MRI is an ideal method for identifying areas of muscle atrophy and fatty infiltration. Studies comparing clinical and MRI features of foot and leg muscle atrophy in Charcot-Marie-Tooth disease type 1A (CMT-1A) duplication are lacking. The aim of this study is to describe clinical and MRI patterns of lower limb amyotrophy in CMT-1A. A total of 10 secondary CMT-1A patients and 1 proband patient with de novo mutation were prospectively evaluated. Ages of patients ranged from 8 to 61 years (median, 24). Disease severity in terms of ability to walk and run was established using a nine-point functional disability scale (FDS). We administered the CMT neuropathy score (CMTNS), based on patient's symptoms, neurological examination and neurophysiological testing. Muscle strength of flexo-extensor ankle and toe muscles was assessed manually with the standard Medical Research Council scale. In all 11 patients, leg MRI study included T1- and T2-weighted spin-echo sequences in coronal and axial planes, and a T1-weighted spin-echo sequence with chemical sift fat suppression before and after paramagnetic contrast agent injection. In seven patients both feet were simultaneously studied in coronal and axial planes. Six patients had pes cavus, an FDS score of 0 (normal), mild CMTNS and normal muscle power of foot flexo-extensors. In these six patients, MRI showed muscle fatty infiltration of intrinsic foot muscles mainly involving the lumbricals, all four leg muscle compartments being preserved. The remaining five patients had FDS scores from 1 (cramps or fatigability) to 3 (walking difficulty), mild to moderate CMTNS and variable weakness of peroneal musculature. In these five patients MRI showed, besides intrinsic foot muscle involvement, variable and distally accentuated fatty infiltration of the lateral, anterior and superficial posterior leg muscle compartments and, to a lesser degree, of the deep posterior compartment. In four patients muscle oedema and post-contrast enhancement was noted. MRI demonstrated fatty infiltration of clinically normal muscles. We conclude that clinical-MRI patterns of lower limb muscle atrophy vary with evolution of semiology. Selective involvement of intrinsic foot muscles is the characteristic pattern of CMT-1A cases with minimal disease signs. Afterwards this pattern usually combines variable involvement of leg muscles. Our findings help to clarity the pathogenesis of pes cavus in the disease.