A Study on Secular Changes of and Factors Responsible for Seasonal Distribution of Births

Abstract

Changes in birth distribution pattern during a few centuries were investigated in both Japan and the USA, using various forms of records. Regional differences in seasonal birth distribution, and their connection with birth rates and marriage seasons were also studied. The following conclusions were drawn. 1) From the second half of the 19th century to the first half of the 1960's, seasonal distribution of births in Tokyo and Osaka showed a fluctuating pattern with a peak from January to March and a trough between May and August. Most variations during that period were in the amplitude of the basic pattern. 2) Old registration books of parishoners known as 'Ninbetsu-cho' of three towns in Osaka City revealed a peculiar pattern with a sharp peak between November and December, ( October and November of the lunar calender) from 1755 to 1867. 3) Japanese birth distribution became less noticeable in the 1960's and the 1970's with very low peak between July and August similar to that formed in the USA. 4) In USA, there is a secular change of shifts of peaks in different periods. The typical pattern with a peak between Aug. and Sep. can be seen from the second half of the 18th century to the first half of the 19th century and again after 1930. During certain periods since the 17th century, different patterns are apparent: e. g. a wide peak between May and September until the 1970's, or a pattern with two peaks in spring and autumn from the second half of the century until the 1920's. 5) Regional differences of the birth distribution in Japan gradually decreased and have reached uniformity throughout Japan. (6) As for the relationships between birth seasonality and regional birth rates, regional birth rates fell in summer, becoming lower as one travels south. 7) As for the relationship between birth seasonality and annual birth rate, a distinct difference was seen in the peak season of Jan.-Mar., and the birth rate in the trough season of May-July changed little until World War II when contraception became popular in Japan. 8) Prior to World War II the seasonal distribution of marriages in Japan was typified by peaks in spring. After the Wax it changed to a pattern with peaks in both spring and late autumn. Seasonal differences of marriages increased, in contrast to birth patterns. 9) Comparing the distribution of marriage ceremonies and registrations after World War II in Japan, seasonal variance was larger for ceremonies than registrations. The marriage registration peak shifted ahead one or two months. This suggests that most marriages were registered several months later than ceremonies. In the 1970's, however, this difference was reduced. 10) It seems worthwhile to consider the possibility that certain seasonal biological suppressing factors varying in period and by region may be responsible for the distributional birth pattern.