Summary 



In this analysis of population change in two samples of small towns which 

 had increased or decreased regularly since 1920, it was developed that: 



1. Population characteristics have become more alike during the period 

 1930-1950. When the population distributions by age and sex are com- 

 pared the trend is consistently in the direction of greater homogeneity. 

 Fertility ratios, dependency ratios, and population density also tend to be- 

 come more alike in value. 



2. Census residence categories are of relatively little value in the 

 analysis of the small northern New England town. In some areas of the 

 United States there is a notable relationship between census residence 

 categories and minor civil divisions. This is not true in northern New 

 England where the pattern of settlement more closely resembles the farm 

 village than in other parts of the country. Individuals in the different 

 census residence categories vary in their characteristics, but the direction 

 of the variation is not consistently one way. The farm and non-farm com- 

 ponents within each sample of towns have different rates of reproduction. 

 In the sample of increasing towns, the farm component has higher fer- 

 tility rates; in the other, the reverse is true. When the sex ratio for resi- 

 dence components within each sample was measured, it was higher for 

 the farm than non-farm in the increasing towns and the reverse in the 

 decreasing towns. Apparently population change in the town affects the 

 value of the ratios. 



These two census residence categories are largely intermixed in the New 

 England town, not separated spatially as in some other areas of the country. 

 Hence generalizations derived from the behavior of these residence cate- 

 gories may be misleading, if applied to rural New England towns since 

 the variation is complementary. 



3. Tliere are some differences in agricultural adjustments betiveen farms 

 in towns losing population and those in toivns gaining population. The 

 value of farms measured in constant dollars has decreased about the same 

 amount in both samples of towns. The average number of acres per farm 

 has gone up in the sample of decreasing towns and has gone down in the 

 sample of increasing towns. Herd size has increased in the sample of de- 

 creasing towns while the reverse is true in the increasing towns. Pro- 

 duction per cow has increased in both samples, but more rapidly in the 

 increasing towns. It was found that these trends differed sufficiently from 

 the corresponding county trends to ascribe them in part to some factor 

 (perhaps population change and quality of resources) other than distance 

 from market. 



4. There appears to be some general difference in the economic ori- 

 entation of the decreasing and increasing toivns. The decreasing towns have 

 almost twice as many retail establishments handling non-staple items as do 

 the increasing towns. The former also have more personal services avail- 

 able and more establishments which can be classified as recreation. The 

 inference from this is that they are catering to the recreation industry 

 since all are too small to be the shopping centers in their areas. The in- 



29 



