The Frequencies of Twins, Relative to Age of Mothers, in American Populations.

H E frequency of twin births varies in different populations. Furthermore, T it is known for several European groups, for the white population of Australia and for Japanese that a relation exists between the age of the mother and the percentage of twin births. An analysis of vital statistics data by various authors has led to the conclusion that dizygotic and monozygotic twins behave differently in this respect (see DAHLBERG 1926; KOMAI and FUKUOKA 1936). In the white populations analyzed the frequency of dizygotic twins rises steadily with increasing age of the mother until a maximum is reached in the 35-40 year age group. The maximum frequency was found to be about three times that of the initial frequency. After the maximum the curve declines again. For Japanese, the data are relatively limited but they indicate a similar trend as in whites, except for the maximum occurring slightly earlier, a t age 34. In contrast to the distinct correlation between age of mother and frequency of dizygotic twins there is little variation between different age groups of mothers in frequency of monozygotic twins. In various populations there appears a slight increase with rising age of mothers but the maximum increase, as seen in data fom Australia, is less than 30 percent over the initial frequency. No extensive analyses of the vital statistics of plural births had been available for the United States until the recent publications by STRANDSKOV and his collaborators (1945-1946). These studies report on the gross plural birth frequencies in the total, the “white” and the ‘Lcolored” U. S. populations, on the sex ratios among them, and, in the last cited report, on the frequencies of monozygotic and dizygotic twin births. Data on the relation between age of mothers and twin births in the United States have been published by the U. S. BUREAU OF THE CENSUS. They are not suitable for a detailed treatment because the figures are neither broken down for different racial groups nor do they give the sex of twins born. The former information is important since it is known that differences exist in total frequencies of twin births for “white” and “colored” U. S. populations, the latter information is needed for an estimate of the partition of gross twin frequencies into dizygotic and monozygotic ones. The BUREAU OF THE CENSUS is in possession of the required raw data, and DR. HALBERT L. DUNN, Chief of the NATIONAL OFFICE OF VITAL STATISTICS, has kindly made available to us a table giving the number of cases of twins by age of mother, sex, and race in the United States for the year 1938. The listing is according to five year age groups