A previously described in situ growth model was used to predict the summer growth of age-1 and age-2 broad whitefishCoregonus nasusin the Sagavanirktok River delta near Prudhoe Bay, Alaska, during 1990–1992. Comparisons of observed versus predicted weekly growth increments over the entire 9-year study (1982 and 1985–1992) yieldedr= 0.789 (N= 77;P< 0.01) for age- I fish andr= 0.522 (N= 70;P< 0.01) for age-2 fish. Based on daily comparisons of observed versus predicted mean fork length, the model was able to predict growth in 1991 and 1992 with mean squared errors (MSE) of 3.0 and 1.8 mm for age- I fish and 6.5 and 21.2 mm for age-2 fish. Results were similar to those previously described for 1982 and 1985–1989. The model failed noticeably in 1990 for age-1 fish (MSE = 48.6), and model fit for age-2 fish (MSE = 29.9) was the poorest of the 9 years examined. Poor model fit for both year-classes in 1990 was attributed to a divergence of observed from predicted growth; observed growth began lagging predicted growth about the third week of July. By the end of August, predicted mean length exceeded observed length by 13.6 mm for age-1 fish and 12.8 mm for age-2 fish. Data suggest that the divergence in growth began at the time of a massive immigration of least ciscoesC. sardinellainto the delta. We speculate that the presence of least ciscoes may have overtaxed the trophic carrying capacity of the delta, thereby curtailing the growth of young broad whitefish.