The combined arm-leg (Cruiser) ergometer is assumed to be a relevant testing and training instrument in the rehabilitation of patients with a lower limb amputation. The efficiency and submaximal strain have not been established and thus cannot be compared with alternative common modes of exercise. A total of 22 healthy able-bodied men (n=10) and women (n=12) were enrolled in four discontinuous submaximal graded exercise tests. Each test consisted of seven bouts of 3 min exercise ranging from 20 to 45 W and was performed on, respectively, the Cruiser ergometer, a bicycle ergometer, a handbike, and again the Cruiser ergometer. Cardiorespiratory parameters were measured and rate of perceived exertion was determined. Gross mechanical efficiency (GE) was determined from power output and submaximal steady-state energy cost. Repeated-measures analysis of variance (P<0.05) was used to evaluate the effects of exercise mode, exercise intensity, and sex. No differences in GE and cardiorespiratory strain were found between both Cruiser tests (GE 45 W: men 13.0%, women 15%) and the bicycle test (GE 45 W: men 13.2%, women 14.6%). GEs of handbiking (45 W: men 11.2%, women 12.2%) were lower compared with the Cruiser and bicycle test results, whereas cardiorespiratory strain in handbiking was consistently higher. Apart from a lower rate of perceived exertion at the second Cruiser test, no differences were found between the repeated Cruiser tests. It can be concluded that GE and cardiorespiratory strain in submaximal Cruiser exercise are comparable with leg cycling, the repeatability was good, and no obvious learning effects were observed. The results of this study form a base for further research in patients with a lower limb amputation.