Objectives. This study evaluated: (1) the marginal fit of copings fabricated from commercially pure titanium (cp Ti) and titanium-aluminum-vanadium alloy (Ti-6Al-4V) after each firing cycle for a recommended porcelain; and (2) the effectiveness of a stress-relieving treatment on the fit. The results were compared with copings made with Pd-Ag alloy submitted to the recommended porcelain firing protocol. Methods. Bovine teeth were machined for metal-ceramic crowns and die casts were obtained for waxing patterns. The firing cycle of titanium porcelain was performed on cp Ti and Ti-6Al-4V copings obtained using the Rematitan System. A stress-relieving treatment (cp Ti treated and Ti-6Al-4V treated groups) was performed on original invested castings at the first porcelain firing temperature (800 degreesC). Fit (mum) was measured in the 'as cast' condition and after each porcelain firing cycle with a traveling microscope. The values were compared using the analysis of variance and Tukey's test (P < 0.05), the Pd-Ag group was compared to titanium groups using Dunnett's test (P < 0.05). Results. The marginal fit values in the as-cast condition (initial) were 84.43 mum for Ti-6Al-4V not treated, 76.35 mum for Ti-6Al-4V treated, 86.09 mum for cp Ti not treated, 97.66 mum for cp Ti treated and 58.11 mum for Pd-Ag group. The Ti-6Al-4V, cp Ti, and Pd-Ag groups exhibit significant differences in marginal fit values when the as-cast condition (initial) was compared with those after the recommended porcelain firing protocol (by Tukey's test at 5% probability level). No significant differences were observed between stress-relieving treated and not treated groups. Significance. The firing cycles used for porcelain build-up may worsen the fit of titanium copings to values that suggest no inferences to the detriment of clinical applications. (C) 2003 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.197686691