The purpose of this study was to document and compare the architectural parameters (fibre bundle length, angle of pennation) of human skeletal muscle in cadaveric specimens and live subjects. The medial (MG) and lateral (LG) gastrocnemius, and posterior (PS) and anterior (AS) soleus were examined bilaterally in 5 cadavers (mean age 72.6, range 65–83 y) and 9 live subjects (mean age 76.3, range 70–92 y). Data were obtained from direct measurement of cadaveric specimens and from ultrasonographic scans of the live subjects. In cadaveric muscle, fibre bundles were isolated; their length was measured in millimetres and pennation angles were recorded in degrees. In live muscle, similar measurements were taken from ultrasonographic scans of relaxed and contracted muscle. For the scans of relaxed muscle, subjects were positioned prone with the foot at a 90° angle to the leg, and for scans of contracted muscle, subjects were asked to sustain full plantarflexion during the scanning process. Fibre bundle length and angle of pennation were compared at matched locations in both groups. It was found that the relationship between cadaveric and in vivo values for fibre length and angle of pennation varied between muscle parts. The cadaveric architectural parameters did not tend to lie consistently towards either extreme of relaxation or contraction. Rather, within MG, PS and AS, cadaveric fibre bundle lengths lay between those for relaxed and contracted in vivo muscle. Similarly both the anterior and posterior cadaveric fibre angles of pennation lay between the in vivo values within LG and PS. In summary, architectural characteristics of cadaveric muscle differ from both relaxed and contracted in vivo muscle. Therefore, when developing models of skeletal muscle based on cadaveric studies, the architectural differences between live and cadaveric tissue should be taken into consideration.