an extremely low figure, and it would have interested me much more if I could have known all the circumstances of the trial, especially the heating power of the coal; but the last element was entirely wanting. So that we are unable to decide whether that triple expansion engine, using 1.23 pounds of coal, was really more economical than the one just referred to using 1.63. I wish that this element might be given so that efficiencies might be more accurately determined. In all cases reference should be made to the quality of the coal when an analysis cannot be reported.

Mr. Wm. H. Odell.--I want to cite a case which has been under my own observation for the past five or six years in the city of Binghamton, in the State of New York, in the mill of Joseph P. Noyes & Co. They have a wonderful supply of water, but from the time that cold weather sets in they invariably run their engine and boiler in preference to using the water. They have found by careful observation that the cost of running the mill is just the same whether they run by water or steam. It requires the same amount of steam to heat the mill if they run by water as if they run by steam exclusively, and use the exhaust from the engine to heat the mill.

Mr. Durfee.-As a further illustration of the point raised by Prof. Wood, I would say that I have lately removed some machinery from one building to another and changed the boilers; under the first boilers we used bituminous coal. Under the new boilers we use "pea and dust," a very much cheaper coal per ton, but I found that the money consumed in driving that machinery was practically the same in each case, the last fuel costing a great deal less per ton, but we were obliged to use more tons of it, so that the actual cost was practically the same under the two conditions.

Mr. Borden.-May I ask Prof. Denton to repeat his question about steam and water which he asked just now?

Prof. Denton.-Given a case that has plenty of water, and another case using steam, why is it that each of them sells his goods equally cheap? Why does not the water mill undersell the steam mill?

Mr. Borden. The selling price of goods is not fixed by the cost of production in the establishment which can produce them most economically, but rather by the cost in a concern that can just sustain itself without making either profit or loss.

Human nature is such that all who can do better than that

cheerfully pocket the profit, and those who cannot do as well go to the wall, leaving the selling price to be determined by those at the foot of the class, rather than by those at the head.

Steam power located near the coal fields and the leading markets may have sufficient advantage in the transportation of fuel and raw materials and its finished product, over a water power located at a distance from both coal fields and markets, to fully offset the advantages which the latter would possess over the former if both were located at the same point. This is especially true of establishments requiring a considerable quantity of steam for other purposes than power.

Prof. Denton.-How as to Fall River?

Mr. Borden.-Fall River is quite accessible both to the coal fields and the markets as compared with points farther east, but the water power there is not of sufficient magnitude to be of much importance, more than seven-eighths of the power used there being from steain.

The large water powers at Lowell, Lawrence, Manchester, Lewiston, Biddeford, Augusta (Me.), Augusta (Ga.), Holyoke, Cohoes and many other points that might be named, if used within their limits of minimum flow, plus reserves available, are decidedly more economical than steam power, on the basis of original cost of development, although on the basis of present charges made by original owners they may not be.

At several of the localities I have named the rivers or their tributaries have their sources in large lakes, the outlets from which are owned and controlled by the companies that own the water privileges on the stream. Very large volumes of water can be held back during the winter and spring, and let down during the summer. If the power attempted to be used on these streams was held within the limits of regulation by those large sheets of water, the power available would be as regular as if produced by steam. Prof. Denton. Then I understand there is no location in New England where large manufactures are carried on where there is plenty of water running all the year round.

Mr. Borden. The large water privileges in New England have plenty of water to drive a large amount of machinery the year round, but if construction ceased at that point there would be a considerable amount of unused water running by during two-thirds or three-quarters of the year. For the purpose of utilizing this water, many manufacturers put in additional machinery, to be run

by water when it is available and by steam during the remainder of the year. The more common reason however why many establishments have outgrown their water power is that the natural increase of their business has required enlargements of their plant. There are important advantages in making the increase in connection with existing plants, rather than by seeking new locations for such increase, although the adoption of steam power therefor is relatively more expensive than would be justifiable in an entirely new establishment.

Mr. W. S. Rogers.—I would state for Prof. Denton that I recall a case in Cincinnati where there is one large cotton mill that is located where they have ample water, sometimes all the year round and sometimes for two or three years; occasionally they have a break every two or three weeks. But they did not use to be that way. I notice that they are not using their water wheel at all, but are using their engines for steaming, heating, drying and running their establishment, and they are located right on the race. The firm have dissolved and one member has started in business for himself, and instead of starting where he had plenty of power and could use the same water, he went, I should judge, four hundred feet from it and put up his mills and uses a steam engine, and his mill runs just as many hours as the other does, and he can go to market and sell just as cheap and compete with the others, and he is no nearer the coal field and has no advantages over the other mill and the other has none over him, while the other has abandoned the water wheel and is using steam. So I think that steam there is the best.

Mr. L. F. Lyne.-Before the discussion is closed I would like to make a remark having reference to the remarks of Mr. Durfee, which bore upon a very important subject, and in corroboration of his statement in reference to the equal money value of bituminous coal as against pea and dust. During the five years I have operated some boiler furnaces and experimented to some extent with those fuels, and I have found that his statement is correct. Another element which places the bituminous coal in a more economical light than the pea and dust is the fact that it is not so destructive to the grate bars and furnaces. In a run of five years we still have the same grate bars that we started with, and all the repairs that had been made to the furnaces during that time is the replacing of four courses of brick just abo e the grate bars, to say nothing of the prevention of the destructive action of the sulphur which you

find in impure anthracite upon the inside of the iron chimneys. which convey the smoke and gases to the atmosphere.

Mr. Chas. H. Manning.-In Mr. Main's first assumption of the cost in coal per horse power of the three types of engines, I think he is hard on the non-condensing engine in charging it with 3 lbs. of coal, even if the feed temperature is as low as 100° Fahr., which is fair for the condensing engine, but unnecessarily low for the non-condensing.

When using a pressure of 100 lbs. per gauge, the benefit of condensing is very small, and with the less initial condensation due to less cooling of the cylinder during the exhaust, the higher temperature of feed water and the saving of power necessary for an air pump, the non-condensing engine with 100 lbs. initial against 5 lbs. back pressure will show an efficiency very close to that of a condensing engine with 80 lbs. initial pressure.

Mr. Main says that "the difference in the amount of condensation by passing the steam through an engine or passing it through pressure regulators and pipes should be charged to the power."

This would give the casual reader the impression that passing through a reducing valve caused condensation, which certainly Mr. Main does not intend to convey, as during the passage of the valve there is free expansion, therefore superheating.

When using all the exhaust steam for heating, etc., he charges 25 per cent. of the fuel to power, wherein I am confident he is again in excess of the facts unless the cylinder volume is larger than it should be for the work done. With an engine running against a back pressure of from 5 to 10 lbs., or in the case of the high pressure cylinder of the compound engine, the cards at the terminal pressure should account for from 85 to 87 per cent. of the water evaporated in the boiler.

In support of this I would cite the case of a pair of non-condensing engines at the Amoskeag Mill, Manchester, N. H. 36" diameters of cylinder, 6' stroke, and running 60 revolutions per minute, which are frequently started at from nine to eleven hundred horse power without making any change in the fire-room, or without its being apparent there in the coal consumption, the increase of coal cost being within the daily variation from other causes. More steam than is required to run them at this power is needed in the dye houses at all times, and its course is merely changed from through the reducing valves to through the engine.

My belief is that it costs less than one-half a pound of coal per horse power per hour, but as all the steam is drawn from one general system it is hard to apportion the costs exactly.

In a large proportion of the New England cotton mills the stean power is auxiliary to the water power, and the amount of power required from the engine is a constantly varying amount. Under these circumstances the exact determination of relative diameters of the cylinders of a compound engine becomes impossible, since what is right with one load is wrong with another, and under these circumstances, I think, a smaller ratio than that of 1 to 4, as suggested by Mr. Main, or from 1 to 3, or 1 to 31, is much better.

The tandem type under these conditions is much preferable, as from the unequal distribution of power between the two cylinders, necessitated by a constant receiver, pressure is less objectionable.

In a marine or a pumping engine the maximum power is required nearly all the time, but even then with a boiler pressure of 100 lbs., I should choose a smaller ratio than that laid down by Mr. Main.

For maximum power it is not well to expand below 10 lbs. absolute, and with 115 absolute initial this would give about 11.5 expansions total, and √11.5 = 3.39, or roughly, the ratio of cylinders should be about 1.4.

Mr. Main.-The first question asked was by Prof. Denton. I would say to him that I have not been able to get any figures that extended over a long enough period to say exactly what the cost of repairs would be, but have established the cost at two per cent., as given in the paper for all cases.

Almost all the questions asked have been answered by other persons.

Mr. Babcock speaks of the case of a woolen mill where all the exhaust steam can be used. I would say that before putting in this compound engine at the Pacific Mills we had a five hundred horse power high pressure engine, and all of the exhaust steam from that engine was used in the dye house. When the engine was shut down the fuel consumption was just as much, and sometimes even more, than when the engine was running, and the way that I explained that to myself was that the engine was a sort of regulator on the dye house, and they could not get any more steam than went through it, and when the engine was not running they could draw more heavily (laughter).

With reference to the cost of power as produced by the appa