It has been suggested that corticomotoneuronal drive to ankle dorsiflexors is greater than to ankle plantar flexor muscles, despite the finding that plantar flexors are no less active than TA during walking and standing. The present study was undertaken to determine whether there was differential involvement of distal lower limb muscles in amyotrophic lateral sclerosis (ALS), to elucidate pathophysiological mechanisms of selective muscle involvement. Prospective studies were undertaken in 52 ALS patients, including clinical assessment, disease staging (revised ALS functional rating scale), Medical Research Council sum score, and a scale of upper motor neurone (UMN) dysfunction. Motor unit number estimates (MUNE) and compound muscle action potentials (CMAP) from ankle dorsiflexors and plantar flexors were used to provide objective measures. A novel 'split leg index' was calculated as follows: SLI = CMAPDF ÷ CMAPPF. In ALS, there was significantly greater reduction of MUNE and CMAP amplitude recorded from plantar flexors when compared to dorsiflexors, suggesting preferential involvement of plantar flexor muscles, underpinning a 'split leg' appearance. The SLI correlated with clinical plantar flexor strength (R= -0.56, p < 0.001). In no patient did the SLI suggest preferential dorsiflexor involvement. In subgroup analyses, mean SLI was greatest in lower limb-onset ALS. In conclusion, the present study has established dissociated involvement of muscles acting around the ankle in ALS. We suggest this reflects underlying differences in cortical, descending or local spinal modulation of these muscles.