The application of Bradley and Crocker, declared to be an interference with that of the Cowles brothers, was for a useful process for reducing and heating ores by electricity. The invention is thus generally described: "In a large number of chemical and metallurgical processes, in which high temperatures are required, and in which it is impossible, on account of the nature of the operation, to heat the materials by the direct action of fire in a reverberatory or blast furnace, it is customary to treat such materials in closed, externally heated crucibles or retorts. For example, the reduction of sodium, potassium, and zinc, and the manufacture of aluminium chloride, are carried on in this way. That this method is very troublesome and expensive. The crucibles or retorts are rapidly destroyed, and, being necessarily small in order that it shall not require too long to heat them through, the labor of charging and managing a large number of small retorts becomes very great. The object of our invention is to overcome these and other difficulties, and to obtain the heat necessary to carry on such operations in a convenient and efficient manner, and to concentrate the heat exactly where it is needed; and, furthermore, our invention has for its object the attainment in commercial processes of temperatures very much higher than have ever been reached before. Primarily, our invention consists in carrying on chemical and metallurgical operations or processes in chambers or furnaces, preferably constructed of or inclosed in nonconductors of heat, in which furnaces the necessary heat is produced and maintained by passing a powerful electric current either through the materials themselves, or through separate conductors near or in contact with said materials, the energy of the electric current being converted into heat by the electrical resistance of the materials or conductors through which it passes. Our invention is applicable to a large number of chemical and metallurgical processes, but, in order to enable others to use the invention, we shall describe in detail two or three of its most typical applications." After describing one means of carrying the current along the sides of the furnace, and developing the heat by the resistance there, the patentee proceeds: "Another form of furnace which we have devised for carrying out our invention is shown in figures 1 and 2, in which A and B are plates of carbon set in refractory brickwork, and covered by an arch, D. This chamber is inclosed in an outer chamber, J, J, J, which may also be of brickwork, or of some porous nonconductor of heat, such as the material known as 'terra cotta lumber.' Between the inner and outer chambers there is a space, as shown, which is filled with asbestos, mineral wool, or other suitable nonconductor. S is a tap hole for withdrawing residue, slag, etc. G is a conduit for taking off the gases or vapors produced. The door, H, is so arranged that the whole of it may be removed in order to remove the carbon plates, etc., or only a plug may be taken out for charging the furnace. Electrical connection is made to the two carbon plates, A and B, by two heavy copper strips or bars, E and F. This furnace may be employed for the reduction of the same metals as we have described in the case of the first form of apparatus (that is, potassium, sodium, and zinc), and the same materials may be used; but instead of the heat being produced by an electric current, in which conductors surround or are in contact with the materials undergoing treatment, the current is caused to pass through the materials themselves, and the heat is produced in said materials. For example, in employing this furnace for the reduction of zinc, we take the ordinary mixture of roasted zinc ore (zinc oxide) and carbon, and charge the furnace, as indicated in the drawings. This material rests upon the carbon plates, A and B, as shown, and the copper strips, E and F, being now connected, respectively, to the poles of a suitable dynamoelectric machine or other source of electrical energy, a current will pass from one carbon plate to the other, through the material resting upon and between them, it being conducted by the carbon contained in the mixture. * * * The zinc, as fast as it is reduced, distills over through the conduit, G, and is collected in the usual manner. Sodium and potassium may be reduced in a similar way. *** We believe that the metal aluminium can be reduced by our process from its oxide (alumina) by mixing the oxide with carbon, and subjecting the mixture to an intense heat in our furnace. This reduction of alumina by carbon directly has often been attempted, but a high enough temperature cannot be attained in the ordinary way by combustion. By our proc

ess, however, an almost unlimited temperature can be produced, and it is probable that in this way we can accomplish this important result. * In an application for letters patent of the United States now pending (filed February 23, 1883, serial number 85,957), Charles S. Bradley, one of the present inventors, has described an electro-metallurgical process in which an electric current is employed to perform two functions: First, to effect the electrolytic decomposition of the materials treated; and, second, to supply the heat necessary to maintain said materials in the fused state while they are being electrolyzed. The present invention resembles the above to a certain extent, but in the present invention the electric current which we employ performs no electrolytic actions, the reaction which takes place being purely chemical, and the function of the current being solely to develop the heat which is a necessary condition of the reaction. For this reason, our invention does not require the use of a continuous current of electricity. An alternating current may be employed if desired, which is an advantage, since large alternating current dynamos may be constructed more cheaply than the continuous current machines, and it is also less trouble and expense to run them."

Among the 10 claims made in the applications just before the interference was declared with the Cowles process were the following: "(2) The hereindescribed process, which consists in mixing together in a finely-divided state two or more materials at least one of which is a conductor of electricity, and in heating said materials or maintaining them at a high temperature by passing an electric current through them, in order to cause them to act or be acted upon chemically." "(9) The herein-described electrical heating process, in which the material to be treated is mixed with carbon, both being in a finelydivided state, and the mixture is heated or maintained at a high temperature by passing through it an electric current. (10) The herein-described process of obtaining aluminium from alumina, which consists in mixing the alumina with carbon, and in heating the mixture or maintaining it at a high temperature by means of an electric current."

The subject-matter of the interference in the patent office was as follows: "The process of reducing ores which consists in mixing the ore with carbon, and subjecting the charge to the action of heat generated by passing an electric current through the same, the ore being kept in contact with the carbon, which forms a continuous part of the electric circuit, substantially as set forth in applicant's 2d, 5th, 9th, and 10th claims, and in the four claims presented in the application of E. H. and A. H. Cowles, of Cleveland, Ohio, entitled 'Electric Furnaces.'"

After the assignment of May 18, 1885, the features of the Bradley and Crocker invention, to which reference has been made, were eliminated therefrom, including the claims declared to be in interference.

R. S. Taylor, for complainant.

Betts, Hyde & Betts, and Loren Prentiss, for respondents.

TAFT, Circuit Judge (after stating the facts). We have in this case to deal with the use of the electric current to perform two functions in metallurgical operations. Whenever a current of sufficient quantity and intensity is passed through a chemical compound in a fluid condition, it will cause a chemical disruption, and one of the elements will go to the anode, or the place at which the current enters the fluid mass, and the other will go to the cathode, or place where the current leaves it. This chemical dissolution by the current is called "electrolysis." If the compound to be treated is metallic, the metal element will gather at the cathode, while the other will go to the anode. If the other element is a gas, as it is when the ores are oxides, the oxygen or other gas will bubble out of the bath at the anode. The effort that the electric current has to make in order to pass through any body or matter produces heat. The greater the resistance, or, what is the same thing, the lower

the conductivity of a material, the more heat will be generated by forcing a current through it. Heat can thus be produced of great intensity. The highest heat known is that caused by the electric arc, where the current forces its way through the air and completes the circuit. The usual way of reducing metals from their ores has been to fuse the ores by the heat of combustion in the presence of what is called a reagent. The reagent is a substance with a stronger chemical affinity for the nonmetallic element of the ore when the ore is melted than the metal sought. The result of the melting and reaction is that the metal is left pure. Whenever the ore is an oxide, carbon will serve as a reagent, because the affinity of carbon for oxygen is very strong. Carbon is a conductor of electricity in which there is sufficient resistance to the passage of the current to generate a very great heat. The Cowles brothers conceived the idea of a furnace for reducing ores by the use of electricity in which granulated ore should be mixed with granulated carbon, so that the carbon should form a continuous conductor from anode to cathode. The passage of the current would generate heat in every particle of carbon throughout the furnace, and would soon fuse the ore, when the carbon, acting as a reagent, would take the oxygen from the fused compound, and leave the ore pure. Whether electrolysis necessarily took place in the furnace I shall discuss later.

Bradley's process of 1883 was a fusion of the ore, unmixed with carbon by the electric arc, an electrolysis of the fused ore, and a maintenance of the fusion from the heat generated by the passage of the current and the electrolytic action. There was no reagent in the Bradley process. The only agent of dissolution, except the fusion, was the current. In 1885, the Cowles brothers were convinced that they had made a great discovery,-one which would revolutionize the art of winning the rarer metals, like aluminium, from their usually refractory ores. As they were forming their company, they learned that some one else had made a similar discovery. They wished to buy peace, as their counsel says, and so they proposed to buy from Bradley and Crocker the invention which threatened to destroy the value of their own, and, to make themselves perfectly secure, they secured the assignment of all other inventions or discoveries of the assignors which would tend to interfere with the monopoly they hoped to enjoy from their own. They were seeking insurance from invasion by the assignors, and so the assignment was made to cover patents, applications, and discoveries which never were, and which the assignors, of course, knew had no existence. Bradley and Crocker were willing to make the language thus broad, because, as it could convey nothing, it did them no injury. The intention of the Cowles brothers was not only to secure the application which had interfered, but also to have a most sweeping guaranty from Bradley and Crocker that they had no other discovery which would interfere with the Cowles discoveries. This is the only explanation possible of the assignment of European patents and applications, and of the reference to inventions, applications, and patents all in the plural, in addition to

the application specifically assigned. It is not claimed that Bradley and Crocker represented that they had other inventions and discoveries which would satisfy the words of the assignment. It is manifest that the general and indefinite language was used only to secure the Cowles Company against possible and undisclosed inventions, and not for the purpose of including inventions known to both parties to be in existence.

With the general purpose of the parties in the framing of the contract well understood, we come to the question whether they intended to include in the assignment Bradley's electrolytic process of 1883. The evidence establishes beyond controversy that when the assignment was drawn and executed the Cowles brothers knew that Bradley had applied for a patent for a process of reducing metals by electrolysis in which the current also effected the necessary fusion. They knew this from the Bradley and Crocker application which they were buying, in which the Bradley process was not only described, but the patent office number was given. It was discussed by them. Had the Cowles Company then intended that it should pass by the assignment, it is not possible that it would not have been specifically mentioned therein. The Bradley and Crocker application was identified in the assignment by number and otherwise. Why not Bradley's? On its face, the assignment purported to convey Bradley and Crocker's applications, patents, and discoveries. This was the stronger reason, if an individual application of Bradley was to pass, to make the reference to it free from doubt. The Cowles Company was using general terms to describe patents and applications in the assignment for fear that the assignors might have others undisclosed. But here was one brought to its attention while the contract was being drafted. Surely the failure to mention it specifically is strong evidence that it did not intend or wish to acquire it. Otherwise, why did not the Cowles Company at once take steps to prosecute the application to a patent as it did in the case of the Bradley and Crocker application? For eight years it made no claim of title to the Bradley application. It is suggested that, at the time of the assignment, it stood rejected. But the Cowles Company did not know this, and, even if it had, rejections are not final in the patent office, as the many rejections of this very application abundantly show. It is fair also to infer that Bradley did not suppose he had included his application of 1883 in the assignment, because from 1885 until the issuance of the patent, in 1892, Bradley pressed for its allowance.

The circumstances present convincing evidence that neither of the parties to the assignment thought that the Bradley application of 1883 was within its terms. But this conclusion does not dispose of the case. It was the purpose of the parties to this assignment that a certain class of discoveries, applications, and patents, if owned by the assignors, should pass to the assignee. The description of them ex industria was made general, so that it might include individuals of the class whose existence was not known. If, now, it appears that there was an application which must be included within the general description to effect the purpose of the

assignment, can the court exclude it from the operation of the grant, because both parties at the time of the assignment knew of its existence, but were ignorant that it was within the description of the class to be conveyed? It does not seem so. The question of construction is, what did the assignors intend to convey? If a class, then all the individuals fairly within it must be included in the assignment. The very object in a class description is to avoid the necessity of mentioning individuals, and to include individuals which might otherwise be omitted. It follows that the mistake of the parties in thinking that a certain individual is not of the class conveyed will not exclude it from the grant. But the fact that both parties do not think the individual to be in the class described may be a pregnant circumstance to show in what sense the words describing the class are used, if those words are capable of several meanings. This, then, brings us to the question whether the Bradley application of 1883 is fairly included within the words of the grant. The important words are:

"Any and all discoveries and inventions relating to electric smelting processes and furnaces, and all patents they have obtained therefor, and all applications now pending and caveats on file in the United States patent office relating to electric smelting processes and furnaces, which do or may interfere with any applications for patents made by Eugene and Alfred H. Cowles, of Cleveland, Ohio, now pending in the United States patent office."

It is clear that Bradley's application of 1883 was not included in the grant, unless it related to electric smelting processes and furnaces, i. e. unless the process described therein was an "electric smelting process," as that term might reasonably have been understood by the parties when the assignment was executed. The meaning of terms often changes from time to time, and the words of a contract are to be construed as of the time when it was entered into. "Smelting," by its derivation, is synonymous with "melting," but in metallurgy and the commercial manufacture it has come to have a more contracted meaning. Thus Prof. Morton, the expert for the defendant, quotes, from a treatise on metallurgy by Frederick Overman, this distinction between "melting" and "smelting":

"When metallic ores are exposed to heat, and such reagents as develop the metal, we call it 'smelting,' in contradistinction from the mere application of heat, causing the ore to become fluid, which is called 'melting.'

Prof. Morton, however, is of the opinion that "smelting" really means nothing more then "melting apart," and that any process in which the melting apart or separation of a metal from its ore is effected by the use of electricity is correctly described as electric smelting, and therefore that the Bradley process was electric smelting.

Prof. Langley, the expert for the complainant, thus defines "smelt": "The word 'smelt' is customarily applied to that class of metallurgical operations in which a metal results, said metal being in a metallic condition, and obtained from an ore or mixture in which the metal originally existed in the form of a chemical compound. In all instances the operation of smelting results in producing something different from the body operated upon, and this change is brought about by the action of heat and chemical force. Usually the chemical action involved is between carbon, on the one hand, and v.68F.no.2-24