the blowpipe regenerative plan, by the use of a strong flow of gas and a steam or other blast driven through the air flues of a regenerative furnace. This produces a furious fire that is pretty to look at, good to stand away from, and which must be difficult to work with, besides being very destructive. The third is the regenerative plan, pure and simple; it consists in relieving the gas from pressure, increasing its volume, splitting it up, so that air can get to it, and then mixing it with a sufficient quantity of air as hot as can be made. The slower and lazier the movements of the gas and air the better, and the result is a beautiful soft, intense heat, that gives us the greatest amount of work with the least wear and tear. These methods are paralleled in using coal by the reverberatory style, the blast style, and the regenerative gas style. Except for the

necessary use of blast furnaces the regenerative gas system is incomparably ahead of the others both in efficiency and

economy.

"Notwithstanding the doubts of the so-called conservative men, it is a fact that 2240 pounds of muck-bar can be made in this way with fifteen bushels of slack; and the record of a whole year's run, including the drowning-out by floods and the falling out of large quantities of very thin steel scraps shows that by the most adverse figuring the cost of fuel per 2240 pounds of product could not be made up to quite sixty cents, and the natural gas men positively refused to include that furnace in an offer to furnish gas for the fuel bill, because they were getting more than twice that sum from the iron mills for their puddling furnaces. If then by the indirect and expensive process of splitting up

solid coal in a gas producer into a gas of the average composition of 70 per cent. nitrogen, 10 per cent. carbonic acid, and 20 per cent. carbonic oxide, we can obtain the great economies which we have already secured, we ought certainly to be able to do much better with natural gas, which is all combustible. To change a regenerative furnace over to the use of natural gas is a very simple matter; it is only necessary to use one-fifth of the volume required of producer gas to relieve it of pressure, split it up and mix it with five times as much air per volume as the producer gas required. This air is obtained in a continuous regenerative furnace by supplying the one-fifth volume of natural gas to the gas ports and using the ordinary supply of air; the equation then reads: Producer gas (one-fifth combustible) + air, one-fifth natural gas (all combustible) + air. Practice shows that that equation is wrong, and the expression should be: Producer gas air, less than onefifth natural gas + air, because there is a large gain in effective heat due to the absence of the four-fifths of noncombustible gases which the producer makes, and which have to be kept up to the temperature of the furnace. In the Siemens regenerative furnace, the natural gas is applied cold into the ports, and all of the chambers are used for heating the air. This looks like a perfect arrangement and it comes very near it in our present state of knowledge, but it is conceivable, if not possible, that an absorbent of nitrogen might be discovered, which would allow us to have a supply of pure oxygen to apply to our natural gas that would make combustion perfect."

All of these different modes of burning natural gas may be seen in operation in the city of Pittsburgh.

As may be supposed, invention is now being severely taxed to discover the most efficient and economical method of burning this gas. The results of a number of average tests show

1. Compared with coal gas, natural gas exceeds it in calorific value, 33 per cent.

2. With crude, ordinary, and best method of combustion, the calorific value of natural gas compared with coal under · best condition is

=

With crude method, 20 cubic feet 1 pound of coal. With ordinary method, 11.29 cubic feet = 1 pound of coal.

With best method, 8.92 cubic feet = 1 pound of coal. 3. The value of 1000 cubic feet of natural

with coal at $1 per ton of 2000 pounds is—

gas compared

With crude method of carbonization, 2.5 cents.

With ordinary method of carbonization, 4.43 cents. These careful tests of the calorific power of gas were made, first, by measuring the number of cubic feet required under ordinary conditions to evaporate a certain number of pounds of water. These experiments were made by Mr. C. E. Hequembourg, of Bradford, Pa., using a boiler 14 feet long, 62 inches in diameter, and having 96 three-inch flues. With this boiler he made a number of six-hour tests, using gas for fuel under all sorts of conditions: then reversing the firing conditions by substituting coal. As a result of these experiments, he found that

he could evaporate 8.55 pounds of water with one pound of coal; the same quantity of water was evaporated with 67.97 cubic feet of gas. A test of a similar character in another establishment went to show that the calorific effect of a ton of Pittsburgh coal equalled that resulting from the consumption of 22,000 feet of Butler gas. With coal at $1 per ton (Pittsburgh price) the gas might be estimated as costing 4 cents per 1000 feet, other estimates make it between 7 and 8 cents per 1000 feet. The committee of engineers referring to some experiments, made under their supervision, make the following observations:

"As the introduction of natural gas in this city has been of such recent date, most of its users consume it in such a crude manner that they fail to get the best results, the difficulty being the expense of making the necessary changes in the burning. There is, however, one notable exception among the large consumers, namely, the Union Iron Mills of Messrs. Carnegie, Bros. & Co., where it is being used with economy in Siemens' regenerative furnaces.

"An experiment was made to ascertain the value of gas as a fuel in comparison with coal in generating steam, using a tubular boiler of 42 inches diameter, 10 feet long, with 4 inch tubes. It was first fired with selected Youghiogheny coal, broken to about 4-inch cubes, and the furnace was charged in such a manner as to obtain the best results possible with the stack which was attached to the boiler. Nine pounds of water evaporated, to the pound of coal consumed, was the best result obtained. The water was measured by two meters, one on the suction, the other

on the discharge. The water was fed into a heater at a temperature of from 60° to 62°. The heater was placed in the flue leading from the boiler to the stack in both coal and gas experiments. In making the calculations the standard 76-pound bushel of the Pittsburgh district was used, 684 pounds of water were evaporated per bushel, which was 60.90 per cent. of the theoretical value of the coal. When gas was burned under the same boiler, but with a different furnace and taking a pound of gas to be equal to 23.5 cubic feet, the amount of water evaporated was found to be 20.31 pounds, or, 83.40 per cent. of the theoretical heat-units were utilized."

The advantages resulting from the use of a fuel so cheap, pure, and easily managed, cannot be over-estimated. In the manufacture of glass, it has been found to be especially valuable. Its freedom from dust, ashes, etc., has improved the quality of the product. One of the great difficulties in the manufacture of glass has been that of properly regulating the heat during the annealing process. This difficulty is removed almost entirely by the regular and graded flow of gas. By this means glass of a really superior quality can be furnished, as the molecules composing it come to their normal positions without strain. This is due to the complete control which is maintained over the heat during the gradual process of annealing the glass.

Natural gas appears to be applicable as a fuel to nearly every purpose where coal has been used hitherto. Success, however, has not followed its use for smelting ores in the