The aim of the present article is to evaluate critically the various methods employed in studies designed to quantify precisely skeletal muscle substrate utilization during exercise. In general, the pattern of substrate utilization during exercise can be described well from O2 uptake measurements and the respiratory exchange ratio. However, if the aim is to quantify limb or muscle metabolism, invasive measurements have to be carried out, such as the determination of blood flow, arterio-venous (a-v) difference measurements for O2 and relevant substrates, and biopsies of the active muscle. As many substrates and metabolites may be both taken up and released by muscle at rest and during exercise, isotopes can be used to determine uptake and/or release and also fractional uptake can be accounted for. Furthermore, the use of isotopes opens up further possibilities for the estimation of oxidation rates of various substrates. There are several methodological concerns to be aware of when studying the metabolic response to exercise in human subjects. These concerns include: (1) the muscle mass involved in the exercise is largely unknown (bicycle or treadmill). Moreover, whether the muscle sample obtained from a limb muscle and the substrate and metabolite concentrations are representative can be a problem; (2) the placement of the venous catheter can be critical, and it should be secured so that the blood sample represents blood from the active muscle with a minimum of contamination from other muscles and tissues; (3) the use of net limb glycerol release to estimate lipolysis is probably not valid (triacylglycerol utilization by muscle), since glycerol can be metabolized in skeletal muscle; (4) the precision of blood-borne substrate concentrations during exercise measured by a-v difference is hampered since they become very small due to the high blood flow. Recommendations are given in order to obtain more quantitative and conclusive data in studies investigating the regulatory mechanisms for substrate choice by muscle.