here is considerable evidence that tobacco smoking by women is strongly associated with a range of adverse reproductive outcomes. ' j 2 Research has demonstrated that smoking lowers birthweight and that there is an increased risk of infertility, with a strong dose-response relati~nship.~.~ A recent study of over 2,500 pregnancies found that smoking nearly doubled the risk of delayed c~nception.~ Smoking has been linked to a significantly increased risk of spontaneous abortion and secondary There is also strong evidence of an association between smoking and perinatal mortality."In contrast, associations reported between smoking and miscarriage have been somewhat incon~istent."*'~ An Italian case control study found a 40% increased risk of miscarriage among current smokers compared to never smokers, and the risk increased with the number of cigarettes smoked per day.l2,I3 A study of assisted fertility found that smokers produced fewer oocytes, had a pregnancy rate less than half that of non-smokers, and those smokers who became pregnant had an increased rate of mi~carriage.'~ However a large ultrasound study reported that early pregnancy loss is not associated with ~moking.'~ The inconsistent findings may be due to confounding by other exposures such as alcohol, cocaine, caffeine, or may reflect the lack of a true ass~ciation.'~~'~ There is also mounting evidence of an association between tobacco smoking and menstrual symptoms in young women, although again the evidence lacks consistency. A US study has reported that the relative risk of developing secondary amenorrhoea in adolescent women who smoked more than 20 cigarettes a day was 2.0 (95% confidence interval 1.2 to 3.1) compared to non-smokers.20 In another US study, decreased duration of bleeding, increased daily amount of bleeding (subjectively scored) and increased duration of dysmenorrhoea was found among smokers aged 37-39 years, with increased duration of dysmenorrhoea being the most significant effect.21 A study of nearly 600 young Swedish women found that the prevalence and severity of dysmenorrhoea was significantly associated with cigarette smoking, and that this relationship was dose-related.22 Another study of 2181 young women in Britain found that regular smokers were significantly more likely than those who had never smoked to experience symptoms associated with menstruation such as pain, depression, irritability and cramps.23 However this evidence is often