At first there is only a blue flame, in which the hand may be held; but wait until the lining becomes white hot, and then throw on a little coal, and you will find a totally different result. It is also seen in the Siemens gas furnace, with which you are doubtless familiar. There is the introduction of gas with its necessary complement of air. Until the furnace and retorts become heated, the air and gas flutter through only partially united, and do little good; but as soon as the retorts and furnace become thoroughly hot, the same gas and air will melt a fire-brick.

These are common phenomena, which are familiar, but apt to be unnoticed; but they logically point to the truth that no furnaces should present a cooling medium in contact with fuel which is undergoing this process of digestion, so to speak. It will be very evident, I think, from these facts that water-legs in direct contact with a fire are a mistake. They tend to check a fire as far as their influence extends, as a thin sheet of ice upon the stomach after dinner would check digestion, and for the same reason, namely, the abstraction of heat from a chemical process. If fire-brick could be laid around a locomotive furnace, and the grate, of course, kept of the same area as before, it is my belief that a very important advantage would be at once apparent. An old-fashioned cast iron heater always produced a treacherous fire. It would grow dead around the outside next to the cold iron; but put a fire-clay lining into it, and it was as good as any other stove.

If I have now made clear what I mean by making heat, we will next consider the steam boiler. What is a steam boiler? It is a thing to absorb heat. The bottom line of this science is the bottom of a pot over a fire, which is the best boiler surface in the world; there is water upon one side of a piece of iron and heat against the other. One square foot of the iron will transmit through it a given number of units of heat into the water at a given temperature in a given time; two square feet twice as many, and three, three times as many, and so on. Put a cover upon the pot, and seal it tight, leave an orifice for the steam, and that is a steam boiler with all its mysteries.

The old-fashioned, plain cylinder boiler is a plain cylindrical pot over the fire. If enough plain cylinder boilers presenting the requisite number of square feet of absorbing surface are put into a cotton mill, experience has shown that they will make a yard of cotton cloth about as cheaply as tubular boilers. If this is so, why do not all put them in? Because it is the crudest and most expensive form of boiler when its enormous area of ground, brickwork, and its fittings are considered. Not all have the money or the room for them. To produce space, the area is drawn in sidewise and lengthwise, but we must have the necessary amount of square feet of absorbing surface, consequently the boiler is doubled up, so to speak, and we have a "flue boiler." We draw in sidewise and lengthwise once more and double up the surface again, and that is a "tubular boiler." That includes all the "mystery" on that subject.

Now, we find among the mills, just as I imagine we should upon the railroads, that the almost universal tendency is to put in too small boilers and furnaces. To skimp at boilers is to spend at the coal yard. Small boilers mean heavy and over-deep fires, and rapid destruction of apparatus. In sugar houses you will see this frequently illustrated, and will find 16 inch fires upon their grates.

We have found that, as we could persuade mill owners to put in more boilers and extend their furnaces, so that coal could be burned moderately and time for combustion afforded, we often saved as high as 1,000 tons in a yearly consumption of 4,000.

Now, when the ordinary locomotive sends particles of coal into the cars in which I am riding, I do not think it would be unfair criticism to say that the process of combustion was not properly carried out. When we see dense volumes of gas emitted from the stack, it is evident that a portion of the hard dollars which were paid for the coal are being uselessly thrown into the air; and it will be well to remember that only a little of the unburnt gas is visible to the eye.

One point I wish to make is this: We find, as I have said, that as we spread out with boilers and furnaces in the mills, so that we can take matters deliberately, we save money.

Now, coming again to locomotives. I think, if we examine the subject carefully, the fact will strike us a little curiously. The first locomotive built in Philadelphia weighed about 14 tons. Judging from the cut I have seen, I should think her furnace might have been 30 inches square. We have gone from that little 14 ton engine to machines of 50 and 60 tons - perhaps more. The engines have been increased over four times, but I will ask you if the furnace areas have been increased (applause) in proportion? Some of the furnaces of the engines are six feet by three, but that is an increase of less than 3 to 1 of furnace, as against 4 to 1 of weight of engine.

When my attention was first called to this matter, I had supposed, as most people do who are outside of the railway profession, that there was something subtile and mysterious about railway engineering that none but those brought up to the business could understand. Possibly it is so, and I am merely making suggestions for what they are worth, but I think the position I have taken in this matter was established by some experiments of three weeks' duration, which I conducted between Milan and Como, in Italy, for the Italian government, in pulling freight trains up grades of 100 feet to the mile. The experiments were made with an engine built by the Reading Railroad.

We competed with English, French, Belgian, and Austrian engines. These machines required the best of fuel to perform the mountain service, and could use coal dust only when it was pressed into brick. We used in the Reading Railroad machine different fuels upon different days, making the road trip of 120 miles each day with one kind of fuel. We used coal dust scraped up in the yards, also the best Cardiff coal, anthracite, and five kinds of Italian lignite, the best of which possesses about half the combustible value of coal.