1. The excretion in the bile and urine of intraperitoneally injected (14)C-labelled monoquaternary ammonium or pyridinium cations was measured in bile-duct-cannulated rats (ten compounds) and in guinea pigs and rabbits (six compounds). 2. Seven of these, namely N-methylpyridinium, tetraethylammonium, trimethylphenylammonium, diethylmethylphenylammonium, methylphenyldipropylammonium, dibenzyldimethylammonium and tribenzylmethylammonium, were excreted largely unchanged in the bile and urine. 3. 3-Hydroxyphenyltrimethylammonium, 3-bromo-N-methylpyridinium and cetyltrimethylammonium were metabolized to an appreciable extent in the rat. 4. In intact rats intraperitoneally injected trimethylphenylammonium (mol.wt. 136) was excreted mainly in the urine, dibenzyldimethylammonium (mol.wt. 226) was excreted in roughly equal amounts in the urine and faeces, and tribenzylmethylammonium (mol.wt. 302) was excreted mainly in the faeces. The faecal excretion of these compounds corresponded to their biliary excretion in bile-duct-cannulated rats. About 3-4% of tribenzyl[(14)C]methylammonium was eliminated as (14)CO(2). 5. In rats the extent of biliary excretion of four cations with molecular weights in the range 94-164 was less than 10% of the dose, whereas that of five cations with molecular weights 173-302 was greater than 10%. These results and other data from the literature suggested that the molecular weight needed for the biliary excretion of such cations to an extent of 10% or more of the dose was about 200+/-50. Studies with six cations in guinea pigs and rabbits suggest that this value applies also to these species. 6. The results suggest that the threshold molecular weight for the appreciable (>10%) biliary excretion of monoquaternary cations is different from that for anions (Millburn et al., 1967a; Hirom et al., 1972b). With rats, guinea pigs and rabbits, no significant species difference was noted, whereas with anions there is a marked species difference.