Six trials were conducted to study the conditions of preparation and treatment of moist forage samples for the in situ measurement of their ruminal dry matter and nitrogen degradations. The following preparations were compared in five trials: Trial I on fresh forages: lacerated fresh before bagging and freezing in liquid nitrogen and then stored at -20 degrees C, forage dried at 60 degrees C and ground to 0.8 mm, forage dried at 80 degrees C and ground to 0.8 mm, forage dried at 60 degrees C and ground to 4 mm. Trial II on lacerated fresh forages: put immediately in bags then in numen, frozen in Liquid nitrogen and used immediately, frozen in liquid nitrogen and stored at -20 degrees C, freeze dried then stored at -20 degrees C. Trial In: on silages: lacerated silage, frozen in liquid nitrogen and then stored at -20 degrees C, silage dried at 80 degrees C and ground to 0.8 mm. Trial IV on hays: undried ground to 12 or 4 mm, dried at 80 degrees C and ground to 0.8 mm. Trial V: after ruminal incubation, bags beaten or not with a 'stomacher' after washing and before oven-drying. In trial VI, particle losses through the bag pores were measured. The main objective of all these trials was to evaluate a mode of sample preparation of moist materials (fresh and silage) in two steps: processing in a 'universal mill' to particles in about 5 mm length, bagging and rapid freezing of the bags in liquid nitrogen. There was no difference between fresh forage placed immediately in the rumen, fresh forage frozen in nitrogen and placed immediately in the rumen, and the same stored and then used several months later. Drying, even at 60 degrees C, lowered effective nitrogen degradability against moist forage; drying at 80 degrees C lowered it by 10 points (P < 0.01) (1 point = 1% on a scale from 0 to 100). Freeze-drying had a weak negative effect (-3.1 points; P < 0.05). The nitrogen degradability of hays increased with decreasing particle size (+7.7 points, P < 0.01, from 12 to 0.8 mm mesh size). Beating with a 'stomacher' is useful for reducing microbial contamination of bag residues (increasing nitrogen degradability by +4.3 points, P < 0.05, for a poorly digestible forage, but only +1 point [not significant] for a digestible one). Finally, losses of particles through bag pores were low, 1.3% of used dry matter. The mode of preparation tested is suitable for the study of nitrogen degradation of moist forages in the rumen. It is therefore recommended that such moist forages be used directly or after freezing in liquid nitrogen without either oven-drying or freeze-drying. ((C) Elsevier / Inra).