Cleveland Trust Co. v. Schriber-Schroth Co.

92 F.2d 330 (1937)

CLEVELAND TRUST CO. et al.
v.
SCHRIBER-SCHROTH CO.
SAME
v.
ABERDEEN MOTOR SUPPLY CO.
SAME
v.
F. E. ROWE SALES CO.

Nos. 7223-7225.

Circuit Court of Appeals, Sixth Circuit.

October 8, 1937.

Newton D. Baker and F. O. Richey, both of Cleveland, Ohio (A. C. Denison, W. C. McCoy, Baker, Hostetler, Sidlo & Patterson, Richey & Watts, and Evans & McCoy, all of Cleveland, Ohio, on the brief), for appellants.

J. H. Bruninga and J. H. Sutherland, both of St. Louis, Mo. (Bert M. Kent, of Cleveland, Ohio, on the brief), for appellees.

Before HICKS, SIMONS, and ALLEN, Circuit Judges.

SIMONS, Circuit Judge.

In a study of the patents and prior art involved in Waukesha Motor Co. v. Willys-Overland, Inc. (C.C.A.) 77 F.(2d) 906, and Perfect Circle Co. v. Hastings Mfg. Co. (C.C.A.) 88 F.(2d) 813, we learned something of the principles of operation and the mechanics employed in the conventional four-cycle internal combustion engine of the modern motor car. It will therefore be unnecessary here to repeat what was there said in reference thereto. In the first we considered the explosion chamber, and in the second the piston ring. In the present infringement suit we deal with the piston.

In the rapid reciprocation of a piston within the cylinder, as in the references described, the desirability of maintaining a close fit was obvious and had long been recognized. This was necessary to avoid waste of power through gases being permitted to pass below the piston from the explosion chamber, a condition known as "blow-by," to prevent lubricating oil passing above the piston into the explosion chamber, a condition known as "pumping oil," and to prevent the slapping of the piston in the cylinder, making for a noisy engine and tending to break its parts. On the other hand a too close fit would, because of the expansion of the piston by the great heat generated in the cylinder, cause it to bind or "seize" therein, thus stalling the engine or perhaps even wrecking it.

Cylinders were, and generally still are, cast of iron, and prior to the period we are here considering pistons were likewise so cast. The disadvantages in this organization were early recognized. Since iron is a poor conductor of heat, the particles of carbon that would form on the piston head would become overheated and cause premature explosion in the combustion chamber. Iron is also heavy, and with the great acceleration of piston reciprocation, due to the ever increasing demand for speed, the industry began to cast about for expedients to lighten the pistons. There were numerous efforts and failures in this respect with the conventional iron trunk piston. Lighter metals were considered, but, since strength and durability to withstand the great force of rapid explosions upon the piston head and the powerful thrust imparted to its bearing faces by the power stroke of the piston as it imparted energy to the crank shaft were prerequisites, the art was limited to metal that was both strong and durable. It turned, therefore, quite naturally and inevitably, to aluminum. A new and for a long time seemingly insuperable problem at once confronted the automotive engineers, due to the fact that the coefficient of expansion differs greatly in the two metals, and that under heat aluminum will expand much more rapidly than iron. Since the desirable clearance is minute, approximating one one-thousandth of an inch, it was at once clear that an aluminum piston, installed when cold with required clearance in an iron cylinder, would, when heated by operation, seize, and that a piston with clearance sufficient to avoid seizing would waste power, pump oil, cause undesirable noises, and lack durability.

Ingenious, numerous, and varied were the expedients resorted to in an effort to achieve the clearly recognized advantages of aluminum pistons, while at the same time avoiding troublesome results, and equally numerous were the failures, by reason of which a prejudice had developed in the industry against them, and it was said they had had a "black eye."

The Cleveland Trust Company, appellant, is the repository of numerous patent rights assigned to it by inventors and by piston and automobile manufacturers, and it is its contention that the aluminum piston problem was completely solved by some of its assignors, and that a structure embodying the disclosures of the patent to Jardine, No. 1,763,523, granted June 10, 1930, but in a more specific and improved form described in Maynard, No. 1,655,968, granted January 10, 1928, but on an application later than Jardine's, had gone into wide commercial use, being accepted by the industry, with the exception of some divisions of the General Motors Corporation, had sold in the tens of millions, and is now infringed by pistons of the appellees. The Jardine and Maynard patents belong to the Trust Company. Conceiving, however, that a Gulick application, which was being prosecuted by the Packard Motor Car Company, would, with its broader claims, dominate both Jardine and Maynard, the Trust Company purchased it and prosecuted it to successful conclusion after numerous interferences, the patent issuing as Gulick No. 1,815,733, on July 21, 1931. The Trust Company also owns Schmiedeknecht patent, No. 1,256,265, granted February 12, 1918, and Mooers, No. 1,402,309, granted January 3, 1922, both for pistons, and seven accused devices are all alleged to infringe some of the claims of some or all of the five patents.

The appellees are three Cleveland dealers who have sold alleged infringing pistons manufactured by the Sterling Products Corporation of St. Louis, Mo. While the manufacturer was named in the original bills of complaint, they were dismissed as to it, it being understood, however, that Sterling Products Corporation was conducting the defense of the three suits. By stipulation they were consolidated below, and all are before this court upon the same record. At the outset only two alleged infringing devices were involved, one a Sterling piston known as Exhibit 1, and the other a Ray-Day piston known as Exhibit 8. During the progress of the trial five additional structures were introduced as infringements. While there is no finding to that effect, it seems to have been assumed that all of the alleged infringing devices were sold by each of the three defendants. The consolidated cause was referred to a master, whose very detailed findings of fact and conclusions of law were affirmed by the district judge without discussion and decrees entered dismissing the bills. While the Chrysler Corporation appears as party plaintiff below and as co-appellant here, it has never entered its appearance, and its presence in the case as a nominal plaintiff is due to the fact that it owns a half interest in the Maynard patent and was therefore joined as party plaintiff. An issue raised below as to the propriety of Chrysler being made a nominal party is abandoned here, since the defendants have not appealed.

The controversy is involved and has been bitterly fought. The record is voluminous, the exhibits, physical and otherwise, numerous, the Patent Office history discloses persistent challenge, and of eighty-odd claims fifty-two are said to be infringed. While we have given careful consideration to the briefs and record, it is manifest that the necessities of the case and the desirable limit of an opinion will permit us to discuss only the more salient features of the controversy, with perhaps summary disposition of its lesser issues, for the challenge to the validity of the patents and to the fact of their infringement includes most, if not all, of the defenses known to the patent law.

We deal first with the patent to Gulick, in agreement with the contention that if valid it is basic. Gulick had been connected with various motorcar and motorcycle manufacturing companies in a number of capacities, including that of factory manager. He was also an inventor, having prosecuted some fifty applications in the Patent Office. As early as 1908 he had a conception of a split and flexible piston, but did nothing more for several years than to make sketches. In 1911 he made some experiments with modified standard iron pistons, giving them block tests in a motorcycle and in an Excelsior motor. The motor was later put into a Renault car and given a substantial road test. These early pistons were scrapped. In 1914 he made his first test with aluminum pistons in an Amplex car, which he used personally for some 4,500 to 5,300 miles. In 1917, along with other patents and patent applications, Gulick assigned his inchoate rights to the Packard Motor Car Company, and the application for the piston patent was filed by it on November 30th of that year. There followed long discussions with the examiners, administrative reviews, bitterly contested interferences, and appeals to the Court of Appeals of the District of Columbia (Long v. Gulick, 57 App.D.C. 98, 17 F. (2d) 686), and to the Court of Customs and Patent Appeals (Hartog v. Long, 47 F.(2d) 369). There had also been amendments to the original application at the instance of the inventor's assignee. It was not until July 21, 1931, that the Gulick patent issued with its forty-three claims.

The Gulick patent is for a combination of elements, many, if not indeed all of which, are to be found in the prior art. Nowhere, however, is found his precise organization, and the question at once presents itself as to whether there was exercise of the inventive faculty in the concept of their combination, for that a new result was achieved and a new mode of operation of an aluminum piston in its iron cylinder brought about is not successfully challenged on this record. Gulick separated the smaller ring carrying head of the piston from the somewhat larger skirt by a groove or air gap. This tends to prevent the passage of heat from the head directly to the skirt and permits the bearing faces of the skirt to yield to pressure independently of the head. The wrist pin bosses are pulled inwardly from the skirt periphery, and instead of being supported by the skirt are carried from the piston head by depending flanges, thus providing a vertically rigid support for the bosses with freedom of lateral motion in a direction at right angles to the load thrusts on the piston. The surface of the skirt where the bosses are connected to the head in conventional pistons was cut away or relieved, so that the bosses are not directly constrained by the cylinder wall. They are held in place by webs extending in the general direction of load thrusts on the piston. One bearing face of the skirt is slotted vertically from the bottom edge up to and connecting with the air gap. That Spillman and Mooers patent 1,092,870, and Franquist patent 1,153,902, showed air gaps between piston head and skirt, or that splitting the piston skirt vertically or combining the vertical and longitudinal splits into the so-called T-slot were old practices, is not conclusive on the question of invention.

Since the piston head is the first to heat, and becomes hotter than the skirt, it had always been machined with greater clearance, the control of oil and gas flow being taken care of in part at least by the piston rings, which function as a packing. See Perfect Circle Co. v. Hastings, supra. The problem of maintaining a close fit between an aluminum piston and an iron cylinder without seizing under heat expansion is a problem involving the skirt, which guides the piston and carries the bearings for the wrist pin, upon which is exerted the thrusts of the connecting rod through the cycle of reciprocation. To decelerate the flow of heat from piston head to skirt by an air gap was perhaps an obvious expedient in the art. To slit the skirt vertically so that pressure of the cylinder wall would compensate for thermal expansion was perhaps not beyond expected skill in the art. Minute relief of the cylinder periphery in the region of the bosses was shown in Franquist. But to combine insulation of head from skirt, retraction of the bosses from the skirt periphery, connection of such bosses to the skirt with webs laterally flexible and yet so carried from the head as to support the load upon the wrist pin with sufficient strength and rigidity, and to utilize the mechanical force of the cylinder wall upon the skirt and the thermal expansion of the bosses so as to compensate evenly and fully for head expansion and to secure a balanced flexibility of the skirt with no bending concentration at any point therein, discloses, we think, a meritorious concept beyond the reach of those skilled in the art. Franquist, of course, does disclose the broad concept of a flexible skirt. But Franquist gives no hint at all of the Gulick invention. His flexibility was obtained by grooves on the accordion principle, which, while they interrupted the outer periphery of the piston skirt, did not destroy its inner unity. Some flexibility no doubt it had, but certainly not in the sense that Gulick disclosed it, and it is doubtful that the Franquist piston could have met the severe tests required by the industry. And of commercial history it has none upon this record.

Probably the most difficult task that confronts a patent court lies in determining whether there is invention in a combination whereof many or all co-operating elements are old, and when concepts, however remarkable when disclosed, have, in a rapidly developing art, already become commonplace, not only to the expert, but to the unskilled. It is not easy after the lapse of years to evaluate advances with the "eye and the understanding" of those in the field when they were made. Inventive concept is an abstraction impossible to define, and so courts have sought for simple rules by which its presence may be detected. These have proved helpful, even though never absolute. Long recognition of an existing problem in any art, and the advantages to accrue to an industry from its solution, coupled with repeated failures to find solution, has ever been considered as highly persuasive of invention when success is finally attained. Sometimes achievement is revolutionary, an abandonment of all previous approaches, and the charting of a new way to the goal. More often the inventor begins where others leave off, and perceives the vital forward step to which predecessors had been blind. Even then success may not be fully recognized until the world gives acclaim.

The present record is conclusive of the long unsuccessful search for a light weight piston sufficiently durable to obviate the well-known disadvantages of the iron trunk piston in high speed engines. There was much research and experimentation by those most highly skilled, including the engineers of the Aluminum Company, Colonel Vincent of the Packard Motor Car Company, one of the designers of the Liberty engine, and the engineering staff of the Franklin Automobile Company, under the direction of the defendants' expert Stellman, its chief engineer. Limited success may here and there have been attained, but the problem in its full sweep was not solved until the Gulick concept, carried forward to some extent by Maynard, with possibly some instruction from Jardine, reached commercial embodiment. Then and then only did the industry, by acceptance of the Aluminum Company's piston answering the Maynard drawing and specification, recognize the complete solution of the problem, and the many millions that have since been sold and used have set the seal of commercial approval upon such solution. It is idle to say in an imperfect world that invention should be denied because no piston, not even Maynard, performs perfectly.

It is, of course, axiomatic that commercial success is not of itself conclusive upon the question of novelty or invention, and where such success is fairly attributable to other causes, such as the reputation of the manufacturer, his extensive advertising and superior salesmanship, it will be wanting in persuasiveness. But, where no alternative inferences are to be drawn, commercial success is highly indicative of invention. This is so in the usual case where the test of success is acceptance by the public. But it must be here noted, if not elsewhere observed, that where success follows acceptance by experts rather than by those unskilled, it is of far greater tribute to the ingenuity of the inventor. It is clearly demonstrated upon this record, if indeed it is not within the field of general knowledge, that the automotive industry, in acceptance of new devices upon which success of highly complicated and delicate machinery in the hands of inexpert users must depend, exercises the greatest caution, and things new, whatever their theoretical virtue, will not be considered until they have been subjected to the most rigorous practical tests for efficiency and durability. This must be so when failure of one year's product may bring tremendous losses or complete disaster.

So also with respect to the presumption that attaches to the validity of a grant by the Patent Office. Where, as in this case, substantially all pertinent prior art has been cited against the patent, where, in hard fought interferences, novelty and invention have been challenged, and where priority of conception has been finally adjudicated only upon repeated review in both administrative and judicial forums, the patent should not be stricken down except upon very clear and convincing proof of invalidity. The weight to be given this presumption has been so recently and so thoroughly discussed, and the rule stated in all the varying formulae by which courts have announced and affirmed it, in Radio Corporation v. Radio Laboratories, 293 U.S. 1, 55 S. Ct. 928, 79 L. Ed. 163, that it is unnecessary here to repeat or to cite additional cases.

In what we have said thus far it is perhaps clear that in respect to novelty and invention over recorded prior art we have concluded the Gulick patent to be valid. But this is not the whole controversy with respect to it. It is the contention of the defendant that the Gulick patent is invalid because abandoned by him prior to his application, and because it had been in public use in this country for more than two years before such application within the purview of Rev.St. § 4886, as amended in 1897 and § 4920 (35 U.S. C.A. §§ 31, 69). The master so held. His findings were based on the fact that Gulick's 1911 and 1914 pistons had been scrapped, and on inference that Gulick's expressed satisfaction with his 1914 experimentation was conclusive of the completion of his invention in that year. That there had been no formal abandonment or disclaimer is clear. The master's findings are based upon constructive abandonment deduced from the above circumstances and delay in applying for a patent.

Questions relevant to actual or to constructive abandonment of inventions are questions of fact, Walker (6th Ed.) § 152; Kendall v. Winsor, 21 How. 322, 330, 16 L. Ed. 165, and much weight must therefore be given to the findings of the master, who saw and heard the witnesses. Consideration should also be given to the rule that concurrent findings of master and judge should not be set aside except for clear error, although this may prove to be an unsafe guide to just decision where exceptions to a master's report are unilluminated by oral argument and a decree entered without elucidation of the reasons upon which it is based. The law, however, in reference to abandonment, requires that every reasonable doubt relevant to any such question should be resolved in favor of the patent, for it does not favor forfeiture. Walker, § 152, and cases there cited. It was said by this court, Gear Grinding Machine Co. v. Studebaker Corp. (C.C.A.) 270 F. 934, 936: "Abandonment depends upon intent, actual or imputed. The actual intent did not exist, and the circumstances do not require that the intent be by law imputed, as against the truth." That Gulick did not intend to abandon the subject matter of his invention seems to us clear. From the beginning of his experimentation in 1911 he made drawings from time to time, tests with slotted iron pistons, and later with slotted aluminum pistons, and was at the same time not only busy in the earning of a livelihood but with plans for the purchase from a receiver of the property of the Amplex Motor Car Company. The scrapping of his experimental pistons was without his knowledge or consent, and the fact that he replaced the experimental pistons of the Renault car with standard pistons, before its sale negatives any inference that the experimental pistons were intended to be abandoned. Distinction must also be made between abandonment of the subject matter of a patent and abandonment of crude experimental devices. While Gulick variously expressed satisfaction with the performance of his experimental pistons, it is clear upon review of his entire evidence that what he had then in mind was that they performed satisfactorily in reference to clearance. They had not demonstrated their strength and durability, or that they were properly designed or machined for commercial use.

The master was largely influenced by the fact that Gulick's clarification of his testimony followed a recess, and gave little credence to it. We have reviewed it carefully. That Gulick's experiments were not completed to the end that a practical piston had been fully designed and tested seems clear. He is in some respects corroborated and in no wise impeached. Mere suspicion as to his good faith is not such clear and convincing proof of abandonment as will satisfy every reasonable doubt and sustain the heavy burden which rests upon those challenging the validity of his patent. The master did not apply to the evidence on the issue of abandonment that standard of proof required to establish it and so the question is one of law, and the finding must be rejected.

There was no public use of the Gulick invention more than two years prior to the application so as to invalidate the patent. So long as use by the inventor can fairly be considered experimental and collateral to the development of the invention in its complete form, the bar of the statute does not begin to run. Reo Motor Car Co. v. Gear Grinding Mach. Co., 42 F.(2d) 965, 968 (C.C.A. 6). The emphasis that is laid upon Gulick's use of a car with experimental pistons for business and pleasure for a considerable period does not impress us. Road tests of many thousands of miles are considered necessary to determine the efficiency of an engine and its parts. That such transportation is not wasted does not of itself constitute a public use. As was said in the second Gear Grinding Case: "We see no principle, in the precedents or in the policy of the statute which would require a valuable invention to be lost by permitting the bar to begin to run while the invention * * * is being gradually improved and developed up to the point of ascertaining whether it has real utility. Nor can it rightly be said that a public use of the developing and incomplete invention — incomplete as finally claimed — can nevertheless operate as a public use bar because adding the final element of perfection did not involve invention as compared with the earlier form to which it was added."

We are not convinced that the Gulick application was broadened to include new matter after intervening rights had arisen. This issue was adjudicated in the Court of Appeals of the District of Columbia, Long v. Gulick, 57 App.D.C. 98, 17 F.(2d) 686, and the Court of Customs and Patent Appeals, Hartog v. Long, 47 F. (2d) 365, 366, 367. With the conclusions there reached we agree. As to the Monckmeier and Schoengarth prior uses, it is sufficient to say, without analysis of the voluminous evidence, that they rest mainly upon the uncertain memory of witnesses testifying long after the event, and this does not rise to the dignity of proof reguisite to establish prior use. Barbed Wire Patent Case, 143 U.S. 275, 12 S. Ct. 443, 36 L. Ed. 154; Eibel Process Case (Eibel Process Co. v. Minnesota & Ontario Paper Co.), 261 U.S. 45, 43 S. Ct. 322, 67 L. Ed. 523.

Jardine, upon an application filed March 11, 1920, claimed an invention of which the chief object was construction of a light alloy piston which would operate satisfactorily at all temperatures and lend itself to economic methods of manufacture. He cut away all the skirt portions at the webs, extending the latter to nearly the length of the skirt parts or slippers. Horizontal slots partially insulated the slippers from the head and these were split vertically to render them yieldable to contact with the cylinder wall. The head was integrally connected to the skirt only in the region above the bosses. His thought was that since heat expansion tends to increase the diameter of the piston in a direction parallel with the wrist pin bosses, when the skirt around the bosses is removed, the central part of the web is free to move while its ends are restrained and resisted by the skirt faces in contact with the cylinder wall. There is insistent challenge and as insistent defense of the validity of Jardine. On the one hand it is urged that Jardine is but the slipper structure of Ricardo patent, No. 1,294,833, and that it involved no invention to merely slit vertically its skirt parts as disclosed by Gulick and others. The response is that Jardine discloses a combination of inter-related operating parts that goes far beyond Ricardo, is economical to construct, and carries the Gulick concept an important step forward. Jardine indicates the forces thermal and mechanical, which bear upon the flexible web to compensate for heat expansion of the cylinder head, pointing to the web as being in effect a cantilever structure, weakest at its supports. Jardine came to the attention of the Aluminum Company and the Maxwell at a time when Ricardo and other English engineers were trying to modify the Ricardo piston to meet the piston problem. It was thought to give great promise. Tests had been made and commercial manufacture planned when the Maynard piston was presented to them. It appeared so much better that the construction illustrated and described by him was adopted. Since it is the Maynard specific form embodying the basic concept of Gulick that made commercial piston history, since Jardine was without substantial commercial recognition and none of the accused devices are slipper pistons, since whatever Jardine contributed to the art, if anything, is said to be incorporated in Maynard, and since the really important infringement is said to be the copying by the defendants of the commercial embodiment of Maynard, we feel under no necessity at this time of ruling upon the validity of the Jardine patent or the fact of its infringement.

Maynard, upon a patent application filed January 3, 1931, embodies the Gulick combination of skirt insulation, skirt flexibility by means of vertical slotting cooperating with longitudinal slotting, and flexible webs in the region of the wrist pin bosses. He also follows Jardine's simplified design to permit economical manufacture and Jardine's boss relief. The Maynard piston, however, is not a slipper piston. While Jardine carried the webs nearly to the bottom of the skirt and omitted all of the skirt on the web faces, Maynard provided webs about two-thirds of the length of the skirt and continued the latter circumferentially below the webs except for a single vertical slot throughout one bearing face of the skirt. This structure increased the effect of heat on retarding the expansion of the piston so as to permit of a narrowed web. The upper part of the skirt, which reaches higher temperature, is provided with greater compensation against heat expansion by having its periphery interrupted by relief about the bosses. The lower part of the skirt requires no interruption of its periphery other than the vertical slot because it receives less heat. This results in a better bearing surface and means additional to the piston rings for closing the passage between the piston and the cylinder wall. It is clear that Maynard, while not departing from the teaching of Gulick in basic combination of elements, discloses a piston lighter and more economical of manufacture than Gulick and one more rugged and durable than Jardine. That the Maynard piston had a very great effect upon the industry and substantially advanced the art, there can be no doubt upon this record. In the precise form disclosed by Maynard and depicted in his drawings, many thousands were made and successfully used. In a somewhat modified form its production and use has run into very many millions.

Little is claimed for Mooers and Schmiedeknecht, and, while some of the claims of each may possibly read on some of the accused structures, they have apparently no commercial history and have not substantially advanced the art. We do not pass upon them, and apprehend that the real controversy here is in respect to validity and infringement of Gulick and Maynard.

Of the devices charged to infringe, the one of prime importance to the plaintiff is undoubtedly the Sterling piston, exemplified by Exhibit 1. The defendants seek to avoid infringement by this structure, which, with slight modification, follows the drawings and description of the Maynard patent, and without change, the commercial structure of the plaintiff's licensee, by the contention that its webs are rigid at the top and thus it is distinguished from Maynard, compensation for rigidity being achieved by reducing the thrust faces of the piston skirt at an angle to the vertical sides of the web, giving to the relief in the region of the bosses the so-called "keystone" configuration. But there are no limitations in the claims in suit as to the region of flexibility in the webs, and, if the keystone relief provides greater flexibility at the union of the webs with the skirt, it is but a difference in degree, and not one of mode of operation or result and will not avoid infringement. In comparing commercial embodiments of invention with the drawings of patents it is not unusual to find that, without departing from the teachings of the inventor, improved operation may result from some modification of the disclosure. This court has indicated, Gordon Form Lathe Co. v. Walcott Machine Co., 32 F.(2d) 55, 61, that superiority of operation, even when brought about by a valuable improvement which perhaps was patentable as such, will not of necessity negative infringement, citing Schiebel Toy, etc., Co. v. Clark (C.C.A.) 217 F. 760, 771, Toledo Machine & Tool Co. v. E. W. Bliss Co., (C.C.A.) 287 F. 443, 447. The Gordon and Redlin patent there considered was in a true sense a pioneer patent and entitled to broad equivalents, but this does not mean that meritorious inventions which substantially advance the art are entitled to no reasonable range of equivalency. As was said in Eibel Process Case, supra, at page 63 of 261 U.S., 43 S. Ct. 322, 328, 67 L. Ed. 523: "Eibel made a very useful discovery, which has substantially advanced the art. His was not a pioneer patent, creating a new art; but a patent which is only an improvement on an old machine may be very meritorious, and entitled to liberal treatment. Indeed, when one notes the crude working of machines of famous pioneer inventions and discoveries, and compares them with the modern machines and processes exemplifying the principle of the pioneer discovery, one hesitates in the division of credit between the original inventor and the improvers, and certainly finds no reason to withhold from the really meritorious improver, the application of the rule `Ut res magis valeat quam pereat' which has been sustained in so many cases in this court." A careful consideration of the patents to Gulick and Maynard, an analysis of the claims of each in suit, and a review of the evidence as to principles and modes of operation, lead us inevitably to the conclusion that the Sterling piston (Exhibit 1) infringes the claims in suit of both, and that such minor differences as appears in the commercial embodiment of Maynard, with which Exhibit 1 is identical, do not fall without the reasonable range of equivalents to which the inventors' contribution to the advancement of the art entitle them.

In the remaining accused devices we find no infringement. True, there is similarity. Here and there are identical elements, singly and in combination, but the complete combination, upon the conception of which we found Gulick and Maynard to denote invention does not in any others appear, and so it must be said here as in Gear Grinding Machine Co. v. Studebaker Corporation, supra, in reference to such situation: "It is the patentees' necessary dependence upon the co-operation of all the elements in one machine in order to make the claims valid that also makes this achievement possible."

We find claims 1, 11, 12, 13, 15, 18, 30, 33 of Gulick and Maynard patent, in respect to all of the claims sued upon, valid. We find such claims to be infringed by the Sterling piston exemplified by Exhibit 1. The decrees below are set aside, and the causes remanded for the entry of decrees granting the usual relief of injunction and accounting in respect to such infringement as has been found, and dismissing the bills with respect to the alleged infringements of Jardine, Mooers, and Schmiedeknecht, without prejudice. Costs of appeal will be borne by the defendants.