Despite their high degree of safety, beta-lactam agents cause adverse reactions. This article deals with the types of adverse reactions to various beta-lactam agents that have been reported and, especially, with the difficulties involved in monitoring and evaluating these reactions. Comparisons of the results of studies of the same drug carried out in various countries show striking differences in the incidence of adverse effects reported. A reason for this variation is the lack of strict definitions of the events that should be regarded as adverse effects. Only rarely are all the unexpected events occurring in patients in clinical studies reported, but the investigators are allowed to make subjective judgments about which reactions are related to the drug tested and are thus reportable. With such procedures there is an inherent risk of overlooking unexpected adverse effects. Also, in the analysis of laboratory adverse reactions the rule that "one finds what one looks for" applies. It seems obvious that if stricter rules for registration of adverse effects had been applied, the toxic effects of various beta-lactams, such as the nephrotoxicity of cephaloridine and the coagulopathies associated with moxalactam, would have been detected much earlier.