In this study the effect of time on the marginal fracture behaviour of amalgam was determined. Six alloys in clinical service for up to 6 years were used as a data base and occlusal photographs were categorized according to a standard set of photographs depicting varying degrees of marginal fracture. The standard set of photographs was assigned a linear numerical scale, and the index of extent of fracture was considered to be an arithmetic mean based on this linear scale. An observed patient effect and unequal sample size were considered in the determination of both the mean and variance. The results of applying these procedures to the data of this study showed that marginal fracture increased with time for all alloys; that the relative ranking of alloys did not change with time; and that, with certain limitations, early time evaluations can predict long-term behaviour.

The relationship of clinical marginal fracture to creep and zinc content of amalgam alloys was examined. Using multiple regression analysis, a predictive equation was found which produced an excellent fit to the experimental data (R2 = .944).

An analysis of the relationship between creep and microstructural characteristics of several gamma2-containing amalgams showed the grain size of the Ag-Hg phase (gamma1) to be a predominant factor influencing creep. When gamma1 grain size is increased, creep is reduced.

In this study of the marginal fracture of clinical amalgam restorations, it was found that restoration class or size did not influence fracture behavior, and a contralateral pair design was not found to be necessary. However, a slight effect was found for tooth type.

No abstract