Components of the zooplankton production cycle (assimilation, respiration, and predation) on Georges Bank were analyzed using three models of increasing complexity, i.e., total biomass (Riley, 1947), size-structure (Huntley and Boyd, 1984) and species population (Davis, 1984a). Other temperate marine areas with seasonal cycles in temperature similar to Georges Bank were identified and available data on zooplankton biomass and species cycles were found. For Georges Bank, the species level model, based on well-defined empirical relationships between temperature, development rate, and fecundity, revealed that highest production occurs by small animals during warmest months (e.g., September), when biomass is low. By contrast, the total biomass model calculated maximal production during June, coinciding with peak herbivore mass, but could not account for high fall production because size-dependent effects were not considered. The size-structured model computed high fall production rates but greatly overestimated June production by failing to consider life-history characteristics of component species (i.e., diapause in Calanus). In general, it was found that the temperature cycle has a much greater influence on production rates than does food concentration. Zooplankton cycles from the North Sea, Japan Sea, and Argentinian Shelf were similar to Georges Bank, suggesting that the modeling results may be generally true for temperate areas having large annual ranges in temperature (> 10°C). The traditional view that temperate zooplankton production is primarily food-limited and occurs mainly during spring/summer coinciding with peak herbivore stock can no longer be considered valid. Instead the cycle appears largely driven by temperature so that production is highest during summer/fall when biomass is low and small warm-water species dominate. The misconception of the production cycle has resulted from use of oversimplified models.