Sources whence the material was derived.-The following tables and diagrams have been prepared to show the growth of the common school system of the United States since 1870. The figures given are based upon data derived from the school reports of the various States and Territories covering the period in question. Those for 1889 are approximate, as has already been observed, and may need a slight future correction.

The files of some of the State reports are incomplete, especially those of the Southern States; and the reports existing contain often very meager statistical information, reporting perhaps only one or two items in the earlier years, and those probably for only one-half the counties in the State, or less. Many anomalies and inconsistencies occur in these reports.

All the available statistical material has been collected, and examined as thoroughly as was practicable. The text of the State reports was examined when it seemed necessary, in the progress of the work, in order to obtain any information bearing upon the subject that would be of assistance in making a trustworthy estimate, in case of official statistics being missing or unreliable.

The Southern States.-The reports of the Gulf States in particular during the first half of the decade 1870-80 are fragmentary and unreliable. The reports of average daily attendance and of the value of school property are almost entirely wanting; while the reports of the average length of school term, when there are any such, and of

other items, are characterized by strongly marked fluctuations and an. omalies, the effects of which are graphically depicted on Diagram IV (p. 19), in the lines of the two Southern Divisions, during the years 870-74. In fact, all the figures for these two divisions, as given in the following pages, must be considered as more or less unreliable until about the year 1875, at which date the school systems of the Southern States had in general passed through their formative period.

Uniformity striven after.-Effort has been made to render the statistics homogeneous. This is especially requisite in the case of school expenditure. What should be reported as "expenditure" is a matter of custom or opinion. In these tables are included only the expenditures for public day schools up through the grade of high school-that is, schools giving primary and secondary instruction. Expenditures of public moneys for evening schools, normal schools, teachers' institutes, superior, special, and professional schools, and schools and institutions for the defective, dependent, and delinquent classes have been deducted when possible, though sometimes it has not been possible, on account of detailed reports not having been made. Also the amount expended for the payment of indebtedness has been excluded whenever known; it has been frequently shown that this is not an expenditure for education.

POPULATION.

Statistics of population.-The population, except for the census years 1870 and 1880, has been estimated. The census of 1890 has been used, in connection with that of 1880, as a basis for computing the population since 1880; the State censuses of 1885 and other years have also been utilized. The figures given in the following table will be found sufficiently reliable:

TABLE 3.- Population of the United States by years, 1870 to 1839.

a The population is given for same epoch as the enrollment, expenditure, ete; hence that for 1889 will need future correction, as the population of some of the States for 1888 enters into this total. See Chap. XXII.

Relative number of school children.-In finding what proportion' the school children 6 to 14 are of the total population, a great difference is observed in different sections of the United States. This is an important fact, and needs to be emphasized. The more children there are in any given number of the population, the fewer will be the taxpayers,

and the greater will be the burden of providing each child with common school education. The accompanying diagram shows graphically the number of children 6 to 14 years of age in each 100 of the population.

L-Diagram showing the number of school children 6 to 14 years of age in every 100 persons in 1880. The shaded parts represent the children 6 to 14, the unshaded all other persons.

It is seen by this that it is in the Southern section of the Union that children most predominate, in the South Central States over 22 out of every 100 persons being children 6 to 14 years of age; in the North Central Division the proportion of such children is less, being 19.5; in the North Atlantic States still less, 17; and in the Western Division, including the Pacific States and Territories, least of all, 15.9. It will be noted that the proportion of children is greatest in those sections where the per capita valuation of property is the smallest; the greater number of children to educate exists where there is the least means for educating them.

In individual States the contrast in the relative number of children is still more striking. Montana has 11.4 children in each 100 persons; Mississippi 22.9. Evidently if Mississippi had the same proportion of her population enrolled in the schools as Montana, she would have only one-half the proportion of her school population enrolled. In order to gain a correct idea of school enrollment in these two States, for instance, it is essential that it should be viewed under both these aspects, i. e., in its relation (1) to the total population and (2) to the school population.

The inequality of burden arising from a small or an excessive number of children is brought more prominently into relief by considering the number of the school population of 6 to 14 years to every 100 adults; i.e., by comparing the number of children to be educated with the number of persons there are to provide the means for education for them.

TABLE 5.-Percentage of the population enrolled in the common schools.

Enrollment.-Tables 4 and 5 exhibit various facts relating to the enrollment of pupils in common schools since 1870; the data of the latter table are also graphically represented in Diagram III (p. 14). While the table gives the facts with greater numerical accuracy, the diagram presents more clearly to the eye the status of the different geographical divisions with respect to each other, and the changes that have taken place from year to year.

The diagram shows what percentage of the total population was enrolled in the common schools. The most noticeable feature presented by this diagram is the growth of common schools in the South. Beginning with an enrollment of less than 8 per cent. in 1870, the two southern divisions rise almost uninterruptedly in the scale. The South Atlantic passes above the Western Division in 1880, and the South Central above the same in 1883. Their course is still upward in the main, until in 1889, with an enrollment of about 20 per cent., they both stand above the North Atlantic Division, which has 17.8 per

cent.

Hardly less significant, though not so much a matter of common knowledge, is the decrease of the proportion of the population reported as enrolled in the common schools in the two northern divisions. This movement of decrease has been going on almost continuously since 1870, until in 1889 the North Atlantic Division has fallen below every division except the Western. The final dip downward (from 1888 to 1889) is very noticeable. The detailed table, given in Chap. XXIX, shows that there was an absolute decrease in the enrollment of Maine, New Hampshire, Vermont, and Rhode Island from 1888 to 1889, while the enrollment of New York and New Jersey was almost stationary, the school population at the same time increasing rapidly.