do not try to force them.

Ambitious attempts at orig

inal research will have such a tendency.

These are in brief the arguments on both sides of the topic under discussion which come to the mind of the writer, who has no desire to appear as an advocate nor to hold such fixed views that they are not subject to change. Other facts will no doubt be added on both sides of the question in the discussion. The fallacy or lack of weight of some of the statements made will be exposed. When the arguments are over and an understanding arrived at, let us hope that something will have been contributed toward the better teaching of engineering in our technical schools.

ORIGINAL RESEARCH BY STUDENTS,

BY R. C. CARPENTER,

Professor of Experimental Engineering, Sibley College, Cornell University, Ithaca, N. Y.

It is, no doubt, an open question as to the amount and extent of work of the above character that can be profitably undertaken by students in a college course. There are many engineers who feel that work of investigation cannot properly be done by men who have not completed the preliminary training for professional work, on the general grounds that immature students are incompetent and inexperienced. These feelings are no doubt founded on good and sufficient reasons in many cases, but it is also no doubt probably true, that much depends upon the individual student and also upon the training he may have had, so that it would be unsafe to lay down any general law or dictum which would be applicable to all cases. What the writer has to say must be understood as limited by certain conditions, these being, First, the wisdom of requiring students to undertake as a part of their educational work, some branch of investigation or research. It is presumed of course that the student has had the training that would fit him properly to undertake the work proposed. Secondly, the probability that the work would be suggested and laid out by the professor so that the part performed by the student will be in some

respects subordinate, although in the remarks that follow it may not be so characterized. Thirdly, the value of this class of work to the student as an educational exercise, although this will not be discussed at length in the paper.

The object of the paper is rather to set forth the writer's own views as influenced or modified by some years of experience with this class of work.

It will be noticed that the term "research" is used in such a sense as to cover nearly the entire field of "experiment," and is not confined to investigations in which the work is entirely along original lines. While no doubt a great deal of this higher class of work can be and is done in colleges, from its very nature more time is required than can generally be spared in an undergraduate course.

The work that is actually accomplished is in nearly every case the execution of experiments which are systematically planned by the head of the department. This work is here called original research, and it is to this class of work that this paper principally applies. This is original research only to a limited extent, and yet this work will afford ample means of exercising the mental powers, the accuracy of observation, and even the handicraft of the student. Many students will never succeed as observers, and all students need systematic training for such work before they can acquire the habits and methods necessary to obtain reliable results. This preliminary training is of value to all students, whether they apply it in original investigation or not, since it gives them a practical acquaint

ance with measuring apparatus, a useful application of theoretical principles, and an opportunity of studying the material or machine itself.

All the students in Sibley college, in both electrical and mechanical engineering courses, have a preliminary training in both the physical and the engineering laboratories, for the last two years of the time spent in college averaging four afternoons per week, and this preliminary training, which is useful in many other ways, is expected to make the men so familiar with methods of research that they can take a subject for investigation and follow it through a series of careful experiments to a logical conclusion. The engineering laboratory experiments, after a series of trials, were found to give the best results when arranged in a formal course, which was prescribed for all the students, not however in the same order. There is a special reason for this formal arrangement of engineering experiments, since it permits the carrying out of a system of instruction, in which all the students participate and in which the methods can be given great prominence.

In nearly all the engineering experiments "methods" are of great importance, since the subject-matter or result sought is rarely a "constant of nature," but on the other hand is a co-efficient which will vary to a great extent with the system adopted. This may be made clear by an example. Suppose strength of cast iron, it will

the subject is the be found that in this case, the form and size of the the specimen and the method of applying the stress

will affect the result, and unless a certain and well understood scheme of testing be adopted, the result will have little value for future reference or comparison. Fixed schemes or standard methods which refer results to well known units, must be largely adopted in engineering experimental work, and students need training in such work before they can reasonably hope to succeed in original investigation.* The list of the experiments in the the regular laboratory course of Sibley college may be of interest, and is as follows:

JUNIOR YEAR.

FALL TERM-Strength of Materials: Tensile and transverse tests. Calibration: Steam guages, indicator springs, weirs and water meters, thermometers, pyrometers, dynamometers. Experiments: Flue gas analysis, air thermometer.

WINTER TERM-Strength of Materials: Compression and torsion. Lubricants: Co-efficient of friction, viscosity. Steam Engine: Calorimetery, indicator practice, heating values of fuels. Calibration: Planimeter, indicator drum springs.

SPRING TERM-Strength of Materials: Impact, Keep's tests of castiron, cement tests. Lubrication: Flash and chill test, durability. Efficiency Tests: Water motor, steam pump, steam engine.

SENIOR YEAR.

FALL TERM-Strength of Materials: Elastic curve, brick, stone and cement. Steam Engine: Valve setting, Hirn's analysis of simple engine, efficiency test of compound engine. Efficiency Test: Gas engine.

WINTER TERM-Efficiency Tests: Injector, triple expansion Corliss engine, hot air engine, oil engine, turbine water-wheel, wind-mill, water engine.

SPRING TERM-Regular work not required. Nearly the entire term is devoted to special research and investigation, which is embodied in a thesis.

Note: Students work as far as possible in groups of two.

*The writer has prepared and published a work on Experimental Engineering (J. Wiley & Sons, N. Y.) in which full directions and standard methods, as followed in Sibley college, are given.