the detailed survival percentages is to employ a number characterizing only one age group, which is higher than all the other age groups represented. From this figure it is clearly evident that the age groups to be employed can only be determined by ascertaining which ones are essentially equal in magnitude. In neither case can a single number derived from the fullest age groups represent the number of pupils beginning at earlier dates, since each succeeding statistical generation is larger than the one which preceded it.

TABLE 8.-Frequency of the ratios between the number of boys and girls who have reached their grade one or more years earlier than the normal age, and the total number of boys and girls enrolled in 80 cities in 1908 and 1918.

CHANGE IN THE PROPORTION OF PUPILS UNDER AGE, OF NORMAL AGE, AND OVER AGE SINCE 1908.

In figure 10 the 80 cities studied have been distributed according to the percentage of under-age pupils. In 1908 it was customary for cities to have only 2 or 3 per cent of under-age pupils. At any rate more cities had these percentages than any others. There seems to be little difference between the percentages of boys and girls who are under age. In 1918 no single percentage seems to characterize

the general practice. The "mountain tops" of 1908 have been demolished.

FIG. 10.-Distribution of 80 cities according to the per cent of pupils who have reached their grade one or more years earlier than the normal age in 1908 and 1918.

TABLE 9-Frequency of the ratios between the number of boys and girls of normal age and the total number of boys and girls enrolled in school in 1908 and 1918.

In figure 11 the 80 cities have been distributed according to the percentage of boys and girls of normal age. In 1908 the largest number of cities (mode) had from 50 to 51 per cent of their boys of normal age. In 1918 the mode for boys had shifted to the percentage group, 62 to 63. For girls in 1908 the mode was in the percentage group 60 to 61, while in 1918 the most usual percentage was 70 to 71. In general in this figure the black curves for 1918 fall to the right of the dotted curves for 1908. These changes unmistakably indicate that school curricula are coming more and more to conform to the chronological if not to mental age of the pupils. Even if children do enter school earlier than they did 10 years ago, such a condition does not vitiate this fact. If earlier entrance at school results in an increase in the percentage of children, both under age and of normal age and reduces the percentage of children over age, the practice is surely commendable. The chronological age of children and the curriculum are becoming more closely harmonized as the percentage of children of normal age increases.

FIG. 11.-Distribution of 80 cities according to the percentage of boys and girls of normal age enrolled in school in 1908 and 1918.

TABLE 10.-Frequency of the ratios between the number of boys and girls retarded one, two, three, four and more years. and the total number enrolled in all cities (80) in 1908 and 1918.

908 1918

Less than 1 per cent...

1 per cent.

2 per cent.

3 per cent.

4 per cent.

5 per cent. 6 per cent. 7 per cent. 8 per cent. 9 per cent. 10 per cent.

1908 1918 1908 1918 1908 1918 1908 1918 1908 1918 1908 1918 1908 1918 1908 1918

2 3 4 5 6 7 8 9 10 11 12 13 14 15

5 23

9 49

8

6 26

22

11

18 27 24 23

1

10

17

16 14 14

1

1 12

1

14 18

14

1

1

57 17