signed readings will be of some value in giving the student familiarity with the literature of a subject, but it falls far short of the seminary in other respects. He believed the seminary method could be readily used not only for minor subjects which are treated fully, but also for portions of more extensive subjects. In his own experience he had used this method very largely in the teaching of water supply and of sanitary engineering and found that, without exception, the the students felt that they got a good deal from the work. In these subjects it has been found best to use the method as supplementary to lectures, giving perhaps one third of the time to the seminary. There is ample opportunity for assigning certain comparatively independent portions a few weeks in advance. These subjects are then as thoroughly investigated as possible, carefully written up, read and discussed in class in regular order. Where a certain order of treatment is essential, and where each part depends on the part immediately preceding, it would be impossible to do this. Papers can be prepared, however, and handed in for examination. The presentation of the papers before the class is a valuable feature, as it adds interest to the work and responsibility to the one preparing the paper. After having been read and discussed, papers should be carefully revised and corrected with the aid of the instructor and, if practicable, printed for the use of the students. While the members of the class are good listeners and usually much interested in the paper and discussion, they do not as a rule carry much away with them in the shape of good notes, and as a result would not get a very clear idea

of the subject as a whole if given entirely by this method. This is indeed the serious drawback to its application. Printed notes would obviate the difficulty and besides give valuable practice in revision and abstraction. In comparatively brief subjects this can be done by the aid of typewriter and mimeograph, with moderate cost; supplemented by such notes the seminary method would be the very best possible way of presenting many professional subjects. In other cases it should be carried out as far as possible, consistent with the acquirement of a fair knowledge of the subject as a whole, for the student doing the work is gaining in this way faster than he can in any other. Knowledge of details of the entire subject is not important. What is required of all students is a knowledge of fundamentals and the bearing of each part on the whole; habits of query and investigation will unlock the details when they are needed. Professional subjects can be made as valuable for training as such subjects as History or Economics, and this fact alone justifies the study of such subjects to a very considerable extent.

PROFESSOR ROBERT FLETCHER found the following, which partakes of the seminar method, to work fairly well: The main subdivisions of a course of civil engineering may be drawn out in the form of schedules or programs; the principle references to illustrations, descriptions, to articles of any sort which are important and valuable as bearing upon that topic, may be affixed to the various sections and subdivisions of this topic, and put into the hands of the student. These simply direct his study of the subject. A student is

up a

all at sea when he is first left to himself to work subject. Indexes may be put into his hands, but it is possible that he may fail to get at that which is most important. If the instructor has already arranged from his own reading and experience some of the most valuable references, and puts them into their proper connection with the program of the course of study, the student has something at hand as a guide or model which serves to start him right; he learns the uses of indexes, and gets some idea of the sources of information and, of course, then, from this beginning he can proceed, as far as opportunity will permit, for himself.

PROFESSOR STORM BULL agreed most heartily with Professor Spalding in this matter. He desired, however, to point out one danger which sometimes might destroy the usefulness of the method. In a comparatively large class there will always be students who either dislike the study or who are naturally indolent; these students will not work up the topic assigned to them until a short time before it is to be presented to the class, and because of the size of the class there may, perhaps, be but one assignment for each term or semester. The consequence will be that the students in question will do but very little work themselves, and the information received by the remainder of the class through the presentation of the topics will be of very little value. If the class is small the students may be called on so frequently that this will not occur, and it was, therefore, the opinion of the speaker that the use of the method should be limited to such classes. He would, in conclusion, state as his opinion

that the method should not be used simply for the gaining of knowledge, but in part for learning to investigate a subject rather than for the subjects themselves.

PROFESSOR C. F. ALLEN said that it was understood, of course, by all, that the same methods are not applicable in small classes as in large classes, yet something can be done with this method in large classes, although it can be used more fully in small classes. A modification of this method had been used by him to a slight extent in a class which is neither very large nor very small. The students go so far as to prepare certain subjects upon which the professor has been accustomed to lecture, with the pupose and the result that the student lectures to the class upon the subject assigned. This serves several good ends. The student, after he becomes a practicing engineer, will be called upon to make a report to some body of men, and the fact that he has previously been upon his feet for a similar purpose will be of value to him. In preparing his lecture he will naturally follow something of the methods of his professor. The exercise should not be carried very far, but is a good one for use in a limited way.

PROFESSOR W. K. HATT thought that Professor Spalding quite properly pointed out that the use of the method was limited to certain subjects, and the speaker had noticed the success of the application of that method. It is a fact also that the student in preparing his paper for the seminary gets very much interested in that and is apt to neglect his other work. Because the student must have had his preliminary

training in mathematics and similar subjects, the speaker thought the seminar method should be limited to the senior year and to certain subjects.

PROFESSOR L. S. RANDOLPH found this subject to be one in which he had been interested for a year or two and he had been reaching out towards it gradually. His attention was first directed to it by a description of the use of this seminar method in literary colleges and he had tried it with one of his post-graduate class. The speaker remembers the first time he assigned a topic to one of them. The instructions were to design a turbine wheel and make the general drawings. The work done was surprising. When asked how he got along, the man said he did not know there was so much to a turbine wheel before. When a student works out a subject, passing from one step to another, he gets the relation of the different parts in a way impossible either in a lecture course or by means of a recitation. A student may drop a recitation; he may study one portion of a recitation better than another; but if he has to work out a subject so that one step depends on the other he will get the idea in a fashion impossible by any other method. When he leaves the college and goes out into practical life he has to take the subject up in that way and handle it, and while it is not necessary always to call it a seminar method, or anything of that kind, that principle is of the utmost value in teaching Once or twice the

speaker had gone so far as to plan out a course in which the whole work was to be handled by problems, not only simple problems, but also complicated ones, involving a number of simpler ones, so arranged as to