AO Smith Corporation v. Petroleum Iron Works Co.

73 F.2d 531 | 6th Cir. | 1934

73 F.2d 531 (1934)

A. O. SMITH CORPORATION
v.
PETROLEUM IRON WORKS CO. OF OHIO.^[*]
PETROLEUM IRON WORKS CO. OF OHIO
v.
A. O. SMITH CORPORATION.

Nos. 6330, 6331.

Circuit Court of Appeals, Sixth Circuit.

November 10, 1934.

George I. Haight, of Chicago, Ill., and A. C. Paul, of Minneapolis, Minn. (Maxwell & Ramsey, of Cincinnati, Ohio, Paul, Paul & Moore, of Minneapolis, Minn., E. A. Andrus and P. O. Dunham, both of Milwaukee, Wis., and M. K. Hobbs, of Chicago, Ill., on the brief), for A. O. Smith Corporation.

A. B. Stoughton, of Philadelphia, Pa. (John Walsh, James A. Richmond, and Louis A. Spiess, all of Washington, D. C., L. B. Coppinger, of Sharon, Pa., and Day & Day, of Cleveland, Ohio, on the brief), for Petroleum Iron Works Co.

Before MOORMAN, SIMONS, and ALLEN, Circuit Judges.

SIMONS, Circuit Judge.

Two causes of action were incorporated in one suit, begun below by the A. O. Smith Corporation. The first was a patent infringement suit alleging infringement of Smith reissue patent, No. 16,865, for an "oil refining still and method of making the same by electric welding." The second was based upon wrongful and fraudulent appropriation of certain secret processes in the field of electric are welding alleged to be owned by the plaintiff, and useful in the fabrication of oil pressure vessels of the kind described in the patent. The second cause of action was, on the defendant's motion, dismissed, for the reason that the two causes were inconsistent. The original suit went forward as a patent suit, and the secret process case was reinstituted by independent bill. The two causes were, however, heard together, and both were referred to a master, who filed a separate report in each. The master held the patent invalid, but, if valid, so limited as not to be infringed. His report was confirmed by the court with respect to the finding of invalidity. From the decree the plaintiff has appealed. In the secret process suit, certain of the processes relied upon were reported by the master to be the property of the plaintiff and wrongfully appropriated by the defendant. Other processes were held not entitled to protection because within the realm of mechanical skill and so within the public domain. The court sustained those findings of the master which were adverse to the defendant, but granted a decree which included relief for infringement of processes rejected by the master. From this decree the defendant appeals.

Both appeals were heard together, though separately briefed, and will be decided by this single opinion. 6330 is the appeal in the patent infringement suit; 6331 is the appeal in the secret process suit.

The Patent Case.

The original patent is No. 1,577,410, dated March 16, 1926. The reissue application was filed November 26, 1927, and the reissue patent granted January 31, 1928. In the language of the specification the invention "relates to stills employed in refining crude oils, and to a method by the practice of the steps of which it is now rendered possible to construct stills embodying the invention in its concrete form." The claims in suit are 3, 5, 6, 14, and 16, and are printed in the margin.^[1]

To understand the plaintiff's relation to the oil industry and its alleged contribution thereto, whether under the ægis of the patent or otherwise, a brief review of the industry's evolution is required. For some years prior to 1920 there had developed a demand for gasoline far greater than could be economically supplied by atmospheric or low pressure distillation from crude oil. The so-called cracking processes then came to be an important feature of refinery operation. By such processes the heavier oils were resolved into lighter products through the removal of carbon, with the result that a much greater amount of gasoline could be recovered from a given quantity of oil than by distillation. The cracking processes, however, required the crude oil to be subjected both to high temperatures and high pressure, and this in turn created a demand for vessels or stills which could successfully withstand both and be of unusually large capacity.

Prior to the development of the plaintiff's still, the low pressure vessels employed in the oil industry were either of riveted or hammer-welded construction, and the relatively new high pressure vessels which were used were forged. Hammer-welded vessels were limited in wall thickness to 1 5/8 inches because of the nature of the process by which they were manufactured. The riveted structures lacked integrality, failed to react to pressure or heat as a single piece of metal, and the failure of the welded seams often resulted in dangerous fires and explosions. Forged vessels, that is, those made in one piece by hammering out a single steel ingot, were very little used because expensive to produce and because extremely limited in their size by the magnitude of the equipment required in their production.

The Smith conception was of a vessel constructed from a plurality of plates much longer and thicker than had ever been used before, and so joined as to create a single integral tubular structure of great size; the plates being fused at their meeting edges by an electric are, and the vessel free from such circumferential seams as constituted objections in prior construction. That the plaintiff succeeded in the materialization of the Smith concept, supplied a long-felt need in the oil industry, and built up a profitable business of great magnitude in the production of large cracking stills, is beyond controversy upon this record.

The problem that confronts us is to determine whether the patent denotes invention, whether the inventive step is disclosed with such particularity as is required by the patent law and its limits fixed by the terms of the grant. It will be noted that of the claims in suit 3 and 5 are for an article of manufacture, while 6, 14, and 16 are for a method by which the article is produced. The relationship and interdependence of coincident method, machine, and product inventions are clearly pointed out in Reo Motor Car Company v. Gear Grinding Machine Co. (C. C. A.) 42 F.(2d) 965, at pages 967, 968. It was there indicated by Judge Denison, correctly, we think, that the underlying inventive thought will usually be found in one of the three, and the other two will be relatively collateral. It is clear from the language of the article claims that here, as in the Reo Case, the method dominates the result. The still described and claimed in claims 3 and 5 is a still made according to the method substantially as set forth in claims 6, 14, and 16. It is sufficient, therefore, in our view, to discuss the method claims. If they fail, the patent fails. Cf. Judge Hickenlooper's discussion in Nestle-Le Mur Company v. Eugene, Ltd., (C. C. A.) 55 F.(2d) 854.

It will perhaps be sufficient to analyze claim 6, the first of the method claims. The steps in the method there disclosed for the manufacture of stills of extreme dimensions in their sequence are: (1) Forming thick plates of great length into longitudinal troughlike sections; (2) assembling a plurality of such sections in circular order with their edges in register; (3) fusing such sections in the meeting line of their edges by an electric are to constitute an integral tubular structure without circumferential joints. Electric are welding was old. The assembly of plates or bars with their edges in register so that they may be fused is a step in such welding both old and obvious. No method of forming thick plates or shaping them into troughlike sections is disclosed. There is neither novelty nor invention in the forming of a tubular body by placing such sections in register in circular order. We are therefore forced to seek for novelty and to discover invention in the utilization of plates of greater thickness and length than had been used before, in the achievement of an integral tubular structure in the sense disclosed in the specifications, and in its production without circumferential joints.

While we recognize that the fabrication of structures of unusual dimensions may often present problems not present in smaller structures and not within the reach of the skilled mechanic, and that in their solution the inventive genius may be exercised, yet it has become sufficiently axiomatic to warrant dispensing with citation that a mere change in proportions, without more, does not amount to invention. The patentee points out very clearly in his specification the limitation upon the size of prior art stills made with curved plates welded together by the hammer or forge method. He then describes his improved method, which is the fusing of the plates along their meeting edges by an electric are, whereby successive layers of welding metal flowing from a fusible weldrod are deposited in a previously prepared welding groove so that thorough amalgamation of the metal takes place and an integral structure, free from any joints, is assured. We search the specification, however, in vain for any step in a method of electric are welding not known to the art. Electric are welding was not the discovery nor the invention of the patentee. Steam boilers and other pressure vessels had been so made, though not of course of the wall thickness, diameter, or length described in the patent. The precise difficulty involved in applying electric welding to plates of greater length and thickness than had been fused before at their meeting edges by an electric are is not pointed out in the specification, nor are the steps that are new in the alleged improved method of are welding indicated. As the master in his report has well said, "The size of the vessel which the patentee envisioned was new in the industry, and nothing else was new in any way so far as the claims cover any process or structure."

So far as achieving an integral structure apart from the claimed novel method, the inventor was certainly not the first to do this, as appears both from the specification and the evidence. Moreover, the integrality of the vessel is a result and not a step in the method. If nothing new appears in the method, either as a single step or in combination with others, or if invention is absent, the result can save neither the method nor the article that is achieved by following it. Prior vessels, arc-welded at the meeting edges of the plates, however limited in length, diameter, or wall thickness, were integral vessels. The integrality of a much larger vessel may be new, but, unless brought about by a new method which discloses inventive thought, is the result of mere increase in size or proportions, and this, as we have indicated, is not invention.

Merely making the joints by are welding, the plaintiff contends, will not necessarily produce an integral structure in the sense that it will react as a single piece of metal, but are welding in the manner set forth in the specification will provide such integral structure. The first illustration relied upon to demonstrate the novelty and ingenuity of the method would seem to indicate that invention consists of welding at the meeting edges, as set forth in the claims, and it is pointed out that an arc-welded joint formed by overlapping the edges, then arc-welding the overlapped edges, would not produce an integral structure such that in the language of the specification "a joint can hardly be said to exist." But the inference that such welding constitutes the inventive concept is immediately repudiated by the plaintiff in its second illustration, wherein it is pointed out that a joint produced by arc-welding the meeting edges, but reinforced by building up weld metal on both sides of the plate so as to produce in effect a band of steel of greatly increased thickness extending along the seam, would likewise not produce an integral vessel "such that a joint can hardly be said to exist," nor would such a structure react as a single piece of metal to the temperature and pressure conditions encountered in oil cracking. Here we apparently have the contention that welding at the meeting edges is not the essential inventive concept, but that welding at the meeting edges without leaving excess metal to produce the reinforcing band of steel (or strap joint) at the seams is. But, if such is the inventive concept, it is likewise abandoned in the third illustration, where it is pointed out that a joint produced by arc-welding the meeting edges (and we assume without leaving excess metal), but made in such manner that the weld metal is "of notably brittle structure," would also not produce an integral structure which would react as a single piece of metal. By what steps new to the art, either singly or in combination, this avoidance of brittleness in the metal of the seams is to be achieved is not indicated, and nothing with respect to it as a step or steps in the method is claimed.

Reluctant to strike down a patent where commercial success is so amply demonstrated, we search for further light upon the inventive concept. The claims tell us nothing but that the plates are welded at their meeting edges by an electric are, though indeed claim 16 avoids even this limitation and covers all welding employing electric current, but nothing of novelty or invention is found in the specification, either with reference to are welding or any other conceivable type of electric welding. The inventor describes what happens in the practice of arc welding; i. e., that successive layers of welding metal flowing from a fusible weld-rod are deposited in the previously prepared welding groove, so that thorough amalgamation of the metals takes place. If it is the welding groove that is the essence of the invention, it is not claimed, nor is any particular configuration of the groove claimed as a step in the method. The drawings show a U-shaped groove, but all the inventor says of that is that the groove may have the form indicated in the drawing. It is neither recited nor claimed that providing a U-shaped groove is an essential step in the method. As a means for conducting the electric current, the inventor prefers to use covered weld-rods, but the use of covered weld-rods is not a step in the claimed method, and their inhering advantages are said by the inventor himself to be "now well known in the metallic are welding art."

If the method of assembling and fusing plates of unusual length at their meeting edges discloses nothing of novelty except in degree, and therefore nought of invention, it is difficult to see invention in the avoidance of circumferential seams. The absence of such seams is but an inevitable and obvious result of the method. The virtue that can in no way be avoided is not a demonstration of invention, and at most the absence of circumferential seams is inherent in the concept of plates of unusual length.

We have gone thus particularly into the specification, though fully realizing that specifications may explain but not expand the claims. Howe Machine Company v. National Needle Co., 134 U.S. 388, 10 S. Ct. 570, 33 L. Ed. 963; McCarty v. Lehigh Valley Railroad Co., 160 U.S. 110, 16 S. Ct. 240, 40 L. Ed. 358; Lehigh Valley Railroad Company v. Mellon, 104 U.S. 112, 26 L. Ed. 639; Pearson v. Uncle Sam Mfg. Company, et al., 11 F.(2d) 603 (C. C. A. 5); Wadsworth Electric Mfg. Co. v. Westinghouse Electric & Mfg. Co., 71 F.(2d) 850 (C. C. A. 6); Chicago Forging & Mfg. Co. v. Bade-Cummins Mfg. Co., 63 F.(2d) 928 (C. C. A. 6). This minuteness is perhaps dictated by the thought that it is not always easy to determine when the specification is merely explanatory and when it really covers some step in the method or element in a structure not claimed, either literally or by necessary inference. The plaintiff brought forth something new in oil cracking stills. Whether its product was the result of inventive thought, the mere excellence of its technical skill, the magnitude and costliness of its physical equipment, or the secret processes hereinafter discussed, we are unable to say. If invention was involved, it is sufficient to say that the patent in suit does not disclose it. "The statute requires the patentee not only to explain the principle of his apparatus (or method) and to describe it in such terms that any person skilled in the art to which it appertains may construct and use it after the expiration of the patent, but also to inform the public during the life of the patent of the limits of the monopoly asserted, so that it may be known which features may be safely used or manufactured without a license and which may not." Permutit Company v. Graver Corp., 284 U.S. 52, at page 60, 52 S. Ct. 53, 55, 76 L. Ed. 163. This the patent does not do, and we hold the claims in suit invalid. It becomes unnecessary, therefore, to discuss the questions of infringement, validity of the reissue as such, or intervening rights.

The Secret Process Case.

The first challenge to the decree in 6331, awarding damages to the plaintiff for the appropriation of its secret processes, is that the dismissal of the second cause of action in the original suit was res adjudicata of the present controversy. It need give us little difficulty. It is well settled that, even though the cause of action and the parties be the same, a prior determination is a bar to a subsequent suit only when it is based upon the merits. If the first suit was dismissed for defect of pleading or parties or a misconception of the form of proceeding or the want of jurisdiction, the judgment of dismissal is no bar to another suit. Hughes v. United States, 4 Wall. 232, 237, 18 L. Ed. 303; Baker v. Cummings, 181 U.S. 117, 124, 21 S. Ct. 578, 45 L. Ed. 776; Southern Pacific Railroad v. United States, 168 U.S. 1, 49, 18 S. Ct. 18, 42 L. Ed. 355; Brown v. Fletcher, 182 F. 963 (C. C. A. 6); Detroit Trust Company v. Dunitz, 59 F.(2d) 905 (C. C. A. 6).

More important is the challenge to the decree awarding damages for appropriation of those of the plaintiff's secret processes for which patents have now been issued, on the ground that the right to secrecy and to patent monopoly are inconsistent, and that both cannot be asserted with respect to the same process. The secret processes relied upon were twelve in number. Secrets numbered 1, 2, 3, 4, 5, 5-a, 9, and 10, are now the subject of granted patents, and applications upon secrets 6 and 8 were pending at the time of the hearing. The argument is substantially this: Since one having rights mutually inconsistent must assert that one upon which he relies, the filing of an application in the Patent Office is an election to rely upon monopoly and an abandonment of secrecy, a disclosure of the secret to the public, and a surrender of all rights thereto except such as are conferred by the patent grant. It is therefore argued that the decree awarding damages and their computation is erroneous in so far as it covers the period between the time of the applications and the issue of the patents. In so far as the contention is directed to inconsistency between the two remedies for the same period of time, the argument is undoubtedly sound. We doubt, however, that the filing of an application for a patent is the dedication of a secret to the public. At the time of the application, the inventor or discoverer has no means of knowing whether the patent will ultimately be granted. If the secret is valuable, the discoverer, conceiving it to be patentable, would by making application hazard both secret and patent. This would defeat the very purpose of the patent law, which conditions monopoly for a limited period upon the complete surrender thereafter of the subject-matter to the public, and we think no such principle can be deduced from the authorities. Macbeth-Evans Glass Co. v. General Electric Co., 246 F. 695, 701 (C. C. A. 6), does not support the conclusion for which contention is made. In that case Macbeth elected to use his invention as a trade secret for some ten years before applying for a patent. It was there said: "If the right to secure protection of the patent laws can be effectively repudiated, it certainly has been here." The inventor's course of conduct was construed as in effect an abandonment of his right to patent monopoly, and inconsistent with the duty of diligence resting upon an inventor desiring to patent his invention. The rule of diligence was early recognized by the Supreme Court in Pennock v. Dialogue, 2 Pet. 1, 7 L. Ed. 327, and Kendall v. Winsor, 62 U. S. (21 How.) 322, 328, 16 L. Ed. 165. That from it should be deduced a rule that during the period in which the patent application is being pursued in the Patent Office there should be no right on the part of the inventor to protect his process by secrecy or that failing a patent he should forfeit equitable protection is unsound. No injunctive relief is available to an inventor between the date of application and the patent, for the duration of the monopoly is measured by the grant. Gayler v. Wilder, 10 How. 477, 493, 13 L. Ed. 504; Rein v. Clayton, 37 F. 354, 3 L. R. A. 78 (C. C. Mich.). Our conclusion is also supported by the established practice in the Patent Office to maintain all live applications in secrecy pending disposition. There is no proof here that the plaintiff was unduly dilatory in seeking the protection of the patent laws for his secret processes. There was no effort to extend monopoly beyond the term provided for by law. There is, moreover, in the case of patentable processes, every incentive to dispatch in seeking protection of the patent law. Prior to patenting the inventor not only risks the loss of his secret through inadvertence or breach of confidence, but likewise through independent discovery. During his period of monopoly he incurs no such hazard. We cannot lightly assume delay inimical to self-interest.

It is next contended that the plaintiff has forfeited its right to equitable protection of its trade secrets since they were necessary to the practice of the patented method, and the deliberate withholding of them from the disclosures thereof was a breach of faith as against the public. Of this contention it is sufficient to say that the proofs are barren so far as indicating that the secret processes were known to Smith at the time of the application either for the original or for the reissue patent. They were not the discovery of Smith, but of Stresau, the plaintiff's engineer. In fact such evidence as there is controverts rather than supports the asserted inference. Nor is our conclusion inconsistent with the suggestion that the secret processes may have been largely responsible for the plaintiff's commercial success, since that was a progressive development that followed not so much the granting of the patent as the inauguration of some if not all of the so-called secret processes.

As already indicated, the master found certain of the secret processes entitled to protection, but reported adversely to the plaintiff with respect to the others, namely, those from 5 to 12, inclusive. These he found to be within the public domain, because within the realm of mechanical skill. The court overruled the master with respect to such processes, and granted relief. This, it is contended, it was beyond the power of the court to do. We have recognized, as have other courts, that there is a presumption of correctness to the findings of a master if based upon substantial evidence, but we know of no rule that gives to such findings the finality of a jury's verdict.^[2] No such sanctity in equity cases attaches even to the decision of the court itself. As was said by this court, Mills Novelty Company v. Monarch Tool & Mfg. Co., 49 F.(2d) 28, 29, "Appeals in equity bring up the whole case (with certain inferences in favor of the decree below), and the decree below should be sustained if it was right for any reason." Cf. Laursen et al. v. Lowe, 46 F.(2d) 303 (C. C. A. 6).

Apart from these considerations, however, we think the overruling of the master's report in respect to those alleged secrets held by the master not entitled to equitable protection was warranted by the master's erroneous application of the law. We do not accept the plaintiff's contention that, regardless of whether processes are novel, those things which it has attempted to hide from the public will be protected against use by any one who obtains knowledge of them through breach of a confidential relation, and the master was right in rejecting it. Nevertheless, in endeavoring to ascertain whether the processes were novel, the master seems to have applied the test of invention recognized by the patent law, and to have held invalid all processes which appeared to him to be within expected mechanical skill. It is agreed, and we think correctly, that processes which are not patentable may yet be the subject of trade secrets. However, if the rule applied by the master be the true one, it would be difficult to conceive of any process whose secrecy is entitled to be protected which is not at the same time patentable, except in so far as the patent law limits the subject-matter of patents.

To entitle one to a patent, there must be invention. The applicant must have exercised some degree of ingenuity, displayed some flash of genius, inspiration, or imagination not within the reach of mere partisanship. That which constitutes invention has, of course, never been successfully defined. There are tests, however, which, responded to, proclaim its absence; the most familiar, of course, being that leading to the determination, always difficult to make, that the thing achieved was the result of ordinary mechanical skill. A process may, however, be maintained in secrecy and be entitled to equitable protection even though invention is not present. The cases which deal with the elements necessarily present in a proprietary process are careful to define such processes as resulting from invention, or discovery. Peabody v. Norfolk, 98 Mass. 452, 96 Am. Dec. 664; Salomon v. Hertz, 40 N. J. Eq. 400, 2 A. 379; Eastman Kodak Co. v. Reichenbach, 79 Hun, 183, 29 N. Y. S. 1143; Herold v. Herold China & Pottery Co., 257 F. 911 (C. C. A. 6); O. & W. Thum Co. v. Tloczynski, 114 Mich. 149, 72 N.W. 140, 38 L. R. A. 200, 68 Am. St. Rep. 469. Quite clearly discovery is something less than invention. Invention requires genius, imagination, inspiration, or whatever is the faculty that gives birth to the inventive concept. Discovery may be the result of industry, application, or be perhaps merely fortuitous. The discoverer, however, is entitled to the same protection as the inventor. In Board of Trade of City of Chicago v. Christie Grain & Stock Co., 198 U.S. 236, 25 S. Ct. 637, 639, 49 L. Ed. 1031, the plaintiff collected stock quotations, which it furnished its customers. Said Mr. Justice Holmes of its tabulation: "It stands like a trade secret. The plaintiff has the right to keep the work which it has done, or paid for doing, to itself. The fact that others might do similar work, if they might, does not authorize them to steal the plaintiff's."

The mere fact that the means by which a discovery is made are obvious, that experimentation which leads from known factors to an ascertainable but presently unknown result may be simple, we think cannot destroy the value of the discovery to one who makes it, or advantage the competitor who by unfair means, or as the beneficiary of a broken faith, obtains the desired knowledge without himself paying the price in labor, money, or machines expended by the discoverer. Facts of great value may, like the lost purse upon the highway, lie long unnoticed upon the public commons. Hundreds pass them by, till one more observant than the rest makes discovery. It is idle to say that, in the eyes of the law, interest may not in such case follow discernment. We think the court below was right in rejecting the master's application of the law in respect to those secret processes held by the master to be invalid for the reason that they were in the public domain, or within the reach of the skilled mechanic in the trade.

We shall not discuss in great detail the evidence that sustains the findings of the master or the court with respect to the secret processes. It is sufficient to say that we find the evidence as to discovery, secrecy, appropriation, and breach of confidence not only substantial but compelling our independent judgment as to the correctness of decision below. The defendant had sought in many ways to obtain the knowledge necessary to enable it to compete successfully with the plaintiff in the making of pressure vessels. It finally engaged Hawthorne, who for many years had been employed in the plaintiff's welding department, and who was under express contract not to divulge any information received by him during his confidential relationship to the plaintiff. The conclusion is inescapable that the defendant knew of Hawthorne's employment in a confidential capacity. Shanor, its superintendent, had been denied access to the plaintiff's welding shop, though shown its other activities. The infringement of the secret processes followed Hawthorne's engagement with the defendant. Hawthorne had represented to the defendant that electric welding was more highly developed, more economically done, and on heavier products by the plaintiff than by others. The defendant largely increased his salary over that paid by its predecessor. Hawthorne's own estimate of the value of plaintiff's secrets and their cost of acquisition are revealed in his address to the National Board of Welding Engineers after he joined the defendant's organization, and the extensive advertising by the defendant of the novelty of its Fluid-Fusion welding process is further corroboration both of the value of the secret processes, their recent discovery, and the fact that they did not lie exposed in the public domain.

One contention remains to be noted. The defendant urges that Hawthorne's employment contract was invalid because, being one of many similar contracts made by the plaintiff with its engineers, it was in unlawful restraint of trade. It is only necessary to say that contracts of this character have frequently been enforced, United States v. Addyston Pipe & Steel Co., 85 F. 271, 46 L. R. A. 122 (C. C. A. 6); Harrison v. Glucose Sugar Refining Co., 116 F. 304, 58 L. R. A. 915 (C. C. A. 7); Thibodeau v. Hildreth, 121 F. 892, 63 L. R. A. 480 (C. C. A. 1). Nims on Unfair Competition and Trade Marks (3d Ed.) 413; that protection is frequently granted against invasion of proprietary secrets through breach of confidence in the absence of express contract, Du Pont de Nemours Powder Co. v. Masland, 244 U.S. 100, 37 S. Ct. 575, 576, 61 L. Ed. 1016 ("The property may be denied, but the confidence cannot be"); and, finally, that Hawthorne's contract is not here in issue. That as between the employer and employee the latter might have complained of the contract is of no avail to the defendant here. Its invalidity has no bearing upon the rights of the plaintiff to its secrets, nor does it justify their invasion by the defendant in violation of the confidence reposed in Hawthorne.

The decree in 6330 holding the patent claims in suit invalid is affirmed. The decree in 6331 awarding an accounting and damages for appropriation of the plaintiff's secret processes is likewise affirmed, with the proviso that if, upon remand to the District Court for further proceedings, it should be found that additional patents have issued on any of the secret processes, the decree may be modified by dissolving the injunction with respect to them. There being but a single record in the two cases, the cost of its printing will be divided equally between the parties.

ALLEN, Circuit Judge (dissenting).

I agree with the decision in the secret process case, but I cannot assent to the conclusion that the claims in suit are invalid. The majority opinion sustains the holding of the District Court that claims 3, 5, 6, 14, and 16 lack patentable novelty. This holding was based upon the conclusion that the only contribution made by this still was an increase in capacity and pressure resistance. Such a contribution, however valuable commercially, does not constitute invention. Powers-Kennedy Contracting Corp. v. Concrete Mixing & Conveying Co., 282 U.S. 175, 185, 51 S. Ct. 95, 75 L. Ed. 278. However, the Smith still does more than that. It solves intricate problems never before solved in the cracking of oil.

Prior to the conception of the Smith still, the vessels which were used for the refining of oils were in general low pressure vessels, either riveted or constructed by hammer welding. There were also vessels forged from a single ingot of steel, which were capable of a higher pressure. All of these vessels were considerably smaller, and hence less useful commercially, than the Smith vessel. The hammer-welded vessels were welded only at the outer and inner edges of the laps which were joined, and were subject to unusual corrosion at the seams. The hammer-welded plates lacked equal tensile strength. The riveted vessels were weakened by the insertion of rivet holes, their efficiency being reduced to 75 per cent. They developed leaks and lacked integrality. The forged vessels which were nearest to the Smith product in performance were limited in size and expensive to manufacture. It was difficult and costly to fit them with the needed manway connections, and, if variations from the purely cylindrical shape were necessary, the cost was very high. The Smith still in great degree solved all of the problems presented by these other types of still construction. Most of all it exhibited a unique integrality.

The Smith still withstands the contortions caused by the extremely high temperature and pressure necessary in the production of gasoline under the oil cracking process. In addition, this still is unique in the industry, for its capacity to endure the stress and strain of unequal expansion and contraction. Owing to the thermal expansion of steel, the diameter of the vessel varies in the oil cracking process with the variation in temperature which is present in different portions of the reaction chamber. In the course of an operation cycle in the Dubbs cracking process there is a wave in the vessel wall which travels down the vessel and then up, causing severe stresses in addition to those created by the extreme internal pressure of the oil cracking operation. The Smith still reacts with oneness as a thing alive, and is able to adapt itself to these unusual stresses.

Thus in a Smith vessel, which was 65 feet in length, 269,000 pounds in weight, with a wall thickness of 3½ inches, and an inside diameter of 7 feet 7½ inches, as the cycle of operation proceeded, the following changes in the shape and form of the vessel were noted: At first, the hot oil, hitting the bottom of the horizontal vessel and covering the lower part of it to about one-third of its volume, heated the lower part of the vessel to a higher temperature than the upper part, the result of which was that the vessel took on a sausagelike shape, that is, it was bent with the lower part larger than the upper part, which curved the ends of the horizontal vessel upwards. The tower which was attached to the end of the vessel therefore leaned in one direction. As the coke deposited and insulated the bottom of the vessel, that part of the vessel became colder than the top and the vessel took a reverse shape. The ends of the vessel bent down, and the tower leaned in the opposite direction. At the maximum deflection, the horizontal vessel was pulled away from its support 1 5/8 inches. In spite of the deflection, this vessel was safe to operate because it reacted as one unit, flexible throughout its entire length and breadth.

It is not shown in this record that any still prior to the Smith still reacted with such integrality under these strenuous conditions.

The master, in his scholarly opinion which held these patents invalid for lack of novelty, states that there had been large welded vessels composed of separate parts fused together long before the idea embodied in this patent was conceived, and cites the steam boiler, the air tank, and other arc-welded articles. The welding of engine parts hardly presents a problem analogous to that solved through the Smith still. The boilers described by Holslag, expert witness for the appellee, were "low pressure boilers," "under a hundred pounds." While the air pressure vessels or air tanks arc-welded were subjected to far higher pressure, none of these air tanks were subjected at the same time to excessive temperature and to marked and sudden shifts in temperature. The Smith stills used in the Dubbs system were compelled to submit to a pressure of from 275 pounds per square inch in the high pressure stills, and to a variation in temperature in the same operation of from 250 degrees F. to 915 degrees F. In the Holmes-Manley system there was a variation of from approximately 900 degrees F. at the bottom of the still to about 420 degrees F. at the top of the bubble tower.

As correctly stated in the majority opinion, the "plaintiff brought forth something new in oil cracking stills." Aside from the fact that this something new has enabled gasoline producers to distill better gasoline more cheaply, because the stills are larger, thicker, and of longer life than other stills used in oil cracking, aside from the fact that the extreme thickness of the walls in these stills and their integrality have greatly reduced the hazards of the process, a new thing was invented, namely, a large vessel composed of sections united in one piece by electric welding and highly adapted through its integrality for the cracking of oils under (1) extreme pressure, (2) extreme temperature, and (3) extreme shifts in temperature in the same cycle of operation.

The elements of combination were old, but the combination and arrangement which attained this end were new. The result was beneficial and had never been attained before. Hence it is evidence of invention. Webster Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L. Ed. 1177.

The elements of a combination may all be old, for in making a combination the inventor has the whole field of mechanics as his source. Diamond Rubber Co. of New York v. Consolidated Rubber Tire Co., 220 U.S. 428, 31 S. Ct. 444, 55 L. Ed. 527; Leeds & Catlin v. Victor Talking Machine Co., 213 U.S. 301, 318, 29 S. Ct. 495, 53 L. Ed. 805; Du Bois v. Kirk, 158 U.S. 58, 15 S. Ct. 729, 39 L. Ed. 895.

The need of the industry for precisely this type of vessel is shown by the fact that, after the Smith still began to be used, it rapidly replaced the prior vessels in the art. 1,088 of them, having an aggregate sale value of $16,000,000, were sold by the appellant prior to the trial. The appellees (the only competitors in the production of these vessels up to the trial) advertised the peculiar qualities existing in the Smith still, and imitated in their own still, as meeting a long-felt demand, saying in their circulars:

"That is the purpose of Fluid-Fusion welded pressure vessels — to meet the demand which has always existed in the Oil Industry for cracking stills and equipment capable of resisting terrific degrees of heat and pressure.

"With Fluid-Fusion welded vessels high pressure cracking is carried to limits unknown before and the Industry is finally rid of the shackles of inadequate equipment."

If a particular result has been long desired and often sought, but never attained, want of invention cannot be predicated of a device or process which first reached that result, on the ground that the simplicity of the means is so marked in view of the state of the prior art, that many believed they could readily have produced it if required. General use is evidence of utility. Gandy v. Main Belting Co., 143 U.S. 587, 594, 12 S. Ct. 598, 36 L. Ed. 272; The Barbed Wire Patent, 143 U.S. 275, 12 S. Ct. 443, 36 L. Ed. 154. The fact that Smith's still was immediately adopted by oil distillers is pregnant evidence of its novelty and utility. Magowan v. New York Belting & Packing Co., 141 U.S. 332, 12 S. Ct. 71, 35 L. Ed. 781.

"Where the question of novelty is in doubt, the fact that the device has gone into general use, and displaced other devices employed for a similar purpose, is sufficient to turn the scale in favor of the invention." Potts & Co. v. Creager, 155 U.S. 597, 609, 15 S. Ct. 194, 199, 39 L. Ed. 275.

In Krementz v. S. Cottle Co., 148 U.S. 556, 13 S. Ct. 719, 37 L. Ed. 558, the inventive concept was that a collar button might be made of a single continuous piece of metal, in which the head was hollow and round in shape. The Supreme Court of the United States in that case reversed a decree which had held the patent on this device invalid for want of novelty.

The case of Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 43 S. Ct. 322, 67 L. Ed. 523, involved a patent which merely changed the pitch of a sieve known as a "wire" from 3 to 12 inches so as to apply the obvious principle that water will run down hill. However, it revolutionized the process of high-speed paper manufacture so that the output of nondefective paper was increased 20 per cent. This called for less inventive genius and certainly the results were less spectacular than in the case at hand. In the Eibel Process Case Chief Justice Taft considered the same questions that have arisen here, namely, novelty of invention, long-felt need, resultant use, retrospective obviousness, vague terms of the patent, failure of the specifications properly to instruct those skilled in the art, equivalents of the same elements in the new combination, presumption of validity of patent, and infringement. All of these questions were resolved in favor of the patentee.

If the application in increased degree of Newton's principle of gravitation to the manufacture of paper producing a new result is invention, how can we say that the application of the old art of are welding to the oil cracking problem producing such a new result as the Smith still is not invention?

Upon the holding of invalidity, the judgment should be reversed.

NOTES

[*] Opinion modified and rehearing denied ___ F.(2d) ___.

[1] 3. A still for use in refining oils, comprising a plurality of thick plates of extreme length and of uniform width curved transversely and assembled edge to edge to constitute a tubular structure, the sections being fused together in the line of their longitudinal meeting edges by an electric are, to form an integral tubular structure of uniform diameter and without circumferential seams in its length.

5. A still for use in refining oils composed of a plurality of transversely curved thick plates of extreme length assembled with their longitudinal edges in register to create a tubular structure having a smooth interior, the said plates being fused at such meeting line by an electric are to form an integral tubular structure without circumferential joints in its length, and heads fused to the ends of the tubular member by an electric are to constitute the whole as an integral structure.

6. In the manufacture of stills of extreme dimensions for use in oil distillation, the method which comprises the steps of forming thick plates of great length into longitudinal trough-like sections, assembling a plurality of such sections in circular order with their edges in register, and fusing such sections in the meeting line of their edges by an electric are to constitute an integral tubular structure without circumferential joints.

14. In the manufacture of stills of extreme dimensions for use in refining oil, the method which comprises the steps of converting metal plates of extreme thickness and great length into longitudinal trough-like sections, assembling a plurality of such sections in circular order with their longitudinal edges in register, and heating and welding the said sections in their meeting lines by an electric current to constitute an integral tubular structure without circumferential joints in the length thereof.

16. In the manufacture of stills of extreme dimensions for use in refining oil, the method which comprises the steps of forming a tubular body of metal of excessive thickness and great length, forming a concaved head of substantially like diameter and thickness of metal, arranging the heads one at each end of the tubular body, heating the parts in their meeting lines by an electrical current, and welding them into an integral structure to complete the closure.

[2] See Equity Rule 61½ (28 USCA § 723), effective September 1, 1932, since trial below. NOTE. — The secret process case having been heard below in camera, and the record having been there sealed, we have refrained from discussing the several processes, and the record here may likewise be sealed at the plaintiff's request.