that such a case cannot arise under this system. This is not profit sharing in which losses in business would disturb the relation, but simply an offer from the employer to give the employee a fraction of the gain or saving accomplished in the cost of work. If any saving is made, the employee gets his fraction of it; if no saving is made, he simply gets his wages. Mr. Parker's case of a loss by fire suggests simply to me the need that exists in all these matters of a clear, definite contract at the commencement. The feasibility of framing such a contract is illustrated by the foundry rules which cover three pages of this paper. They are brief; but I think they meet all the contingencies which would naturally arise. The case of piece-work operations with prices for unfixed periods has been touched upon by several of the speakers, and that is a subject which I have referred to at our previous meetings, and which I have always looked upon as a blunder on the part of managers. If you give contract or piece work to a man and tell him that you reserve to yourself the right to reduce his rate at any time, you are simply taking away from him the stimulus to reduce the cost. Workmen know well enough that if they make large wages the employer will cut down their piece rate, and that in order to make larger earnings they must then work harder. The result is that where that system obtains, the workman gauges the point at which he thinks the employer will let him alone, and regulates his work so as not to produce more than that limit. In my experience I have found it exceedingly beneficial to make contract or piece rates for definite and usually for pretty long periods, always for twelve months, and in the case of older jobs, where the work is well understood, the rates are fixed for two, three and sometimes five years. The workman then has an inducement to do the best he can during that period, and at the end of it the reduction of cost has sometimes been surprisingly large. In answer to Mr. Nason's question, in regard to a varied product, I would say, that in my case that difficulty is overcome by dividing the product into grades or classes, each of which has a graded pricing, and the foreman determining which grade the work belongs to at the time it is finished. Want of faith on the part of the men in any system of this kind is a fact.that has to be recognized and which is very apt to continue during the first year. As stated in the As stated in the paper, the best possible argument wherewith to meet it is a cash dividend. In starting this system in the first instance, I encountered that difficulty very generally. The men were either indifferent or else hostile to it, believing that it was some scheme whereby the Company was to get more from them without paying for it. And in cases of that kind all you can do is to simply wait, and perhaps to reason a little with your more intelligent men. Induce them to use their influ ence to carry the thing into effect fairly, and at the end of the year pay them a dividend if it has been earned. Doubts and difficulties will disappear very promptly after the men have received the first dividend in cash. Prof. Webb.-If each man watches the other, and each man notices that the eleventh man is not doing his share, would not they want to get rid of the eleventh man? In some cases, might they not even go so far as to make the proprietor aware of that fact? The President.-I can say that the latter effect might not obtain in some cases, although it has not happened in my experience. The other effect is very marked, that the men are interested in the efficiency of the others about them, and that the men are all interested in economy in avoiding the waste of materials. The tendency of the system is unquestionably to raise the morale of the whole force, so that it acts beneficially in that respect as well as others. CCCXLII. SOME PROPERTIES OF AMMONIA. BY DE VOLSON WOOD, HOBOKEN, N. J. (Member of the Society.) AMMONIA is so extensively used in engineering processes, especially for refrigerating purposes, that it seems desirable to have tables of its properties in English units. For my part I desire to represent the relation between the pressures and temperatures of saturated vapors by Rankine's formula -whenever it can be done with sufficient accuracy. The first two terms of this formula were deduced by Rankine from his hypothesis of Molecular Vortices, but, finding that it did not represent Regnault's experiments throughout their range with sufficient accuracy, he added a third term analogous to the second, containing the inverse square of the absolute temperature. It is well known that this equation represents the results of Regnault's experiments upon several saturated vapors with much accuracy. In order to test it with ammonia, I tabulated the three sets of experiments given in Regnault's Relations des Expériences, Vol. II., pp. 598–607. The results, after reducing them to degrees Fahrenheit and pounds per square inch, are given in the first and second columns of the following table, and are represented by dots in Fig. 130. Having selected three points through which the trial curve should pass, the constants A, B, C, in Equation (1) were determined. The value of C was 0.00178, which, after being divided by 72 gave results so small as to make the value of the third term inappreciable, and it was omitted in all the later investigations. The constants were then found for the locus passing through the points T1 = 435.66°, p1 = 15.8 and 72 = 592.51, p1 = 335.93, and the results compared with the experiments, after which the constants were changed arbitrarily, but slightly, so as to obtain better results. The formula finally adopted was: Com. log p = 6.2469 2200 τ (2) and the pressures corresponding to the temperatures given in the first column were computed and entered in the third column. The Relation between the pressure and temperature of saturated ammonia. differences between the observed and computed results are entered in the fourth column, the + sign being prefixed to those differences in which the computed value exceeds the observed. I have not only had these results verified, so as to be certain of their correctness, but have tried equations containing other values for the constants, to see if the results would agree more nearly with those of experiment; but the equation here given is the most satisfactory. |