Use of seeds from simulated herbaceous canopy by western harvest mice (Reithrodontomys megalotis), deer mice (Peromyscus maniculatus) and white-footed mice (Peromyscus leucopus) was examined in laboratory trials. Each mouse was tested on both simulated full-moon and new-moon nights but with only one of three abundances of surface seeds (high, low or zero abundances of seeds in trays on arena floor). White-footed mice are more total seeds (surface plus canopy seeds) during dark-bright than bright-dark trials and with high than low or zero abundances of surface seeds. However, neither the order of dark and bright conditions nor availability of surface seeds influenced total seeds eaten by deer mice and harvest mice. All three species readily foraged in the canopy in the absence of surface seeds (>90% of trials on both dark and bright nights). When surface seeds were abundant, interspecific differences in canopy foraging were evident as canopy foraging occurred during 54% of trials on both dark and bright nights for harvest mice, 58% and 21% of trials on dark and bright nights, respectively, for white-footed mice and 8% and 0% of trials on dark and bright nights for deer mice.