In 1987 the Cannabis Expiration Notice scheme decreased penalties for the personal use of cannabis in South Australia. Data from four National Campaign Against Drug Abuse (NCADA) household drug-use surveys covering the period 1985 to 1993 were analysed to measure the effect of the decriminalisation on cannabis use. The main outcomes used were the self-reported prevalence rates of having ever used cannabis and current weekly use. Logistic regression was used to control for the potentially confounding effects of age and sex. Other outcomes were rates of having ever been offered cannabis and willingness to use cannabis if offered it. Between 1985 and 1993 the adjusted prevalence rate of ever having used cannabis increased in South Australia from 26 per cent to 38 per cent. There were also significant increases in Victoria and Tasmania, and to a lesser extent in New South Wales. The increase in South Australia was not significantly greater than the average increase (P = 0.1). Adjusted rates of weekly use increased between 1988 and 1991 in South Australia, but did not change through 1993. Although the effect was in the direction of a greater increase in South Australia, this was not statistically significant when compared to increases in the rest of Australia (P = 0.07). The greatest increase in adjusted weekly use occurred in Tasmania between 1991 and 1993, from 2 per cent to 7 per cent. Although the NCADA survey data indicate that there were increases in cannabis use in South Australia in 1985-1993, they cannot be attributed to the effects of partial decriminalisation, because similar increases occurred in other states.