IMPORTANCE OF STATISTICAL IDEAS. 113 



increases disproportionately. In the last census especially it was found 

 that the overflow of population over the far Western States seemed to 

 have been checked, the increase of population being mainly in the older 

 States and the towns and cities of the older States. The phenomena 

 in England and Germany and in other Continental countries are 

 accordingly not singular. The older countries, and the older parts even 

 of a new country like the United States are becoming more and more 

 the centers where populations live and grow, because they are the most 

 convenient places for the general exchange of services with each other 

 among the component parts of a large population, which constitutes pro- 

 duction and consumption. A small expenditure of effort in proportion 

 enables such communities to obtain from a distance the food and raw 

 materials which they require. Migration is no longer the necessity that 

 it was. 



Decline in Rate of Growth of Population. 



I come now to another idea appearing on the surface of the census 

 returns when they are compared for a long time past, and the con- 

 nected returns of births, marriages and deaths, which have now been 

 kept in most civilized communities for generations. Great as the 

 increase of population is with which we have been dealing, there are 

 indications that the rate of growth in the most recent census periods 

 is less in many quarters than it formerly was, while there has been a 

 corresponding decline in the birth-rates; and to some extent, though 

 not to the same extent, in the rate of the excess of births over deaths, 

 which is the critical rate of course in a question of the increase of popu- 

 lation. These facts have suggested to some a question as to how far 

 the increase of population which has been so marked in the past cen- 

 tury is likely to continue, and speculations have been indulged in as to 

 whether there is a real decline in the fecundity of population among 

 the peoples in question resembling the decline in France, both in its 

 nature and consequences. I do not propose to discuss all these various 

 questions, but rather to indicate the way in which the problem is sug- 

 gested by the statistics, and the importance of the questions thus raised 

 for discussion, as a proof of the value of the continuous statistical 

 records themselves. 



The United States naturally claims first attention in a matter like 

 this, both on' account of the magnitude of the increase of population 

 there, and the evidence that recent growth has not been quite the same 

 as it was earlier in the century. Continuing a table which was printed 

 in my address as president of the Statistical Society, in 1882, above 

 referred to, we find that the growth of population in the United States 

 since 1800 has been as follows in each census period : 



VOL. LX. — 8. 



