The in sacco nitrogen digestibility of 7 hays of different qualities was measured by the mobile nylon bag technique (MNBT) in fistulated horses in the precaical part and the total digestive tract. The comparison between in sacco result and in vivo digestibility measured by the conventional technique (total faeces collection), of the same hays was used to validate MNBT. Four adult horses fitted with a caecum cannula were used. They were fed at maintenance and received in 2 equal meals (8:00,

more »

... 0) a ration composed of 30 % barley given first and 70 % hay. Thirty cylindrical nylon bags (f! 1 cm, length 6 cm, porosity 48 N m) containing 200 mg of ground hay (3 mm mesh) were introduced into stomach of each animal through a nasooesophagal probe during the morning feed after the concentrate, at each measuring period. These were organised into 2 successive 4 x 4 latin squares (2 x 4 hays including 1 control). About 15 bags were recovered on arrival in the caecum 2 h to 6 h after introduction (mean retention time, MRT = 4.2 h ± 0.6). The other bags were recovered in faeces between 18 and 60 h after introduction (MRT = 38.8 h t 2.7). After washing, the bags were dried (48 h at 60°C), weighed, and their contents pooled by animal, collection site and hay. Nitrogen content (micro-Kjeldahl), NDF and nitrogen in NDF (N-NDF) were assayed in these bag residues and for the same hays in faeces collected during in vivo digestibility studies. The in sacco digestibility was calculated for precaical part and total digestive tract. In precxcal part, it affords an estimate of the true digestibility of feed in the stomach and small intestine. In total digestive tract, the N true digestibility of feed was calculated for the 2 methods in vivo and in sacco after assuming that only dietary N reaching the faeces was N-NDF (Glade, 1984, JAS, 58(3), 638-645). In the total tract, digestibilities of dry matter (not reported) and nitrogen, and estimated true digestibility measured by MNBT correlated significantly with the corresponding in vivo parameters (Rz= 0.99, 0.77, 0.75 respectively). As expected, the in sacco digestibility of N was higher by 15 to 20 points than its in vivo digestibility. In contrast, the true digestibilities were similar. In the precaecal part, the in sacco N digestibility was high (63.5 ± 2.6 %), and varied little among hays. It decreased from 67 to 61 % when the N-NDF/N ratio increased. The proportion of feed N digested before the caecum varied respectively from 103 to 76 % and from 90 to 69 % when expressed in total % digestible N or in % of truly digestible N. In conclusion, the MNBT results are consistent with the results obtained with the conventional method on the total tract of the horse. Then it would afford an evaluation of the precaecal digestibility of nitrogen in feed.