Sixteen ruminally cannulated yearling beef cattle (British x British; average BW = 295 +/- 25 kg; three steers and one heifer per treatment) grazing dormant intermediate wheatgrass (Thinopyrum intermedium Host) were allotted to four treatments: 1) no supplement (CON); 2) alfalfa hay (ALF; .52% of BW); 3) cottonseed meal (CSM; .22% of BW); and 4) corn gluten meal-wheat bran feed (CGMWBF; .36% of BW). Supplements were formulated (DM basis) so that intakes were isonitrogenous and were provided once daily (0700). Sampling periods were in February (FEB; .96% N in masticate), March (MAR; 1.06% N in masticate), and April (APR; 1.12% N in masticate) following an initial 21-d adaptation period. Daily grazing time was 1.1 to 1.5 h longer (P < .05) for CON cattle than for supplemented cattle. Forage OM intake (OMI) was not altered (P > .15) by supplemental protein; however, total OMI was greater (P < .04) for supplemented than for unsupplemented cattle. Harvesting efficiency (grams of OM intake-kilogram of BW-1.minute spent grazing-1) was greater (P < .05) for cattle fed CSM than for those fed CGF or CON; cattle fed ALF were intermediate in harvesting efficiency. Ruminal fluid kinetics, in situ rate and extent of NDF digestion, and total VFA concentration were not influenced (P > .10) by type of supplemental protein. Ruminal NH3 N concentration exhibited a treatment x sampling time interaction (P < .05). Type of protein supplement did not seem to affect most digestion measures; however, harvesting efficiency was influenced by supplement type.