be, except by his compensation in salary. This sometimes is liberal, but often otherwise. It is but seldom that the superintendent owns much of his company's stock, and he generally remains so poor that no matter what may be the prospective or actual profits, he can get none of it.

Now, in view of these facts, it seems to me that there ought to be some rule or custom established allowing a certain percentage of the value of results, after they reach a certain limit, to go to the superintendent; and while this would be only simple justice, it would stimulate men to make extraordinary efforts, and the man who failed to accomplish something under such circumstances might be truly pronounced a failure.

The organization, then, will be as follows:

The superintendent to have charge of everything until the gas leaves the burner, and the secretary to have charge of everything connected with his office after the gas leaves the burner -both to be responsible to the president, and neither to interfere with the duties of the other.

If either fails to meet the requirements demanded by his office, the president has but to show this, and hold each to a strict accountability.

Of course the performance of these duties will depend largely upon the capacity of the president to determine when they are properly executed, as he ought not to accept less than agreed upon, or expect more.

On motion, a vote of thanks was tendered to Mr. McIlhenny. MR. CARTWRIGHT-There seems to be some misunderstanding in regard to the phrase used by Mr. McIlhenny, in the paper which he has just read, and it may, perhaps, be well to have it explained. He spoke of the control of the several officers which he has named, ending with the burner. I think it would be a little more satisfactory if he would make that expression a little more explicit. Does he mean that the control ceases at the end of the burner, or when the gas leaves the burner?

MR. MCILHENNY-The language used is that the secretary should have charge of everything after the gas leaves the bur

ner, which includes all the bills, and that the superintendent should have charge of everything before the gas leaves the burner, which includes everything up to the time the bills are made-meters, street mains, the works, and all the details incident to the production and distribution of the gas.

Mr. LittleHALES-I presume it does not come within the scope of the secretary to trace the gas from the burners. I suppose his duty ends at the meter rather than at the burner. I presume that those who copied the paper did not copy it right, and the term meter" should occur where the term "burner" occurs.

66

A MEMBER-The presumption is that there is no loss between the meter and burner.

MR. MCILHENNY-Bills, of course, are made out at the meter; but the consumer knows nothing about the gas until it reaches the burner. That is the proper point of designation between the time it leaves the holder, and the time it is registered on the bills, or on the meter, which is the same thing. I make the burner the dividing line between the duties of the two officers, the superintendent and the secretary. Of course I am not laying this down as a rule. I do not suppose there are many who will follow it; but it seems to me that, in order to simplify these things, and in order to get the best results, it is necessary to divide the responsibility in some way; and I have done it by making the secretary responsible after the gas is registered, and the superintendent responsible before it is registered.

MR. HERZOG-There is one thing to which Mr. McIlhenny refers and which he desires should be introduced here, that has been put into successful operation in the large Continental gas companies, and that is giving the superintendent a share in the profits to a certain extent. Some of these large companies pay their superintendents a small salary, and, at the same time, give them a percentage of their earnings. I believe that is the just and fair way for the companies to do, and it has worked very successfully there. In one place in Germany they have about 15 or 18 gas works, and they give their superintendent a certain portion of the net earnings of the company; and the

talents and ability of the superintendent contribute to the prosperity of the concern.

MR. NEAL-I really do not think some of the members understand what Mr. McIlhenny means by the secretary to have charge of the gas after it passes the burner. I do not think that is exactly what he means, because, after the gas passes the burner or the meter, what control has the secretary over it? He has sold it, and it passes out of his hands. I apprehend that he means this: that the secretary is to have control of the meter after the gas passes it. After the gas passes the meter I do not see how he has any control over it.

MR. MCILHENNY-What I said was that the secretary should have control of everything after the gas leaves the burner. MR. NEAL-What is everything?

MR. MCILHENNY-All the details.

MR. NEAL-The gas has passed the meter and burner, and now everything, as Mr. McIlhenny says, is in charge of the secretary. He makes out the bills, he collects the bills from the consumers, and the money is paid over to the treasurer. So it seems that the secretary does not have control of quite everything. I think the money that is paid over is quite an important element.

MR. MCILHENNY-I stated in the paper that the secretary had charge of the clerks and meter-takers, and all the details incidental to the collection of the bills; and that the money was to be paid to the treasurer, and that he was to deposit it and make a correct return of it to the company.

MR. HARBISON-I think there is one point in connection with Mr. McIlhenny's paper that the Association did not take in, and that is this: as he was reading his very interesting paper, and as the points were being brought out, it was very clear to me, as he looked around the room, that he must have had a very intimate acquaintance with the personal merits of the members of this Association. In defining the qualifications that are necessary for a first-class president, he had in view, I am sure, half a dozen gentlemen who are present; and when he was defining the duties and qualifications of a secretary, he also cast a meaning glance around the room. I want

to call attention to this point, because it shows that his acquaintance is not simply limited to gas making, but he is also well acquainted with the qualifications and the calibre of the members of this Association. [Laughter.]

DR. SLOANE then read a paper as follows:

RELATION OF THEORY TO PRACTICE IN THE GAS FLAME.

It is well known to us all that heat, in its industrial applications, is wasted to a great extent. The old problem of the conversion of heat of low into corresponding heat of high temperature, has to be solved, and to its non-solution the waste in most cases is due.

Theoretically, one pound of coal should melt forty pounds of iron, yet this result is never even approached in metallurgical practice.

I wish to examine very cursorily the loss of heat in a gas flame. Light is here produced by intense heat; the combustible gases are peculiarly fitted to produce such heat, because, as they are already in the gaseous state, they do not in their combustion have to make the step from solid to gaseous, as solid combustibles do.

The light given by the flame is due to the ignition of particles of solid matter. Carbon is present in the unignited gas in combination with hydrogen, as constituent atoms of gaseous hydrocarbon. When the gas burns, the heat of the flame decomposes these compounds; solid carbon, in a state of almost molecular division, is liberated; and becoming white hot or incandescent, emits light. It travels upward through the flame, until, reaching the top and outer zones, it is burnt into carbon dioxide, commonly called carbonic acid gas.

All the light of gas is due to the ignition of this carbon. The proof of this fact is this: while gases and liquids ignited may give light, it can always be determined by spectroscopic analysis whether light is due to an ignited gas or not. Theoretically and in fact, the spectrum of light derived from the ignition of a gas is discontinuous, while liquids and solids ignited give a continuous one. On subjecting the light of illuminating gas to this examination, it is found that it is due to the ig

nition of solid or liquid matter, because its spectrum is continuous. As there is no reason for the belief or probability that there is a liquid there present, we find that its light is derived from an ignited solid; and by analysis we find that carbon is the only substance present that remains in the solid state at elevated, or even ordinary temperatures.

This proves the fact. A quasi proof is found in the deposition of carbon, in the form of lampblack, upon a cold body immersed in the flame. Therefore, as a light producer, all the work of a flame is the heating of carbon to a white heat.

'I shall now proceed to consider the arithmetic of the process, to show what waste of heat takes place, and what a sacrifice of economy to convenience the whole operation is. I will start with a gas composed of the following principal constituents:

This represents about such a gas as is now delivered in New York. For ease of calculation, I will use the analysis by weight, and refer everything to 100 parts of the gas.

The olefiant and marsh gases supply the flame with its solid carbon; and, to make every allowance, we will assume that all the carbon contained in them is dissociated, and plays its part in producing the light of the flame. They contain 56.67 parts of this carbon. The heat developed by burning each of the combustible gases, has been determined by physicists. If we multiply the amount of hydrogen and other combustible gases, each by its own heating power, we shall have the total heating power of the flame. Thus, one part (by weight) of hydrogen will raise 34,462 parts of the water one degree centigrade; therefore, 34,462 X5=172,310 is the number of parts of water which the five parts of hydrogen in the gas under consideration will raise one degree centigrade in temperature.

Performing a parallel computation for all the constituents.