460 Of the Checks to Population Bk. ii. 



ment be just, I should conceive that each marriage 

 must have yielded, or would yield, in the course 

 of its duration, as many as 9 or 10 births. 



When from any actual survey it appears, that 

 there are about 3 living children to each marriage, 

 or 5 persons, or only 4a to a house, which are 

 very common proportions, we must not infer that 

 the average number of births to a marriage is not 

 much above 3. We must recollect, that all the 

 marriages or establishments of the present year 

 are of course without children, all of the year 

 before have only one, all of the year before 

 that can hardly be expected to have as many 

 as two, and all of the fourth year will certainly, 

 in the natural course of things, have less than 

 three. One out of five children is a very unusu- 

 ally small proportion to lose in the course of ten 

 years ; and after ten years, it may be supposed 

 that the eldest begin to leave their parents ; so 

 that if each marriage be supposed accurately to 

 yield 5 births in the course of its duration, the 

 families which had increased to their full comple- 

 ment would only have four children; and a very 

 large proportion of those which were in the earlier 

 stages of increase would have less than three ;* 

 and consequently, taking into consideration the 

 number of families where one of the parents may 

 be supposed to be dead, I much doubt whether in 

 this case a survey would give 4^ to a family. In 



* It has been calculated that, on an average, the difference of 

 age in the children of the same family is about two years. 



