Objective: To characterize the torsional structural properties of the feline femur and design a bone model surrogate for mechanical testing of feline orthopedic implants. Study design: Experimental. Sample population: Paired feline femurs (n = 30) and bone models (8 materials, n = 4/group).Methods: Femurs were cyclically tested nondestructively in torsion and then loaded to failure. A generic femoral model was then designed from native femur dimensions and tested similarly by using 1 of 8 materials that were 3-dimensionally printed or machined. Outcome measures consisting of torsional compliance, angular deformation (AD), and torque to failure were compared by using Student's t test (P < .05). Failure modes are reported as descriptive statistics. Results: Torsional compliance (1.6 AE 0.3 /Nm, 2.0 AE 0.1 /Nm), AD (3.1 AE 0.6 , 3.8 AE 0.2 ) and torque to failure (7.8 AE1.2 Nm, 8.1 AE 1.3 Nm) did not differ between feline femurs and short-fiber epoxy (SFE) models. Conversely, most printed materials displayed excessive TC and failed by plastic deformation (AD > 15-fold that of native femurs) rather than by fracture. Feline bone and SFE both failed by spiral fractures. Conclusion: None of the outcome measures differed between the 4th generation SFE model and cadaveric femurs, but differences were identified between feline bone and printed materials. Clinical impact: Machined SFE can be used to create a surrogate bone model with torsional structural properties similar to those of feline femurs. In contrast, common printable materials appear unsuitable to produce a realistic feline bone surrogate.