For moderate work rates (i.e. below the lactate threshold, theta), oxygen uptake (Vo2) approaches the steady state mono‐exponentially. At higher work rates, the Vo2 kinetics are more complex, reflecting the delayed superimposition of an additional, slow component. The mechanisms of this ‘slow’ component are poorly understood. It has been demonstrated, however, that while a prior bout of supra theta L cycling (with a 6 min recovery) does not significantly affect the V02 time course for a subsequent sub‐theta L bout, it significantly speeds the V02 response to a subsequent supra‐theta L bout (Gausche, Harmon, Lamarra & Whipp, 1989; Gerbino, Ward & Whipp, 1996). These investigators proposed that this speeding was a result of improved muscle perfusion during the exercise transient, possibly related to the residual metabolic acidaemia still present at the start of the subsequent exercise bout. To determine whether speeding of the V02 kinetics could also be induced by a bout of prior high‐intensity exercise performed at a remote site (e.g. the arms), subjects each performed two 6 min bouts of high‐intensity cycling (leg exercise: LE) at a work rate equivalent to 50% of lambda le' (the difference between maximum V02,LE and theta L,LE). On one occasion this was preceded by a 6 min period of cycling at 50% lambda LE and, on another, by a similar period of arm‐crank exercise (arm exercise: AE) at 50% lambda LE in each case, the work bouts were separated by 6 min of unloaded pedalling. Pulmonary gas exchange variables were derived breath‐by‐breath. During unloaded pedalling and at minute 6 of each work bout, arterialized venous blood samples were drawn from the dorsum of the heated hand or at the wrist for analysis of PH, lactate, pyruvate, noradrenaline (NAdr), adrenaline (Adr), and potassium (K+). The difference in V02 between minute 6 and 3 of each work bout (lambda V02 inverted question mark6‐3] and the ‘partial’ O2 deficit (O2 Def) provided indices of the slow phase of V02 kinetics. The initial AE and LE bouts resulted in similar degrees of metabolic (lactic) acidaemia; the residual acidaemia at the end of the subsequent 6 min recovery phase was also similar for the two protocols, as were [K+], [Adr inverted question mark and [NAdr]. The subsequent LE bouts were associated with reductions in both lambda V02[6‐3] and O2 Def, relative to control, with the effect being more marked when the work was preceded by a prior LE bout than a prior AE bout: lambda V02[6‐3] averaging 32 and 56% of control, respectively, and O2 Def 71 and 81%. Consequently, the increase in [lactate] and decrease in PH induced in this second LE bout were smaller when preceded by prior leg exercise than prior arm exercise. It is therefore concluded that while metabolic acidaemia induced at a site remote from the legs is associated with a less prominent slow phase of the V02 kinetics for high‐intensity leg exercise, a component specific to the involved contractile units appears to exert the dominant effect. The mechanisms underlying this response are, however, presently uncertain.