in the United Stales in Fifty Years. 91 



Thus showing a gradual decrease in the proportion of children 

 during forty years of G.29 per cent ; which, allowing for the ordi- 

 nary difference between the number of males and females, is equiv- 

 alent to something more than 3 per cent of the whole popula- 

 tion. So, if the children under ten, be compared with the females 

 of the preceding census, we see a correspondent diminution of 

 ratio, viz : 



1790. 1800. 1810. 1820. 1830. 



No. of females, 1,556,839 2,100,008 2,874,433 3,871,647 5,171,115 



No. of children at the succeed- 

 ing census 1,489,315 2,016,479 2,625,790 3,427,730 4,485,130 



Prop, of children, per cent,.... 95.C6 96.02 91.35 88.53 86.73 



But these proportions are also affected by immigration. In the 

 first case, in which the comparison is made between the children 

 and the females of the same census, the proportion of children is 

 lessened by reason of the greater proportion of adults in the immi- 

 grating class than in the whole population. But in the last case, in 

 which the children of the succeeding census are compared with the 

 females of the preceding, the proportion of children is increased by 

 immigration. 



The first source of error is, however, inconsiderable. The in- 

 crease of immigrants in ten years, we have seen, may be estimated 

 at 20 per cent of the whole number ; and to such increase we 

 must add the portion of immigrant children under ten at the time 

 the census is taken. Now, if we suppose the females to constitute 

 one-third of those who migrate hither, and the children one-sixth, (as 

 seemed to be the proportion in Canada,) and if we further suppose 

 that, one-tenth of those children who arrive in the first year of the 

 decennial term would be under ten years of age at the succeeding 

 census, two-tenths of those who arrive in the second year, three in 

 the third, and so on throughout the term, we shall find, after mak- 

 ing a fair deduction for the intervening deaths, that the proportion 

 of children to females in such immigrants will be little inferior to the 

 proportion in the indigenous population. Let us, however, as- 

 sume it to be 3 per cent less, or 30 per cent on the whole number 

 of immigrants and their increase, and to adapt our estimates to this 

 supposition, we must in the first comparison add 3 per cent of the 

 whole number of immigrants to compensate for the excess of adults, 

 and in the second comparison deduct 30 per cent to correct the 

 excess of children gained by immigration. With these corrections 

 the proportion of children will be as follows : 



