FOLLOWING the original observation of Rapkine (1931) that the level of trichloroacetic acid (TCA)-soluble sulphydryl (SH) compounds in dividing sea urchin eggs fluctuated during cell division, considerable evidence has been accumulated to show that protein and non-protein SH groups play an important role in cell division (Needham, 1950 ;Barron, 1951 ;Mazia, 1954Mazia, , 1959Stern, 1959Stern, , 1960. Although it is not known how SH groups participate in this process, attempts have been made to relate carcinogenesis to a disturbance of this function. Thus Crabtree (1944, 1945, 1946) postulated that a primary carcinogenic disturbance is the binding of carcinogen to SH-containing cell constituents, and that anti-carcinogenic compounds which react with SH groups prevent this. On the other hand, Calcutt (1961) suggested that a rise in total SH level is essential for carcinogenesis and that anticarcinogenic substances inhibit this rise. Thus, the question of how SH groups take part in carcinogenesis has not been answered.The question of whether it is the protein or non-protein SH groups which play a role also remains open. This is not surprising since there is a lack of agreement by different workers on the effects of carcinogens on SH levels. In particular, apparently conflicting effects of hepatocarcinogens on the non-protein SH content of the liver have been reported. Among the workers who gave a single dose of carcinogen, Boyland and Mawson (1938) noted a rise in the TCA-soluble SH level in the liver of mice for more than 20 days following an intraperitoneal injection of 3,4: 5,6-dibenzocarbazole; Kennaway, Kennaway and Warren (1944) found that intraperitoneal injection of dimethylaminoazobenzene (DAB) into mice did not affect the metaphosphoric acid-soluble SH level; Roy, Miya and Carr (1958) observed that 5 hours after oral administration of DAB or /3-naphthylamine to rats the sulphosalicylic acid-soluble SH level in the liver was lower than in the controls; while Neish andRylett (1960, 1961) reported the appearance of an acidic thiol peptide in the liver of rats after an intraperitoneal injection of hepatocarcinogens, in contrast to non-carcinogenic compounds. In the case of continuous feeding of carcinogenic aminoazo dyes to rats, Roy, Miya andCarr (1957, 1958) reported slightly increased sulphosalicylic acid-soluble SH concentrations in the liver after 4 and 8 weeks followed by a depletion after 12 weeks if DAB was incorporated in a stock diet, and no increase if a low-protein diet was used; Fiala (1958) and Fiala and Fiala (1959) concluded that feeding 3'-methyl-DAB caused a late decrease of liver glutathione, but an increase of sulphosalicylic acid-soluble non-protein SH; and finally Calcutt, Doxey and Coates (1960) noted a slight SULPHYDRYL GROUPS AND ASCORBIC ACID increase of both metaphosphoric acid-soluble SH and total SH groups after feeding DAB, but in a later paper (1961) this effect was considered to be due to dietary factors rather than to the azo dye.To tell which effects of carcinogens are si...