The complainant is the owner of patents Nos. 967,578 and 967,579, both granted August 16, 1910. No. 967,578 was based upon an application filed June 3, 1909, by Siemund, who is shown to have been a German citizen, then residing in New York, and in the business of electrical welding and repairing of marine boilers and machinery for a number of the steamship and railroad companies around the harbor. This business of electrical welding had been developed by Siemund in Germany, and he was induced to come to the United States to establish the business here, by persons familiar with the German work. In the month of June, 1908, he did the first job of repair work at New York, upon the steamer Altai of the Hamburg American Dine.
At that time the art of welding by electricity — -that is, by the voltaic arc- — had become, to a certain extent, well known, through the teach*284ings and practice of what is called the Benardos method, and certain improvements or developments thereof.
This Benardos method was described in the United States patent to Benardos & Olszewski, No. 363,320, granted May 17, 1887, on an application filed December 3, 1885. The letters patent recite a French patent, No. 171,596, October 10, 1885; a Belgian patent, No. 70,569, October 20, 1885; an English patent, No. 12,984, October 21, 1885; a German patent, No. 38,011, November 1, 1885; a Swedish patent, No. 726, November 6, 1885; a Russian patent, No. 11,982, December 31, 1886; and a Spanish patent, No. 10,267, January 5, 1887.
Benardos and Olszewski were Russians, and their patent is so simple and has been so generally recognized that one sentence from the specifications will describe all that need be quoted:
“Our invention contemplates the formation or production of the voltaic arc between the metal .to be operated on and a conductor which is brought for said purpose into proper proximity to that point on the metal which is to be operated on, the' conductor forming one pole, while the metal to be worked constitutes in itself the other pole.”
This Benardos method or invention' is recognized in text-books, a number of which have been introduced in evidence, and in technical articles and descriptions, explaining its practical application and resultant conditions, such as the necessity for using shields over the eyes,, ways to economically and easily manipulate and operate the device, the current necessary therefor, and the phenomena caused by the process.
Benardos says that the conductor which is to form, one pole, usually the positive pole, is to preferably consist of a rod of carbon, but may be of other material. In addition to the process of welding, Benardos claimed the separation or. working of the metal, by continuing the process of melting under the voltaic arc, until the fluid metal could be removed by the perforation of the plate, or by allowing the fluid metal to flow away from any but a horizontal surface. He also provides for adding metal or minerals when needed, and suggested the use of carbon pieces to hold the molten material against a vertical surface, or a magnetic current to draw the melted portion, if paramagnetic, against the underside of an overhead plate. See description of Benardos experiments, at Tegel, near Berlin, page 427 of the English and American Mechanic, by Van Cleve & Edwards, 1890.
It will thus be seen that the Benardos process contemplated and described the broad and variable or comprehensive application of the voltaic arc with a movable conductor, to every operation dependent upon the melting of the metal at the point of formation of the arc, either for welding or working of the surface by the process of melting.
On September 18, 1888, C. L. Coffin, of Detroit, applied for a patent, issued on January 8, 1889, No. 395,878 which claimed an improved process of electric welding by subjecting a metal conductor to the effect of the.voltaic arc, which at the other pole fuses the work; that is, the two ends of the joint or parts to be welded.
Thus the Benardos method of fusing the work at the joint and the material to be melted is accomplished by the fusion of the positive con*285ductor, which is to he of metal, and which is, when molten, to fall upon the weld and furnish additional material to form the same.
The idea of “dripping metal onto the work from a metallic pencil'’ connected with the positive main was used as if well understood in the Unwin & Howard patent, No. 480,794, of August 16, 1892. The Coffin patent, No. 405,345, of June 18, 1889, shows a modification or improved form of device by means of which extra material, to be united with the fluid metal produced by the Benardos method, can be supplied between the points forming the ends of the arc.
Another Coffin patent, No. 507,419, of October 24,1893, seeks to improve upon the Benardos method by applying magnetic force at the extremities of the voltaic arc, and the Coleman patent, No. 650,124, May 22, 1900, presents again the Benardos idea of reversing the current, in order to assist the removal of the metal made fluid under the voltaic arc. This patent provides means for observing the work and for adding agents, such as oxygen or sulphur, to assist in the process of fusion, to make an easier flow, and to change the composition of the metal:
All of these patents are based upon the Benardos process, and the art of welding seems to have been practiced along these general lines until the needs of the art in welding thin surfaces and in working up- • on the vertical or overhead surfaces of marine boilers and such structures (where repairs had to be made without removing the article to be repaired to the repair shop, or where the working space was extremely circumscribed) were recognized and supplied.
In doing this kind of work, Siemund found that a metallic conductor of small diameter (% to % of an inch) with a current of from 50 to 70 volts and of about 200 amperes,1 could be brought to the point of fusion, and the voltaic arc (which in all the previous patents and processes had, when produced by a carbon-electrode, spanned a space of 2 to 4 inches) could be shortened to such an extent that the molten particles from the electrode could be deposited upon the superficially melted surface of the work, in a series of waves or layers, along the line of the movement of tile arc, and that the resultant distortion, from cooling or change of substance, was so slight as to avoid injury to the structure, while leaving a stronger weld.
Siemund therefore filed an application in the Patent Office, upon the 3d day of June, 1909, in which he set forth the circumstances under which work could be performed to advantage by his application of voltaic arc welding. He stated the materials apd instruments which he employed, giving the general relation of the varying elements, viz., the current, the size of the electrode, and the length of the arc. He included descriptions of the locations in which such wrork could be performed to advantage, and provided for the use of what he called means for maintaining a constant load upon the dynamo; that is, by arranging for an artificial or alternative resistance (to be brought in if the voltaic arc were terminated) such as a two-way switch. He described the use of certain fluxes, as previously disclosed by Coleman and Unwin & Howard, and stated that his method was of advantage whenever it was necessary to apply the welding metal “downward, up-*286w.ard or laterally, by having the metal follow the arc, and.being deposited at exactly the point of attachment desired. He states that the ar.c will “bring the free end of the welding wire to a condition sufficiently molten that it will progressively detach itself from the main body portion of the wire, yet will be only sufficient to raise to a like melting point that portion of the surface of the work.” He states that the new metal and the old metal are “at the same degree' of temperature and in the same molten condition at their points of contact.” He says that in the preferred form, practice of his claimed invention will bring the welding metal under the influence of the magnetic steel, in such manner that as it melts it will follow the lines of magnetic force leading outwardly through the electric arc to the point of desired dép-í osition. ' "
To accomplish this last result Siemund provided for an armature wiring of the welding clamp, to furnish a magnetic current, which hi stated would repel the molten particles of the electrode. Upon comment by the Patent Office to the effect that molten metal was nonmagnetic, and that the tendency of the magnetic current would be to attract rather than to repel, the language of the application was changed to provide as above for “bringing the metal under the influence of the magnetic field.”
Siemund recited 10 claims for the method of welding,, and 7 claims for an insulation or apparatus to accomplish that result. The Patent' Office cited certain patents as preliminary matters to be considered, and pointed out the apparent joinder of methods relating to distinct and separate inventions (if patentable) and the inclusion also of claims for devices to carry on those methods.
To meet this objection, Siemund filed, on the 19th of January, 1910, a second application, with substantially the same description arid specifications, but asking for a separate allowance of the method claims then withdrawn from the prior application. The Patent Office thereupon granted a patent upon each application, on August 10, 1910. The present action charges infringement of these two patents, and the complainant has specified claims 1, 2, and 5 of the method patent, and claim 2 of the apparatus patent, as infringed by the defendants. These claims are as follows: Claim 2, patent No. 967,578:
“2. An installation for electric welding or repairing, comprising a source of direct current supply, an arc welding circuit connected, therewith and including a clamp adapted to hold a piece of the welding or repairing metal, and means for maintaining the' tension of the current during temporary interruptions of the arc; substantially as described.”
Claims 1, 2, and 5, patent No. 967,579:
“1. The method of electric welding or repairing, which consists in establishing an electric are between the metallic object to be welded or repaired ■and the welding or repairing metal, the heat of ‘the arc being such as to raise the temperature of those portions of the object to be welded or repaired and also the temperature of the welding or repairing metal to incipient fusion and also that the new metal as it melts will become detached and unite in small increments with the parts to be welded or repaired, the intense heating of the object to be welded or repaired being restricted to the place of *287application of the new metal, thereby avoiding objectionable deformation and tension strains: substantially as described.
•‘2. The method of electric welding or repairing, which consists in establishing an electric welding arc between the metallic object to be welded or repaired, and the end of a rod of the welding or repairing metal, the heat of the arc being sufficient to raise the parts to be welded or repaired to incipient fusion and to cause the end portions of the rod to' melt and unite with substantial homogeneity to the fused iiortions of the piece being welded or repaired; substantially as described.”
“5. In the art of eleciric arc welding or repairing, so correlating the current employed for the production of the arc, to the welding or repairing metal and the metallic objects to be welded or repaired as to melt off in small increments the end portions of the welding or repairing metal and to likewise fuse the immediate parts to be welded or repaired but without causing objectionable deformation or tension strains in neighboring or distant parts of the piece to be welded or repaired; substantially as described.”
The device upon which claim 2 of patent No. 967,578 is drawn was stated in the specifications and drawings to consist of what is known as a two-w'ay switch or shunt system for the diversion of the current, if the arc used for the welding was disconnected or ceased to be formed. Under certain circumstances or installations, what is called by Siemund “a certain load for the dynamo when in operation” consists merely of a certain amount of resistance or opposition to the passage of the electric current by another path than through the welding arc, so that the changes of potential will not be so abrupt and the current wrould not fluctuate to such an extent as if no switch or no alternative resistance wrere provided.
This idea was stated by the experts for both parties, and is plainly shown by the patents introduced, to comprise a part of what is generally described as the generator or system of wiring for the supply of the current, and is not at all a part of the welding device. In other words, it makes for economy and preservation of condition, but does not enter into the apparatus or method of welding itself, and was old in the art, even when applied to electric welding.
The Unwin & Howard patent (supra), the Edison patent, No. 264,668, September 19, 1882, as well as the Burton & Angelí patent, No. 488,468, December 20, 1892, and the Burton & Angelí patent, No. 537,008, Aprjl 9, 1895, have to do with what are called in the testimony merely details in the formation of the generator or in the apparatus supplying the current, and in arranging the current for various, methods of electric welding.
These patents had some hearing upon claim 2 of patent No. 967,578, and show the lack of invention in the device described in that claim, but this has uo bearing upon the remaining claims of patent No. 967,579.
These patents just mentioned did, however, have to do with the various claims of the Siemund patents which were amended in the Patent Office as showm by the file wrapper, and correspond to certain of the devices explained by Siemund in his specifications. But after taking the testimony, it became apparent that no invention could be based upon the application of this particular method of operation to the use of the welding arc with the apparatus described by Sie-mund.
*288The testimony also showed that the defendants have never used the two-way switch, and get the desired result from a “compound wound” generator, a well-known device. Hence the charge of infringement of claim 2, and even the claim for this installation to supply current as described therein, was therefore abandoned.
[1] .This takes the question of claim 2 of patent No. 967,578 out of the case, but would seem to give the defendants the right to have a determination that this claim is invalid, from the standpoint of patentable invention. This would make it unnecessary to consider further patent No. 967,578, were it not for the fact that both this patent and patent No. 967,579 described, as has been said above, the same conditions in the welding art, the same improvements as claimed by Siemund, and exactly the same specifications and details of application or illustration as in patent No. 967,579.
Another patent, No. 948,764, issued to Kjellberg, on February 8, 1910, upon an application filed August 13, 1907, presents several questions which must be considered separately. The date of this patent makes it too late to be treated as an anticipation of Siemund. Westinghouse Mach. Co. et al. v. General Electric Co. et al. (D. C.) 199 Fed. 907; Union Typewriter Co. v. L. C. Smith & Bros. (C. C.) 173 Fed. 288; General Electric Co. v. Allis-Chalmers Co. (C. C.) 190 Fed. 165. But it was so treated in the Patent Office, for the date of granting preceded the consideration of the Siemund application. No interference was declared and no issue as to date of invention was there determined, for the Kjellberg patent was treated as if inoperative, or at .least as not showing the Siemund method. But the Kjellberg application wás earlier than any use by Siemund in this country, and we must consider whether Kjellberg was the real inventor if any patentable novelty was disclosed by him which would have infringed Siemund (if later) or would have.been an anticipation if the date of issue had preceded the Siemund application. Sundh Electric Co. v. Interborough Rapid Transit Co., 198 Fed. 94, 117 C. C. A. 280.
Kjellberg described a method and device for using electric welding in limited spaces, as on ship boilers and for overhead work, by the use of a metal electrode held “at a sharp angle to the line of movement.” He deposits the metal in successive layers by the. movement or manipulation of the arc. His metallic electrode is held in a handle which is the equivalent of the holder or clamp of Siemund, and his arc is plainly shorter than in the Benardos method as practiced or as used by Coffin and the others above named. He heats the work only at the point of welding and says that the metal of the electrode is “pulled or sucked by the electric current from the positive electrode to the negative.” In other words, he manipulates the voltaic arc in such a way as to allow the deposit and adhesion of the molten-metal.
But Kjellberg also claims an electrode with a cover of noncombustible, normally nonconducting material, which will form compounds or alloys with the molten metal, and which, he explains, will have the effect of causing the end of the metal electrode to melt, in the shape of a cup, thus holding the molten material in place while being distributed in the line of the weld and also attached upward, if the welding be done upon the under surface of the work.
*289A great deal of contention has arisen in the testimony of the various witnesses about the operation of this Kjellberg patent. The complainant’s witnesses’ charge that the Kjellberg patent is inoperative and point out their statements to the Patent Office when Kjellberg was cited as a basis for criticism of the Siemund application. The complainant’s expert also testifies that the Kjellberg device, when used as described in the patent, will not produce a welding or brazing, but will merely cause a melting of the electrode when applied to an overhead weld", and that the melted portion will drop to the floor, instead of becoming affixed to the point of the material or overhead surface to be welded, where the voltaic arc is formed.
The testimony shows experiments by experienced welders, who were instructed, and who, on the face of the experiments, endeavored to produce the results stated by Kjellberg. The good faith and sufficiency of these experiments is attacked by the defendant Enderlin, who testifies that any competent and experienced welder can cause the molten metal from the electrode to drop to the floor, and not to be deposited upward against the surface to be welded, in such a way that a spectator, looking on at a distance of from three to four feet, with the usual glass shield which is necessary when the voltaic arc is employed, could not detect that the operator was not actually carrying out his instructions to make a weld.
On the other hand, Mr. Kenyon, the defendants’ expert, testifies that the Kjellberg patent is inoperative, ánd even goes so far as to say that the Siemund and Enderlin methods of welding are inoperative, when they seek to apply the molten or partially melted extremity or particles from the electrode upward against the under surface of the material to be welded, in the location of the voltaic arc. Mr. Kenyon testifies that no such effect could be produced unless there is a coating (even though this be extremely thin) upon the outer surface of the wire electrode, which coating will diminish the fluidity of the melted portion of the electrode, and produce wliat Mr. Kenyon calls a “pinching” effect toward the center where the fluidity is greater. He thinks that the flux used in certain instances by Siemund and by the defendants, and in a way furnished by an electrode cover of chalk, magnesia, and potassium, as described by Kjellberg, is sufficient to produce this pinching effect in any instante where welding is actually accomplished.
It is made apparent by the testimony of the complainant’s witnesses 'and of the defendant Enderlin himself that the overhead process of welding can he accomplished with or without the flux coating, and with or without a cover of substances which will form alloys or furnish an electrode of secondary conductivity to the metal wire; and one witness testifies that potassium or magnesia would have no effect at all in forming such an alloy.
But the necessary conclusion from this conflict seems to be that the Kjellberg patent is inoperative only when the electrode is held in such a position and so applied that the fluid metal is not in fact caused to be deposited at the other pole of the voltaic arc and to adhere thereto. If the Kjellberg electrode be of the proper proportions and handled in the proper way, the question would be, not whether Kjellberg was op*290erative, but whether the device used was really that disclosed by the Kjellberg patent, and was actually the device used by the complainant and by the defendants.
The' Patent Office allowed the Siemund patents over the objection based upon Kjellberg. An examination of the Kjellberg patent shows plainly an electrode of such diameter that the operation would be that of the Benardos method, and which would be effective only for welding materials in such position that the brazing or soldering metal could fall upon the joint (unless by the experience of the operator, that is by experiment, the metal electrode should be made of such slight diameter fhat the phenomenon described by Mr: Kenyon as “pinching,” and by the complainant and defendants in this case as the actual projection of the small particles from the electrode by means of the voltaic arc, could take place). Kjellberg states that his electrode is to be made of solder, and that the electrode metal is to be pulled or sucked from the positive electrode to the negative electrode. He indicates that the path in which the electrode is to be moved is about the distance of the diameter of the electrode from the work, and he seems to have had in mind, and when operating (provided the operator has solved the mechanical principles of successful manipulation) to be following, the same method which the complainant has attempted to patent and which the deiendants use.
But the Kjellberg patent, in claims 2 and 3, states that the invention lies in the use of the cover or outside material forming the cup in which the molten part of the electrode shall be held. If the idea of such a device was invention, it has nothing to do at least with the patent in suit, and whether or not it was operative for overhead work does not affect the present question. Claim 1 of the Kjellberg patent is as follows:
“The herein described method of welding, brazing or soldering with an electrode containing soldering material. consisting in passing a current to form an are between the electrode and the parts to be welded and then moving the electrode in a line parallel to the part -where the solder is to be applied while held at a sharp angle to the line of movement, so that the soldering material will pass from the electrode in a continuous stream and be applied in rows and layers to the parts to be welded.”
In this he would seem to be describing the result which he obtained when welding as is done by the complainant and the defendants in this action. But in this claim 1, Kjellberg does not describe nor state a method of performing a welding operation on an overhead surface, as' distinguished from the Benardos method. He describes rather the way in which he manipulates the appliances which will accomplish overhead welding, if the method of constructing and arranging these has previously been understood.
One of the witnesses in the present case is said to operate under Kjellberg’s patent and to perform successful welding on overhead work. But this witness in fact uses the Siemund methods rather than those claimed by Kjellberg. This again contradicts the contention of the experts that none of the methods described will do what they are supposed to accomplish, and indicates further that the welding operation itself is not the result of any method claimed by Coffin, Kjellberg, *291or Siemund, and that these patents are all attempts, by certain successful operators, to explain what they have succeeded in accomplishing by manipulating a metallic electrode, of small diameter, when actuated by an electric current, sufficient only in voltage and amperage, to produce a voltaic arc of extremely short length, and causing a melting or incipient fusion at the immediate point to which the arc is sprung upon the work.
The defendants’ witnesses further show that the angle at which the electrode is held, with reference to the surface of the work, depends upon the manipulation of the individual workman, and that the result is not because of any particular angle which may be formed, but rather that the angle is but a measurement of the location of the parts when the workman is successfully carrying on the process.
The conclusion, therefore, seems necessarily to be that Coffin, Kjell-berg, Siemund, the defendants, and the workmen operating under the Kjellberg patent, have all, through these various times, been applying the teachings of the Benardos method, as improved upon by Coffin and Kjellberg through the use of a metal electrode, which can be made to supply additional material by melting at the positive pole of the arc. They have all found by experience that, under certain relative conditions of size of parts and strength and quantity of current, an experienced workman can weld upon an overhead surface. They have all realized that the use of a flux may assist the process of melting, and that in some instances the presence of the flux or coating material may have the “pinching” effect, or may cause the voltaic arc to operate more directly upon the fluid metal in the center than at the surface of the electrode. They have also learned that, in the absence of such coating or fluxing material, the proper proportion and relation of the parts will allow the operation of overhead welding to be accomplished. But Siemund was the first man who described to the Patent Office, or who expressed in writing, a definite description of the proportion and arrangement of the entire apparatus and the method of manipulation of the parts thereof, when making a successful overhead weld.
[3] It would not seem that such a description of particular arrangement or adaptation to a particular purpose was the invention of a method; nor can it be held that an arrangement of the parts of a device (described in prior patents and used for a purpose and in a way shown in those patents, but, coupled with the skill of the operator, made to be effective only under certain conditions, which will produce results desirable under certain circumstances) can be patentable as a new invention, when the earlier patents show both an understanding of the possibility of these results, the existence of such an arrangement of parts and of the conditions produced, together with an understanding of what is necessary for a skilled operator to get these results, even though the reason assigned, or the explanation of the cause of the results themselves, be mistakenly stated and attributed to incorrect factors or parts in doing the work.
[2] The functions performed by certain apparatus, when arranged under certain conditions, cannot be patented as a new method of producing the result, if the so-called “method” is merely a description in *292new terms of one of the forms of the old process, as carried out by the previously disclosed necessary elements of the device, and where the so-called new method i^ but a description of an equivalent experimentation with the old device, under conditions recognized as possible, within the knowledge of any mechanic but not previously stated in language. Berardini v. Tocci (C. C.) 190 Fed. 329.
Nor is mere adjustment patentable. The testimony in the case and the patents cited show that, throughout the entire development of the art of electric welding, the necessity of avoiding undue heating, immediately adjacent to the point of fusion, and to a disturbance or change in the chemical and physical structure of the material around the place where the actual welding is taking place, necessitated a careful use of the voltaic arc, and a localization or an attempt at confining the influence of the heat and the process of melting to -as little of the substance or surface as would be sufficient to make a homogeneous weld.
In this way, by experiment, Siemund has stated in his application for patent'the general limits as to amperage and voltage within which he can supply a current through a metallic pencil, which will produce a voltaic arc of extremely short length and sufficient only to melt the extreme end of the pencil and the immediate points where the voltaic arc is attached to the work. The length of this arc is to be kept within such limits that the flow of the detached fluid particles, under the influence of the current passing through the arc, will cause them to pass' to the other extremity of the arc; that is, to come in contact with the fluid or incipiently melted surface of the work. A careful manipulation so as to shorten the arc to the requisite length, and then to remove the electrode so as to produce a fresh melted portion without having the electrode itself unite with the work upon cessation of the arc, or without destroying the arc by removing the electrode too far from the work, is recognized and stated by Siemund as the real explanation of how he was able to successful^ weld, an overhead surface. But this is not in his claims of invention. He cannot patent the function which the elements perform, nor can he claim every sort of device which produces the result because he has described the forces and the correlation of the parts in producing the result. Manhattan General Const. Co. v. Helios-Upton Co. (C. C.) 135 Fed. 785; National Hollow Brake-Beam Co. et al. v. Interchangeable Brake-Beam Co., 106 Fed. 693, 45 C. C. A. 544; Fuller v. Yentzer, 94 U. S. 288, 24 L. Ed. 103; Westinghouse v. Boyden Power Brake Co., 170 U. S. 537, 18 Sup. Ct. 707, 42 L. Ed. 1136; Telephone Cases, 126 U. S. 1, 8 Sup. Ct. 778, 31 L. Ed. 863.
The Benardos device showed the qualities of the voltaic arc and the effect of bringing into conjunction metal rendered fluid by that arc; the Coffin patent and Coffin experiments showed the understanding of how to produce the same result through tire use of the metallic electrode, and an understanding of proportioning the current, the material, and the size of the arc, to the necessities of the work, that is, so as to accomplish a weld; while Kjellberg described the application of these methods to overhead welding, explained the use of fluxes and of the *293necessary manipulation of the device, so as to cause a deposit of the metal through the effect of the arc, and indicated an arrangement (which he did not fully understand and appreciate but which he did describe) where the proper relation of the parts and adjustment of the electric current is so apparent that any workman, skilled in the application of electricity, would consider it as governed by the relative size and requirements of the work, even though that statement is not expressed.
When we come to the Siemund patents, we have the first definite expression of the general limits necessary in practical work. We have a definite statement of just how the workman would operate in successfully welding by the electric current, to an overhead surface. We have the statement that this is actually accomplished, most successfully, by the use of an additional coil of wire or system of winding around the handle of the electrode, through which a magnetic current can be applied, during the process of welding, and we have a statement as to the joint use of the coating or flux, for the double purpose of improving the material after welding, and of making the melting process easier at the exact point of the springing of the arc. Cut, strange as it may seem, Siemund stated that he preferred to make his invention depend upon the use of this magnetic current, and in his application to the Patent Office, as shown by the file wrapper introduced in evidence, not only stated that successful welding to an overhead surface could not be accomplished unless the magnetic current was added around the electrode, but he also stated that without the use of such magnetic current and fluxes or coating, the melted material would fall to the ground and could not be made to attach to the overhead surface.
It will be noticed that Benardos used the electrode for either the positive or negative pole, and in some of the other patents, such as Coleman, the electrode is made the negative pole: the reason there being, as was recognized in the Benardos patent, that for the purpose of cutting metal or withdrawing the molten metal, a reversal of the electric current would have a tendency to cause the fluid material to move away from the point of fusion.
With this in view, and recognizing the proposition that a magnetic current will not affect molten iron positively, the Patent Office took exception to Siemund’s claim that his invention consisted either in the employment of the magnetic current or in the possibility of using it to dispel the molten material, and suggested that the current was more likely to attract the same, or to cause the arc to revolve.
Siemund thereupon confined his claim for the device and method with the magnetic'winding to but one claim of the patent, and stated that the voltaic arc and the-electrode were merely subject to the magnetic influence, without saying that it attracted or repelled, and, in fact, without saying whether it was influenced at all by the magnetic current, but still retaining the claim of invention for a magnetized apparatus. The other claims, covering a .structure without this magnetic winding, were subsequently added and were the result of the effect of the Kjellberg and Coffin patents, or of further experimenta--*294tión, in which Siemund apparently realized that his original conclu-' sions were not correct, and that welding could be done without the use of the magnetic coil. 'He also relinquished some of his claims as to adjustment of the current, and upon this trial the balance of these claims have been withdrawn, and, as has already been stated, claim 2 of patent No. 967,578 is apparently invalid.
For all of these reasons it seems to the court to follow that Sie-mund was making an expert mechanical application of the Benardos method, as modified by the Coffin and Kjellberg devices and methods, in such a way as to successfully apportion and manipulate the structure for the purpose of welding to an overhead surface. Fie attempted to describe these mechanical adjustments, in connection with what he thought was an invention, through the use of a magnetic coil, and what •is left, after the withdrawal of the claims about the means of regulating the current and of attempting to magnetize the arc, does.not show patentable invention.
The defendants have raised the objection that the language of the claims is so broad that they cannot stand independent of the statement contained in the specifications. That is, that they are not patentable except as limited by the specifications, and that they do not describe in terms the matters which the specifications show Siemund intended to be claimed as his real invention. Siemund’s claims contain nothing new except as contained in the words “substantially as described,” and this reference to the specifications is only available to show invention where the application of old methods is thus made to apply to a new use which of itself constitutes invention. Hohmann & Maurer Mfg. Co. et al. v. Charles J. Tagliabue (C. C.) 175 Fed. 87 at page 91.
This contention need not be considered in detail, for the preceding discussion answers the proposition. If the invention is patentable, the claims as limited to the nature of the invention disclosed by the patent as a whole would not be objectionable. On the other hand, if there be no invention, then the claims are merely a statement of the .teaching of the prior art, and are not valid, but are no more invalid because general in form.
The conclusion that the patent is invalid is rendered much less debatable by the general evidence as to prior use. It appears from the testimony that the defendants, in attempting to practice the Benardos method but with a metallic holder, accidentally found that the formation'of the voltaic arc between the holder and the work produced a welding of the holder to the work, and by further experimentation .reached the use of a device, substantially equivalent to the Siemund apparatus as described in the patent, and operated under the exact method set forth in Siemund’s method claims. The testimony does not show when the defendants first used-the apparatus for overhead : welding. But Siemund claims a method and device for welding in all positions, and does not claim invention because the device can be used for overhead welding also. ITis specification shows this merely as an illustration of scope and practical utility, and this was taught by the earlier patents and also invented by others in so far as it involves patentability.
*295The defendants use a barrel of water as an artificial quieting resistance. They have never attempted the magnetic winding around the electrode holder, and have never attempted the elaboration of alternative methods for supplying the current so as to maintain a more or less constant tension, whether the voltaic arc be in use or temporarily broken during the progress of the work, except as furnished by the use of a compound generator.
But these differences have nothing to do with any of the claims under discussion, nor with the method and devices of the Siemund patent proper, and the defendants are therefore, as was said at the outset, using substantially the identical method of the Siemund patent, and have been for such a period as the testimony shows the same devices to have been employed by them in New York harbor. The testimony shows that Siemund was brought over from Germany by inducements made to him around the years 1906-07, and because of news of his success in repairing boilers upon the German steamships. The various government authorities and large steamship companies around the harbor of New York gradually learned of the Siemund methods, as well as of those of Enderlin, and, shortly before the issuance of the patent to him, his employment of these methods was well known to the steamship trade. The persons who employed Siemund seem to have had no knowledge of the work done by Enderlin. But, on the other hand, a number of other individuals either employed or knew of the work done by the defendants, and the evidence shows conclusively that, for seven or eight years prior to the Siemund experiments, Enderlin was welding on a small scale, but still publicly, and in such a way that it is strange the matter did not become better known, by the method of electric current, of substantially the proportions needed for the Siemund method and by the use of a metallic electrode of such small size as to produce a voltaic arc, manipulated in almost the identical way which Siemund later patented.
Such use makes it apparent that Siemund was not at ,the time of obtaining the patent the inventor in the sense which the statute requires; and even if the particular improvements upon the Benardos method were patentable, or if the particular device showed patentable novelty, the American patent must be held invalid when tested from the standpoint of the defendants’ prior use.
The defendants may have a decree.