Objectives: To describe a novel canine castless partial carpal arthrodesis plate (par-CA) and its ex vivo biomechanical comparison with T-plate and cross pinning techniques for canine partial carpal arthrodesis. Methods: The three implant systems were applied to three cohorts of six forelimbs from Greyhounds euthanatized for reasons unrelated to the study. Intercarpal and carpometacarpal palmar fibrocartilage and ligaments were sectioned. Potentiometers were applied between the radial carpal and third metacarpal bones to measure micromotion, and limbs were loaded at 30% of bodyweight at 1 Hertz for 10,000 cycles on a servo-hydraulic universal testing machine. Following assessment of micromotion, limbs were loaded to failure at 20 mm/s and ultimate strength, ultimate displacement, and stiffness were measured. Results: The T-plate (p <0.01) and par-CA (p <0.01) had reduced micromotion relative to the cross pin constructs but there was no significant difference between the control, T-plate and par-CA constructs. There was no significant difference in ultimate strength between constructs. Ultimate displacement was reduced in the plated constructs. Stiffness did not differ between constructs. Clinical significance: The novel par-CA construct was biomechanically similar to the T-plate and both were superior to cross pins in resisting micromotion. There was no difference in load at failure between constructs. The par-CA plate permits radial and ulnar carpal bone compression, a more distal location of the plate to limit impingement, and placement of screws in two metacarpal bones; features which may offer clinical benefits over T-plate fixation.