A total number of four hundred day-old unsexed broiler chicks of Pure Line Strain were obtained from INMAA local hatchery. The birds were reared using deep litter of wood shavings in a closed-side poultry house from day-old to the end point set for slaughter. Throughout the experimental period light was provided for 24 hrs.

The experimental diets were formulated to be approximately iso caloric (3.2 kcal/kg and iso nitrogenous (22.5%).

The chicks were fed the pre-starter diet from day-old to seven days of age before being shifted to the starter diet from 8 to 17 days of age. They were then fed the grower diet from 18 - 35 days of age and the finisher diet from 36 days of age to the end of the experimental feeding. The percent inclusion rate of ingredients of the pre-starter, starter, grower and finisher diets were shown in Table 1 while their percent chemical composition as shown in Table 2. The experimental birds were offered feed and water in a sufficient number of feeders and drinkers. The birds have ad libitum feed and water.

Feed samples were analyzed according to the methods of AOAC [5] .

At 29 days of age 180 male birds were chosen randomly, labeled and divided into six groups each group contains 30 birds. The experimental birds were allowed to have a period of three days as adaptation period before the onset of the experiment (at 32 days of age).

At 32 days of age, treatment was beginning into six groups. Group A fed ad libitum (control) while group B and C fed 120 g, 60 g/bird/day for eight days, respectively. Group D and E fed 120 g, 60 g/bird/day for four days, respectively followed by zero feeding. Group F deprived of food during the whole experimental period (8 days).

Day-old birds at the rearing phase were weighed using a balance of maximum 5 kg. Live body weight and feed intake were taken every week, but during restriction phase birds were weighed individually every 48 hrs. Mortality during both phases was recorded daily as it occurs.

Note: Pre-starter diet form day old to seven days of age, Starter: from 8 to 17 days, Grower: 18 - 35 days, Finisher: from 36 days of age to the end of the experimental according to feeding system of Sudanese Kwaiti Company Farm.

Metabolizable energy values were calculated by the modified equation of Lodhi et al. [7] .

At 34 days of age, every 48 hrs all birds of each group were weighed. Then four birds from each group were randomly selected to slaughtering. Slaughtering was done according to Islamic Shariah by severing right and left jugular and carotid cervical vessels just caudal to the larynx using a sharp knife. After complete bleeding, the head at its articulation and shanks at the hock joint were separated and weighed. The birds were then plucked after soaking in warm water. Evisceration was done by ventral medial incision. Liver and gizzard organs were separated and weighed. Carcasses were finally weighed, soaked in ice water and sealed in polythene bags.

Completely Randomized Design (CRD) was used in this experiment, and the difference between the groups for all parameters was determined by analyses of variance (ANOVA) according to Gomez and Gomez [6] . The coefficient of Variation (C.V) was calculated from the ANOVA table following the formula:

C .V = Residual .m .s overall mean × 100

As Table 3 shown final live body weight was not significantly (P > 0.05) different between experimental groups during period 32 to 34 days of age. This indicated

*: P < 0.05, **: P < 0.01, NS: Not Significant; Means in a row with the same letter designation are not significantly (P > 0.05) different; L.S: Level of Significant; C.V: Coefficient of Variation.

that the experimental birds were not affected statistically by feed restriction program during this period. But numerically the more sever feed restricted groups (C, E, and F) shows the lowest reading for final live weight. This finding previously was reported by Abdul and Friani [8] who showed lower body weight at 21 - 35 days of age in restricted birds allowed 75% and 60% of ad libitum intake.

Weight gain was significantly (P < 0.01) different among the experimental groups. Group B and D gained less than control group while groups C, E and F showed a loss in weight. This decrease in weight gain could be due to decreased feed intake of these groups, and also may be due to the evacuation of the gastrointestinal tract. This result is in line with Ismail [9] who reported that feed restriction subject the birds to a period of undernutrition, due to reduced intake of nutrients, which results in slow growth and loss of body weight. Many previous studies found that broilers kept under a restricted feeding program gained less weight than those kept under ad libitum feeding [4] [10] [11] . In the same period, carcass weight and dressing percentage were not differed significantly (P > 0.05). This result may be related to the fact that carcass weight is the reflection of the body weight [12] .

Effect of feed restriction program on the performance of male broiler chicks from 32 to 36 days of age is shown in Table 4. Feed restriction created a significant effect in the final live weight of experimental groups. Control group showed the highest value for final live weight (1808 ± 253 g) and group F that subjected to a longer period of feed removal showed the lowest value (1208 ± 124 g), which supports the statement that improvement in body weight of birds is highly correlated to feed consumption and duration of feed Restriction [9] [13] - [18] .

*: P < 0.05, **: P < 0.01, NS: Not Significant; Means in a row with the same letter designation are not significantly (P > 0.05) different; L.S: Level of Significant; C.V: Coefficient of Variation.

During the same period carcass weight was significantly (P < 0.01) lower in restricted groups than that fed ad libitum. This finding was expected because carcass weight depends on live weight as stated by Kokoszynski and Bernacki [19] .

The dressing percentage was not significantly (P > 0.05) different between the experimental groups and ranged from 72% for group F to 75% for group B. The same range of dressing percentage was achieved by Janocha et al. [20] .

During the period 32 to 38 day of age, the control group was highest for final live weight, weight gain, and carcass weight, while other groups showed the lowest values than control (Table 5). The probable explanation of the lower values of the feed-restricted groups is that with an increase of birds age, the effect of feed restriction program will be more severe. This result agreed with Santoso et al. [21] who restricted broilers to 75%, 65%, 55% and 45% of ad libitum intake for 10 days and found that body weight of severely restricted (65% or less) male and female broiler chicken was significantly lower than ad libitum fed birds at 42 days.

The dressing percentage was also significantly (P < 0.05) different between groups in an irregular pattern. This may be due to fact that birds of similar weight may not always yield the same amount of edible carcass [22] .

Table 6 shows the performance of male broiler chicks 32 to 40 days of age. Group A was superior for final live weight, weight gain, carcass weight and dressing percentage followed by group B while group F showed the lowest value

*: P < 0.05, **: P < 0.01, NS: Not Significant; Means in a row with the same letter designation are not significantly (P > 0.05) different; L.S: Level of Significant; C.V: Coefficient of Variation.

*: P < 0.05, **: P < 0.01, NS: Not Significant; Means in a row with the same letter designation are not significantly (P > 0.05) different. L.S: Level of Significant; C.V: Coefficient of Variation.

for final live weight, loss in weight, carcass weight and dressing percentage. All the previous performance values differ significantly (P > 0.01) among the experimental groups. It is clear that these values vary with feed intake as stated by Urdaneta-Rincon and Leeson [16] and Leeson and Summers [22] that increased feed intake leads to increased live weight, weight gain, and dressing percentage.

This result indicates that the loss in weight increased with increased duration of feed restriction. This agrees with Omosebi et al. [18] who reported that live weight loss increased with the increased duration of feed withdrawal. In our experiment, group F which was fasted 8 days did not differ significantly from groups C and E. The probable explanation of this result is that birds of group F who were subjected to severe feed restriction tended to eat litter. Ismail [9] noticed that severely restricted birds ate litter mixed with feed and so they increase their feed intake and reduced feed wastage.

It is generally clear that severe feed restriction will lead to decreased live weight and hence decreased dressing percentage, which can be noticed in that the dressing percentage during the first 4 days of restriction was not significantly different, while in the other 4 days of restriction period was significant, and these findings agreed with Omojola et al. [23] who stated that dressing percentage increases as live weight increase and vice versa.

Effects of feed restriction on non-carcass components were shown in Table 7. With few exceptions, the general trend of increase in the percentage of head, shank, liver, and gizzard when the birds were subjected to severe feed restriction. Omojola et al. [23] reported that edible viscera as a percentage of live weight decrease slightly with increasing body weight. In our study, this was true with the control group. Increase in the percentage of organs when birds subjected to severe feed restriction was due to the decrease live weight of feed restricted birds.

Means ± Standard Deviation; *: P < 0.05, **: P < 0.01, NS: Not Significant; Means with the same letter in the same row are not significant (P > 0.05); L.S: Level of Significant; C.V: Coefficient of Variation.

The present result is in close agreement with that found by Ocak and sivri [24] who restricted broilers to 75% of ad libitum intake for 10 days of age and found that feed restriction at late age resulted in an increase in the relative weight of edible inner organs than fed ad libitum.