The impact of skin blood flow changes on near-infrared spectroscopy (NIRS)-derived measures of muscle oxygen saturation (SmO(2)) and blood volume has not been fully established. We measured SmO(2) and total hemoglobin concentration ([tHb]) responses of the right vastus lateralis during rest and dynamic knee extension exercise in ten young, healthy males. The protocol was repeated four times: twice without thigh heating for reliability, and twice with different grades of thigh heating for assessing the impact of cutaneous vasodilation on SmO(2) and Δ[tHb]. The reliability of our SmO(2) and [tHb] measurements was good. Thigh heating at 37 and 42°C caused marked increases in cutaneous vascular conductance (CVC) during rest and exercise (P < 0.001 between each condition), and small increases in SmO(2) during rest (from 69 ± 8% to 71 ± 7% and 73 ± 6%, respectively; P < 0.05 between each condition), but not during exercise (e.g. 1 min exercise: 51 ± 11% vs. 51 ± 11% and 52 ± 11%, respectively; P > 0.05 at all time points). In contrast, heating-induced increases in %CVC(peak) were accompanied by increases in [tHb] at rest and during exercise and a decrease in Δ[tHb] during exercise (all P < 0.05). Our findings suggest that NIRS-derived measures of SmO(2) and blood volume are differentially affected by skin blood flow at rest and during exercise. The findings from this study should be considered in NIRS experiments where skin blood flow can change markedly (e.g. high-intensity and/or prolonged exercise).