The bone-implant bond strength to hydroxyapatite-coated cylindrical Ti-6A1–4V implants was evaluated in torsional tests using a canine model. Experimental hydroxyapatite (HA) coatings were deposited using pulsed laser deposition (PLD) under conditions that yield highly crystalline, phase-pure coatings on Ti-6A1–4V. Commercially available plasma-sprayed (PS) HA coatings of lower crystallinity and purity served as positive controls (for bone bonding) and uncoated polished and grit-blasted Ti-6A1–4V samples as negative controls. The torsional shear strength of the PLD HA-coated specimens implanted in cancellous bone was substantially lower than that of the PS HA-coated implants, at both 2 and 6 weeks post-implantation. Within statistical limits, the shear strength of the PLD HA-coated samples was similar to that of the polished Ti-6A1–4V control implants and lower than the grit-blasted Ti-6A1–4V control implants. Examination of PLD HA-coated samples after mechanical testing revealed many areas from which the coating had detached. Previous results from laboratory solubility studies showed that (a) PLD films released less calcium and phosphate than the PS controls, (b) rates of dissolution were much lower and (c) the PLD coatings contained fewer decomposition products. Taken with these previous findings, the present results suggest that the surface morphology and/or grain size of implants may be more important for bone bonding than the phase purity of hydroxyapatite coatings. Furthermore, the local chemical environment resulting from the dissolution of the more soluble phases present in the PS HA coatings may enhance bone bonding.