particularly LN chicks through 23 days of age, although no difference between HN and LN birds occurred in this trait at 105 days of age.

Selection for body weight at eight weeks of age. 17. Overfeeding

Female chicks from lines selected for high and low body weight and their reciprocal F1 crosses were overfed via force feeding to crop capacity from hatching to 21 days of age. Low-weight line chicks could be overfed at an earlier age to a greater extent than those from the high-weight line and F1 progeny. The degree of overfeeding appeared to be associated with the relative size of certain gastrointestinal components. Differences among lines in their capacity to be overfed were reflected in concomitant changes in body weight and carcass fat. Relationships among feed consumption and growth are discussed within the context of selection for body weight and heterosis under various feeding regimens.

Selection for body weight at eight weeks of age. 16. Restriction of feed and water

The consumption of feed or water of chickens selected for high body weight was restricted to that of a line selected for low body weight. In addition, reciprocal crosses of the two lines underwent feed or water restriction to that of the low body weight parental line. Restriction of water depressed feed intake to a greater degree in the high line than in the crosses but depressed body weight by approximately 10% in all groups through 28 days of age. Feed restriction caused a greater depression in body weight in the high line than in the crosses at 28 days of age. The effect of feed or water restriction on the feed:water ratio was dependent upon mating type, suggesting the absence of a simple relationship among growth, body composition, and feed efficiency.