Measurement of progression in amyotrophic lateral sclerosis/motor neuron disease (ALS/MND) is useful for charting the natural history and assessing efficacy in drug trials. Common measures are maximal voluntary isometric contraction (MVIC) which mainly focuses on proximal muscles, and electrophysiologic measures of the compound muscle action potential (CMAP) and motor-unit number estimation (MUNE) which focus on distal muscles. We have undertaken a study to compare the relationships between MVIC, CMAP and MUNE recorded in intrinsic hand muscles of 10 subjects with ALS/MND. Grip and pinch (between the first and fifth digits) MVIC were recorded. The CMAP and MUNE (determined by the multipoint stimulation technique) were recorded from thenar and hypothenar muscle groups. To facilitate comparisons between strength and electrophysiologic measures, the MUNE values and the CMAP values from the thenar and hypothenar muscle groups were each summed. Test-retest correlations were high for all measures (r = 0.75-0.99). Pinch and grip MVIC were highly correlated (r = 0.83). However, MVIC measures showed weaker correlations with summed MUNE values (pinch r = 0.56, grip r = 0.64) and summed CMAP values (pinch r = 0.58 and grip r = 0.65). The weak correlations between MVIC and electrophysiologic measures are due to two factors. First, grip and pinch MVIC are correlated through co-activation of agonist muscle groups, for we recorded strong concomitant muscle activity from forearm flexor muscles during pinch. In contrast, CMAP and MUNE reflect measurement from isolated muscle groups in the hand. Second, MVIC and the CMAP are affected by collateral reinnervation and only indirectly assess motor-unit loss, while MUNE is uninfluenced by reinnervation and directly addresses the degree of motor-unit loss. These factors determine the information available from endpoint measures. We offer guidelines for choosing useful measures based upon the goals of a given study.