Movements of Rhinolophus ferrumequinum in and out of the nursery roost at Stackpole (West Wales, U.K.) were monitored automatically from 1994 to 2018 with simultaneous measurements of roost and external air temperatures. Pups were counted manually in June–July and mean birth dates calculated. Maximal foraging times of the population between 16:00 h and 08:00 h and temperatures at midnight showed three types of activity. These types of activity explained why warmer springs were followed by earlier birth dates. When April was warmer the number of degree days, linked to the activity of night-flying insects, was higher so the maximal foraging times were longer. Hence, mean birth dates were earlier due to faster gestation. The indirect effect of degree days on the birth date, measured by the partial regression coefficient (ß = -0.321), was weaker than the direct effect (ß = -0.628) and the mediating effect of maximal foraging time was significant (P < 0.001). During May–June and June–July bats foraged mainly from dusk to dawn so there was little variation in the maximal foraging times of the population, and it did not significantly mediate the effect of temperature on birth date. Birth dates were later when the external temperatures in June–July were higher (ß = 0.309), but the effect was small (R2 = 9.5%). Path analysis further revealed that longer maximal foraging times of the population in April predicted the year-to-year changes in the number of births and subsequently the number of adult females. Maximal foraging times of the population in April were a major influence on birth timing and ultimately determined whether the population grew or declined.