The stability of the biological reagents used in routine immunological procedures is a matter of practical interest to the public health worker, and an attempt has been made in the following research to estimate the period during which the diluted toxin used in the Schick test for the estimation of susceptibility to diphtheria remains potent. In the early days of Schick testing, although Leete (1920) reported that a diluted toxin retained its potency for as long as 5 weeks, it has hitherto been considered inadvisable to use a toxin at a longer interval than 48 hours after it had been diluted to its proper strength. Glenny, Waddington and Pope (1928) and O'Brien, Okell and Parish (1928) have introduced a new diluent consisting of a solution of a combination of crystal borax 57, boric acid 84 and sodium chloride 99 parts. A 1·5 per cent, solution of this mixture is now used as a diluent and has the effect of sustaining the potency of the diluted toxin over a greater period of time. In our work over 400 persons have been tested with toxins made up according to this formula, a comparison being made in each individual between the reactions elicited by an aged toxin, stored for varying periods at room temperature or in the cold room, with those produced by a fresh toxin. Enough toxin was made up at the beginning of the experiment to last over a considerable period of time, samples being sent to us by post periodically together with batches of freshly diluted toxin. The testing of reagents was carried out immediately on receipt, the lower age groups being selected in any series of individuals as being likely to give the largest number of reactors. The actual testing was performed practically throughout by one of us (J. C. J. McEntee) so as to maintain uniformity of technique. Each pair of toxins was controlled by a single injection of inactivated toxin and the results were scrutinised at the 4th and 10th days, a record of the comparative intensities of the positives being made at the same time. The results are recorded in Table I.