Experiments aimed at elucidating the identity of antigenic metabolites of practolol, using in vitro generated practolol metabolites and sera from practolol patients, were unsuccessful as none of the sera tested contained measurable concentrations of antimetabolite antibodies. Collaborative experiments with workers who had originally established the technique also failed to detect such antibodies. It was concluded, after follow up studies, that the active sera must have been used up or damaged during the original investigations. Rabbits and guinea-pigs, injected with the protein bound practolol metabolites did not respond by synthesising antimetabolite antibodies. The probable reason for the lack of response was the low concentration of hapten binding (1 metabolite molecule/6 protein molecules). A ratio of at least 10:1 is normally used in such experiments and ratios of greater than 100:1 are not uncommon. Neither in vitro generated metabolites of practolol nor chemical analogues, had any effect on human skin fibroblast growth or collagen synthesis in vitro. In contrast, practolol, propranolol and paracetamol all inhibited these fibroblasts functions in a dose related fashion. Cells from uninvolved skin of a psoriasis patient were more sensitive to the inhibitory actions of the two 3-receptor blocking drugs than fibroblasts obtained from a control volunteer but were less sensitive to paracetamol indicating a variation in the response to 3-receptor blocking drugs and that such changes in sensitivity need not be paralleled by chemically related compounds such as paracetamol. Practolol was not taken up by the fibroblasts to any great extent indieating that its site of action was the plasma membrane. Uptake of leucine was inhibited to the same extent as collagen synthesis suggesting that practolol may interfere with protein synthesis by limiting substrate availability. In vitro morphological studies are consistent with the idea that the three drugs act by different mechanisms although further studies are necessary to confirm this. The following conclusions can be drawn from the experimental findings. - The side effects of practolol are more likely to have been due to the parent molecule than to a metabolite. - The action of practolol is likely to have been an inhibitory, rather than a stimulatory, one. - Susceptible patients have an increased sensitfvity to practolol which could be reflected in the response of fibroblasts from such patients in vitro.