The cross-cultural study of secular trends in twinning has been of int
erest to students of human biology. Although recent vital statistics a
re likely to be more reliable, older vital records allow a long-term s
tudy of twinning trends and of the effect of epidemics on twin inciden
ce. Here, I report the results of an investigation of the secular tren
d of twin maternities from 1851 to 1901 in Escazu, Costa Rica, The res
earch also investigates whether epidemics affected twin incidence in a
consistent manner. Because the church records did not state the mothe
r's age, it was not possible to standardize the raw twinning rate for
maternal age, Thus the moving averages of the raw rates for every 33 y
ears (the average length of a generation) were computed to minimize th
e effect of changing average maternal age on twin incidence. No consis
tent response to epidemics is apparent in the twinning rates. However,
both the raw twinning rates and their moving averages indicate that t
he frequency of twinning decreased in Escazu during the second part of
the nineteenth century. A decline in the twinning rate in other popul
ations is usually attributed to environmental changes caused by the pr
ocess of industrialization. How ever, the decline in the twin maternit
y incidence in Escazu could not be the result of such an environment b
ecause the population was rural and nonindustrialized. This study indi
cates that a decline in twin maternities may result from more diverse
causes than was previously thought.