be an encyclopædia of terms giving such information as will enable anyone to express the meaning of the term clearly in his own words. While the form of words used in a definition need not be always the same, it is of great importance that the same idea should be conveyed by them. By this is meant that the sense in which terms are used and laws or axioms stated, should be uniform. There comes up at once the difficulty as to how this uniformity is to be obtained. Not only is the question of authority as to the meaning involved, but also that of a perfectly unmistakable mode of presentation. One society such as this might appoint a committee, and after great trouble and considerable expense adopt and publish definitions of the more important terms, but the chances are great that almost everyone else would simply ignore the whole thing. As a matter of fact, the work of a society committee has often less weight than that of a single individual. That there is need for some such compilation of definitions, covering the terms used in the different branches of engineering, is evident from the frequent discussions which arise as to the meaning of certain terms and the unsatisfactory characters of much of the information given in even our best dictionaries. That the dictionary fails to be perfectly satisfactory is mainly due to the fact that special knowledge is in many cases required to understand the definitions given, and space cannot be afforded to supply the requisite discussion even if a perfectly safe guide were to be found on the staff compiling the work.

There comes into discussions on this subject a curi

ous mixture of ideas as to what ought to be, and what is, the meaning of a term. This is more particularly the case, perhaps, with single words such as stress and strain, resilience, coefficient, modulus and so forth. We find preconceived notions as to the meaning of the word, or a knowledge of usage in other branches of literature, or simply the unprecise popular usage coming in to make difficulties. Take the terms stress and strain just mentioned. As a result of the fact that in popular language, and often also in technical literature, strain means either the deformation of a body or the system of forces which produces or is produced by it, Rankine was led to introduce the word stress to represent the system of elastic reactions caused by a change of shape in an elastic substance. Strain was then used to signify deformation. Rankine's lead was followed by others, one of the first and more important cases being in Thomson and Tait's treatment of elasticity. In this case, however, it may be remarked, the word stress represents the applied force equal and opposite to Rankine's elastic reaction. There may possibly be room for some difference of opinion as to whether the sense in which these words are used is perfectly consistent with their previous meaning, but that is a small matter. The most important question is, do we require two terms, and if so do those given satisfactorily fill the bill? So far as the necessities of the case are concerned, it may be possible that, so long as we keep to such applications as are common in practical work, the double meaning of the word strain gives little difficulty, and the beautiful indefiniteness of the word may even be advantageous. When we try to discuss prob

lems in the theory of elasticity, however, it becomes absolutely necessary to make a clear distinction between quantities of different kinds.

The words stress and strain have been taken for the purpose of illustration, partly because nearly everyone is familiar with the discussions which frequently arise concerning them, and also with the very loose way in which they are used. There are many other terms used in as indefinite a manner, but it is not necessary to multiply examples. Neither does it seem necessary to bring into this discussion any question as to the actual meaning of particular terms. A slight or even considerable difference between the meaning of a word in a particular branch of science and its ordinary everyday meaning is not a serious matter so long as we maintain perfect uniformity in its scientific use. When introducing new terms it is, of course, well to avoid, as much as possible, the use of common words in any but their ordinary meaning, but there are also some advantages in using words which from their ordinary application suggest at least the character of the physical quantity to which they are applied. One great difficulty in this subject is just this introduction of new terms, and in my opinion it is better to go to a little trouble to use language already familiar, even if it be necessary to use a sentence where a word would do if we had one to stand for that sentence. A special vocabulary and a list of definitions at the beginning of a book is an abomination to most people and, although it saves space, it is apt to render the book much less attractive and harder to read them if it were twice as long.

Standard works on engineering agree fairly well in their use of terms. There are, of course, a number with regard to which there is diversity of usage; for example, we sometimes find in one work the term coefficient of resilience, while in another the same thing is called modulus of elasticity. Again, there is difficulty with terms like isotropy and homogeneity and so forth. A little care would avoid such trouble, because in most cases the words are made to appear synonymous, not so much from want of knowledge as from carelessness. Generally, however, in the better class of books, a careful reader will find no difficulty in understanding what the terms imply, although there may be some annoyance from the fact that in his own vocabulary a different term is used. Societies of teachers can do a great deal to produce uniformity in the use of terms and also in the simplification of definitions and methods of treatment, if only some guide can be got, and together with that, a determination to bury personal prejudices.

When we turn to popular engineering literature and to journal articles we do not find anything like agreement in the use of terms. In fact, a great deal of this kind of literature is more an obstacle than a help to progress. We suffer from a superabundance of journalistic literature. So much is printed that, in order to keep up the quantity, almost any quality has to be used. It is doubtful whether anything but slow development can be looked to for a cure in this direction. There is no doubt but that the rapid spread of scientific education will quickly have its effect in raising the standard of popular literature. It is probable,

however, that it will always be those that know least who are most anxious to write, and therefore that the standard will have to be raised by the discouragement of poor publications.

There is one direction in which a good dictionary of scientific and engineering terms would be of great service. That is in the writing and interpretation of specifications. If it were possible to refer for the interpretation of all terms to a standard work, a great many troublesome disputes might be avoided, and a great deal of time might be saved which is now used in carefully safeguarding the statements made in the specifications. Such an application would, besides the convenience of making uniformity, bring the terms in their proper meaning before practical men and hence, through the workman, into practical use.

DISCUSSION.

PROFESSOR J. GALBRAITH remarked that the subject of this paper overlapped, to some extent, the work allotted to the Committee on "Uniformity of Symbols for Engineering Text-Books," the meanings of terms having to be considered in connection with the question of notation. The paper instanced stress and strain as examples of words whose usage in scientific works is in a very unsettled state. The speaker might mention another word in the same category, a word which of all words one might expect to be used in one sense only, viz., the word unit. In the expressions "unit length," "unit mass," "unit velocity," meaning, e. g., a foot, a gramme, a mile per hour, respectively, the word unit is used in the correct sense. Can this be