testing materials, we are not liable to fall into those errors to which Mr. Goss. Professor Hibbard has referred. I am sure it is true that in all well-regulated laboratories the students in this department gather inspiration, their intellectual processes are quickened, they see purpose in the methods employed, and the working out of natural consequences in the results obtained. The whole tendency of the work is to arouse interest or enthusiasm.

I wish most heartily to commend the thorough manner in which the courses described have been systematized.

MR. WILLIAM METCALF.-The thought that entered my mind Mr. Metcalf. just now in connection with the education of the engineer is the one prominently brought forth in the President's address, and that is, the importance of teaching a student to think. I think that is all that education is worth. I want to say a word here now that may seem proud, but I want to say a word for my alma mater. The one particular characteristic of that school throughout the whole course was, in my day, and I believe practically is now, that the student was taught only one thing at a time. He didn't have two hours a week at this, two hours a week at that, and two hours a week at another subject, and was not fed on mental hash all the way through so that at the end he would know everything: but every day and every week just one plain simple thing until he got through with that subject. The result was said to be tiresome by critics of the school. It was nothing of the kind. There was a little study in the morning, a little scientific recreation in the afternoon-the afternoon always given to some practical work, some laboratory work, drawing, etc., sufficient to teach the student at least the fundamentals of the tools he had to work with. He was always at one subject until he got through, and when he got through with it he generally knew it pretty well.

MR. MANSFIELD MERRIMAN.-In listening to these very inter- Mr. Merriman. esting discussions regarding the conduct of laboratory work, I have been impressed with the large amount of labor spent by instructors to systematize the work and prepare the necessary plans in order that the student may derive benefit from them. It may, indeed, be thought by some that in many instances the greater part of the mental work is done by the instructor and that there is not much left for the student. There are blanks for every stage of the work, and it is sometimes carried so far that there is a

Mr. Merriman. blank at one machine on pink paper and at another machine on brown paper. On each piece of paper there are lines half-printed and the student regards it as an academic exercise to fill out the dotted portions. This is probably an extreme view to take of the matter. If the student should come into the laboratory with the old-fashioned, small engineer's field book with horizontal and vertical lines on opposite pages, and be required to take down notes from the experiments and bring in the next day his own report in the way he thinks it ought to be arranged, it might be more advantageous to him in developing his power of thought. In large classes, however, such a plan would be imprac icable, and it may be said in favor of carefully prepared blanks that they give training in system and order. The papers that have been read hence present excellent solutions of the problems forced upon us in handling large numbers of students, and it must be said that great credit is due to instructors in laboratory work for the systematic methods which they are developing.

Mr. Lanza.

Mr. Stoughton.

MR. GAETANO LANZA.-Only two of the matters mentioned in the interesting and valuable paper of Professor Hatt will be referred to here. First, every competent teacher must constantly put forth his best efforts to make the students think for themselves, and the more completely he can accomplish this object, the greater will be the success attendant upon his teaching; as the student who only memorizes and does not think, is not competent to undertake the solution of such new problems as are constantly liable to arise in the practice of his profession. Second, the object of a graduating thesis is, to my mind, to teach the student how to make investigation. I prefer that the student should, if feasible, select the subject himself, the professor, of course, advising him, and not allowing him to take up a subject in which it is not possible for him to do good work. Then the student should, with the advice of the professor, map out a plan for carrying on the investigation, which, again, the professor should pass upon, and when the plan is properly arranged the student should perform the work.

MR. BRADLEY STOUGHTON.-One thing which you, Mr. President, have so well said at this meeting already, interested me very greatly, and I desire to refer to it again. That is, the propriety of studying, not the details of any practice, but the principles upon which that practice rests. It seems to me that, in this way,

we should find a solution of the various questions we have been dis- Mr. Stoughton. cussing here this evening. A student mind is a limited vessel, and if one tries to fill it too full, and especially too full of details, it begins to spill and leak very fast, so that the amount of knowledge which escapes is altogether disproportionate to the amount which remains. In our over-crowded laboratories it is a great temptation to adopt printed laboratory blanks, routine methods of instruction and systems which enable us to put a number of students through the motions of many tests in a limited time. But is this the best method of concentrating their attention on the principles which we desire them to absorb? True, they become momentarily acquainted with many details, some of the least important of which are repeated several times, but it gives them neither the encouragement nor the time for independent thought which alone impresses them with the underlying principles of their occupation, but which almost every man finds at first both tedious and unattractive.

MR. WILLIAM KENT.-We have heard a good deal lately of Mr. Kent. teaching principles and not details. In order to get a correct knowledge of principles close attention should be given to details. One of my students had a knowledge of the principles of the connecting rod, but about the details all he knew was that it is a bar with two holes in it. He didn't know anything about the strap, key, the shape or thickness. His powers of observation or his habit of attention to details had not been cultivated enough. What he needed was more knowledge of details, as a foundation for his knowledge of principles.

Our professor of electrical engineering recently complained that his Junior students had not been sufficiently grounded in the fundamentals of electricity, information which they ought to get in the Sophomore year. I told him to write out fifteen or twenty questions covering the points in electrical physics which every student ought to be able to pass one hundred per cent. on before he comes into the Junior class. He wrote out the questions covering all the information a man actually needed before entering the Junior work, and there is no reason why every one of these questions ought not have been answered one hundred per cent. They were the fundamentals and went into details all right. The professor of physics was glad to have the list of questions and asked

Mr. Kent. me to give him fifteen or twenty questions on heat, the answers to which should be known thoroughly by every Junior student in mechanical engineering. Physics should be taught with reference to its application in engineering work, and the same might be said of other fundamental studies, such as mathematics and even English.

Mr. Hatt.

MR. W. K. HATT (by letter).-Every one agrees that every effort should be used to prevent an unthinking performance of work on the part of students, and, if necessary, even efficiency of administration and the inculcation of method should be sacrificed to this end. In the paper of the speaker certain practices are described which lead the student to a thinking consideration of his work. It is also true that by properly prepared instruction sheets describing manipulation, etc., the instructor's time can be given to the desired end, and by properly prepared blank forms the student is relieved of purely clerical work and his time can be devoted to the thing in hand. It is in favor of the use of proper forms, that it is part of the duty of the instructor to inculcate system and good habits with respect to recording data. All this general discussion which has followed the paper is of interest but scarcely touches the point. The problem of the paper is one of meeting practical conditions of administration of a course so that the largest use of the equipment may be obtained without allowing the work to degenerate to a mere mechanical performance.

A LARGE HYDRAULIC TESTING MACHINE FOR

UNIFORM LOADS.

By ROBERT A. CUMMINGS.

The following is a brief description of a testing machine and its method of application. It was especially designed and constructed by the author during 1903, for the purpose of investigating shearing stresses in beams of reinforced concrete. These beams were freely supported and of variable spans. A uniformly distributed load was applied by this testing machine, the maximum being 5,000 pounds per lineal foot.

Many mechanical problems presented themselves in connection with the design and construction of this novel apparatus, partly on account of the temporary nature of its employment, but mainly because of the necessity for precision in the recorded results.

Hydraulic power was selected as the most practical for obtaining the large load requisite for the tests. It was decided that the best method for the distribution of the applied load was by means of a series of beams arranged as shown in the accompanying illustrations.* Short lengths of I beams were placed along the specimen beam, the points of application of the load being 12 inches apart longitudinally. Above the shorter lengths of I beams were arranged longer beams, the supported end of each one resting upon the center of two of the shorter, and above these again were still larger beams similarly arranged. This method afforded an absolutely uniform distribution of load upon the specimen under test.

The various specimens tested were 10" x 18" in cross section, the spans varying from 8 feet to 20 feet. It was assumed that the maximum deflection producing failure of the specimen would not exceed 2 inches. A large-scale diagram of the distribution beams was made to show the distortion of the loading with 2-inch deflection at the center. The curve of assumed deflections of specimen was plotted as a parabola, and the outlines of the lower beams

*Acknowledgment is made to the Engineering News and Engineering Record for the cuts used in this paper.