This study tests five mechanisms for the coexistence of red squirrels (Tamiasciurus hudsonicus) and eastern chipmunks (Tamias striatus). These mechanisms, based on optimal foraging theory, depend on differences between the two species in habitat over space and time. The mechanisms are related to: 1) habitat; 2) microhabitat; 3) foraging and travel costs; 4); seasonal variation in predation risk; and 5) seasonal variation in thermoregulatory costs. Each species' foraging efficiency was estimated, using GUDs (Giving-up-densities) from February to July 1995 at the Morgan Arboretum, Sainte-Anne-de-Bellevue, Quebec, Canada. GUDs were measured in 36 aluminum seed trays, each containing 20 g of sunflower seeds, mixed with 2.25 l of kiln dried sand. Trays were set in three habitats (coniferous, deciduous and mixed forest), separated by at least 300 m with 6 pairs of trays per habitat. In general, the chipmunk was more efficient (left lower GUDs) than the red squirrel. The red squirrel was more efficient than the chipmunk in the coniferous forest during the spring and had exclusive access to food during the winter when the chipmunk was in hibernation. The chipmunk was the more efficient forager in the mixed forest in both spring and summer. This spatio-temporal trade-off in foraging efficiency constitutes a mechanism which can promote coexistence between these two diurnal rodents.