Performance of Two Male Broiler Breeder Strains Raised and Maintained on Various Constant Photoschedules1

Performance of Two Male Broiler Breeder Strains Raised and Maintained on Various Constant Photoschedules 1 J. A. RENDEN,2 S. S. OATES,2 and M. S. WEST3 Departments of Poultry Science and Research Data Analysis, Alabama Agricultural Experiment Station, Auburn University, Alabama 36849-5416 (Received for publication December 17, 1990)

1991 Poultry Science 70:1602-1609 INTRODUCTION

Laying hens are generally provided 14 to 15 h of light/day (Harrison et al, 1969), although as little as 9 to 13 h of light has been suitable for egg production (McCluskey and Parker, 1963). In most cases, roosters are housed with hens and receive similar photoschedules. Lamoreux (1943) reported that photostimulation of mature roosters occurred within 4 wk and semen production required 9 to 12 h of light/day (1 h of light was insufficient). The light intensity requirement was approximately 2 lx, and selection for fecundity decreased light requirements. Parker and McCluskey (1964,1965) found that Delaware males raised and maintained on either 1, 3, 9, or 13 h of light/day did not differ in semen production or fertility. The mortality rate was higher and sexual maturity was delayed in birds on 1 and 3 h of light compared with longer light periods; all males were fertile by 25 wk of age.

'Alabama Agricultural Experiment Station Number 12-902797P. department of Poultry Science. department of Research Data Analysis.

These authors concluded that semen production was not adversely affected in males reared and maintained on short artificial photoperiods and that a photoschedule of 3 h of light (L):21 h of darkness (D) would be sufficient for rooster fertility. Reduction of the photoperiod (from 14 to 6 to 8 h) during puberty (11 to 14 wk of age) decreases testes growth and differentiation of germ cells and delays sexual maturation of Single Comb White Leghorn cockerels (Ingkasuwan and Ogasawara, 1966; Harrison et al., 1970). However, males with delayed sexual maturity maintained on 8L:16D ultimately have larger mature testes and increased and extended spermatozoa output compared with males kept on 14 to 16 h of light/day. Proudfoot (1980) placed meat-type roosters on either 2, 6, 14, or 15.5 h of light at 18 wk of age. There were no significant effects on BW or semen volume. It was suggested that 6 h of light would be adequate for semen production. Reduction of light after sexual maturity in roosters results in increased abnormal spermatozoa and decreased live spermatozoa and semen production (Lamoreux, 1943; Bajpai, 1963).

1602

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

ABSTRACT The purpose of the present study was to examine the interaction of constant photoschedules and genetic background on performance of male broiler breeders. Day-old cockerels from two BW strains were placed on litter floors in light-controlled chambers. Light treatments (LT) (60 lx) consisted of 2,4,8,16, and 24 h light/day. At 9 wk of age, birds were individually caged and evaluated biweekly for semen production. Venous blood samples were collected at 16,32,48, and 64 wk of age. Data for testes weight, histology, and morphometry were obtained at 64 wk. Age at first semen production showed a cubic response in the levels of LT with earliest semen production from 4 and 8 h light (187.0 and 188.2 days, respectively). The BW was linear in the levels of LT within week; average BW was generally greater for birds on short LT than for birds on longer LT. Semen concentration was also linear in the levels of LT within week; increased semen concentration occurred with short LT. Changes in semen weight and spermatozoa count per ejaculate across the levels of LT differed for strain. A larger percentage of males produced semen in the
PHOTOSCHEDULES FOR ROOSTERS

MATERIALS AND METHODS

Day-old cockerels of two strains from the same primary breeder4 differing in BW (43.7 versus 40.0 g for Strain 1 and Strain 2, respectively) were wing-banded and brooded on litter floors in light-controlled chambers (2.4 x 3.2 x 2.3 m). Light treatments (LT) (60 be5 at bird height) consisted of either 2, 4, 8, 16, or 24 h light/day, with lights on at 1300 h for all treatments. The LT were initiated at the first day of age and continued through 64 wk of age. Each LT was replicated with three chambers, and each chamber initially contained 10 birds of each strain. At 9 wk of age, six birds of each strain were randomly selected and placed into individual cages (32 x 53 x 61 cm) within the chambers, and remaining males were removed. Birds were full-fed during their 1st wk of age and thereafter provided feed according to the breeder's recommendations. Both strains in all LT groups received the same amount of

''Peterson Industries Inc., Decatur, AR 72722. ^Minolta Illuminance Meter, Model 1519-206, Ramsey, NJ 07446.

feed, adjusted to obtain suggested BW of the heavier BW strain (1) in the 2-h LT group. This was done to standardize feed intakes of all LT groups and minimize confounding of feed intake and LT effects. The rations were formulated to meet requirements of the National Research Council (1984) and were as follows: 0 to 3 wk, starter ration (22.6% CP, 3,146 kcal ME/kg feed); 4 to 8 wk, Grower 1 ration (19.9% CP, 3,190 kcal ME/kg feed); 9 to 21 wk, Grower 2 ration (16.0% CP, 3,012 kcal ME/kg feed); 22 to 64 wk, breeder ration (15.5% CP, 2,878 kcal ME/kg feed). Chicks were vaccinated for Marek's disease at hatching. Newcastle and bronchitis vaccines and Gumboro's disease vaccines were given at 14 and 18 days of age, respectively. All birds were weighed weekly through 28 wk of age and at 28-day intervals thereafter. After caging (9 wk of age), males were massaged twice weekly for semen production. Semen weight (1 g = 1 mL; Brillard and de Reviers, 1981), concentration, and spermatozoa intactness were measured (Bilgili and Renden, 1984) at first ejaculation (age at sexual maturity) and at 14-day intervals thereafter. Fertility was tested at 31, 47, and 63 wk of age. For each chamber, semen was pooled within strains and insemination dose was adjusted to provide 100 x 106 intact spermatozoa. Three hens were inseminated per LT and strain replicate. Eggs were collected for 21 days beginning the 2nd day after the single insemination, incubated for 7 days, and examined for true fertility. At 16, 32, 48, and 64 wk of age, venous blood samples were collected from all birds. Blood sampling was done during 1300 to 1500 h. Within 1 min of handling, 2 to 3 mL of blood was collected into EDTA-coated syringes using 21-gauge needles. Blood samples were kept on ice until removal of plasma by centrifugation (1,000 x g for 5 min at 2 C). Plasma samples were frozen (-20 C) for later analysis of testosterone. At the termination of the experiment (64 wk), the males were killed by electrocution. The testes were removed, weighed, and fixed in neutral-buffered formalin (10% solution). Left testes from three birds per strain per chamber were randomly chosen for examination. Tissue blocks (1 cm2) were processed with standard histological techniques, sectioned at 6 Jim, and stained with hematoxylin and eosin (Humason, 1972). Seminiferous

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

De Reviers (1974, 1975, 1977) and De Reviers and Williams (1981) performed a series of experiments examining photoschedules for roosters. Raising and maintaining males under short (8L:16D) or decreasing photoperiods decreased rate of rapid testes growth during puberty, delayed onset of rapid testes growth and sexual maturity by 4 to 8 wk, increased and supported final adult testes weight, and increased and sustained semen production throughout the production period to 51 wk of age. Highest daily spermatozoa output was found with constant photoperiods compared with intermittent photoschedules (e.g., 8L:16D versus 7L:8D:1L:8D). Other work with chickens has suggested that there may be interactions between body weight or genetic strains with photoschedules (de Reviers, 1980; Harris et al., 1984). The purpose of the present experiment was to examine the utility of short constant photoschedules for broiler breeder males. The influence of genetic background on reproductive responses to light treatment was studied by using two strains of broiler breeder males differing in BW.

1603

1604

RENDEN ET AL. TABLE 1. Scoring system for seminiferous epithelium1

Score Criteria 1 2 3 4 5 6 7 8 9 10

Spermatogonia present with some cell division. Spermatogonia, primary and secondary spermatocytes present. Same as number 2 plus round and maturing spermatids. Same as number 3 plus many maturing spermatids. Same as number 4 plus few spermatozoa in lumen. Same as number 4 plus many spermatozoa in lumen. Same as number 6 plus some degree of seminiferous epithelium degeneration and slight variolation. Pyknotic (condensed and reduced) nuclei, variolation, moderate loss of epithelium. Marked loss of general epithelium and presence of few spermatozoa. Marked loss of epithelium with only spermatogonia and Sertoli cells present 'Adapted from Basurto-Kuba et al. (1984) and Wilson et al. (1988).

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

tubules were scored for functional activity on a for plasma. Spiking recovery was determined by 10-point scale (Table 1) adapted from Basurto- adding 1.6,3.2, and 6.4 ng testosterone propioKuba et al. (1984) and Wilson et al. (1988). nate10/mL to five plasma samples, resulting in Morphometric measurements were obtained by recovery rates of 92,92, and 85%, respectively. digitizing the cross-sectional image of a tubule Parallelism was demonstrated by diluting aliwith the use of a video digitizing board6 quots of a sample 2, 3, and 4 times and interfaced to a microcomputer7 and displaying comparing the slope of the line for the diluted the image on a visual monitor. The digitized sample to that of the standard curve (concentraimage was then traced and measurements tion range of .0625 to 16.0 ng testosterone computed with JAVA software.8 Measure- propionate/mL of charcoal-stripped male ments were made on five nearly round tubules chicken plasma); slopes for the diluted sample per slide. Roundness, required for calculating and standard curve were -1.67 and -1.72, epithelial area, was estimated by comparing respectively. Within assay CV for high (3.0 ng/ the longest diameter of the tubule to its mL) and low (.71 ng/mL) pooled plasma perpendicular diameter. Tubules having di- standards were 3.8 and 7.3%, respectively, and ameters within 85% of each other were between-assay CV for the two pools were 3.0 accepted as nearly round. The tubule area was and 5.0%, respectively. The sensitivity of the defined as the area enclosed by outlining the assay (concentration at 2 SD from counts per basement membrane of the seminiferous epi- minute at zero binding) was .06 ng/mL. thelium. The luminal area was the area enclosed by outlining the luminal edge of the Statistical Analyses seminiferous epithelium. The seminiferous epithelial area was calculated by subtracting Data for BW, semen characteristics, fertility, luminal area from tubule area (Basurto-Kuba et and plasma testosterone were analyzed by repeated measures multivariate ANOVA with al, 1984). LT, strain, and age as main effects (Gill, 1981; Looney and Stanley, 1989). Curvilinear LT Radioimmunoassay effects included in the overall model were of Testosterone calculated using orthogonal polynomials with 9 A commercial kit (Number TKTT5) adapted unequally spaced levels. When appropriate, and validated by Sexton et al. (1989) for serum means were compared with Tukey's test Data testosterone of male chickens was revalidated for age at sexual maturity, and testes weights and morphometry were analyzed by ANOVA with LT and strain as main effects. Categorical data for mortality rates and serniniferous tubule Truevision, Inc., Indianapolis, IN 46256. scores were analyzed with chi-square. Signifi'Zenith Data Systems PC, Chicago, IL 60600. cance was accepted at K.05. "Jandel Scientific, Corte Madera, CA 94925. Data for the percentage of males producing ^Diagnostic Products Corp., Los Angeles, CA 90045. 10,Sigma Chemical Co., St Louis, MO 63178-9916. semen were analyzed as follows. For each week

1605

PHOTOSCHEDULES FOR ROOSTERS 300

250 200

!

150

Body

3

u

-- -&-- " — -fl---§L__

--..

100

r\

50

-fr-t": 2

4

6

8

10

12

14

16

18

20

22

24

Light treatment (h)

2

and for each strain, the proportion of birds in each chamber that showed eversion with semen was calculated. The resulting proportion was used as the response. This resulted in 30 responses per week for the 15 chambers and two strains. The data covered Weeks 12 through 63. Treatment averages were plotted over this period and inspected for trends and suggested treatment effects. The scatterplots of the average proportion of birds producing semen appeared to be easily explained by a low-order polynomial. It was also noticed that the curves were consistently higher for short LT (2, 4, and 8 h) than for long LT (16 and 24 h). For each chamber, polynomial regressions were then fit to explain the relation between the average proportion of birds found with semen and age for each strain. Two models were considered: Mt = Bj age + B 2 age 2 + B 3 age3

[1]

ut = Bj age2 + B 2 age3

[2]

where «t represents the true proportion of birds observed with semen at time t. In both models, the intercept was omitted based on me consideration that at time t = 0, ut the average proportion should be 0. After inspecting the 30 regressions obtained for each model, it was observed that the linear term in Model [1] was not significant for the majority of the regressions, whereas the quadratic and cubic terms were significant in all. hi view of the latter observation, Model [2] was

6

8

10

12

14

16

18

20

22

24

Light treatment (h)

FIGURE 2. Response of BW for broiler breeder males at 14 wk (•; y = 1.91 - .02x + .0004x2, R2 = .95); 16 wk (+; y = 2.12 - .03x + .OOlx2, R2 = .96); 24 wk (A; y = 3.29 - .06x + .002x2 R2 = .96); 32 wk (•), 40 wk (O; y = 5.41 - .09x + .003x2, R2 = .84); 48 wk (+), 56 wk (o), and 64 wk of age (V; y = 5.61 - .09x + .003x2, R2 = .77).

chosen. The coefficient of the resulting regressions from fitting Model [2] were then analyzed using a multivariate analysis (Morrison, 1967). RESULTS

Mortality Rates There were no differences in mortality rate between LT groups or strains. There were no deaths until 10 wk of age, and after this time losses occurred randomly during the experiment. Mortality rates for the 2,4, 8,16, and 24 LT groups was 8.3, 11.1, 2.8, 11.1, and 5.6%, respectively. Body Weight At 1 wk of age, there was a linear LT effect on B W with the 2-h LT group having smallest mean weight; a difference between strains existed (Figure 1). At 2 wk of age, there was a quadratic LT effect with the 4-h LT group having smallest mean weights; differences between strains still existed. The BW from 14 to 64 wk is shown in Figure 2. Body weight increased with age, the difference between strains was absent, and there was an overall age by linear LT interaction. However, within the ages of 14,16,24,40, and 64 wk, the LT effect showed quadratic responses with smallest BW in the 16-h LT group from 14 to 24 wk and in the 8-h LT group thereafter.

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

FIGURE 1. Response of BW of Strain 1 and 2 broiler breeder males at 1 and 2 wk of age. Both strains showed a positive linear relationship between BW and hours of light per day at 1 wk [Strain 1 (•): y = 89.20 + .74x, R2 = .97; Strain 2 (0:y = 78.77 + .83x, R* = .96]. Both strains showed a quadratic response to hours of light per day at 2 wk [Strain 1 (0): y = 245.72 - .56x + .07x2, R2 = .95; Strain 2 (o): y = 228.27 - 1.3lx + .10x2, R2 = .98].

4

1606

RENDEN ET AL. TABLE 2. Plasma testosterone concentration of broiler breeder males

Semen Characteristics

Testosterone (ngAnL)

Age (wk)

.13D

16 32 48 64

1.68" 1.57a 1.42° SE = .129

"•"Means within the column without common superscripts were different (K.05).

than in the 16- and 24-h LT groups (Figure 6). There was no difference between strains nor an interaction of LT by strain. The percentage of males producing semen in the various LT showed a cubic response over time. There were neither differences among main effects nor interactions for percent intact spermatozoa or fertility (data not shown). Plasma Testosterone The effect of strain was not significant. There was an overall age effect with lowest testosterone level occurring at 16 wk compared with the other ages (Table 2). There was a positive linear relationship between hours of light and plasma testosterone across ages (i.e., highest testosterone concentration occurred in the 16- and 24-LT groups) (Figure 7).

8

10

12

14

16

18

20

22

24

Light treatment (h) 8

10

12

14

16

18

20

22

24

Light treatment (h)

FIGURE 3. Response of age at sexual maturity for broiler breeder males to hours of light (y = 208.66-7.80x+ .81x2 - .02x3, R 2 = .96).

FIGURE 4. Semen concentration (spermatozoa x 109 per milliliter of semen) for broiler breeder males provided various hours of light during 25 to 32 wk(+); 33 to40wk(A: y=6.81-.05x,R 2 =.80);41to48wk(»;y=7.27-.07x,R 2 =.88); 49 to56wk(O;y = 7.50-.10x,R* = .77); and 57 to 64 wk of age (V).

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

Age at sexual maturity was not different between strains. Age at sexual maturity showed a cubic response to LT, with earliest semen produced with 4 and 8 h LT (Figure 3). Data for semen concentration are shown in Figure 4. Differences between strains did not exist for concentration. From 33 to 64 wk of age, semen concentration was increased with decreasing hours of light. Semen concentration was linear in the levels of LT, but the degree of linearity changed with time. During 41 to 56 wk of age, a stronger negative association between semen concentration and LT existed than at the other ages. There were differences among ages for concentration within LT 2,4, and 8 h. For LT 2 h, semen concentration was greater during 33 to 56 wk compared with 25 to 32 wk of age. For LT 4 h, semen concentration was greater during 33 to 64 wk compared with 25 to 32 wk. For LT 8 h, semen concentration was greater during 49 to 56 wk than from 41 to 48 wk, which was greater than 25 to 32 wk of age. There were no age differences in semen concentration for LT 16 and 24 h. There was an interaction effect between LT and strain on semen weight and number of spermatozoa (concentration by weight; 1 g equivalent to 1 mL) per ejaculate. Strain 1 males showed greatest semen weight and spermatozoa number per ejaculate with 4 and 8 h LT, and Strain 2 males showed greatest weight and number with 2 and 24 h LT (Figure 5a and b). There were no age effects on semen weight or spermatozoa number. A greater percentage of males produced semen per ejaculation in the 2- to 8-h LT groups

1607

PHOTOSCHEDULES FOR ROOSTERS

Testes Characteristics

// // 13-/-.

.15

•

V- -•

.10

10

12

14

16

18

20

22

24

22

24

// :

16

: / • •

20

24

28

32

36 40 44 Age (wk)

48

52

56

60

FIGURE 6. Percentage of broiler breeder males producing semen provided 2 ( ), 4 ( ), 8 ( ), 16 (— ), and 24 h light ( ).

DISCUSSION

Maximum growth rates in broilers occur with constant, continuous light, and the effects of light on growth result from its influence on activity patterns and intake of feed (Morris, 1967). The BW of males in the present study were positively related to the total hours of light at 1 wk of age. The quadratic response of BW to light duration at 2 wk of age may have been associated with initiation of feed restriction at 1 wk of age. Chicks on 6 h or less of light/day will learn to eat during the dark period and use feed more efficiently (Morris, 1967). Variations in BW among the treatments at later ages may have been related to activity patterns as all treatments were fed the same amounts of feed. Morris (1967) reported that a cubic curve best described the relationship between age at

Light treatment (h)

8

10

12

14

16

18

20

Light treatment (h)

FIGURE 5. Semen weight (a) and number of spermatozoa per ejaculate (b) for Strain 1 (O) and 2 ( • ) broiler breeder males provided various hours of light

6

8

10

12

14

16

18

20

22

24

Light treatment (h)

FIGURE 7. Linear response of plasma testosterone for broiler breeder males to hours of light per day (y = .97 + .02x, R 2 = .47).

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

There was a cubic relationship between testes weight per 100 g BW and hours of light per day (i.e., testes weights were decreased in the 16- and 24-h LT groups compared with the shorter LT groups) (Figure 8). Distribution of seminiferous tubule scores were not affected by LT. All of the seminiferous tubules from testes of the 4- and 8-h LT groups received a score of 8. In the 2-h LT group, 22% of the testes received a score of 7, and the remaining testes were scored as 8. In the 16-h LT group, 6% of the testes were scored as 2,18% scored as 5, and the remaining testes scored as 8. In the 24-h LT group, 6% of testes scored as 2, 6% scored as 7, and the remaining testes scored as 8. There were no significant differences in seminiferous epithelial area or tubule diameter between strains or among LT. The range of seminiferous epithelial area for the straintreatment combinations was 22,796.89 to 28,750.00 um2, and the tubule diameter range was 273.31 to 319.35 urn.

1608

RENDEN ET AL.

.80 r

! .60

:

.40 '

' 2

' 4

' 6

' 8

' 10

' 12

' 14

' 16

' 18

' 20

' 22

' 24

Light treatment (h)

FIGURE 8. Cubic response of testes weight per 100 g BW for broiler breeder males (64 wk) to hours of light per day (y = .5802 + .0463x - .0054x2 + .OOOlx3, R 2 = .98).

ACKNOWLEDGMENT

The authors with to thank F. F. Bartol, Department of Animal and Dairy Sciences, for assistance with histological procedures. REFERENCES Bajpai, P. K., 1963. The effect of photoperiodicity on semen characteristics of poultry. Poultry Sci. 42: 462-465. Basurto-Kuba, V. M, E. Heath, and W. A. Wagner, 1984. Spermatozoa and testes in the boan Correlative analysis of sperm morphologic features, seminiferous epithelial area, and testes weight. Am. J. Vet. Res. 45:1328-1332. Bilgili, S. F., and J. A. Renden, 1984. Fluorometric determination of avian sperm viability and concentration. Poultry Sci. 63:2275-2277. Brillard, J. P., and M. de Reviers, 1981. Testis development and daily sperm output in guinea fowl raised under constant daily photoperiods. Reprod. Nutr. Dev. 21:1105-1112. Culbert, J., P. J. Sharp, and J. W. Wells, 1977. Concentrations of androstenedione, testosterone and LH in the blood before and after the onset of spermatogenesis in the cockerel. J. Reprod. Fertil. 51:153-154. de Reviers, M., 1974. Le developpement testiculaire chez le coq. m . Influence de la dupree quotidienne

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

sexual maturity for pullets and the length of the photoperiod. It was concluded that maximal rate of sexual maturation would occur with 10 h of light/day using constant photoschedules. A cubic relationship of age at sexual maturity and photoschedule was observed for males in the current study with earliest maturity occurring with 4 to 8 h light. Semen concentration was increased by decreasing hours of light in the present study, and there were strain by LT interactions for semen weight and number of spermatozoa per ejaculate. Differences in actual BW between the two strains were removed by restricting feed and providing equal amounts. Data in the present study suggest that genotype be considered during development of photoschedules for breeder males, as proposed by de Reviers (1980). The major advantage of short photoschedules may be increased percentage of males producing semen upon artificial ejaculation. Plasma testosterone increases after 20 wk of age in the rooster (Sharp et al., 1977; Culbert et al, 1977; Driot et al., 1978, 1979). In these cited studies, the males were raised and kept on 14 to 16 h light/day to a maximum of 44 wk of age. In the present study, testosterone was increased after 16 wk of age, and there was a positive relationship between testosterone levels and hours of light. The behavioral response of male breeders to artificial massage for ejaculation of semen was favorably influenced by short LT (< 8 h), which may have been dependent upon decreased aggression through reduced testosterone levels or its metabolites (see reviews by Ottinger et al., 1984; Harding, 1986).

Testes weights were increased in the present and other studies (Ingkasuwan and Ogasawara, 1966; de Reviers, 1975, 1977; de Reviers and Williams, 1981) in males on (8 h) light (short) compared with males on 16 h (long) light per day. De Reviers (1980) reported that males receiving 16 h of light/day show rapid testes growth during puberty but size decreases with age, whereas males receiving 8 h of light show delayed adult testes size but weight is maintained for an extended time (e.g., 51 wk of age). The physiological mechanism for photoschedule effects on testes weights may be through negative feedback of testosterone or its metabolites on the hypothalamo-pituitary system (Sharp et al, 1977; Follet and Robinson, 1980; Williams and de Reviers, 1981; Massa, 1984). It has been suggested that seminiferous epithelial area or tubule diameter may be direct measures of testes spermatogenic activity (Tanaka and Yasuda, 1980; Tanaka and Fujioka, 1981; Wilson et al, 1988). However, manipulation of either dietary protein concentration (Wilson et al., 1988) or photoschedule as in the present study did not alter seminiferous epithelial area or tubule diameter.

PHOTOSCHEDULES FOR ROOSTERS

Looney, S. W., and W. B. Stanley, 1989. Exploratory repeated measures analysis for two or more groups. Review and update. Am. Stat 43:220-225. Massa, R., 1984. Patterns and biological significance of steroidal hormone metabolism in birds. J. Exp. Zool. 232:531-537. McCluskey, W. H., and J. E. Parker, 1963. The effect of length of daily light periods on reproduction in female chickens. Poultry Sci. 42:1161-1165. Morris, T. R., 1967. Chapter 2. Light requirements of the fowl. Pages 15-39 in: Environmental Control in Poultry Production T. C. Carter, ed. Oliver and Boyd, LTD. London, England. Morrison, D. F., 1967. Curve fitting for repeated measurements. Page 216 in: Multivariate Statistical Methods. McGraw-Hill Book Co., San Francisco, CA. National Research Council, 1984. Nutrient requirements of poultry. National Academy Press, Washington, DC. Ottinger, M. A., E. Adkins-Regan, J. Buntin, M. F. Cheng, T. DeVoogd, C. Harding, and H. Opel, 1984. Hormonal mediation of reproductive behavior. J. Exp. Zool. 232:605-615. Parker, J. E., and W. H. McCluskey, 1964. The effect of the length of daily light periods on the volume and fertilizing capacity of semen from male chickens. Poultry Sci. 43:1401-1405. Parker, J. E., and W. H. McCluskey, 1965. The effect of length of daily light periods on sexual development and subsequent fertilizing capacity of male chickens. Poultry Sci. 44:23-27. Proudfoot, F. G., 1980. The effects of dietary protein levels, ahemeral light and dark cycles, and intermittent photoperiods on the performance of chicken broiler parent genotypes. Poultry Sci. 59:1258-1267. Sexton, K. J., J. A. Renden, D. N. Marple, and R. J. Kemppainen, 1989. Effects of dietary energy on semen production, fertility, plasma testosterone, and carcass composition of broiler-breeder males in cages. Poultry Sci. 68:1688-1694. Sharp, P. J., J. Gilbert, and J. W. Wells, 1977. Variations in stored and plasma concentrations of androgens and luteinizing hormone during sexual development in the cockerel. J. Endocrinol. 74:467—4-76. Tanaka, S., and T. Fujioka, 1981. Histological changes in the testis of the domestic fowl after partial adenohypophysectomy. Poultry Sci. 60:444-452. Tanaka, S., and M. Yasuda, 1980. Histological changes in the testis of the domestic fowl after adenohypophysectomy. Poultry Sci. 59:1538-1545. Williams, J., and M. de Reviers, 1981. Variations in the plasma levels of luteinizing hormone and androstenedione and their relationship with the adult daily sperm output in cockerels raised under different photoschedules. Reprod. Nutr. Dev. 21:1125-1135. Wilson, J. L., L. M. Krista, G. R. McDaniel, and C. D. Sutton, 1988. Correlation of broiler breeder male semen production and testes morphology. Poultry Sci. 67:660-668.

Downloaded from http://ps.oxfordjournals.org/ at UCSF Library on April 27, 2015

d'eclairement sous photoperiodes constantes. Ann. Biol. Anim. Biochim. Biophys. 14:591-600. de Reviers, M, 1975. Le developpement testicnlaire du coq. m. Etude quantitative des cellules germinales du coq place sous des photoperiodes croissantes appliquees a des ages differents. Ann. Biol. Anim. Biochim. Biophys. 15:633-641. de Reviers, M., 1977. Le developpement testiculaire chez le coq. V. Action de variations progressives de la duree quotidienne d'eclairement. Ann. Biol. Anim. Biochim. Biophys. 17:179-186. de Reviers, M., 1980. Photoperiodism, testis development and sperm production in the fowl. Proc. 9th Int. Cong. Anim. Reprod. Artif. Insemin. Madrid 2: 519-526. de Reviers, M, and J. Williams, 1981. Predicting the adult daily sperm output after the first ejaculates in cockerels raised under different photoschedules. Reprod. Nutr. Dev. 21:1113-1124. Driot, FJ.M, ML de Reviers, and J. Williams, 1979. Plasma testosterone levels in intact and hemicastrated growing cockerels. J. Endocrinol. 81:169-174. Driot, FJ.M, D. H. Gamier, and M Terqui, 1978. Development and validation of a "direct" radioimmunoassay for plasma testosterone in the fowl (Gallus domesricus). Gen. Comp. Endocrinol. 36: 244-249. Foliett, B. K., and J. E. Robinson, 1980. Photoperiod and gonadotrophin secretion in birds. Prog. Reprod. Biol. 5:39-61. Gill, J. L., 1981. Repeated measurement and split plot designs. Pages 169-243 in: Design and Analysis of Experiments in the Animal and Medical Sciences. Vol. 2. The Iowa State University Press, Ames, IA. Harding, C. F., 1986. The importance of androgen metabolism in the regulation of reproductive behavior in the avian male. Poultry Sci. 65:2344-2351. Harris, G. C, Jr., J. A. Benson, and R. S. Sellers, 1984. The influence of daylength, body weight, and age on the reproductive ability of broiler breeder cockerels. Poultry Sci. 63:1705-1710. Harrison, P. C, J. D. Latshaw, J. M. Casey, and J. McGinnis, 1970. Influence of decreased length of different spectral pbotoperiods on testis development of domestic fowl. J. Reprod. Ferul. 22:269-275. Harrison, P. C, G. Schumaier, and J. McGinnis, 1969. Reproductive development and response of White Leghorn pullets subjected to increasing day-lengths at different ages. Poultry Sci. 48:1021-1026. Humason, G. L., 1972. Pages 156-159 in: Animal Tissue Techniques. W. H. Freeman and Company, San Francisco, CA. Ingkasuwan, P., and F. X. Ogasawara, 1966. The effect of light and temperature and their interaction on the semen production of White Leghorn males. Poultry Sci. 45:1199-1206. Lamoreux, W. F., 1943. The influence of different amounts of illumination upon the production of semen in the fowl. J. Exp. Zool. 94:73-95.

1609