the same for each group of cities. The general practice, therefore, is for half the cities in each group to have a teaching load varying from 29 to 39, or coming within a range of 11.

PUPILS.

In all city public schools there are 8,586,601 pupils enrolled, of whom 4,239,467 are boys and 4,347,134 are girls. The city schools enroll, therefore, 41.2 per cent of all the public school children in the country. In addition to the figures given above, it is estimated that 1,021,127 children were enrolled in city private and parochial schools. The cities of Group I enroll 43 per cent of these public school pupils; of Group II, 16.8 per cent; of Group III, 16.5 per cent; of Group IV, 11.8 per cent; and of Group V, 11.9 per cent. Only in the cities of the first group does the proportion of boys exceed that of girls. In the cities of the first four groups about 81.1 per cent of all pupils are enrolled in the elementary schools, 12.2 per cent in high schools, 5.3 per cent in kindergartens, and 1.4 per cent in special schools. The proportion registered in high schools is significant, since for the whole country only 9.3 per cent of the total enrollment is found in high schools, the rural statistics bringing down the average. As the cities of Group V did not report high-school enrollment the exact rural-school statistics on this point can not be ascertained. TABLE 3.-Distribution of cities according to number of days the public schools were actually in session, 1917–18.1

This table does not include one city in Group IV reporting a session of 223 days, nor two cities in Group V reporting sessions of 80 days and 78 days, respectively.

9365-20-2

FIG. 2.-Distribution of 1,762 cities which reported the number of days the public schools were actually in session, 1917-18.

LENGTH OF THE CITY SCHOOL TERM.

In Table 87 the length of the school term in individual cities for each type of school maintained is given. From the detailed data a general conception as to the length of the school term can scarcely be obtained. Accordingly, the results have been summarized in Table 3 in the form of a distribution table for each group of cities. The average length of the school term for all types of schools in a city school system has been used in compiling this table and in constructing figure 2. As little difference exists in the length of school term for the cities of the different groups, no attempt has been made

to present the distribution separately for each group of cities in graphic form. The diagram shows the distribution of 1,762 cities according to the average length of the school term in each. It is found that 39 per cent of these cities had a term varying from 176 to 180 days. As this percentage greatly exceeds that for any other similar group, it is evident that a nine-months' school term is maintained by most cities. Over 25 per cent of the cities have a term varying from 181 to 190 days, indicating a nine-and-a-half months' term. Almost 8 per cent of the cities have a longer term, ranging from 191 to 200 days, indicating a 10 months' school term in these cities. A very small percentage of cities have a shorter term than 8 months or 160 days.

The length of the school term is determined by dividing the aggregate attendance, i. e., the total number of days attended by all pupils, by the average daily attendance. According to this method of computation, all cities have been put on the same basis, and the diagram probably represents conditions as they actually are. If the total aggregate attendance for all cities reporting the data is divided by the total average daily attendance in the same cities, an average of 182 days is secured. This number represents the average length of the city school term in the United States. This average is considerably higher than the corresponding average for all public elementary and secondary schools in the United States, which is almost 161 days. In other words the average city school term is about one month longer than the average term for all schools combined.

The average pupil enrolled in the city public schools attends only 143 days out of the 182 days of schooling provided. He wastes, therefore, 39 days or almost 2 months of schooling annually. This percentage of waste in city schools, 21.4 per cent, is not quite so large as the corresponding percentage of waste for the whole country, 25.4 per cent.

JUNIOR HIGH SCHOOLS.

In 1918, 292 city school systems maintained junior high schools. Altogether, 431 junior high schools are in existence-76 in Group I, 90 in Group II, 93 in Group III, 102 in Group IV, and 70 in Group V. In these schools, 1,077 men and 4,468 women teachers are employed. Altogether, 158,108 children are enrolled in these 431 junior high schools, including 75,013 boys and 83,095 girls. The chapter on public high schools includes a report from 557 junior high schools, some of them being maintained as a part of rural or village school systems. These 557 schools enrolled only 116,197 pupils, 54,513 boys and 61,684 girls. Considerable editing was done on the highschool reports to eliminate the schools organized only for departmental teaching. As the city report does not contain much de

tailed information about junior high schools, no attempt has been made to reject those schools offering only departmental teaching in the elementary grades. The interpretation of the term "junior high school," as used in this report, has rested, therefore, with the various city school superintendents. A uniform rigid construction of the term might have reduced considerably the statistics of city junior high schools appearing herein. It is possible that the special highschool report does not include the statistics of all city junior high schools, as only those junior high schools were reported which were more or less closely connected with senior high schools in the same school system. Probably many separately organized junior high schools were not included in that chapter of the Biennial Survey. It is believed, therefore, that the statistics of city junior high schools are fairly reliable and complete, although strict uniformity was not possible in editing the reports.

TABLE 4.--Distribution of 295 cities as to number of students per teacher in junior high schools, 1917-18.

Number of students per Group I. Group II. Group III. Group IV. Group V.

teacher.

The five groups combined.

THE "TEACHING LOAD" IN JUNIOR HIGH SCHOOLS.

It is of special interest to school administrators to know the teaching load in junior high schools. The average teaching load is determined by dividing the total junior high-school enrollment by the total number of teachers in such schools. It has little significance as to the average size of junior high-school classes. This teaching load has been determined for each group of cities and for all groups combined, as shown in Table 4. Only the totals for all cities

have been used in the construction of figure 3. It is seen in the figure that the usual number of students to a teacher varies from 25 to 35. A few cities have unusually high teaching loads, even exceeding 50 pupils to a teacher. On the other hand, some cities have fewer than 20 pupils to a teacher in such schools.

25

FIG. 3.-Distribution of 295 cities showing the number of pupils to a teacher in junior high schools, 1917-18.

As shown in Table 4, half the cities in the first group have from 20 to 32 pupils to a teacher; in the second group from 24 to 29; in the third group 26 to 37; in the fourth group 26 to 38; in the fifth group from 24 to 36; and in all groups combined from 25 to 36 pupils to a teacher. These middle halves probably best characterize the distribution of the teaching load.