This completed specification shows the process claimed by Slade. It is impossible for me to resist the conclusion that it is fatal to Slade's application, and to any other application claiming the same invention, as of a date subsequent to that of the completed specification.

Josiah M. Heath's second English patent, No. 10,798, August, 1845, describes the use of iron sponge for the decarburization of the metal on the hearth of a reverberatory furnace. The following extracts from his specification show this too clearly to admit of dispute or doubt:

The pig-iron to be converted into steel may be melted in a cupola or in a reverberatory furnace, or it may be run direct from a blast-furnace used for smelting iron ore.

The fluid pig-iron should be run from the cupola or other furnace into a receptacle made of any material capable of withstanding an intense heat, similar in form to a common refinery, or to the shallow well of a reverberatory furnace for smelting pigiron.

#

In order to decarburate the fluid iron in the receptacle to the degree necessary to form steel, I mix with it a certain portion of malleable iron, more or less, as I wish to make the steel softer or harder.

The malleable iron to be mixed with the fluid pig-iron in the receptacle may be in scraps or in any convenient form; but by far the most economical, convenient, and purest state in which the malleable iron can be used is in the granular form, produced by reducing any perfectly pure oxide of iron to fragments, and then submitting them to the well-known process of cementation in a common converting-furnace, such as is used for converting bar-iron into blistered steel. The iron ore, in small fragments, is mixed with just that proportion of carbonaceous matter which is sufficient to combine with its oxygen at a red heat in a closed vessel, antl when the process is finished malleable iron is obtained in the purest possible state. Before adding the malleable iron to the fluid pig-iron the former must be brought to a white heat, and this may be done in a separate furnace. But I find the most convenient method is to place the malleable iron on a bed between the receptacle which holds the fluid pig-iron and the chimney up which the waste heat from the combustion of the gas passes.

B is the bed between the receptacle for the fluid metal and the chimney. The malleable iron or deoxidated iron ore is placed upon this bed on either side of the channel K.

D is the door by which the malleable iron or deoxidated ore is introduced upon the bed B, and by which it is raked into the receptacle.

What I now claim is

The exclusive right of preparing cast-steel by decarburating pig-iron to the degree required to form steel by mixing with the pig-iron, run from a cupola or other furnace into a separate receptacle, malleable iron in the proportion necessary to form steel, and running the mixture of pig and malleable iron, while still fluid and in the state of cast-steel, into molds from the receptacle.

Heath's process consisted of three branches: (1) melting the pig-iron; (2) preparing the pure iron or sponge, and (3) mixing with the molten pig-iron enough of the pure iron or sponge to dilute the carbon of the mass to the point at which it became steel. For his third step-ending at that point, Slade substitutes two steps, viz, one beyond and another back to it, making four instead of three branches to his process. If, therefore, Slade claimed broadly the process of decarburizing iron surcharged with carbon by the use of iron sponge on the open hearth of a reverberatory furnace, instead of claiming its use in the "four-act" Siemens-Martin process, he would be anticipated by Heath.

If it should be shown that the decarburization of the bath after the introduction of the sponge in the Siemens-Martin process placed the product in a better condition as to silica or phosphoric acid or any other ingredients, except pure iron, carbon, and oxygen, than that in which the use of the sponge in Heath's process placed it, then Heath's patent might not be fatal to this claim for the use of the sponge in the SiemensMartin process. But nothing of that kind is shown. It would seem, therefore, that since the English patent of Heath was granted in 1845 it has not been possible for Slade, or for Blair, or for any one else, to secure a valid generic claim for the use of iron sponge in the SiemensMartin process for the manufacture of steel, for the use of the sponge in this process seems to be, not merely analogous to its use in Heath's "three-act" process of manufacturing steel, but identical with that old

use.

The function of the sponge in each case seems to be the dilution of the carbon of the cast-iron-nothing more and nothing less. The circumstance that in Heath's process the operator mixed with the cast-iron just enough sponge to dilute the carbon to the point at which the mass becomes steel, while in the Siemens-Martin process the operator added more sponge, so as to carry the dilution further, and then by the addition of some kind of cast-iron brought the mass back to the condition of steel, while it does make a difference between the Siemens-Martin process and the Heath process of manufacturing steel, makes no difference whatever in the function performed by the sponge in the two cases.

The only valid patent for the use of iron sponge in the Siemens-Martin process, in view of Heath's patent, would be a patent for the use of an improvement on Heath's sponge. Heath's patent, then, defeats Slade's claim altogether. It defeats Blair's claim so far as a generic claim for the use of iron sponge in the Siemens-Martin process is involved in his application.

The American patent of Emile and Pierre E. Martin, No. 72,061, December 10, 1867, embraces a description of a process of manufacturing cast-steel consisting of four branches: (1) melting cast-iron; (2) preparing roasted and cemented ore; (3) mixing roasted and cemented ore with the cast-iron; (4) adding cast-iron to the mass. Blair insists that ore so roasted and cemented is iron sponge. Slade insists that the patentees meant ore roasted, reduced, and cemented, which, he says, is melted iron, and did not mean merely ore roasted and cemented, which Blair says is iron sponge.

Slade assigns as a reason for his construction of the patent the fact that these patentees, in their French patent of July 5, 1867, said that the process might be carried on by "simply adding to the bath, as was said at the beginning, ore that has been roasted, reduced, and cemented." He is of the opinion that this reduction was a process of conversion into the iron preceding or additional to cementation, and that the whole involved the fusion of the iron. What had been said at the beginning

was that ore was added "simply roasted, but preferably roasted and cemented." Now, reduction of ore, in its strict chemical sense, certainly does consist in the separation of the metal, with or without fusion, from the substances with which it may be combined. But the word in French, as in English, is often applied to mere pulverization. In common language bodies are said to be reduced to powder. The words "grillé, reduit et cémenté" mean here, I think, just what the words 66 'grillé, concassé et cémenté” mean in the addition of February 26, 1866, to the Martin (French) patent of July 28, 1865, if they mean anything more than mere roasting and reduction by cementation. They refer to the pulverization of the ore after it is roasted and before it is cemented. The cementation of the ore may consist in its reduction by surrounding it with carbonaceous substances and heating the whole mass to a degree not sufficient to cause fusion, in order to deprive the ore more or less completely of its oxygen and carbon. Cementation of the iron itself may consist in a similar treatment of iron with carbonaceous substances, in order to increase instead of diminishing the charge of carbon, and thereby to produce steel.

The use for which the ore was cemented by the Martins would seem to indicate that, although the process of cementation might consist in imparting carbon to the ore by heating it in presence of carbonaceous matter, just as it consists, when applied to wrought-iron, in imparting carbon to the iron itself, yet the object was not to communicate carbon to the iron in the ore, so as to form steel, but to communicate it to the ore itself in quantities just sufficient to liberate the oxygen and produce iron comparatively free from oxygen as well as carbon, in which condition it would be available for the dilution of the carbon in the molten cast-iron with which it was to be mixed.

It seems clear that, while the Martins may have applied to ore the process of cementation often used to carburize iron, their ultimate object must have been, not to carburize the iron in the ore, but to decarburize it, and thereby produce sponge of a higher or lower degree of purity.

It is evident that the Martins, if they did not, by the use of the terms "roasted" and "cemented" ore, mean iron sponge, must have meant carburized iron or steel. If they meant carburized iron or steel, their process was to manufacture steel by (1) melting cast-iron; (2) preparing carburized iron or steel; (3) mixing cast-iron and carburized iron or steel, and (4) adding cast-iron. In that case they meant to dilute the carbon in the cast-iron to the steel-point by adding, not iron freed from carbon, but iron charged with carbon. They proposed to make steel by mixing cast-iron and carburized iron or steel, and not by mixing castiron and sponge.

The application of the term "cemented" to the ore, according to the letter of the patent, which describes it as roasted and cemented ore, would seem to be more in harmony with the object of the process, which is the dilution of the carbon of the cast-iron below the steel-point by the

mixture of iron comparatively free from carbon and oxygen, and the return to that point by the addition of more cast-iron.

Slade insists that the use made of the term "cemented" by the Martins in certain earlier French patents and additions thereto indicates that it does not refer in their American patent of December 10, 1867, to an operation which produces iron sponge.

Those patents and additions contain the following statements:

This ore had been roasted with gas, pulverized (concassé) and partially cemented in cast-iron crucibles [February 21, 1866]. Thus for 1,000 kilograms of cast-iron will be added 250 kilograms of rich ore, prepared by roasting and by a previous cementation, and from 50 to 100 kilograms of peroxide of manganese. [March 3, 1866.]

Use of ores.-We submit the ore to a previous treatment before taking it to the castiron bath indicated in the patent. This treatment consists in pulverizing the ore (when possible rich oxidized manganiferous ore) and mixing three parts of this ore with two parts, in weight, of pulverized coal. A part, in weight, of silecious or aluminous earth is to be added, according to the aluminous or siliceous nature of the gangue of the ore, the whole in balls of variable size, from about 500 to 1 kilogram; and it is to be dried and melted on the bath of iron, cast or directly without an addition of cast-iron, each ball performing the part of a brasque crucible. The result is that, in proportion as the ball is heated and the coal reacts on the ore, there is a disengagement of carbonic acid. In consequence of the lowering temperature, this disengagement, which is produced during the whole time of the reaction, prevents the ball from melting before the complete reduction of the ore. For 1,000 kilograms of cast about 500 kilograms of prepared ore will be added, this quantity varying according to the course of the operation, which is to last 5 hours. The prepared ore may be used without an addition of cast-iron; but then it will always be advisable to commence the operation, as above described, with the bath of cast-iron, but to continue it for a second "run", with the ore alone.

The advantage we find in this use of the prepared ore instead of raw ore is a diminution of waste in the product by reducing the quantity of the black cinders to be taken out of the furnace during the treatment and replacing the same with clear slag ; and, besides, greater profit may be derived from the high temperature more easily developed. [December 19, 1866.]

This bath being thus heated, ore is thrown into it, simply roasted, but preferably roasted, then cemented in small parts, from 10 to 20 kilograms. [July 5, 1867.] The final addition of cast-iron is not absolutely indispensable, but it has the advantage of producing steel having body and not red-short. [July 5, 1867.]

When the process is so conducted as to make the product by the use of ore, this ore is prepared by pulverizing it and mixing it with coal, manganese, lime or a convenient flux, according to the gangue of the ore, then forming balls or bricks, each of which performs the part of a crucible for the cementation of the ore; or it may suffice, as said at the beginning, to throw into the bath ore roasted, reduced, and cemented. [July 5, 1877.]

It seems to me to be clear that these statements point to the introduction into the bath of molten cast-iron, roasted ore, roasted and cemented ore, and roasted, pulverized, and cemented ore. It may not be clear whether the ore prepared in the form of balls, as shown in the addition of December 19, 1866, to the patent of July 28, 1865, and in the patent of July 5, 1867, is to be introduced into the bath before cementation, or is to be cemented in the bath. But, in either view, there seems to be nothing in any of these statements, or in any other statements of these patents, to support the conclusion that the application of the term

2

cemen.

tation" to the preparation of the ore for the bath implies anything more than the process which reduces the ore by freeing its carbon and oxygen without fusion.

It is to be observed that Slade describes no process for producing the sponge, but in his application says:

My improvement consists in the use of deoxidized ore, commonly known as “iron sponge," introduced either hot or cold, in the place of the scrap-iron or other materials above mentioned, for effecting the decarburization of the bath.

I would here state that it is immaterial in my improved process by which method the sponge is produced, or even whether the deoxidation of the ore is absolutely complete; and I therefore lay no claim in this specification to any particular method of manufacturing the sponge.

In Blair's application he says:

My invention relates to the manufacture of wrought or malleable iron and steel from iron sponge prepared by the deoxidation without fusion of iron ore.

He does not describe his process for deoxidizing the ore, but refers for a description to another specification. His Patent No. 126,922 contains the following claim:

1. The process of reducing oxides of iron to the metallic state by subjecting them to a red heat in contact with carbonaceous matter in a chamber constructed substantially as hereinbefore described, so that the material under treatment shall supply the packing, which excludes the air therefrom.

In his Patent No. 126,924 he says:

The iron sponge which I manufacture is the product obtained by treating oxides of iron in contact with carbon at a suitable heat, and with practically complete isolation from the atmosphere, in the manner hereinafter described.

Slade refers to the use made of the term "cementation" by Chenot and others as proof that it was not applied by the Martins to the preparation of iron sponge.

Chenot (Crepusc Nouv. Syst. Métallurgie) states the ordinary acceptation of the term "cementation" to be the penetration of carbon into wrought-iron by action carried on in a closed vessel at a high temperature, long continued in the manufacture of steel. But he also applies the term to the saturation, without heat, of metallic sponge by liquids holding in solution carbon, or an alkali, or a different metal, or other substances, and the deposit around each atom of the sponge of some of the carbon, alkali, metal or other substances. The difference between these two uses of the term "cementation" is wider than the difference between the ordinary use of the term and its use when applied to the conversion of ore into sponge.

Phillips, in his "Elements of Metallurgy" (pp. 343 and 344), describes the process of "cementation" in which, by the use of iron-stone, metallic oxides, or lime, an effect is produced on the cast-iron exactly opposite to that which is produced on wrought-iron by the ordinary process of cementation. He says:

Partial decarburization of cast-iron by cementation.-The fact that articles of cast-iron become softened if embedded in ferric oxide and maintained for a considerable time