The young rabbits were housed in wire mesh cages equipped with a drinking trough and a feeder. They were divided into four (4) batches of 32 rabbits. Each batch was then subdivided into two subgroups of 16 rabbits. Two subgroups of the same batch were fed, one with a complete feed (feed with incorporated fiber source) served all the day (at 9:00 am and 4:00 pm); and the other, with the same feed without incorporated fiber source, served only in the morning (at 9:00 am) and supplemented with the fiber source (identical to the one used to prepare the complete feed) served only in the evening (at 16:00). This constitutes two feeding strategies for each fiber source; a continuous feed and sequential feed. A total of four complete feeds were then served for continuous feeding and four types of feeds (feeds supplemented with fiber sources) were served sequentially.

Complete feed Gliricidia sepium: CFG; Complete feed Leucaena leucocephala: CFL; Complete feed Moringa oleifera: CFM; Complete feed palm nut fiber: CFF.

The different rations served during the experiment for the eight (8) subgroups, were:

- complete feed Gliricidia sepium (CFG);

- supplemented feed Gliricidia sepium (SFG);

- complete feed Leucaena leucocephala (CFL);

- supplemented feed Leucaena leucocephala (SFL);

- complete feed Moringa oleifera (CFM);

- supplemented feed Moringa oleifera (SFM);

- complete feed palm nuts fiber (CFF);

- supplemented feed palm nut fiber (SFF).

Four 15-week-old rabbits per feed treatment were selected according to the average weight of each batch for slaughter. The rabbits bled after being stunned by a sharp blow to the occipital region. The two-man dressing method described by [11] was used in this study.

The physico-chemical composition of the rabbit feeds was determined through the levels of dry matter, ash, fat, and fiber. The Biceps femoris and Longissimus dorsi muscles from the different meat samples were dried at 65˚C for 72 hours. Subsequently, the contents of dry matter, ash, fat, and protein were determined according to standard procedures. The moisture content was determined in two replicates using the thermogravimetric method of the AOAC standard [12] . The determination of protein content was done according to the Kjeldhal method [13] . Lipids were determined by direct Soxhlet extraction and according to the AACC method [14] . The ash content was determined by the AACC method [14] . The crude fiber content of the feed was determined by the method of Weende [15] .

Across feeding strategies for the same fiber source, both in saddle and thigh muscle, no significant difference (p > 0.05) was found for the dry matter content of rabbits fed Leucaena leucocephala leaves or palm nut fibers. The dry matter content of the meat of rabbits fed the Gliricidia sepium complete feed was higher (p < 0.05) compared to that of rabbits fed the feed supplemented in the evening with Gliricidia sepium leaves. For the Moringa oleifera fiber source, the dry matter of the thigh was higher (p < 0.05) in favor of the rabbits fed the supplemented feed. However, the saddles of the same subjects had recorded similar dry

NS: p > 0.05; *: p < 0.05; **: p < 0.01; ***: p < 0.001; mean in the same column followed by different letters differ significantly at the 5% threshold.

matter levels (p > 0.05). Therefore the supply of dried leaves of Gliricidia sepium or Moringa oleifera in the evening would have had an impact on the overall nutrient content of the rabbits’ meat. This was not the case for fiber sources based on Leucaena leucocephala leaves or palm nut fibers, whose dry matter content remained unchanged regardless of the distribution method.

From one fiber source to another, at the saddle, the highest dry matter content (p < 0.05) was recorded by continuously distributed Gliricidia sepium (28.65%) compared to all other fiber sources which recorded similar content (p > 0.05). In the thigh, the lowest (p < 0.05) dry matter content (23.12%) was obtained by Moringa oleifera and the highest (p < 0.05) by the same fiber source distributed sequentially (27.33%). Thigh dry matter rates were equal (p > 0.05) for sequentially distributed Moringa oleifera, continuously distributed Gliricidia sepium and Leucaena leucocephala and for both sequentially and continuously distributed palm nut fibers. Similarly, the thigh dry matter rates recorded by the continuously distributed Moringa oleifera fiber sources were similar (p > 0.05) to those obtained by the sequentially distributed palm nuts and Gliricidia sepium fibers and continuously and sequentially distributed Leucaena leucocephala. These similarities observed between the dry matter contents of meat from different feed treatments show that the amount of nutrients in rabbit meat is not affected by the different fiber sources used. Thus, different fiber sources can be used as substitutes in rabbit feed without changing the overall nutrient content of the meat. The dry matter contents recorded in this study are close to those obtained by [20] (25% - 27.5%), and [21] (24.5% - 25.3%) on 15 - 18 weeks old rabbits fed commercial diets. These results are also similar to those of [8] (24.96% - 27.30%) on rabbits fed rations containing 30% of C. populnea or S. nodiflora leaves.

The mineral fraction of rabbit meat varied significantly (p < 0.05) at the saddle level for the Gliricidia sepium fiber source in favor of the continuous strategy and for Moringa oleifera in favor of the sequential strategy (p < 0.001). For Leucaena leucocephala and palm nut fiber sources, no significant difference (p> 0.05) in the ash content of the shavings was found between the distribution strategies. At the thigh level, there was no variation (p > 0.05) in ash levels between distribution strategies for Moringa oleifera and Leucaena leucocephala fiber sources. However, a significant difference (p < 0.05) was found for Gliricidia sepium and palm nut fiber sources in favor of continuous feeding. For rabbits fed Leucaena leucocephala leaves, ash levels were not affected by the mode of distribution regardless of muscle. Thus, the mineral fraction of the meat of the rabbits was not affected by the feeding method of Leucaena leucocephala leaves. Only the supply of dried leaves of Gliricidia sepium in the evening would have affected the mineral content of the rabbit meat. This is not clearly established for palm nut fibers and Moringa oleifera leaves.

Across fiber sources, ash levels were high (p < 0.001) in the saddles (3.49%) and thighs (3.56%) of rabbits fed the Gliricidia sepium leaf-based complete feed. The saddles of the rabbits fed the continuous palm nut fiber had the lowest (p < 0.001) ash values (0.91%), which were equal to those of the continuously fed Gliricidia sepium and Moringa oleifera fiber sources. In the thigh, low levels (p < 0.05) of minerals were recorded in sequentially distributed Gliricidia sepium (1.10%), which were similar to sequentially distributed palm nut fibers and both continuously and sequentially distributed Leucaena leucocephala and Moringa oleifera fiber sources. This shows that the substitution of these different fiber sources in the rabbit diet will not have a margining effect on the mineral fraction of the meat. Thus, the different fiber sources can also be substituted for each other in the sequential feeding of rabbits without changing the mineral content of the meat. The mineral levels found in this study are high compared to those obtained (1.03% - 1.23%) in rabbits fed rations containing 30% of C. populnea or S. nodiflora leaves [8] . These differences could be explained by the diet, in this case, the presence of different kinds of forage. Indeed, Combes [22] mentioned that it is very likely that feeding through supplementation is the main factor of variation in the mineral fraction of meat.

Across feeding strategies and for the same fiber source, no significant difference (p > 0.05) was observed for the fat content of rabbits fed Leucaena leucocephala and Moringa oleifera leaves. However, in general, fat was low when the rabbits were fed the fiber source in the evening. This difference was highly significant for rabbits fed G. sepium leaves and palm nut fibers. Fat levels were indeed high in the saddles (p < 0.05) and thighs (p < 0.001) of the rabbits fed the G. sepium leaf-based complete feed. For rabbits fed palm nut fiber, it was in the thigh that the lipid content was high (p < 0.001) in favor of continuous distribution. Thus, the distribution mode of Leucaena leucocephala and Moringa oleifera leaves did not affect the quality of the rabbits’ meat, in this case, its fat content. On the other hand, feeding rabbit with G. sepium leaves or with palm nut fibers on a continuous feeling basis increased the fat content of the meat. However, this effect on meat quality seems to be attenuated when the fiber is used in the evening.

Thigh meat from rabbits fed palm nut fiber had the highest fat levels (p < 0.001) in continuous and sequential distribution (5.12% and 4.18% respectively). At the saddle level, the fat levels of these same subjects were also high (p < 0.05) and similar (p > 0.05) to those recorded by G. sepium and M. oleifera in a continuous distribution, which is similar (p > 0.05) to all other fiber sources. The fat levels obtained in this study for thighs are high compared to the margin of 2.90% - 3.22%, obtained on labeled rabbits [23] . The high-fat content of meat from rabbits that were fed palm nut fiber as a source of fiber may be explained by the fat content of the palm nut fiber itself. Palm nut fibers, which are residues of palm oil extraction, give the feed a high-fat content, which relatively increases the fat content in the thigh meat. Therefore, the use of low-fat extraction residues will allow the production of rabbit meat with lower fat content. Indeed, Combes [22] mentioned that diet is one of the most important factors influencing the amount of lipid fraction in rabbit meat. Thus, to substitute palm nut fiber for other sources of fiber in rabbit diets, residues with low fat levels should be used.

The distribution strategy of the different fiber sources had no influence (p > 0.05) on the protein contents of the rabbits in both the thigh and the saddle. Therefore, these different fiber sources can be used continuously or sequentially without affecting the protein content of the rabbit meat.

From one fiber source to another, the highest protein contents in the saddle were obtained by the fiber source of G. sepium (22.56% and 20.91% in continuous and sequential distribution, respectively). For the other fiber sources, the protein contents were equal (p > 0.05). Protein contents were also similar (p > 0.05) for the shreds in sequential distribution. In the thigh, palm nut fiber gave the highest protein contents (24.46% and 23.05%) (p < 0.05) and similar (p > 0.05) to the G. sepium fiber sources which in turn were equal to all other fiber sources. The protein levels found were close to those obtained for rabbits at commercial slaughter ages and weight (19.5% - 22.5%) [22] . The similarities in protein content show that these different fiber sources can be substituted in the rabbit diet without changing the protein content of the meat. Moreover, the protein levels in the rabbits’ thighs were higher than in the saddle. This is in line with Combes [22] who reported that among the cuts, the edible part of the leg is the richest in protein.