Because of the large number of men worldwide who smoke and the fact that cigarette smoke contains known mutagens and carcinogens, there has been concern that smoking may have adverse effects on male reproduction. A review of the epidemiological literature indicates that cigarette smoking is associated with modest reductions in semen quality including sperm concentration, motility and morphology. The associations between male smoking and sperm concentration and motility are stronger among studies of 'healthy' men (e.g. volunteers and sperm donors) than among men from infertility clinic populations. Smoking has also been associated with alterations in hormone levels in males, for example increases in the levels of oestrone and oestradiol. Despite modest reductions in semen quality and altered hormone levels among smokers compared to non-smokers, studies have not shown a reduction in male fertility in association with paternal smoking. There is some evidence to suggest that paternal smoking is associated with congenital anomalies and childhood cancer (with ORs in positive studies generally < 2.0). Smoking has not been shown to be mutagenic to human spermatozoa, although studies have been small and have had methodological shortcomings. It is concluded that, although smokers as a group may not experience reduced fertility, men with marginal semen quality who wish to have children may benefit from quitting smoking, since several small studies indicate the potential for improved semen quality after quitting smoking. More research on the benefits of quitting smoking among men with marginal semen quality is needed. Methodological improvements in studies aimed at clarifying the association between paternal smoking and reproductive and developmental outcomes in offspring should include obtaining accurate paternal smoking dose information, evaluating smoking exposure information in relation to various time windows (e.g. prior to conception, during gestation), and controlling for potential confounders and modifying factors such as age and maternal smoking habits. More sensitive and specific laboratory assays and increased sample sizes are required to establish whether smoking induces mutations in human spermatozoa.