Changes in the twinning rate in Canada 1926-70.

Large variations in the 'twinning rate', i.e., the ratio of twin confinements to total confinements, between different countries and between different time periods have been reported by many authors. Apart from well documented variations in twinning rate with maternal age and parity, little is known of the causes of these variations, which mainly affect dizygotic twinning. James (1972) has reported rapid falls of 20-25% in dizygotic twinning rates in nine European countries, Australia, and New Zealand, starting abruptly between 1956 and 1964. He does not assess the effect of age and parity changes but quotes a survey in Aberdeen James suggests that food additives might be involved. They also reported a dizygotic and monozygotic peak, in all age groups, in 1946. They were able to show that these changes were largely independent of maternal age and parity changes. The USA rate in 1964 showed a further fall 12.5 2-5 E 12.*0 .