A series of experiments were conducted to measure and study the factors affecting the rate of outflow of dietary proteins from the rumen of sheep and of lactating dairy cows. Cr-treated protein supplements were used to measure their outflow using the marker concentration in faeces. Cr-treatment had no effect on the particle size or specific gravity of protein supplements. The amount of N in fish meal flowing into the duodenum of sheep fitted with re-entrant cannulae was closely predicted from the combined estimates of outflow and degradation in the rumen. The rate of outflow of Cr-treated protein supplements from the rumen of sheep and cows was significantly increased by increasing the feed intake. An increase in the proportion of hay in a high concentrate diet (0.25 hay:0.75 concentrate) significantly (P40.05) increased the outflow from the rumen of cows, but the further addition of hay (above 0.5:0.5) had no significant effect on outflow. Fine grinding (5 mm) and pelleting of grass significantly (P<0.001) reduced the outflow from the rumen of cows, but grinding through 40 or 20 mm screens had no significant effect on outflow. More frequent feeding of a completely mixed hay and concentrate diet or the concentrate fraction had no significant effect on the outflow from the rumen of cows. The addition. of NaCl to a completely mixed hay and concentrate diet significantly increased water intake (P40.001) with no significant effects on outflow from the rumen, milk yield or milk composition in cows. The excretion of Cr-treated fish meal in the faeces of sheep was best described by an equation with two rate constants, and that of Cr-treated grass was better described by an equation with 3 rate constants. Our results suggested that it is better to describe the kinetics of outflow of particulate matter through the alimentary tract of ruminants by the use of retention times to avoid the errors in the interpretation of different rate constants.