stick together. But the small particles of a charge of ore mixed with salt are exactly in such a condition while roasting as to have the greatest possible inclination to sinter and adhere to the shelves. They would thus soon obstruct the whole shaft, and prevent any further work. This has been demonstrated by actual experiments on a working scale. It is apparent, therefore, that the application of the Gerstenhöfer furnace, even for desulphurizing purposes, is very limited, and that certain classes of ore must be entirely excluded from it. This is especially the case with galena ores, which are the most expensive to roast in reverberatory furnaces.

In Stetefeldt's opinion, the shelves in the Gerstenhöfer furnace are perfectly superfluous, and all ores, even galena, can be desulphurized by dropping them through a plain shaft heated by fire-places below, if they are reduced to a sufficient degree of fineness. The escape of unroasted dust from the shaft is of no consequence, as a separate fire-place is constructed for the roasting of these suspended particles in the Stetefeldt furnace. Furthermore, the feeding machinery of the Stetefeldt furnace is based upon a principle entirely differing from that used with the Gerstenhöfer furnace.

That a furnace without shelves is cheaper and easier to construct, more durable, less liable to get out of order, and that it requires less labor and skill to run it, must be conceded by everybody.

Much difficulty was experienced to provide suitable feeding machinery for the Stetefeldt furnace. Gerstenhöfer's apparatus, consisting of fluted rollers, which force the ore through slits in the top of the furnace, would not answer at all. The ore fell down in lumps, and arrived at the bottom of the shaft almost raw. The reason for this behavior is simply the tendency of the particles of all finely-pulverized mineral substances to adhere to each other if a slightly compressed mass of them falls through the air. It is, therefore, necessary to introduce the ore pulp so finely divided, that all the particles can be penetrated by the heat within the short time of their fall through the shaft. To feed the pulp with a blower, as it is done in Keith's desulphurizing furnace, was not considered desirable for the following reasons:

1. The fall of the ore would be accelerated.

2. The draft of the fire-places would be impeded by the downward current of the air from the blower.

3. The formation of dust would be considerably increased.

The feeding machinery in its present shape can be briefly described as follows:

A hollow cast-iron frame, kept cool by a small stream of water, rests on top of the furnace. In this frame is inserted a cast-iron grate, which is covered by a punched screen of Russia iron, No. 0, for wet crushing, of the trade. Close to the punched screen moves, inside of the hopper, a coarse wire screen, No. 3 of the trade, which is fastened to a frame. The frame has flanges resting upon adjustable friction rollers outside of the hopper, and receives its motion from a crank, with 13-inch eccentricity. To avoid the motion of a stratum of pulp with the coarse screen, a number of thin iron blades are so arranged across the hopper that their lower edges reach close down to the coarse screen and keep the pulp in place. When the crank is set in motion, the meshes of the coarse wire screen cut through the pulp, and drive it through the openings of the punched screen. In this way the ore is introduced in a continuous stream into the furnace. The motion of the crank-shaft was variably tried in Reno at from 30 to 70 revolutions per minute.

Construction of the furnace at Reno.-The accompanying drawing will give a correct idea of the construction of the furnace at Reno: A, shaft through which the ore falls.

B, top of shaft upon which the feeding machine is arranged. C, damper, which is inserted when the screens of the feeding machin ery are exchanged.

D, door through which the roasted ore is discharged upon the cooling floor.

E, fire-place.

F, flue through which the gases escape near the top of the shaft.
G, triangular flue-bridges, of cast-iron.

H, grate of cast-iron plates, forming the bottom of flue F, allowing the dust, a small part of which settles here, to drop into the chamber I. K, discharge door.

L, fire-place which heats the lower part of flue F, and roasts the dust. M, canal connecting with dust-chamber.

N, discharge door.

O, dust-chamber.

The main dust-chamber of the furnace at Reno is 24 feet long, 8 feet wide, and 10 feet high. From there the gases pass under a dry kiln, 39 feet long and 7 feet wide. The two flues under the dry kiln are 3 feet wide and 4 feet high. A flue, 3 feet 4 inches wide and 4 feet 6 inches high, and about 180 feet long, leads from the dry kiln to an iron chimney of 2 feet 6 inches diameter on a hill-side. The top of the chimney rises about 40 feet above the top of the furnace.

The fire-places and arches are built of the best fire-brick, the rest of common brick. All the walls are built double, with a space between. The furnace is well anchored with iron rails and 7-inch rods.

The following changes are contemplated in the construction of the furnace:

1. The use of oxide of carbon as fuel; the gas to be made out of charcoal in generators. The construction adopted for the latter will be similar to that of the copper-refining furnace at Mansfeld, Prussia. In this way a much more uniform heat can be obtained than by using wood, and labor will be saved, as the generators have to be charged only every three or four hours. Where wood must be hauled a considerable distance, charcoal will be even a cheaper fuel than wood.

2. The chamber I will be abandoned, and the flue F brought down directly on the side, R R, (see ground plan,) of the shaft.

3. A more extensive system of dust-chambers will be connected with the furnace.

Manipulation.-The ore is mixed with the necessary amount of salt on the dry kiln, and crushed by a dry crushing battery through a No. 40 wire screen. A conveyor takes the pulp to a revolving screen, to keep out coarse particles, which may be caused by the breaking of a battery screen. The screened pulp is then taken by an elevator to the top of the furnace and discharged into a bin, which keeps the hopper of the feeding machine filled.

The fire is kept in all the fire-places as uniform as possible, and such a degree of heat is maintained that the roasted ore at the bottom of the shaft is red-hot, but does not sinter or stick together. The ore is discharged when a charge of 1,000 pounds to a ton has accumulated, and cooled in the usual manner. At the same time, roasted ore is discharged through the door N, where most of the dust settles, which is roasted by the fire-place L.

Chemical process in the Stetefeldt furnace.-Küstel describes the chemical proceedings of the chlorination in the Stetefeldt furnace as follows:

"At the first glance, it would seem that, considering the short time of two seconds, in which the falling ore is exposed to the flame, a perfect chlorination could not take place, especially if compared with the known facts apparent in the common roasting furnace-that is, that sulphurous acid is first formed under influence of a dark-red heat, by aid of the H. Ex. Doc. 207———48

oxygen of the air, while the metal, deprived of its sulphur, becomes an oxide. The oxygen of the air and of the oxide act on the sulphurous acid, converting it into sulphuric acid, which again combines with the metal oxide to a sulphate. The sulphate reacts now on the salt, setting the chlorine free, and the formation of chlorides begins.

"This reaction and transformation requires time, which is not offered in Stetefeldt's furnace, but the chlorination is effected nevertheless, and very perfectly, with less salt and in a few seconds. The chemical action in Stetefeldt's furnace is as follows: As soon as the ore enters the furnace each sulphuret particle ignites, being surrounded by a glowing atmosphere, evolving at the same time sulphur, which, in presence of atmospheric air entering undecomposed through the grates, is converted into sulphurous acid, and the metal into an oxide. In contact with ore particles and oxygen the sulphurous acid becomes sulphuric acid, acid does not combine with the metal oxide to a sulphate, as is the case in a common furnace; or if so, only to an insignificant degree, on account of the temperature, which, nearly from the start, is too high. The sulphuric acid, therefore, turns its force directly against the glowing salt particles, setting free the chlorine. All these reactions are, so to say, in statu nascenti. From the burning fuel steam is present among the gases, giving rise to the formation of hydrochloric acid. This hydrochloric acid not only originates directly by decomposition of the salt, but also from the chlorides of the base metals, which are formed in the upper part of the furnace, and again decomposed to oxides and hydrochloric acid in passing through the hot flame. The whole space of the furnace is then filled with glowing gases of chlorine, hydrochloric acid, sulphuric acid, sulphurous acid, oxygen, steam and volatile base metal chlorides; all of them acting on the sulphurets and oxides with great energy. The chlorine decomposes the sulphurets directly, forming chloride of metal and chloride of sulphur. It decomposes and combines also with oxides and sulphates. The hydrochloric acid does the same. The sulphuric acid decomposes the salt and oxidizes the sulphurets, while the oxygen creates sulphurous and sulphuric acid and oxides. The red-hot ore falls down, and, accumulating, continues evolving gases of chlorine, &c.

"Considering now a minute particle of ore (for only as such, not as a mass, can the ore be considered in falling) in a red-hot state being attacked contemporaneously by all those gases which have free access from all sides: the principle of the Stetefeldt furnace is, that the chloridizing result must be effected before the particle reaches the floor. The dust which passes the flame of the small fire-place is even in a better condition for chlorination, being surrounded and acted upon longer by all the chloridizing gases which are formed in the main shaft."

Practical results of the Stetefeldt furnace in chlorination.-A great number of tests were made during the first weeks of running the furnace at Reno. Between 88 and 923 per cent. of the silver contained in the ore was found to be chloridized, all of which is easily extracted in amalga mation. The roasted dust discharged through the door N is generally one per cent. better chloridized than the ore discharged from the main shaft. With an improved system of firing, the chlorination should never be less than 90, and we have no doubt that much higher figures will be obtained. Only very skilled roasters achieve such results in the reverberatory furnace. With ordinary care, a charge cannot be burned in the Stetefeldt furnace, and the roasted pulp is in a splendid condition for barrel amalgamation, as it contains no lumps or sintered

matter. Ores of the most various characters have been roasted with equal success. Even ore containing nothing but silver-bearing galena was treated without any difficulty. In this respect the furnace is admirably adapted to roast ores with large amounts of antimonial and leadbearing minerals.

Amount of salt. In reverberatory furnaces ten per cent. of salt is generally used. This amount may be safely reduced to six per cent. for. very rich ores, and to three and four per cent. for low grade ores, in the Stetefeldt furnace. No experiments have as yet been made to determine if this percentage can be reduced still more. The difference in the percentage necessary is explained by the fact that in the Stetefeldt furnace all the salt is decomposed and utilized, while in the reverberatory furnace a large percentage remains in lumps and entirely unchanged.

Fuel.-The amount of fuel necessary to heat the shaft depends very much upon the character of the ore. The more sulphurets an ore contains the less fuel is required to roast it. The furnace in Reno uses on an average about two cords in twenty-four hours. With this amount between 12 and 15 tons of ore are roasted daily, which is as much as the battery crushes. But the same fuel would just as well roast 20 tons of mainly sulphuret ores, which increase the heat in the shaft when introduced in larger quantities. How many bushels of charcoal a furnace with gas generators would require we are not able to estimate reliably at present; but for most localities in Nevada charcoal will be as cheap if not cheaper than wood.

Labor.-At the mill in Reno eight men are employed to run the furnace. Of these three are firemen, three discharge and cool the roasted ore, and two watch the feeding and elevating machinery. If gas generation with charcoal is used, the labor of three men can be saved, and the two men who watch the feeding machinery can very well charge the generator shafts. In this way only five men will be required to run the furnace. Taking these facts in consideration, it is easy to estimate how much the Stetefeldt furnace cheapens the expense of roasting. In a mill of 20 tons capacity, in twenty-four hours, at least ten reverberatory furnaces would be required. The labor needed in twenty-four hours is, 2 carmen, 2 pulp-coolers, 2 oremen, 30 roasters; total, 36 men; fuel, at least 10 cords of wood in twenty-four hours; salt, 10 per cent. or 4,000 pounds. In the Stetefeldt furnace these 20 tons are roasted with 8 men, (or 5 men if charcoal is used as fuel,) 2 cords of wood, and 2,000 pounds of salt; saving by use of Stetefeldt furnace, every twenty-four hours, 28 men, (now, and 31 men with charcoal furnaces,) 8 cords of wood, 2,000 pounds of salt. Besides this, the original cost of the Stetefeldt furnace is less than that of a corresponding number in capacity of reverberatory furnaces. It requires less repairing and does better work. I may mention, furthermore, an experiment made in Reno to try the capacity of the furnace. One ton of pulp was accumulated and put through the furnace in thirty minutes, indicating the furnace to have a capacity of 48 tons in twenty-four hours. Ninety-one per cent. of the roasted ore was found to be chloridized.