To assess the possible effects of tobacco and cannabis smoking on lung function in young adults between the ages of 18 and 26. A group of over 900 young adults derived from a birth cohort of 1037 subjects born in Dunedin, New Zealand in 1972/73 were studied at age 18, 21 and 26 years. Cannabis and tobacco smoking were documented at each age using a standardized interview. Lung function, as measured by the forced expiratory volume in one second/vital capacity (FEV1/VC) ratio, was obtained by simple spirometry. A fixed effects regression model was used to analyse the data to take account of confounding factors. When the sample was stratified for cumulative use, there was evidence of a linear relationship between cannabis use and FEV1/VC (P < 0.05). In the absence of adjusting for other variables, increasing cannabis use over time was associated with a decline in FEV1/VC with time; the mean FEV1/VC among subjects using cannabis on 900 or more occasions was 7.2%, 2.6% and 5.0% less than non-users at ages 18, 21 and 26, respectively. After controlling for potential confounding factors (age, tobacco smoking and weight) the negative effect of cumulative cannabis use on mean FEV1/VC was only marginally significant (P < 0.09). Age (P < 0.001), cigarette smoking (P < 0.05) and weight (P < 0.001) were all significant predictors of FEV1/VC. Cannabis use and daily cigarette smoking acted additively to influence FEV1/VC. Longitudinal observations over 8 years in young adults revealed a dose-dependent relationship between cumulative cannabis consumption and decline in FEV1/VC. However, when confounders were accounted for the effect was reduced and was only marginally significant, but given the limited time frame over which observations were made, the trend suggests that continued cannabis smoking has the potential to result in clinically important impairment of lung function.