During the winters of 1990, 1991 and 1992, a field study of stability parameters for forecasting slab avalanches was conducted in the Cariboo and Monashee mountains of western Canada. In a level study plot at 1900 m and on nearby slopes, the shear strength of the weak snowpack layer judged most likely to cause slab avalanches was measured with a 0.025 m2 shear frame and a force gauge. Based on the ratio of shear strength to stress due to the snow load overlying the weak layer, a simple stability parameter and a more theoretically based stability index which corrects the strength for normal load were calculated. These stability parameters are compared with avalanche activity reported for the same day within approximately 30 km of the study plot. Each stability parameter is assessed on the basis of the number of days that it successfully predicted one or more potentially harmful avalanches and the number of days that it successfully predicted no potentially harmful avalanches. Both parameters predicted correctly on at least 75% of the 70 days they were evaluated. The simpler empirical stability parameter worked as well as the one that corrects strength for normal load. For large-scale forecasting of dry-snow slab avalanches, shear frame stability parameters appear to be a useful addition to meteorological data, snowpack observations and slope tests.