“…For an isotropic solid in tension or compression, the decay length is usually equivalent to the maximum cross-sectional dimension; however, the decay length is not determined by macroscopic specimen dimensions alone. Shorter decay lengths occur in thin-walled cylinders (Stephen and Wang, 1992) and piezo-electric materials (Borrelli and Patria, 1999), whereas longer lengths are reported for anisotropic materials (Arridge and Folkes, 1976), composites (Horgan, 1982) and negative Poisson's ratio foams (Lakes, 1992). Since the present study showed that the extensometer can be placed much closer to the loaded ends than would be expected based on the macroscopic specimen diameters, it appears that the very short decay length found in cancellous bone is controlled by a much smaller characteristic dimension, reported here as about four trabecular units.…”