This study evaluates the estimates of seasonal snowmelt runoff in the Sutlej, Indus, Kabul and Chenab rivers derived from the model of snow cover area vs. runoff against those obtained from cross correlation of concurrent flows in the rivers. The concurrent flow correlation model explains more than 90 per- cent of the variabililty in flow of these rivers. Compared to this model, the model of snow-cover area vs. runoff explains less of the variability in flow. However, unlike the snow-cover model, the concurrent flow correlation model cannot be used for operational forecasting procedures. Where the strength of correlation is high, the concurrent flow correlation model has potential for use in retrospective analysis of flow for estimating missing data, extending time series and for evaluating estimates derived from other models. In the Himalayan basins under study and at least for the period under observa- tion, the concurrent flow correlation model provides a set of results with which to compare the estimates from the snow cover model.