that have been made of late years, in the public press and in legislative debates, regarding educational conditions.

Age for free attendance.-The age for free school privileges, as will be seen by the table, In an attempt at classification of these ages, it is very different for the different States. can not be found that they follow any very well defined law under the different aspects under which they may be viewed. In most cases they were presumably determined by local or temporary causes, and were not based upon any general consideration, else there would not be seen the diversity that actually exists in the case of contiguous and similarly circumstanced States.

The number of different school ages for free attendance is fifteen; of these, the age six to twenty-one has been selected by fourteen States; five to twenty-one comes next, adopted by ten States; and six to eighteen by seven States. Of the remaining ages, no one has been adopted by more than two States.

As regards duration, the Southern Central States have the shortest average period, extending over 13 years; then follow (excluding in the computation those States whose age limits are undefined) the South Atlantic States. 14 years; the Western States, 14 r years; the North Atlantic States, 14 years; and lastly the North Central States, 151 years.

It will be seen that the differences in the duration of the free school age in the differIn 4 States ent sections of the country are so small as to be practically without significance. In 1 State (Massachusetts) there is no inferior limit of the free-school age. it begins at 4 years of age, in 15 at 5 years, in 23 at 6 years, in 3 at 7 years, and in 1 (Texas) at 8 years. As regards the superior limit of free attendance, in 2 States (Connecticut and Texas) it stops at 16 years, in 8 at 18 years (including the District of Columbia, where the free-school age is 6 to 17 inclusive), in 6 States at 20 years, in 28 at 21 years, and in 3 States (Massachusetts, Rhode Island, and Michigan) the free-school age extends indefinitely upwards.

In the preceding paragraphs the word "States" is to be considered as including also the Territories.

Compulsory school age.-Seventeen States, seven Territories (including Alaska), and the District of Columbia have a compulsory attendance law; twenty-one States and two The compulsory school age of New Mexico has not been reTerritories have none. ported to this Office, though it is known that a law establishing one was recently enacted.1

It is The compulsory school age, unlike the free school age, has one well-defined characteristic, which the geographical classification adopted in the tables clearly exhibits. entirely wanting (except in the District of Columbia, which stands in a peculiar relation to the rest of the country) in the two southern divisions. Every North Atlantic State has compulsory attendance, except Pennsylvania, and the same is true of every North Central State, except Indiana, Iowa, and Missouri.

In many instances, however, the compulsory attendance law, if not actually a dead letter, is practically so, as will be exhibited in another portion of this Report.

The duration of the period of compulsory attendance ranges from 4 years (in Dakota) to 9 years (in Wyoming). Six years of age is the lowest inferior limit and 10 the highest (the latter also in Dakota). The superior limit ranges from 14 to 16 years of age, in no case going beyond the latter.

Increase of school-census population.-Although the school-census populations may not be compared with each other, nor aggregated, on account of their heterogeneity, the perif we assume that the population between centage of increase of these populations may, any two limits of age in any State increases in the same ratio as that between any other two limits (i. e., that the proportion of the population of any given age remains constant in each State), an assumption that may be made as regards the increase of a few years It is on this assumption that the percentage of inwithin very narrow limits of error. crease or decrease of school-census population has been aggregated by geographical divisions in Column 8 of Table 2.

The largest percentage of increase is found, as might be expected, in the Western Division, which contains the newer States and Territories that are being opened up for settlement, and where a small absolute increase sometimes causes a large relative one. Taking the five States and Territories of that division that furnish the necessary data, the average increase per cent. of the school-census population is found to be 5.68, which would cause it to double in about twelve and a half years.

The next largest rate of growth of school-census population is found in the South; the percentage of increase in the South Atlantic Division (I ased on two States only) is 3.21,

'From information received at this Office since the tables were compiled, it appears that the law does not define the age at which children may be compelled to attend school in New Mexico, thereby rendering it applicable to the whole "school population"-i, e., those from 5 to 20 years.

while in the South Central (based on five States) it is 4.21; these rates of increase would cause a doubling in the two divisions in twenty-two and in seventeen years respectively. Of the individual States of these two last named sections, Alabama shows an increase of 7.2 per cent., equivalent to a doubling of her school population (and also of her total population, upon the assumption referred to above) in about ten years. This growth, if correctly reported, is a truly phenomenal one, and emphasizes the plea for more funds for the schools of Alabama, made by Superintendent Palmer of that State in his last report. The average increase of the school-census population for all the States and Territories affording data to be used in the computation (27 in number) is 2.61 per cent., which would cause it to double in twenty-seven years. States and Territories from all parts of the country and under all conditions of growth are represented in this average, and it may be considered as expressing with a tolerable degree of correctness the actual present rate of growth of the United States in population.

TABLE 2.-SCHOOL AGE AND SCHOOL CENSUS.

e Inclusive.

g Approximately.

h School census taken every fifth year.

i Distributed according to attendance.

d Distributed in part according to total popula- j School census every tenth year, beginning 1888.

k School census taken every fourth year.

State or Territory.

TABLE 2.-SCHOOL AGE AND SCHOOL CENSUS-Continued.

The total public school enrolment of the United States, as made up from the latest data received and supplemented by estimates in two cases (Arkansas and Montana), is 11,805,660.

Growth of enrolment.-The yearly increase for the United States can not be accurately determined, as ten States and Territories do not furnish the necessary data. Assuming, however, that the States and Territories so lacking have made the same progress as the others in the same divisions, the yearly increase would be 305,772, or at the rate of 2.66 per cent. per annum. The increase over the amount reported for 1885-86 is greater than this, a circumstance which results from the fact that the returns of some of the States are two or more years later than those given in the preceding Report.

In only five States (New Hampshire, Vermont, South Carolina, Ohio, and Nevada) and one Territory (Arizona) has the enrolment decreased. The largest relative decrease, 5.37 per cent., is found in New Hampshire. Dakota furnishes the largest per cent. of increase, 11.70, followed by Indiana with 9.20. This large development of the school registration of Indiana is a notable circumstance, when compared with the nearly stationary condition of that of the contiguous States, Ohio and Illinois.

The enrolment of Mississippi is considerably less than that reported for last year, as is also the average daily attendance. State Superintendent Preston writes to the Office that there has probably been no actual decrease, but that the attendance has heretofore been fraudulently exaggerated in the reports of certain teachers who were paid according to the number of pupils they taught. The law has been changed so that this method of determining teachers' wages is no longer practised.

Considering the different geographical divisions, all show a satisfactory growth in enrolment, unless it may be the North Atlantic Division, where the rate of increase (.71 per

Increase or decrease.

cent.) is less than one-half the rate of increase of school population (1.62 per cent.); this last quantity, however, is computed from such defective data that entire reliance can not be placed upon it.

Proportion of children enrolled.—In the proportion of children enrolled in the public schools, the North Central States are far in the lead, having 121 pupils in the public schools for every one hundred children six to fourteen years of age. That this result should be a matter of congratulation, considering the relatively low density of population of these States, has been already noted.

Notwithstanding the tremendous strides that have been taken in the development of the school systems of the Southern States during the last ten years, as exhibited in Table 17, they are still far behind the Northern States in regard to proportion of children enrolled in the public schools. In the South Atlantic States only 89, and in the South Central States only 79 children out of every 100, 6 to 14 years of age, are enrolled as pupils in the public schools. This results in a great degree from the excessive proportion of children to grown persons met with in the Southern States; for if we compare the proportion of total population enrolled, the disparity which appears to the prejudice of the Southern States almost disappears, and in one case is quite reversed, i. e., the proportion of total population enrolled is actually greater in the two Southern divisions (being 19.06 and 17.49 in the cases referred to) than it is in the Western Division (16.86).

No account is taken here of the duration of attendance at school, but only of the circumstance that the pupils were on the school registers. If the element of time is taken into consideration, the South appears much more at disadvantage.

The great advantage which the Northern States possess is the much larger number of tax-payers in proportion to the number of children to be educated. Even if the relative wealth of the North and South were equal, which is far from being the case, each tax-payer of the former section would have a far less burden to bear in the work of getting all the "schoolable" children within the schools.

e Only the States for which the increase or decrease is given are represented in this summary. ƒExcluding Alaska.

AVERAGE DAILY ATTENDANCE.

[Table 4, Page 62.]

According to the most recent returns, the number of pupils daily attending all the public schools of the United States while they are in session is, on an average, 7,571,416. As in the case of enrolment, it is not possible to determine the exact yearly increase, but a very fair approximation places it at 218,500, or at the yearly rate of 2.89 per cent. The greatest increase in average attendance is observable in the South; in both of the Southern divisions it is not only remarkably large, but it is to be noted that it exceeds the increase of enrolment; in other words, not only more pupils are going to school there, but also the attendance of those who do go is more regular. This is an evidence of increased appreciation of public schools not to be overlooked.

Florida shows the greatest growth of average attendance, viz, 13.94 per cent.; in the District of Columbia, Virginia, and Georgia the growth is also exceedingly noteworthy. Indiana stands in the same relation to the States on her east and west borders as in the case of enrolment.

The average attendance has decreased in six States, so far as reported, viz, New Hampshire, Vermont, Rhode Island, Connecticut, New York, and South Carolina. The greatest decrease, 3.64 per cent., took place in New Hampshire.