Besides the differences in nationality or place of birth, it is useful in statistics of population to classify according to sex and age; the teacher often finds it convenient to classify the pupils in that way, and is sometimes required to do so.

Nature sees to it that the numbers of the two sexes are nearly equal. A difference of more than two or three per cent is due to social causes. The chief cause of inequality is migration, as the males tend to move first. In most European countries the censuses show the females to be in excess from one to six per cent. In the United States the only states having an excess of females by the census of 1910 were Massachusetts, Rhode Island, Maryland, and the two Carolinas. The males were heavily in excess throughout the West - over 20 per cent in nine states, and over 30 per cent in six, the greatest being in Nevada, where it was 79 per cent. Of course such extreme disparity brings serious results to all forms of social organization, most especially to cultural institutions such as the schools.

1 National Society for the Study of Education, Thirteenth Yearbook, Part I, pp. 76, 79.

In regard to age, it is usual to think of the population in generations, a generation being "the mass of beings living at one period,,, "the average lifetime of man, or the ordinary period of time at which one rank follows another, or father is succeeded by child." This period is about thirty or thirty-five years, three to a century. But with the school population a generation means something different. In a room containing children of a single grade the population changes almost completely in one year. In a high school four years make a generation, and so also in a college. If we put together the eight years of elementary school, the four years of high school, and the four years of college, with a professional course perhaps added to or replacing college, we get sixteen to twenty years as the duration of a school generation for the lawyers, physicians, clergymen, teachers, and others composing the small fraction of the population who have higher education. For the rest of the population, and the great bulk of it, the length of a school generation would scarcely average half as long, though scattering freely between six years and twelve. It may be conveniently near the truth, therefore, to assume for computation in round numbers that there are in a century six school generations of the highly educated and twelve of the lesser educated. Then we may reckon that the principal

Distribution or White Population of the United States by Sex and Age Periods.

Per Cent

Per Cent

Adapted from Thirteenth Census Abstract, pp. 122-124.

Distribution or White Population of the United States by Sex and Age Periods.

(Native population of foreign parentage omitted.) who has charge of the same elementary school for twenty-five years supervises the education of three generations of children.

The distribution of population by age depends chiefly on the death rate at the successive age periods. The number of children born varies but little from year to year. Among those born in a given year the death rate is high during infancy, very slow in later childhood and youth, and increasingly rapid after the age of 25. Therefore in a genetic population the number of persons of a given age is always greater than the number of any higher age. The negroes of this country are so short-lived that only 14 per cent are over 45 years of age, while 18 per cent of the native whites of native parentage are over that age.

With age distribution as with sex distribution the great disturbing factor giving rise to variation from the normal is migration. That has caused the male population of 20 to 24 years of age in the United States to exceed the males of 15 to 19 years by 53,000. If we take the foreign-born males by themselves we find that 11 per cent of them are 20 to 24 years of age, to only 4.7 per cent who are 15 to 19, and still smaller proportions of those of younger ages. In the state of Nevada the children under fourteen years of age make only twenty per cent of the population as compared with thirty or forty per cent in many of the other states.

A newly settled agricultural community, however, has a large percentage of children, larger perhaps than it will ever have again. This is shown in the following table. The names on the left-hand side are of counties in Wisconsin which made the greatest increase in population from 1900 to 1910; on the right-hand side are the names of the four counties which showed the greatest decrease in the same period.

Counties Increasing in Population

Names

Increase per cent

Per cent children

6-9 years

Forest . . .

386

10.

Price . . .

51

IO.5

Washburn . .

48

IO.9

Rusk . . .

New

IO.4

Counties Decreasing in Population

Names

Decrease per cent

Per cent children

6-9 years

Pierce . . .

7.8

9.

Adams. . .

5.9

8.9

Crawford . .

5.8

9.2

Juneau . .

5.1

8.1

If the table had been extended to include counties showing a smaller increase or decrease, the disparity in the proportion of children would have continued. Every county, except those containing large cities, that increased in population over twenty per cent had a larger proportion of children than any county that decreased over two per cent. Evidently the subduing of new land is an undertaking which attracts young families, so that the first generation of children is often larger than the succeeding one. As a result of this tendency we get such experiences with rural schools as these:

In a district near my home the number of children increased from forty to sixty in fifteen years. The district provided for the increase by building a new schoolhouse twice as large as the old one. In another twenty years the attendance dropped to forty-five.

Our home school had an enrollment of about fifty when I attended there as a child. It has now decreased to eighteen.

This fluctuation in the proportion of children doubtless leads to mistakes in the creation of school districts and so aggravates the difficulty inherent in the rural school problem wherever the population is sparse, sometimes even causing difficulty where the population is not particularly sparse. The report of the Wisconsin state superintendent of public instruction for 1908-10 gives eighty-seven schools as having an enrollment of five or less. Thirty of these schools are in eleven counties which were settled early and are now declining in rural population. Eight schools are in the three counties having less than six inhabitants to the square mile, and forty are in the eleven counties having from six to eighteen inhabitants to the square mile.