End-constraint is needed when performing tensile and compressive testing of cancellous bone so that frictional and structural end-effects are minimized and the mechanical properties are not underestimated. Since many end-constrained testing techniques are limited to certain sites and species by the large specimen sizes required, reductions in overall specimen length would be advantageous. The following study examines the effect of specimen gauge length and aspect ratio on the elastic tensile testing of cancellous bone from the bovine femoral condyle. Three different nominal gauge lengths (10, 16, 22 mm) were examined at two different diameters (5.1, 10.7 mm) which bound the range typically used for cancellous bone testing. No effect of gauge length or aspect ratio was observed, but evidence of a diameter effect was noted. Three-dimensional finite element method (FEM) models confirmed the experimental observations and demonstrated that an overestimation of modulus begins to occur for specimens shorter than 10 mm.