3

177

205

216

230

207

62

.77

.20

.80

.06

Explanation of the Cambridge Table.-The three classes taken in the Cambridge table are those that graduated from the high schools (or reached the corresponding stage in the Latin school, whose course is five years long) in the years 1896, 1897 and 1898. These classes entered the primary schools thirteen years earlier, that is, in the years 1883, 1884 and 1885, numbering 1,756, 1,956 and 2,046 respectively, and gradually diminished to the numbers 153, 182 and 164 respectively. In each case the number is the entire membership of the class for the year specified, as determined by careful count during a particular week in December, and is exceptionally trustworthy. The averages based on these three classes yield a first primary class of 1,919 that in thirteen years fell off to 166. The loss from each year to the next of this average class is numerically given. The percentage columns have the following significance:

1. The first column shows what percentage of the number in the grade or year immediately below passed up to the next higher.

2. The second column shows what percentage of the first or lowest primary class reached each of the several grades above.

3. The third column shows what percentage of the original class had fallen out in each of the several years or grades above.

4. The fourth column shows what percentage of the original class was lost in passing from each year to the next.

It should be noted that the population of Cambridge in 1885 was 59,658; in 1895, 81,643. There were some accessions, therefore, to the table from other sources than the lowest primary, though not so many as the increase in population might lead one to expect. Indeed, so far as this increase is due to the excess of the local birth rate over the local death rate, it would not affect the table at all, for no child born in the city after 1885 could possibly have been a member of any one of the three classes considered. So far, also, as this increase in population is due to young children from outside, it has brought no outside accessions to the upper classes of the table. Such children belong to later classes.

The explanation of the Cambridge table applies also to the Somerville table. The Somerville data, like the Cambridge, are based upon trustworthy December counts of actual membership. Such counts have been made for twenty-five years past. The population of Somerville in 1885 was 29,971; in 1895, 52,200.

Inferences from the Cambridge and Somerville Tables. - It is remarkable, in the first place, how closely the two tables agree in the trends revealed. Prepared originally to show how many of those starting in the primary school reached the high, they are found to bring out other facts that merit study for both educational and financial reasons. The tables, for example, agree in showing:

1. A somewhat startling loss (32 per cent. for Cambridge and 31 for Somerville) - whether real or nominal remains to be seen in passing from the first grade to the second.

2.

3.

Another large loss (10 per cent. for Cambridge and 7 for Somerville) in passing from the fourth grade to the fifth. A third large loss (18 per cent. for Cambridge and 17 for Somerville) in rising to the seventh and eighth grades. And a fourth large loss (10 per cent. for Cambridge and 14 for Somerville) in the movement from the ninth grade to the high school.

4.

5. That the least losses are between the third and fourth grades (2 per cent. loss for Cambridge and 6 per cent. gain for Somerville) and between the fifth and sixth grades (2 per cent. for Cambridge and 1 per cent. for Somerville).

6. And, finally, that a much larger percentage than is commonly supposed of those who enter the primary school succeed in entering the high (20 per cent. for Cambridge and 26 per cent. for Somerville), a percentage that needs to be considerably increased for these cities, for reasons to follow.

If the losses apparently suffered by the primary, grammar and high school classes in the journey through the thirteen grades are tabulated for these groups respectively, the showing is as follows:

If it is generally true that of a class entering the primary school one third fall out before reaching the grammar school and another third before reaching the high, obviously the popular impression that high school losses are the most serious in the system is not well grounded. The public should turn its scrutinizing eye to the weightier but more easily overlooked losses of the lower grades.

The surprising loss in passing from the first grade to the second (32 per cent. for Cambridge and 31 for Somerville) is not due to transfers to private or parochial schools. It can be explained only in part by the greater death or sickness or disability rate of young children. The falling back of pupils admitted to the lowest grade in the spring, or at any time after the count has been made, does not affect the matter at all. The true explanation is that, for a variety of reasons, concern for the health of little children, their special susceptibility to certain diseases, the larger proportion of indocile or irresponsive cases among them than in the sifted grades above, the indisposition of parents to subject them to unremitting discipline, the freedom of parents to detain children at home before the compulsory years begin, the not infrequent overcrowding of this grade to the detriment of the pupil's progress, children that have once been counted in the lowest grade frequently drop back so as to be counted in the lowest grade a second time before they pass to the next higher.

The numerical effect of this process is to increase the number in the lowest grade, thus widening the gap between it and the grade above. The point is an important one, since it has also to do with some of the losses higher up. It can be brought