The usual method for observing the anti-arsenical activity of dithiols has been to inject two groups of animals with a lethal dose of the arsenical, to treat one group with a dithiol and to compare the mortalities in the two groups (Stocken and Thompson, 1946;. As Box and Cullumbine (1947) have pointed out, the use of a quantal response (percentage mortality) for quantitative studies of this sort has considerable disadvantages: the dose range in which there is a suitable mortality is small; large numbers of animals must be used to give accurate estimates of mortality; the accuracy of the estimates depends on the observed mortalities and must be weighted accordingly; and it is difficult or impossible to introduce a number of factors into the experiment and to detect interactions between them. Box and Cullumbine suggested that similarly useful information might be obtained much more efficiently by using the times of survival after a lethal dose of poison instead of the mortalities. Weatherall (1945, 1949) observed that animals poisoned with oxciphena-sine and treated with doses of dithiols insufficient to save life lived longer than control animals given no dithiol, and showed that in the controls the log survival time was roughly linearly related to the log dose of oxop.>enarsi ie. It was therefore decided to investigate the relation between dose of arsenical and survival time more fully, and to study the effect of dithiols on this relation, in order to see whether a more satisfactory method of measuring anti-arsenical activity could be so obtained.
MATERIALS AND METHODSSolutions of phenylarsenoxide were prepared by dissolving crystalline phenylarsenoxide (m.p. 159-161 "C.) in 3N sodium hydroxide, neutralizing with 6N hydrochloric acid as nearly as possible without precipitation, and diluting with water until isotonic and then with 0.9 per cent (w/v) sodium chloride solution. Solutions of oxophenarsine (mapharside) were prepared in distilled water from pure material kindly provided by Messrs. Parke Davis and Co. The dithiols used are described in *Present address-Department of Pharmacology, London Hospital Medical School, London, E.I. Table I. The preparation and chemical properties have been or will be described elsewhere Evans, Fraser, and Owen, 1949). Solutions of 1: 2-dimercaptopentane-3: 4 : 5-triol, 1: 2-dimercaptohexane-3: 4: 5: 6-tetrol, dimercaprol glucoside (BALIntrav) and 3(2': 3'-dimercaptopropyl)-mannitol were prepared by dissolving their barium salts in water, adjusting the pH to 6 with ION sulphuric acid, removing any remaining barium ions with saturated sodium sulphate, and removing the barium sulphate by centrifuging for 15 min. at 2,500 r.p.m. The complete removal of barium ions was checked by adding a further trace of saturated sodium sulphate and observing the absence of a precipitate. The concentrations of these solutions were estimated at the time of experiments as follows: sufficient hydrochloric acid was added to make the solutions of normal acidity, and aliquots were then titrated at 00 ...