Fifty-six Australian soils were examined for their ability to produce sulphate from sulphur and sulphide from added sulphur, sulphate and cystine. Selective liquid media were used to obtain initial counts and progressive activities of some of the principal organisms involved. Attempts were made to correlate soil properties, initial microbial counts, classes of micro-organisms detected and metabolic activities. Almost half of the soils oxidized sulphur very slowly or not at all, most probably due to absence of thiobacilli, particularly Th. thiooxidans . Even soils which oxidized sulphur rapidly never gave initial counts greater than 10 4 cells/g. The majority of soils produced sulphide from sulphur, between one-half and two-thirds of them formed sulphide from cystine but sulphate reduction was very rare. Sulphide escaped more readily from aerobic than from waterlogged soils. Numerous genera were isolated including Desulfovibrio and Desulfatomaculum capable of reducing sulphur, both aerobically and anaerobically, particularly sporing and non-sporing mesophiles which were sometimes present at 10 6 cells/g of soil. Sulphate reduction by the genus Desulfovibrio was uncommon, and by the genus Desulfatomaculum extremely rare. Many genera of sporing and non-sporing mesophiles seemed capable of producing sulphide from cystine when first isolated, but they quickly lost this ability on subculturing. Initial counts of 10 6 cells/g of soil were occasionally found but usually much lower. No significant correlations, only trends, were found suggesting that any soil properties measured affected the occurrence or activities of organisms capable of producing sulphate or sulphide.