tion of population residing in the smaller cities exercises more influence than the proportion of whites in determining the proportion of literates. The following figures show the returns of literacy at the censuses of 1860 and 1887 as compared with the present:

These figures show that the proportion of literates in the total population has increased from 8 to 16 per cent in forty years. Notwithstanding this slight increase, the proportion of illiterates in Porto Rico is higher than in any of the states of this Union or any of the other West Indian islands. Guatemala is the only country in the region for which statistics are obtainable, where the proportion of illiterates is higher than in Porto Rico.

SCHOOL ATTENDANCE.

[See Tables XVII and XX.]

The total number of persons attending school in Porto Rico during the year preceding November 10, 1899, was 26,212, or between 2 and 3 per cent of the total population. But in the discussion on age it was shown (p. 49) that the children in Porto Rico between 5 and 15 years old, and so at the age when school attendance is most common, were unusually numerous. Hence it is better to compare the children attending school with those of school age. From Table XX (p. 267) it appears that only 414 children under 5 or over 17 years of age attended school-that is, about 1.5 per cent of the entire number. The school age may therefore be assumed to be 5-17, and the slight proportion of persons over or under these limits neglected.

Number and per cent of children attending school.

It has already been shown that the proportion of persons able to read, and probably also the proportion of children attending school, was much higher in the cities of Porto Rico than in the rural districts. In the following table the facts for the two cities included in Table XX are given.

If the

On the average in the two cities less than one-fourth of the children. 5 to 17 attended school during the year preceding the census. figures for these two cities be subtracted from all Porto Rico, the following results are reached:

In the rural districts of Porto Rico less than one-fifteenth of the children between 5 and 17 years of age attended school during the year preceding the census, but in the large cities the proportion was a little more than three times as great.

In the following table the figures are given for the seven departments after the two large cities have been excluded:

Number and per cent of school children in rural districts, by departments.

The conspicuous position of Mayaguez agrees with the results of the analysis regarding ability to read (p. 74), but in the present table the position of this department is influenced by the inclusion of the city of Mayaguez. Here, too, one notices that the proportion of children attending school, like the proportion of literates, is lowest in the two northwesterly departments, where the proportion of whites is highest, while the two departments of Humacao and Guayama come next to Mayaguez in the proportion of children attending