much fuel is properly consumed in a given time. And it is a mistake to suppose, as some of the witnesses have done, that the mere extension of flame through the flues, or even to the tops of the chimneys, is always evidence of the goodness of the draft. On the contrary, this extension of flame is frequently occasioned by the want of a proper draft to obtain sufficient oxygen, in the furnace, for the perfect combustion of the carbon of the fuel. The flame which is sometimes seen in the flues and britching, is often nothing but a mixture of common air and carbonic oxide; which oxide is generated near the top of the coal upon the grate, for want of a sufficient draft to supply a large quantity of fuel, laid deep upon the grate bars, with the requisite oxygen for its perfect combustion.

Professor Bartlett, in his answer to the 21st cross-interrogatory, alludes to this generation of carbonic oxide, where the requisite quantity of air is not communicated to the whole mass of coal, so as to convert all its carbon into carbonic acid gas. But if there could be a communication of the requisite quantity of air to every part of the fuel, so as to render the whole mass of coal upon the grate incandescent, no carbonic oxide would be formed at the surface of the coal; and the flame would seldom be seen to extend half way to the entrance of the flues of a thirty feet boiler.

It has been attempted to be shown, by the defendants' counsel, that the requisite amount of chimney draft can be procured by increasing the diameters, or the cross sections of the chimneys; without the necessity of increasing the height. And several unprofessional witnesses have expressed such opinions. But here again, I think, the weight of testimony, even of witnesses who are not acquainted with the principles of natural philosophy which regulate chimney draft, as well as the deductions of science, is the other way.

I believe all the witnesses agree that for the purpose of obtaining the best draft, where the chimneys are limited in height, their cross sections should never be less than the aggregate of the cross sections of the several flues, or other passages, through which the products of combustion pass to the base of the chimneys. But upon the question how much larger the sectional area of the chimneys may be made, with advantage to draft, there is a great diversity of opinion among the witnesses examined on both sides, in this case.

Some of the witnesses think the aggregate sectional area of the chimneys should be a little more than that of the flues; while others suppose it should be two or three times as much. Curry, an engine builder at Louisville, in his testimony before me, says he has adopted the rule of making the area of the chimneys twice and a half that of the flues; without any reference whatever to the extent of grate surface upon which the fuel is consumed.

Professor Bartlett, of the Military Academy at West Point, allows a slight increase of the cross section of the chimneys beyond that of the flues. Professor Renwick, of Columbia College, in New-York, says the sectional area of the chimneys should be only a little more than sufficient for the passage of all the air, necessary to consume the

requisite quantity of fuel, with the velocity due to the height of the chimneys; the object of such increase being merely to diminish the friction of the air in its passage through the chimneys.

Another witness thinks the cross section of the chimneys should be about ten per cent greater than that of the flues. And Mr. Byrne, formerly professor of mathematics in the Institution of Civil Engineers, in London, says the diameters of the chimneys are never increased except for the purpose of decreasing the friction of the products of combustion; and that when the increase of cooling surface more than counterbalances the decrease of friction, any further increase of the cross section of the chimneys will injure the draft.

Russell, in his treatise on the steam engine, does not allow for any increase of the area of the chimneys beyond the aggregate cross section of their flues at the smallest ends. He says that, from an examination of the best boilers, it appears to be decided that flues of one-fifth of the area of the fire grate, gradually diminishing to a chimney which shall have one-tenth of the area of such grate, is an excellent proportion; and that the chimney should be of the same diameter throughout its interior. That if the chimney is 40 feet high and has a cross section of one-tenth of the area of the fire grate, it will give an abundant draft. And Cresy, in his Encyclopedia of Engineering, copies this rule of proportion given by Russell.

It is evident, however, that this rule is given by Russell in reference to low pressure boilers; or where an unrestricted grate surface can be obtained. For he requires that the area of the grate should be one square foot to each horse power of the engine. In other words, a grate surface sufficiently large to burn the coal at but little more than three inches deep upon the grate bars, to evaporate a cubic foot of water to each square foot of grate surface per hour.

If the grate surface is restricted, so as to require the coal to be placed much deeper upon the grate, as it is upon boats navigating the Ohio, it follows, that the size of the flues and the diameter and height of the chimneys must be increased as the grate surface is diminished, in order to carry off the products of combustion more rapidly; so as to allow the requisite quantity of air to be pressed through the passages between the bars of a diminished grate. And as this greater depth of the coal, upon the grate, will also increase the friction of the air in its passage through such coal, the height of the chimneys must be still further increased, to overcome that friction.

The following is Boulton and Watt's practical rule for ascertaining the cross section of the chimney of a land engine; as given by Bourne in his treatise upon the steam engine: "Multiply the number of pounds of coal consumed under the boiler per hour, by 12, and divide the product by the square root of the height of the chimney in feet; and the quotient is the area of the chimney in square inches in the smallest part. A factory chimney suitable for a 20 horse boiler is commonly made about 20 inches square, inside, and 80 feet high; and these dimensions are those which answer to a consumption of 15 lbs. of coal for each horse power per hour, which is

a very common consumption in factory engines. It will not answer well to increase the height of a chimney, of this area, to more than 40 or 50 yards, without increasing the area; nor will it be of utility to increase the area much without also increasing the height."

It may be proper to remark, in this connection, however, that in the English steam engine furnaces, as well as in the furnaces of some of the steamboats upon the Ohio, the aggregate area of the passages for the air through the grate, are but about one-third of the whole grate surface; whereas in the furnaces of the Pittsburgh and Cincinnati packets, the thickness of the grate bars and the width of the spaces between them are about equal.

The quantity of air pressed through the fire grates of those packets, by the same amount of draft, would therefore be much greater than in the furnaces of the English engines. That, of course, admits of a greater area of chimney, in proportion to the height, than would be admissible according to Boulton and Watt's practical rule; even if that rule was applicable to steamboats on the Ohio, where the very restricted grate surface, rendered necessary by the peculiar regimen of the river, requires the fuel to be placed so much deeper upon the grate than on land engines.

I have made a calculation of the ratio between the aggregate sectional areas of the chimneys and of the flues, on six of the Pittsburgh and Cincinnati packets, and I find them to be as follows, or nearly so: On the Messenger No. 2 as two to one; on the Brilliant, as two and a half to one; on the Clipper No. 2, and the Buckeye State, as three to one; on the Keystone State, as three and a quarter to one; and on the Cincinnati as three to one. And, according to Mr. Dickenson's computation, the ratio on the Hibernia No. 2 is also about as three to one.

It will be seen, then, that the area of the chimneys of each of these packets, with the exception of the Messenger No. 2, is as large in proportion to the area of the flues even as Mr. Curry thinks is necesary for the benefit of draft; and is much larger than the scientific witnesses, and various writers upon the steam engine, think advisable.

Testimony was also introduced, on the part of the defendants, to the point that a greater area of grate surface, and a greater heating surface for the generation of steam, might be obtained upon these large boats, by the addition of another boiler.

If the boilers were shortened and there were more of them, they probably would generate more steam in the same time. For it would subject a greater portion of the most effective heating surface of the boilers, to the direct radiation of caloric from the incandescent coal upon the grate. And the fire surface of the boilers, even beyond the bridge wall but adjacent to the furnace, is much more effective than that which is nearer to the back ends of the boilers; where the temperature of the flame and gases is much less.

The lightness of water draft required for the large boats navigating the Ohio, especially upon the upper part of the river, renders it improper to increase their weight by the addition of more boilers,

with their necessary fire beds and other appendages. And the necessity of a free passage way, from the bow to the stern of the boat, on each side of the nest of boilers, for receiving and discharging freight, would render the increase of the width of the nest inconvenient; as the boats land with their bows on, and take in freight only at the bow. A little more grate surface might probably be obtained, with advantage, by lengthening the grate bars upon the Pittsburgh and Cincinnati packets six or eight inches; and in this way also, more of the bottoms of the boilers would be subjected to the direct radiation of heat from the burning fuel. But the bars could not be much lengthened without rendering them too weak to sustain the necessary weight of coal upon them when they were very hot; especially near their back ends, where they are exposed to the most intense heat. Perhaps it might be found beneficial to make the face of the grate bars a little wider, or thicker, at the back ends of the bars where the heat is the most intense; so as to leave the spaces between them a little smaller, where the draft of air through them is the strongest, and a little larger at the other end, where such draft is the weakest. I see by the table of measurements, that the area of the passage, for the smoke and gases, over the top of the fire or flame bridge, upon the Pittsburgh and Cincinnati packets, is from 50 to 100 per cent greater than the sectional area of the flues. It probably should be somewhat larger than the passage through the flues; because the gases, at that point in their passage, are more expanded by heat than they are after they have passed beyond the fire bed. But I am induced to think the difference upon these packets, or at least upon some of them, is too great. If so, more steam would be generated by raising the top of the bridge wall, so as to bring the flame and heated gases nearer to the bottom of the boilers.

Where there is too much space between the top of the bridge and the bottom of the boiler, the products of combustion, which are at their highest temperature at this point, instead of passing close to the bottom of the boiler, its most effective heating surface, and imparting their caloric to the water, are drawn off into side currents, between and at the sides of the boilers; where less heat is radiated laterally. The fire bed upon all of these western boats, should be constructed more in the shape of an inverted arch, under each boiler, rising considerably between the boilers, and at the sides of the nest; so as to spread the flame and heated gases more equally over the whole bottom of each boiler. And the top of the bridge wall should be constructed in the same shape, so far as it can conveniently be done. It should be raised so high as to leave the area for the passage of the products of combustion less in proportion to the aggregate sectional area of the flues than it now is on some of the packets; so as to bring the flame, at that point, as near to the shell of the boiler as practicable without injuring the draft. The principles of science, in this respect, are in accordance with the results of actual experiment. For some of the witnesses show that they increased the quantity of steam generated, on two or three boats, by raising the top of the bridge wall some three or four inches.

The peculiar regimen of the Ohio, so often referred to, requires the boilers of boats navigating its waters to be so constructed as to have great strength, and great evaporative power, combined with the least practicable weight. And this has led to the adoption of the present form and dimensions of steamboat boilers on that river. To give sufficient strength to the boiler, to enable the boat to run with a great pressure of steam, the cylindrical form has been adopted. And to diminish the weight of the boiler and its contents, as well as to enable the engine builder to use iron for the shell and flues thin enough to radiate heat with the greatest attainable rapidity, the diameter of the boiler has been limited to about 42 inches; and the thickness of the boiler iron to about one fourth of an inch. With boilers of that size and strength the Pittsburgh and Cincinnati packets, and some other boats on the Ohio, are enabled to carry a pressure of steam of from 140 to 160 pounds to the square inch, in their usual running.

In the construction of these boilers, to give them the greatest. practicable evaporative power, the heated gases, and other products of combustion, after they have communicated as much as possible of their caloric to the water, through the bottom and sides of the boiler, are made to traverse two cylindrical flues within the shell of the boiler, surrounded on all sides by the water; so as to transfer another portion of their remaining heat, to the water, before they are discharged into the britching through which they afterwards ascend to the base of the chimneys.

This arrangement gives a very large fire surface to the boilers, in proportion_to the area of the restricted grate on which the fuel is burned. But as it is difficult to make heat pass downwards, and as caloric is not so readily communicated to the water in the boiler laterally as it is where the bottom of the boiler, or top of the flues, is exposed to the upward action of the flame and heated gases, only about one half of the inner surface of the flues is considered, by Russell and by Armstrong, as effective heating surface. And for the same reason one half of the area of the vertical ends of the boilers should be deducted; and the other half only allowed as effective heating surface.(a)

To add still more to the fire surface of the boilers, on some of the Ohio boats, the connecting walls between the several boilers, and between the two outer boilers and the sides of the nest, have frequently been raised from two to six inches above the centre of the boilers. And to provide a larger area for the passage of the products of combustion to the britching, at the base of the chimneys, so that the draft may not be impaired, these internal flues have been made very large in proportion to the diameters of the boilers.

In all these improvements for rendering the capacity of the boilers (a) In computing the heating surface of the boilers of the Buckeye State, in the note appended to Mr. Dickenson's table, p. 639 of the testimony, I took hut one end of the boilers as heating surface; and not both ends as is there stated. I believe the engineer, in computing the heating surfaces, also took but one end of the boilers; but he added something to the whole heating surface, for the part of the boilers, above their centres, which is exposed to the flame.