indicating not only the number of students of classics in educational establishments, but also their comparative records of progress as tested by various standards; these latter tending to show that in non-classical subjects candidates with a classical training show a marked superiority over those who lack it.

It is from this point that the inquiry was taken up in the investigation of which the first report is now before us. Under the guidance of an Advisory Committee (of which Dean West was chairman) a Special Investigatory Committee (under the same chairmanship) and eight Regional Committees were set up, with a total membership of fifty-nine persons, and with the assistance of forty-eight Professors of education and psychology. The following quotations from the Report indicate the exhaustive nature of the inquiry:

'It has been our great good fortune to secure the voluntary unremunerated help of 8595 teachers . . . who have given much time to marking, checking, and accounting for the experimental work in all parts of the country. Such a freewill offering is unmatched in the history of any educational inquiry thus far conducted in our land. The investigation has been carried on throughout the two academic years 1921-22 and 1922-23 in every State in the Union. . . . The total number of secondary schools enlisted in the investigation in 1313, and the total number of pupils tested is approximately 150,000. The total number of individual tests given is approximately 750,000...

'The plan of procedure has been adhered to rigorously; namely, first to find the facts, then to make an analysis and impartial criticism of the facts, and finally to prepare a progressive constructive programme for the teaching of the classics in our secondary schools. It is clear to us that there are human values involved in learning the classics which are not measurable in mechanical terms, and that there are also certain processes and results which can be so measured with fairly close accuracy. We have endeavoured to test these processes and results by definite scientific experiment. The many tests and controlled experiments employed for this purpose have been devised with the utmost care on the basis of the most recent improved methods of measurement. The statistical and historical studies and the collections of expert opinion have been made with equal care.

Statistical investigation as applied to educational

processes has, no doubt, a much greater place in America than in our own country. English teachers are inclined to look askance at mechanical tests of intellectual achievement, and to be guided rather by general impressions than by tables of figures. It may be admitted also that such tests when confined to a limited area, such as a single school, are not very reliable. But it does not follow that they are not reliable when the field of investigation is widened. The strength of the American investigation lies in the width of the field which it covers. Averages and results derived from over a thousand schools, covering all the area between the Atlantic and the Pacific, cannot be ignored; and this inquiry demands the respectful consideration of all impartial students of education.

The first conclusions, as to which no controversy can arise, are those which give the actual statistics with regard to classical education in America.

'The total enrolment in Latin in the secondary schools of the country for the year 1923-24 is estimated by the United States Bureau of Education at 940,000, slightly in excess of the combined enrolment in all other foreign languages. It is approximately 27 per cent. of the total enrolment of pupils in all secondary schools, including the seventh and eighth grades of junior high schools, or 30 per cent. if these grades are not included. The enrolment in Greek is only about 11,000, but shows some signs of increase. . . . About 83 per cent. of the 20,500 secondary schools of the country offer instruction in one or more foreign languages. Of this number, 94 per cent. offer Latin, a slightly larger percentage than in the case of all other foreign languages combined....

...

"The Latin enrolment in the colleges in the country in 1923-24 was approximately 40,000, and the Greek enrolment about 16,000. . . . Of the 609 colleges in the United States . . . 234 offer courses in beginning Latin, 470 in beginning Greek. . . . Thirty-nine of the forty-eight State superintendents of public instruction state that their attitude towards Latin is sympathetic or distinctly friendly. Seven express themselves as neutral, and two as unsympathetic or distinctly unfriendly. As regards Greek, eight are sympathetic or distinctly friendly, twenty-four are neutral, and sixteen are unsympathetic or distinctly unfriendly.'

So much with regard to the actual numbers of

students of the classics in the schools and colleges of America. It may be added that of every hundred pupils studying Latin in the first year of their four-year course at a secondary school, 69 study it for two years, 31 for three years, and 14 for four years; and of these 14 scarcely 5 may be expected to continue it at college. The next section of the report undertakes an examination of the aims or 'objectives' in the teaching of Latin at the secondary stage. The objectives are classified as 'ultimate,' including the qualities which are expected to become part of the permanent intellectual equipment of the pupil, and 'temporary,' which mark stages in the process of acquisition of the ultimate aims, necessary as stages but not necessarily to be retained as permanent possessions (e.g. the power to translate a Latin passage). The relative importance of these 'objectives,' and the extent to which they are attained by pupils, were made the subject of a number of scientific tests and of a questionnaire addressed to teachers. The results are too long to summarise, but it is interesting to note the answers given by the graduates who were asked to indicate the results from the study of Latin which they believed to have been most valuable in their own experience. The seven values receiving the highest number of votes were: (1) the understanding and use of English words derived from Latin; (2) the understanding of English grammar and language-structure in general; (3) the understanding of Latin words, quotations, etc., occurring in English; (4) the development of a historical perspective and a general cultural background; (5) assistance in learning other foreign languages; (6) general discipline resulting from the cultivation of habits of accuracy, thoroughness, orderly procedure, perseverance, and achievement; (7) the understanding and use of Latin technical terms, and terms derived from Latin, employed in the professions and vocations. Of these it may be observed that only Nos. 4 and 6 can be regarded as ultimate aims, valuable in themselves, the others being subsidiary and ancillary to further objects.

A long section follows on the content of the Latin course at secondary schools. The existing curricula are analysed, criticisms are invited from teachers, and a syllabus is drawn up which is recommended for adoption.

The defects found in the existing system are summarised as follows: 'Congestion arising from introduction into the course of too many formal elements, especially during the first year; too early introduction of the first classical author to be read; failure to include in the course abundant easy reading material for the purpose of developing early the pupil's ability to read Latin as Latin; prescription of too large an amount of classical Latin to be read intensively; lack of sufficient variety in the choice of reading material; and failure to give adequate emphasis to attainment of the ultimate objectives.'

It is observable that the supply of easy reading material (composed of 'made' or adapted Latin), which is particularly desiderated, is said to be much better met by English publishers, and much more extensively used in English schools, than is at present the case in America. In this connexion it is instructive to notice (perhaps as an example of the tendency of the human mind to criticise whatever it has got) that at the recent meeting of the Classical Association at Bangor more than one speaker criticised adversely the use of what was described as predigested' food, which pupils were said to regard as fraudulent and unreal; but the Prime Minister's Committee recommended the postponement of the first classical author to the third year, and the use of 'made' material in the earlier stages.

The general recommendations for the reorganisation of the course are stated as follows:

1. That the formal study of the elements of language during the first year be reduced by the postponement of many forms and principles until later in the course.

2. That the vocabulary, forms, and principles of syntax to be learned in each successive year of the course be selected in such a way as to provide conditions most favourable for developing progressive power to read and understand Latin, and for attaining the ultimate objectives which teachers regard as valid for their pupils.

3. That not less than 80 pages of easy, well-graduated and attractive Latin reading material be introduced into the course, beginning at the earliest possible point and continuing at least through the third semester.

4. That this material should be such as to contribute to the aims stated in recommendation No. 2.

5. That practice in writing Latin be continued throughout the first, second, and third years. It may well be omitted from the work of the fourth year in order to allow full time for the reading.

6. That the amount of classical Latin authors to be read in the standard four-year course shall be not less than 35 pages of Teubner text in the second year, 66 pages in the third year, and 100 pages in the fourth year.' (N.B. the existing course includes in all 80 pages of Cæsar, 82 of Cicero, and 128 of Virgil, which is regarded as excessive.)

7. (Advocates more freedom of choice of authors, which is said to prevail in England.)

'8. That such additional material of instruction be introduced into the course as will provide for fuller attainment of various ultimate objectives of the study of Latin.'

Specific recommendations as to books to be read, and as to the method of studying grammar, are added at considerable length; and it is interesting to note the statement that the results secured in the four-year English schools' (from 36 of which statistics appear to have been obtained) 'show that a grade of scholarship much higher than is commonly attained in the schools of our country can be secured on the basis of a considerably smaller amount of intensive reading of the classical authors.'

Recommendations follow with regard to methods of teaching, which are of more interest to the professional teacher than to the general reader. With reference to these, as to so much else in educational theory, one is at times tempted to murmur to oneself :

'For forms of government let fools contest;
Whate'er is best administered is best.'

The teacher matters so much more than his theory. Nevertheless, the teacher may pick up useful hints for his own application from the experience of others, so long as he does not regard the method as a fetish.

The final chapter reverts to and extends that comparison of the records of classical and non-classical pupils which was noted as a feature of Dean West's earlier book. We give the conclusions of the American investigation without prejudice.