OBJECTIVE Muscle forces are not directly measurable without invasive methods (i.e., tendon force transducers). Techniques such as dynamic electromyography are therefore required to obtain insight into the role of muscles during motion. There is controversy about the choice of recording electrode type. Surface electrodes are noninvasive and allow recording over a large area yet may allow intramuscular crosstalk. Indwelling electrodes also have been used in gait analysis. This electrode type is able to analyze the electrical activity of small or deep muscles. Despite the advantages of intramuscular electrodes for some applications, this electrode type is often rejected because of the need for skin penetration and for specialized personnel to acquire the data. The reliability of the wire electromyographic signal has also been questioned. The objective of this study was to determine if there were differences in the test-retest reliability of surface vs. intramuscular electrodes in gait analysis. METHODS Surface and intramuscular electromyographic data were obtained from the soleus muscle in 18 adults with no history of neuromuscular disease as they performed self-selected speed walking. A statistical criterion (variance ratio) was used to measure the reproduction of phasic patterns of muscle activity with both wire and surface electrodes on repeat-day testing of the soleus muscle. RESULTS Each electrode type was remarkably consistent. Mean variance ratio values for wire electrodes (mean = 187) were slightly lower than mean surface electromyographic variance ratio values (mean = 199). These differences did not approach statistical significance (P = 0.768). CONCLUSIONS The study results show that the dynamic electromyographic signal obtained with wire electrodes is, at minimum, as repeatable as surface electromyograms.