GARRETT, Associate Judge.

This is an interference proceeding relating to four counts, Nos. 1 and 4 being mechanical, and Nos. 8 and 9 being for the method. We here quote Nos. 1 and 4:

"1. A polyphase are electric furnace comprising a plurality of arcing electrodes, and means for unbalancing the current or voltage of one or more of the polyphase arcs."

"4. A polyphase electric furnace, comprising a plurality of arcing electrodes, a polyphase circuit connected with said electrodes, and means for unbalancing the voltage supplied to one of said electrodes."

The history of the case to date is as follows:

On January 3, 1913, appellee filed an application for a patent (serial No. 740,034) for improvements in electrical induction furnaces. There was a renewal application on December 19, 1917 (serial No. 207,966), and patent was issued thereon August 16, 1927, being No. 1,639,340, entitled "Combination Induction Furnace." On September 8, 1919, he filed another application for patent for "Electric Arc Furnace Control and Apparatus Therefor," in which he said: "This application is a continuation, in part of my application for patent, * * * Serial No. 740,034, filed January 3, 1913. * * *"

In the meantime appellant had, on August 9, 1918, filed an application which, on July 8, 1919, ripened into patent No. 1,309,045, for "Electric Furnace and Process of Operating Same." This patent contained the claims which constitute the counts at issue.

Appellee copied these and other claims of the patent, seeking an interference which was declared. The Examiner of the United States Patent Office awarded priority to appellant, Moore, upon six of the counts, and to appellee, Greene, upon seven. The Board of Appeals reversed the Examiner as to three of the counts awarded Greene and awarded them to Moore, but affirmed as to the four in issue, and Moore appealed to this court.

It is the claim of the appellee, sustained by both tribunals of the Patent Office, that the disclosures of the Greene application of January 3, 1913, comprehended the subject-matter of the counts in issue, thereby entitling him to make same as a part of his application of September 8, 1919. Appellant denies that same was disclosed in the early application, and this constitutes the issue.

The mechanism of the Moore patent is one for melting and refining metal by means of heat generated by an electric current, or electric currents, being passed through the mass of material, including a current through the bottom of a polyphase are electric furnace.

It seems that prior to the Moore patent the system, as used in the art, supplied the electric current to all of the electrodes at substantially equal voltages. The Moore patent disclosed means for "unbalancing the current or voltage of one or more of the polyphase arcs" (count 1), or (count 4), "means for unbalancing the voltage supplied to one of said electrodes."

In the final analysis the question before us as to counts 1 and 4 is quite simple, and we have only to enquire whether Greene's application of January 3, 1913, discloses such means as are recited in the Moore patent.

It is a general rule, frequently announced by the Court of Appeals of the District of Columbia and other courts, that the meaning of counts in an interference, if there be any ambiguity, is to be determined by the specification of the application or patent in which the claims originated. McIntyre v. Dodge, 58 App. D.C. 273, 29 F.2d 861; General Electric Co. v. Steinberger (C.C.A.) 214 F. 781. In Brogden v. Slater, 40 F.2d 988, 989, 17 C.C.P.A. 1240, 1241, 1242, this court said: "All of the counts in issue originated in appellee's application, and it is well established that the meaning given to the counts, if there be any ambiguity, must be that disclosed in the specification of the party first to make the claim. Seymour v. Molyneux, 49 App. D.C. 216, 263 F. 468."

It is also the well-established rule that in an interference proceeding the claims of a patent involved therein will be given the broadest interpretation which they reasonably will support. Stern et al. v. Schroeder et al., 36 F.2d 515, 17 C.C.P.A. 670. But this rule does not warrant disregard of expressly defined limitations. In re Joseph Bijur, 40 F.2d 999, 17 C.C.P.A. 1134, citing Slattery v. Larner, 36 F.2d 298, 17 C.C.P.A. 725.

The instant case is to be considered in the light of these two general rules.

In the Moore patent specification it is said: "A further object of my invention is to provide a novel process or method of varying the current and thereby regulating the heat, by unbalancing one or more of the polyphase arcs. As will be seen hereinafter, this may be done in a number of ways, the effect in each case being to regulate the current flow through the bottom of the furnace and through the bath."

There is then a description of operation which it is not deemed necessary to quote.

In the specification of Greene's 1913 application, it is said: "* * * Taps are shown as taken off the four lower turns which constitute the heavy portion of the primary winding and any number of these turns may be connected to the top electrode 5 by means of the switch 10. In this way it is possible to apply any desired voltage to the arc furnace circuit between the bottom electrode 6 and the top electrode 5."

The specification also contains the following: "In the use of 3 phase current, the primary windings may be connected in delta instead of star and the terminals from the heavy sections of the primary may be led to suitable electrodes for generating an arc or other form of resistance heat. The several voltages may be controlled independently of each other as shown in Fig. 1."

The Examiner in his decision, referring to Greene's figure 2, says:

"The figure shows a three phase electric induction furnace including three star connected primary windings. The neutral point of these windings is connected to an electrode in the bottom of the furnace. Three electrodes are provided in the top of the furnace, each of them being connected with one of the windings, at a point intermediate the ends thereof. The original specification states that these last named connections may be adjustable so that "the several voltages may be controlled independently." It seems clear that, by changing the point at which one upper electrode is connected to its winding, while leaving the other two electrode connections unchanged, an unbalancing of the voltage of one of the arcs would be effected. * * *

"Moore contends that, since said application does not expressly state that the voltage is to be unbalanced, the subject matter of the counts is not disclosed therein. However, since it is expressly stated that the several voltages may be varied independently, and since any variation of one independently of the others would produce an unbalancing, it is thought that there is a full disclosure of means for unbalancing the voltage of one or more arcs."

We also quote the following from the decision of the Board of Appeals:

"It is thought there is no real dispute as to the construction disclosed in Greene's earlier case or that he shows a three-phase arc electric furnace in his figure 2 with which a switch such as shown at 10 and 11 in figure 1 may be connected to the separate phases of the polyphase auto-transformer of figure 2. With this construction we are unable to reach any other conclusion than that the presence of the independent switches was intended to result in independent adjustments, otherwise they would not have been independent of each other. Whenever one such switch is adjusted, the function described in these counts necessarily results. We agree with the examiner of interferences that the statement `the several voltages may be controlled independently' in Greene's application, means that by changing one of these switches only, or differently from the others, an unbalancing of the voltage of one or more of the arcs would be affected.

"We are unable to accept the view of Moore that this statement of Greene concerning independently controlling the voltages has relation to controlling the arc voltages `independently of those of the inductive part of the furnace,' as contended by Moore in his brief."

Appellant has not convinced us that the holdings of the tribunals of the Patent Office as to counts 1 and 4 are erroneous. There does not appear to be any ambiguity in the counts, and they are sufficiently broad, it seems to us, for the disclosures of the Greene application to read upon them. There is no limitation in the counts themselves which confines them strictly to the disclosures of the patentee, Moore. Unbalancing seems inherent in controlling the voltages supplied the different electrodes independently of each other.

The method claims, Nos. 8 and 9, read as follows:

"8. The herein described process of controlling the current flow through the bottom of a polyphase arc electric furnace of the neutral star connected type, which consists in unbalancing the current or voltage of one or more of the polyphase arcs.

"9. The herein described process of controlling the current flow through the bottom of a polyphase arc electric furnace of the neutral star connected type, which consists in unbalancing the voltage in one or more of the arcs while maintaining the length of the arcs equal or balanced."

It will be observed that they differ only in the fact that No. 9 calls for controlling the current "while maintaining the length of the arc equally balanced."

It is noted that the Board of Appeals awarded priority to Moore on counts 2, 3, and 5 of the interference as it originally stood. These were article claims and called for a "conducting bottom." The Board held that Greene's application of January 3, 1913, did not disclose a bottom which was itself of conductive material, or conducting, but that his means consisted of an electrode inserted in the bottom. Moore had previously been awarded priority by the Examiner upon article claims, Nos. 6 and 7, and process claims, 10 to 13, inclusive, and Greene did not appeal.

Discussing the process claims here in issue, the Board says: "As to claims 8 and 9, we do not construe the terms "controlling the current flow through the bottom" as restricted to a furnace having a bottom which is conducting. These are method counts and are realized in Greene's construction in which the current is conducted through the bottom by means of the electrodes which project therethrough."

In so far as count 8 is concerned, we agree with the Board. It is a method count and the term "controlling the current flow through the bottom," by the means recited, is the essential feature or element to be looked to, and it does not, as regards the method, appear to be important whether it is control of a current passing directly through a conductive bottom or control of one passing through an electrode inserted in the bottom. It is true that the Moore specification seems to limit his device to a conducting bottom, but this limitation of the specification when the count itself is expressed simply in the words "through the bottom," without limiting it to a conducting bottom, does not seem to us to make applicable the rule of interpreting the count in the light of the specification in which it originated. The broad rule of interpretation above stated applies.

This reasoning applies to the similar elements of count 9, and our holding would be the same relative thereto, but for the fact that the count calls for the feature of unbalancing the voltage in one or more of the arcs while maintaining the length of the arcs equal or balanced.

The Board makes no reference to this additional feature. The Examiner in his decision said: "While it is not specifically stated in the specification of [Greene's] application No. 207,966 that the arcs are to be maintained of equal length, it is thought that such an arrangement may fairly be implied, this being the normal arrangement and the entire disclosure relative to the device shown by Figure 2 of said application being such as to lead to the conclusion that the arc structures are identical."

It seems to us that, in view of the fact that this is a method claim, there should be, under the authorities, a clearer showing of anticipation than we are able to discover in Greene's disclosure. Greene presented no process claims in his original application, and we are loath to hold that the count is met by a mechanism application which, while it might carry out the process, yet contains nothing to indicate any understanding or appreciation of it by Greene in 1913, and gives no intimation of any directions for accomplishing it.

In Deppe et al. v. General Motors Corporation (D.C.) 15 F.2d 419, 428, it was said: "* * * As most of the claims in issue are process claims, it follows, as already said, that the showing of an earlier apparatus, which might carry out the process, cannot constitute anticipation, unless accompanied with directions in the patent for the same. `A process patent can only be anticipated by a similar process. It is not sufficient to show a piece of mechanism by which the process might have been performed.' Carnegie v. Cambria, 185 U.S. 424, 22 S. Ct. 707, 46 L. Ed. 968."

The decision of the Board of Appeals is modified, being affirmed as to counts 1, 4, and 8, and reversed as to count 9, priority as to the latter being awarded to Moore.

Modified.

LENROOT, Associate Judge, dissents as to count 9.