When performing an adjustment on metal-ceramic restorations, opaque porcelain may become exposed, particularly on the lingual surface of maxillary anterior teeth. It is generally believed that exposed opaque porcelain can be more abrasive and destructive to an opposing dentition than body porcelain, and that these teeth may require restoration as a consequence. This study compared the debonding strengths of 2 types of porcelain, with and without opaque porcelain, to 2 types of dental casting alloys. Two porcelain systems, Ceramco3 and Vita 900, and 2 metal alloys, a high noble (Encore) and a base metal (Duceranium U), were used to fabricate and test 56 flexure bars in accordance with ISO 9693:1999(E): Metal-Ceramic Dental Restorative Systems. Half of the bars received opaque porcelain prior to body porcelain additions, and the other half did not. The metal-ceramic debonding strength was determined by using a 3-point flexure apparatus and a mechanical testing device (Instron). A center load was applied at a crosshead speed of 1.5 mm/min(-1) until debonding occurred. In addition to the load-versus-displacement curve, a precision measurement microphone was used to assist in ascertaining the point in time when debonding occurred. Since the sound analysis and the mechanical test were started simultaneously, the debonding load could be more accurately determined. Data were statistically analyzed using 3-way analysis of variance, and all pairwise multiple comparisons were made with the Tukey HSD test (alpha=.05). The difference in mean debonding strength values for opaqued and non-opaqued flexure bars were statistically significant (P=.028). The mean debonding strength values (MPa) for each metal-ceramic system were as follows: Encore-Opaque-Ceramco3 (EOC), 31.43 +/- 6.92(a); Encore-Opaque-Vita 900 (EOV), 30.37 +/- 3.25(a); Duceranium U-No Opaque-Ceramco3 (DNC), 29.20 +/- 6.97(a); Duceranium U-Opaque-Ceramco3 (DOC), 26.61 +/- 4.98(a); Duceranium U-Opaque-Vita 900 (DOV), 26.15 +/- 4.29(a); Encore-No Opaque-Vita 900 (ENV), 25.45 +/- 4.04(a); Encore-No Opaque-Ceramco3 (ENC), 23.96 +/- 4.14(a); Duceranium U-No Opaque-Vita 900 (DNV), 22.88 +/- 6.15(a). Identical superscript letters denote no significant difference among groups. The precision measurement microphone resulted in selection of a debonding strength/crack initiation load that was lower than the peak load recorded during strength testing. Initial debonding during crack initiation strength testing may not always correspond to the peak load recorded, but rather to the point on the load-versus-displacement curve beyond which the relationship is no longer a straight line. Presence of opaque porcelain generally increased the debonding strength for metal-ceramic systems; however, opaque porcelain may not be necessary for a clinically acceptable metal-ceramic bond for some metal-ceramic systems.