Insulin-mediated glucose clearance (GC) is diminished in type 2 diabetes. Skeletal muscle has been estimated to account for essentially all of the impairment. Such estimations were based on leg muscle and extrapolated to whole body muscle mass. However, skeletal muscle is not a uniform tissue and insulin resistance may not be evenly distributed. We measured basal and insulin-mediated (1 pmol min-1 kg-1) GC simultaneously in the arm and leg in type 2 diabetes patients (TYPE 2) and controls (CON) (n=6 for both). During the clamp arterio-venous glucose extraction was higher in CON versus TYPE 2 in the arm (6.9+/-1.0 versus 4.7+/-0.8%; mean+/-s.e.m.; P=0.029), but not in the leg (4.2+/-0.8 versus 3.1+/-0.6%). Blood flow was not different between CON and TYPE 2 but was higher (P<0.05) in arm versus leg (CON: 74+/-8 versus 56+/-5; TYPE 2: 87+/-9 versus 43+/-6 ml min-1 kg-1 muscle, respectively). At basal, CON had 84% higher arm GC (P=0.012) and 87% higher leg GC (P=0.016) compared with TYPE 2. During clamp, the difference between CON and TYPE 2 in arm GC was diminished to 54% but maintained at 80% in the leg. In conclusion, this study shows that glucose clearance is higher in arm than leg muscles, regardless of insulin resistance, which may indicate better preserved insulin sensitivity in arm than leg muscle in type 2 diabetes.