Five ruminally cannulated Hereford steers (average weight 340 kg) were used in a 5 x 5 Latin square design experiment to examine the influences of dietary sodium bentonite (SB), McDougall's buffer salts (MB), cement kiln dust (KD) and clinoptilolite (CLN) on rumen liquid dilution rate, fermentation patterns and roughage intake. Treatments were administered as 350-g additions to a basal protein/energy supplement of 2.62 kilograms. The basal diet of cottonseed hulls was offered ad libitum. Mineral analyses of the four treatment substances revealed fairly high levels of potentially toxic trace minerals (e.g., Pb, Cd and F) in the KD. Intake of cottonseed hulls was not significantly affected by treatment but tended to be higher for MB-supplemented steers (7.19 kg) than for those given the control or other treatments (average 6.72 kg). Some rejection of supplements containing MB and KD was noted. At e hr after supplement feeding rumen pH of MB-treated steers (6.68) was higher (P less than .05) than that of steers fed KD (6.50) and CLN (6.44) but was not different from that od control or SB-treated steers (6.56). Seven hours after supplement was offered rumen pH of MB steers (6.74 was higher (P less than .05) for steers fed MB, KD and CLN than for SB-an control-fed steers. Volatile fatty acids, rumen ammonia, osmolality and protein and ash contents of a crude bacterial fraction were not significantly influenced by mineral compounds, nor was rumen liquid volume. Values for rumen liquid volume were as follows: MB (60.27 liters), SB (50.43 liters), CLN (45.53 liters), control (43.14 liters) and KD (41.81 liters). Liquid dilution rate (percentage per hour) was high for steers on all treatments (average of 9.11) and was not significantly altered by mineral supplementation.