Trunk muscles receive corticospinal innervation ipsilaterally and contralaterally and here we investigate the degree of ipsilateral innervation and any cortical asymmetry in pairs of trunk muscles and proximal and distal limb muscles. Transcranial magnetic stimulation (TMS) was applied to left and right motor cortices in turn and bilateral electromyographic (EMG) recordings were made from internal oblique (IO; lower abdominal), deltoid (D; shoulder) and first dorsal interosseus (1DI; hand) muscles during voluntary contraction in ten healthy subjects. We used a 7-cm figure-of-eight stimulating coil located 2 cm lateral and 2 cm anterior to the vertex over either cortex. Incidence of ipsilateral motor evoked potentials (MEPs) was 85% in IO, 40% in D and 35% in 1DI. Mean (+/- S.E.M.) ipsilateral MEP latencies were longer ( P<0.05; paired t-test) than contralateral MEP latencies (contralateral vs. ipsilateral; IO: 16.1+/-0.4 ms vs. 19.0+/-0.5 ms; D: 9.7+/-0.3 ms vs. 15.1+/-1.9 ms; 1DI: 18.3+/-0.6 ms vs. 23.3+/-1.4 ms), suggesting that ipsilateral MEPs were not a result of interhemispheric current spread. Where data were available, we calculated a ratio (ipsilateral MEP areas/contralateral MEP areas) for a given muscle (IO: n=16; D: n=8; 1DI: n=7 ratios). Mean values for these ratios were 0.70+/-0.20 (IO), 0.14+/-0.05 (D) and 0.08+/-0.02 (1DI), revealing stronger ipsilateral drive to IO. Comparisons of the sizes of these ratios revealed a bias towards one cortex or the other (four subjects right; three subjects left). The predominant cortex showed a mean ratio of 1.21+/-0.38 compared with 0.26+/-0.06 in the other cortex ( P<0.05). It appears that the corticospinal control of IO has a strong ipsilateral component relative to the limb muscles and also shows hemispheric asymmetry.