As it has been suggested that blood glucose might play a role in the action of some cognitive enhancing drugs, the influence of glucose containing drinks on human memory was examined. In a double-blind study the influence was examined of a drink containing 50 g glucose, or a placebo, on the ability to recall a word list. There was a significant correlation between blood glucose values and the number of words recalled. Those whose blood glucose levels were increasing remembered significantly more words than those whose blood glucose levels were falling. No relationship was found between blood glucose and performance on a test of spatial memory. In a second study blood glucose levels were raised for 2 h by taking a series of glucose-containing drinks. The number of words recalled from a word list correlated significantly with blood glucose levels but not with recall of a Wechsler story. The glucose-induced improvement in memory did not occur only in those whose blood glucose levels were initially low; rather it occurred irrespective of initial blood glucose level.