Performance of Female Broiler Breeders Submitted to Different Feeding Schedules

1 Embrapa Suínos e Aves Eng. Agr., DSc. vavila@cnpsa.embrapa.br 2 Universidade Federal do Rio Grande do Sul Eng. Agr., PhD. ampenz@conex.com.br 3 Embrapa Suínos e Aves Méd. Vet., DSc. pbrum@cnpsa.embrapa.br 4 Embrapa Suínos e Aves Zootec., MSc. prosa@cnpsa.embrapa.br 5 Embrapa Suínos e Aves Eng. Agr., DSc. antilog@cnpsa.embrapa.br 6 Embrapa Suínos e Aves Zootec., PhD. elsio@cnpsa.embrapa.br * Scholarship granted by CNPq. Author(s)


INTRODUCTION
Differences in diet composition and in temperature might adversely affect egg production of broiler breeder hens (Robbins et al., 1988).Kohne et al. (1973, reported that feeding time should be considered at high environmental temperature due to the caloric increment produced by exothermic reactions of nutritional metabolism.Heat increment was higher 5 h after feeding in birds fed at 6:00 am than in birds fed at 2:00 pm when indoor temperature increased (Wilson et al., 1989).
Usually, female broiler breeders are fed once a day, in the morning.Cave (1981) and Bootwalla et al. (1983) questioned if nutrient requirements are fulfilled in this feeding system.Therefore, Cave (1981) evaluated broiler breeder hens from 24 to 63 weeks of age submitted to different feeding schedules and no differences were seen for egg production.On the other hand, more frequent feeding decreased weight gain and increased egg mass, indicating that this strategy enhanced nutrient availability for egg production and regulated excessive body tissue deposition.Hens selected for medium and light body weight showed an increase in egg weight and production when meal was offered in the afternoon (Balnave, 1977).However, feeding time had no effect on egg production and weight (Brake, 1998).
Some brazilian poultry companies feed broiler breeder parents in the afternoon, without considering the consequences of this strategy.This work aimed to evaluate the effect of time of feeding on the performance of female broiler breeder during the period of egg production.

MATERIAL AND METHODS
A total of 1,296 Arbor Acres hens were mated with 144 roosters.Birds were raised until 18 weeks of age in a commercial poultry company according to the management techniques recommended by Sadia (1993).Then, they were transferred to an experimental poultry house at Embrapa Suínos e Aves.Twenty-seven hens and 3 males were allocated per pen.Between 19 and 20 weeks of age (pre-experiment period), birds were fed at 12:00 am in order to standardize feeding.Four treatments were tested: T1=100% feeding at 6:30 am; T2= 50% feeding at 6:30 am and 50% at 3:30 pm; T3=100% feeding at 11:00 am and T4=100% feeding at 3:30 pm.Each treatment had 12 repetitions allocated in a completely randomized design.From the 20 th to the 24 th week of age, birds were adapted to the different treatments; performance was evaluated from the 25 th to the 66 th week of age.
Throughout breeding, all treatments were given similar management and amount of feed (g.bird -1 .day -1 ).In order to reduce the stress that occurs just before daily feeding, feeders were controlled by a mechanic system and were filled with pellet and mashed feed one day before the meal was offered.Feed amounts were adjusted based on female body weight and egg production and on male body weight.Birds were weighed at every two weeks.For both sexes, the reference weight was the mean weight of birds from four pens of each treatment.
Diets were formulated according to the nutritional requirements of Arbor Acres guide, according to Sadia (1993).Percentage composition of experimental diets in the pre-laying (18 to 23 weeks of age), laying I (24 to 47 weeks of age) and laying II (48 to 66 weeks of age) phases are shown in Table 1.
Eggs were counted daily in each pen and average percentage of total egg production per hen per day (EPP) was obtained from 25 to 66 weeks of age.This variable was analyzed using GLM (SAS, 1996) according to the following model: y ijk = m + t i + e ij + s k + ts ik + e ijk .Where: i=1,..,4 treatments; j=1,...,12 pens; k=1,..., k weeks of evaluation; and y ijk is the mean value of the i th response from j th pen within the k th week; m is the mean value observed in the experiment; t i is the i th treatment effect; e ij is the experimental error according to a normal distribution with mean zero and constant of variance s 2 ; s k is the effect of the k th week; ts ik is the interaction effect between the i th treatment and the k th week; e ijk is the error of weekly evaluation considering a normal distribution.Experimental error e ij was used to test the week (W) effect and the interaction between treatment and week.Treatment averages were compared by Students t-test.Production equations for each pen were estimated by NLIN (SAS, 1996) and Fialho & Ledur (1997) model, considering age at production peak (XP), production at peak (P), production decrease per week after peak (S) and period from onset to peak of production (Tp).
The eggs were classified according to weight (lighter than 46 g and heavier than 85 g) in non-hatchable (NHA) and hatchable (HA).The percentage of hen mortality (MOR), hen cost (R$28.00)and egg sale cost (R$0.15,0.16, 0.17, 0.18, 0.19 and 0.20) were used to calculate the gross profit margin (GM) per hen and treatment using the following formula: Where: GM = Gross profit margin per hen; NE = Number of available eggs; PE i = percentage of eggs; i = 1Þ Non-hatchable eggs and i = 2 Þ hatchable eggs; P 1 = price of non-hatchable eggs = 0.25*P 2 ; P 2 = price of hatchable eggs; CM = Cost of hen (buying and rearing); MOR = Percentage of hen mortality.MOR, GM, parameters of the production curve (XP, P, S and Tp) and production traits (TEP, AFE, AG50, AG60, AG70, AG80 and DAP80) were used to compare treatments by GLM (SAS,1996) according to the ANOVA model: y ij = m + t i + e ij ; where: i=1,,4 treatments; j=1,...,12 pens; y ij is the value of the i th parameter within the j th pen; m is the overall mean of the parameter in the experiment; t i is the effect of the i th treatment; e ij is the experimental error according to a normal distribution with mean zero and constant of variance s 2 .Means of treatments were compared by Students t-test.

RESULTS AND DISCUSSION
Average percentage of egg production per hen per day throughout the production period is shown in  2, as well as the means for the parameters (XP, P, S and Tp) estimated using the production equation for each treatment.Egg production was different among ages and treatments and there was a significant interaction (p<0.01) between the sources of variation.However, T3 and T4 production curves were slightly below the curves of other treatments (Figure 1), that might explain the better results of T1 and T2 for PPE (Table 2) and TEP (Table 3).
Figure 1 -Percentage of egg production.hen - .day - in laying hen breeders submitted to different feeding schedules (from 25 to 66 weeks).

Variables
Average mortality was 3.6% and values were not different (p>0.10)among treatments.Similar results were observed when broiler breeder hens were fed as Performance of Female Broiler Breeders Submitted to Different Feeding Schedules 200 recommended by the strain guide or ad libitum, or an association of both (Robbins et al., 1988).However, mortality was high in a group of hens fed with low or high protein levels (Cave, 1981), but a non-significant mortality was found in the control group from 24 to 63 weeks of age, which was fed protein levels similar to those of the present experiment.
Only XP and P were different (p<0.05)among treatments.For XP, hens of T4 reached production peak at 33 weeks of age while hens from other treatments were earlier and reached the production peak at 32 weeks of age.P was low in T3, intermediate in T4 and high in T1 and T2.This result explained differences in PPE and TEP among treatments.
PPE, SOE, HA, NHA, AFE, AG50, AG60, AG70, AG80 and DAP80 of each treatment are shown in Table 3.All variables, except NHA, were different (p<0.05)among feeding times.Similar production results was reported by Harms (1991).Changes in feeding time affected egg production because it was observed that hens fed at the end of the day had smaller egg production.On the other hand, our findings are different from the results reported by Cave (1981), Bootwalla et al. (1983), Brake (1988) and Samara et al. (1996), who reported that different feeding times had no effect on egg production.
Hens of T1 and T2 showed higher egg production (TEP, SOE, HA), whereas birds fed at 11:00 am.(T3) had lower production results when compared to the other treatments.We can speculate that this result could be due to heat stress condition caused by the increase in ambient temperature, which occurred at the same time of the caloric increment (metabolic heat production during nutritional metabolism), i.e., at approximately 5 to 6 hours after feed intake.Other consequences were decrease (p<0.05) of the period in which production was above 80%, delay of sexual maturity and impairment of other production parameters.Similarly, Harms (1984) has reported previously that hens with adequate weight gain reached sexual maturity and egg production equal or higher than 80% earlier than those with inadequate weight.
AFE was earlier in T2, and AG was at least similar to hens of other treatments at the different ages (AG50, AG60, AG70, AG80), probably because dual feeding enhanced nutritional efficiency and requirements for maintenance and egg production were fulfilled.This fact agrees with results reported by Robbins et al. (1988) and Katanbaf et al. (1989).They observed that sexual maturity was 14 and 60 days earlier, respectively, and that hens fed ad libitum had improved body growth and egg formation when compared to restricted birds.
Treatments were different (p<0.05) in relation to gross profit margin (Table 4).In T2, gross profit margin was positive when an egg selling price of R$ 0.16 was considered.There were no differences between T1 and T2, which where more efficient than T3 and T4 due to differences in hen mortality and in the production of eggs (available, not hatchable and hatchable).Such results showed that traditional and dual feeding systems were superior to the other feeding schedules.

Figure 1 .
Figure 1.PPE and MOR are shown in Table2, as well as the means for the parameters (XP, P, S and Tp) estimated using the production equation for each treatment.Egg production was different among ages and treatments and there was a significant interaction (p<0.01) between the sources of variation.However, T3 and T4 production curves were slightly below the curves of other treatments (Figure1), that might explain the better results of T1 and T2 for PPE (Table2) and TEP (Table3).

Table 3 -
Total egg production (TEP), sound eggs (SOE), hatchable eggs (HA), non-hatchable eggs (NHA), age in weeks at different percentages of egg production (AG) and number of days with production equal or higher than 80% (DAP80), evaluated in broiler breeder hens from 25 to 66 weeks of age.

Table 4 -
Gross profit margin of different feeding time according to egg price ranges.Values expressed as Reals (R$).