:The present study was undertaken to evaluate the effects of faecal collection interval (using a settling column) and the levels of dietary meat and bone meals (MBM) on the apparent digestibility coefficient (ADC) for dry matter , protein , energy and phosphorus in gibel carp . Six iso2nitrogenous (crude protein : 410 g/ kg) and iso2energetic (gross energy : 18 kJ/ g) diets were used in which fish meal (FM) protein was gradually replaced by MBM at 0 % , 20 % , 40 % , 60 % , 80 % and 100 %. These diet combinations were fed to juvenile gibel carp for 11 weeks. The faecal samples were collected two weeks after the beginning of the experiment using a settlement column starting 1 min after excretion started , or 4h and 16h after feeding. The results showed that the ADCs of dry matter , protein , energy and phosphorus increased significantly as the time from excretion to faecal collections in 2 creased while not being significant for the high MBM inclusion diets. ADCs of dry matter , crude protein , energy and phosphorus decreased linearly or almost linearly with the increase in dietary MBM levels. These findings suggest that faeces should be collected soon after they had settled under the conditions of this study for digestibility determination while rapid leaching is apparently re 2 sponsible for this false reading of digestibility with increasing sampling interval. It is also apparent that the digestibility is one of the problems affecting the use of MBM in juvenile gibel carp . Determination of the apparent digestibility of feed2 stuffs is required to formulate economically optimized feeds to meet also the nutritional demands of the animal and to minimize environmental impact derived from excre2 tion [1 ,2 ] . The measurement of digestibility for aquatic species is technically more difficult than that for terrestrial animals because nutrients from the faeces can quickly leach into the water , depending on the contact time in the water [3 ,4 ] . A number of techniques have been used to overcome this problem , including stripping of the contents of the hindgut [5 10 ] , removal of faeces after intestinal dissection [6 ,7 ,9 ,11 13 ] , and anal suction of faeces using a specific device [7 ,14 ,15 ] . However , some partially digested food , urine , sperm , eggs or intestinal epithelial cells can contaminate the samples and result in an underestimation of the apparent digestibility coefficient ( ADC) [16 19 ] .Furthermore , such methods require frequent handling of the fish which is stressful and is not suitable for obtaining faeces from small fish [20 ] . Also , these methods can not be used for continuous assessment of digestibility in com2 bination with growth trials [18 ] .Other researchers collected faeces in the water by pipetting immediately after release [21 ] , netting [7 ] , de2cantation column [22 ,23 ] , continuous effluent filtra2 [24 ,25 ] and the use of metabolic chambers [26 ] . These methods can avoid the problems caused by collection of un2defaecated intestinal contents. However , ...