Bumblebees, when foraging for nectar in flowers of Aconitum columbianum, do not consistently drain them of reward. The amount of residual nectar varies among sites, times and species of bee. At times, residual nectar can be a significant percentage of total 24 h production strongly suggesting that, if nectar secretion rates are to be studied, flowers must have their standing crop of nectar drained by hand; assuming that a recently visited blossom is empty can lead to erroneous nectar production values. The fact that residual nectar exists is consistent with predictions of optimal foraging theory but more detailed work is necessary before it can be concluded that the volumes left behind in A. columbianum flowers are optimal.