HOELTKE
v.
C. M. KEMP MFG. CO.

*913 Frank Keiper, of Rochester, N. Y. (Robert R. Carman, of Baltimore, Md., on the brief), for appellant.

Arlon V. Cushman, of Washington, D. C., and Lewis W. Lake, of Baltimore, Md. (Cushman, Darby & Cushman, William M. Cushman, and C. Willard Hayes, all of Washington, D. C., on the brief), for appellee.

Before PARKER and SOPER, Circuit Judges, and HAYES, District Judge.

PARKER, Circuit Judge.

This was a suit instituted to enjoin infringement of patent No. 1,822,402 issued September 8, 1931, to William F. Hoeltke on an application filed April 16, 1929. Hoeltke, the patentee, was complainant. The defendant was the C. M. Kemp Manufacturing Company, of Baltimore, a corporation engaged among other things in the manufacture and sale of apparatus for mixing and distributing a mixture of air and gas for industrial purposes in factories. The complaint, in addition to asking damages for infringement of the patent and injunction against further infringement, prayed for an accounting by defendant of profits made on the sale of infringing apparatus prior to the grant of the patent, on the ground that defendant had obtained from complainant a disclosure as to the nature of the patent, and, in fraudulent breach of the confidence so reposed, had made and sold apparatus embodying complainant's invention and attempted to obtain a patent thereon. Defendant denied both the infringement and the fraud alleged, and pleaded in addition that the patent was anticipated in the prior art and was lacking in invention. The court below decided the issues so presented in favor of the defendant, and complainant has appealed.

The patent in suit covers an automatic fire check for use in connection with gas burners in such way as to prevent the ignition of the gas and air mixture in the pipe lines with consequent explosion and *914 wrecking of the apparatus. Prior to complainant's invention, the only fire check in use consisted of a nipple, which contained a perforated refractory metal disk, and behind this a screen of copper wire gauze, and which was inserted between the end of the gas pipe line and the burner. This disk and screen permitted the mixture of gas and air to flow to the burner, but in case of backfire, arrested the flame and prevented its igniting the mixture of gas and air in the pipes for a considerable period. The nipple was painted with linseed oil and lamp black; and when a backfire occurred, the heating of the nipple caused these to give off an odor which warned the person in charge of the apparatus of the approaching danger, so that he could shut off the gas by means of a hand valve located behind the nipple. This fire check with the manually operated valve was satisfactory and sufficient if workmen were near when the backfire occurred; but, if no one was near enough to be warned by the odor, the flame burned through the disk and screen and ignited the gas and air mixture in the pipes, resulting in costly and dangerous explosions. The only other device in use for preventing backfires was the large "backfire preventer" of the Kemp patent, which we shall describe later. It was located near the mixing machinery, and, while efficient in preventing the wrecking of the machinery by backfires, it did not prevent costly explosions in the pipe lines in cases where the backfire was allowed to burn through the fire check.

Complainant was a mechanic or millwright employed by the American Can Company in its plant in Brooklyn, N. Y., in which apparatus manufactured by defendant had been installed. An explosion resulting from a backfire occurred in that plant on April 24, 1928, which caused considerable damage and resulted in the shutting down of the plant for the day. Complainant made inquiry of an employee of defendant (through whom defendant later made application for patent covering complainant's invention) as to why defendant did not provide some device to prevent such occurrences, and was told by this employee that it was not possible to provide such device, and that if this had been possible, defendant would have provided it. Complainant thereupon went to work to evolve such a device and after a number of experiments invented an automatic valve to be used in connection with the fire check described. It consisted of a spring valve held open by a lever, which, in turn, was held in tension by a pin placed in front of the refractory disk of the old fire check. This pin was made in two parts and was fastened together by solder, or other fusible metal, so that a backfire would melt the fusible metal and cause the pin to part. This would release the lever, and the spring would thereupon close the valve. The figures and the pertinent part of the specification contained in the patent issued to complainant are as follows:

"It sometimes happens that the flame of the burner strikes back through the burner into the line, and the flame will then burn at the refractory disk 15 and after a while this disk and nipple and the reducer will get hot enough so that the flame will strike back into the copper wire mat and will then melt out the copper and may even travel further back in the line and cause an explosion. To prevent this I perforate the reducer 10 at the top and bottom and pass through it a pin or rod 20, which pin is shown in Figure 3. This pin makes a close sliding fit in the reducer and is made with an enlarged head 21 at one end and is threaded at the lower end 22. The upper end 21 is forked and a screw passes through the forked end 21 and through the slot 6 in the end of the lever 5 and connects the lever and the pin together. The *915 pin 20 is made of two parts which overlap each other and form a joint as indicated at 23 and the parts are soldered together at this place with a solder that has a rather low melting point. When the pin is in the position shown in Figure 1, a nut is placed thereon and screwed up to make contact with the reducer and hold the pin 20 so that the soldered joint 23 stands substantially centrally of the reducer 10. When the flame strikes back in the reducer, the solder quickly melts long before the refractory disk 15 can heat up and opens the joint in the pin 20. The spring in the valve housing 1 then expands pushing up the pin 2 and the lever 5 which is released by the melting of the joint 23 in the pin 20 and the valve then closes thus shutting off the flow of gas upon which the flame at once goes out and the line cools off. This prevents any danger from fire or explosion, and also prevents the waste of gas.

"It will be understood that the brass rod 20 can be replaced by any metal having a low melting point, or a wood or vegetable fiber can be used that will quickly burn through.

"It will be seen that if a backfire occurs the flame will pass through the nipple 11 into the reducer 10, where for the time being the passage of the flame is arrested by the refractory element 15 at which point the flame will burn without passing through the element until the element has heated up. This takes some time, and during this time the flame is heating the soldered joint 23. This joint quickly reaches the melting point of the solder, and then the joint lets go and permits the valve to close. If the refractory element 15 were omitted the copper mat by itself would hold the flame in position to play on the soldered joint until the copper mat had heated up sufficiently so as to no longer cool the gas below the burning point. Thereafter the flame would gradually work its way through the copper mat, taking some minutes in so doing, but long before this could happen the soldered joint 23 would let go and permit the valve to close, thus cutting off the supply of gas and causing the flame to go out."

The claims of the patent relied on in this suit are claims 1, 6, 8, 14 and 16. Claims 1 and 14 which may be taken as typical, are as follows:

"1. A safety gas cock for a gas line comprising a valve, a spring adapted to hold the valve closed, a lever adapted to normally hold the valve open against the spring, a pin adapted to hold said lever in position to hold said valve open, said pin extending into the gas line and being adapted to be separated intermediate its ends by a rise in temperature in the gas line, thereby releasing the lever and permitting the spring to close the valve."

"14. A safety gas cock for a gas line comprising a valve and valve housing, a spring adapted to hold the valve closed, a lever pivoted on the housing, a pin extending into the housing between the valve and the lever, said lever being adapted to hold the valve open through said pin against the spring, a rod adapted to hold said lever in position to hold said valve open, the portion of said gas line between the rod and the valve having a flame check therein that will permit the gas to pass through it freely and will retard the passage of any flame that may get into the gas line, the lower end of said rod being located at the discharge end of said flame check and where it will be quickly heated by any flame burning thereon and being held in place in the gas line at normal temperatures by fusible material, which material will melt with a rise in temperature and release the rod and lever and permit the spring to close the valve before a flame can pass through the gas line from the rod to the valve."

After complainant had perfected his device, he wrote defendant under date of December 4, 1928, calling attention to the explosion which had occurred at the Brooklyn plant of the American Can Company, stating that he had perfected a device that would overcome that trouble which he had tried out, but had not shown to any one, and asking whether he could interest defendant in this safety device. Defendant replied under date of December 10th that it was interested in complainant's invention and would send a representative to confer with him. In February, 1929, a representative of defendant did call to see complainant, and, without looking at the device or permitting complainant to describe it to him, advised complainant to apply for a patent, and when papers were filed to send them to defendant. On April 16, 1929, complainant filed his application for patent and on April 27th sent a copy of his drawing and specification to defendant. On May 15th, defendant returned these with the following letter:

*916 "Your letter of the 27th April duly received. The writer has deferred reply until he could give the matter some study and thought.

"He is rather disappointed in your patent application, as your earlier letter caused him to think you had developed an improvement in our back fire preventer, while we find the device disclosed in your patent application is a fire check.

"It seems that you have incorporated no new feature in this fire check, but, on the contrary it comprises all the features of our back fire preventer which is now in common use on our gas apparatus with which you are familiar and which is described in our Patent No. 688,863 to the late C. M. Kemp, dated December 17, 1901.

"The features common to both this back fire preventer and the fire check which you describe are

"A valve at the inlet side

"A fuse at the discharge side

"A screen positioned between the fuse and the valve. The fuse holds the valve open against the tension of a spring.

"The melting of this fuse by flame burning back in the device permits the spring to close the valve.

"Your patent also describes and shows the screen arrangement of our present fire check, namely, a refractory perforated disc in the end toward the burner and a rolled gauze disc back of same.

"The writer cannot see that there is any invention in your device and certainly no patent claims of any consequence can be procured in view of the present art and particularly the patent to C. M. Kemp No. 688,863.

"The writer also considers the fire check inoperative. There is one very objectionable feature which will prevent its acceptance by the insurance authorities and prevents it being dependable.

"Our Mr. Dunkak will be in New York within the next few days and we have asked him to call on you and discuss this matter personally.

"We return herewith your patent, drawing and copy of specifications.

"We appreciate and thank you for submitting the device to us and remain

Notwithstanding this letter, defendant sent a representative to call on complainant to negotiate with him for the purchase of his patent; but nothing came of these negotiations. Defendant claims, and has introduced evidence to show, that it had been working on an automatic fire check and had developed same at the time of the receipt of the letter from complainant; but this fact was not mentioned by its representatives in conversations with complainant and was not referred to in the letter which we have quoted. Within a few months after receiving copies of complainant's drawings and specification, however, it began the manufacture and sale of an automatic fire check which it claims to have perfected in its plant before seeing them, but which was substantially the fire check of the patent except that the soldered pin which holds the lever in tension in complainant's patent was replaced by a manifest equivalent, a wire fuse in a sleeve, with a soldered joint directly in the flame of the backfire. On August 23, 1929, without any notice whatever to complainant, it caused two of its employees to file application for a patent on this device, with assignment of the rights in the patent to be obtained thereunder to defendant. Not until an interference was declared in the Patent Office on October 16, 1930, between the patent application of complainant and that of the employees of defendant, did complainant learn of defendant's effort to obtain a patent on the device which was in substance the device of his invention. This was nearly eighteen months after complainant had disclosed his drawings and specification to defendant, and nearly fourteen months after defendant's employees had filed application for patent for defendant's benefit. After the interference was declared, the attorneys for defendant made two efforts to purchase complainant's rights in his invention, offering him first $3,500 and finally $5,000 therefor; but these attempts to purchase same came to naught, as complainant was demanding a price far in excess of these figures.

After defendant's attorney had failed in these efforts to purchase complainant's rights, its attorneys, on December 20, 1930, made a motion to dissolve the interference in the Patent Office on the ground that the subject matter thereof was not patentable, citing eight prior patents including the Kemp patent No. 688,863 heretofore mentioned, to which we shall refer more at length hereafter. This motion was denied and the invention held patentable on *917 March 28, 1931. Thereupon, attorneys for defendant took no further action, but allowed judgment to be entered in the interference proceedings; and patent was duly issued to plaintiff on September 28, 1931. Defendant's contention is that it took no further interest in the proceedings after its motion to dissolve the interference was denied because it considered the patent worthless, and also because it had found that the automatic fire check which it had been making was subject to grave objections and had decided to make a different type of fire check. It appears, however, for reasons which we shall go into more fully hereafter, that this latter type of fire check, while different in appearance, embodies the substance of complainant's invention with a mere rearrangement of parts.

Three questions are presented by the appeal: (1) Whether complainant's patent is valid; (2) whether it is infringed by the automatic fire checks manufactured and sold by defendant; and (3) whether defendant should be required to account for profits and damages with respect to infringing devices sold prior to the grant of complainant's patent because of the confidential relationship existing between the parties resulting from complainant's disclosure to defendant of his specification and drawings. We think that all of these questions should be answered in the affirmative.

Defendant has cited 33 patents as basis for its contention that complainant's invention is lacking in novelty; and this in itself is some evidence of the weakness of the contention. Such a citation of so many prior patents almost inevitably means either that none of them is nearly like the invention of the patentee and that the attempt is being made to invalidate the patent because the patentee has brought together for the purposes of his invention devices to be found in prior patents of different character, or that prior attempts to solve the problem with which he was confronted have not met with success. Scott v. Fisher Knitting Mach. Co. (C. C. A. 2d) 145 F. 915, 916; Mahony v. Malcom (C. C. A. 7th) 143 F. 124, 125; Forsyth v. Garlock (C. C. A. 1st) 142 F. 461, 463; Mallinckrodt Chemical Works v. E. R. Squibb & Sons (D. C. Mo.) 6 F. Supp. 173, 175; York Ice Machinery Corporation v. L. & K. Ice Corporation (D. C. N. Y.) 6 F. Supp. 544, 546; Handy v. American Flyer Mfg. Co. (D. C. N. Y.) 44 F.(2d) 633, 635; Gillette Safety Razor Co. v. Clark Blade & R. Co. (C. C. N. J.) 187 F. 149, 152. We have carefully examined all of the patents cited, however, and none of them, in our opinion, anticipates the invention of complainant. While each of the devices which complainant uses is to be found in one or more of the patents cited, none of these patents covers any such combination as the automatic safety fire check which complainant has evolved. The thing nearest to it in the prior art is the backfire preventer of Kemp patent No. 688,863, to which we have referred; but this is far from an anticipation of complainant's device. It covers a large tubular structure 18 to 20 inches in diameter and about the same height, designed to prevent backfire from reaching the mixing machinery and utterly unsuited for use as an automatic fire check near the burner. While it contains a valve closed by a spring, a refractory screen and a fusible member for holding the valve open, neither the arrangement nor the use of these suggests the device which complainant has evolved for automatically shutting off the gas at the burner in case of backfire. That what we have in mind may more readily appear, we give here the drawing of the Kemp patent and the explanation of the working of the device as contained in defendant's brief.

"The old Kemp backfire preventer (expired Patent No. 688,863), as stated above, is adapted to cut off flow of the combustible gas-air mixture in the event of a flash *918 back or backfire in the line. It discloses a valve 5 adjacent the inlet 4; a flame arresting means 12 below the valve; a fuse 9 adjacent the outlet 19 which projects into the line and is removably mounted, and is operatively connected by rod 6 extending from the fuse to the valve for the purpose of release on backfire in the line. In the event of a flash back traveling rearwardly (from left to right, Fig. 1 (Kemp patent) in the line, the flame will be directed downwardly toward the fuse 9 and will be arrested in its rearward travel by the flame arresting means (screen) 12 in the casing. The fuse 9 between the burner and screen or stop 12, will be melted by the heat of the backfire, releasing the valve 5, which will immediately be closed by the spring 8. Thus, the flow of gas will be cut off on the side of the flame arresting means remote from the flame so that the flame will be promptly extinguished and cannot go back into the line.

"In addition to the fusible support 9, which serves to permit the valve to close in the event of a flame burning in the casing and to furnish a further safeguard, the said backfire preventer of the old Kemp patent is provided with pressure responsive means to release the valve if an explosion occurs in the line and the pressure is thereby raised. Such pressure responsive means takes the form of a weak bottom wall 11 which consists of a thin metallic diaphragm, and in response to excessive pressure said diaphragm 11 is forced downwardly and the support for the valve operating rod 6 is thereby removed and the valve closes."

The evidence clearly shows that the device of the Kemp patent could not be relied on to check the backfire as a result of the melting of the fusible bridge. On the contrary, it was not unusual for the bottom of the device to be blown out; and it was constructed as indicated above so that this would happen if the valve should not close as the result of the melting of the metal of the bridge. In complainant's device, the thing to be melted is a small fuse, not a metal bridge strong enough to support the pressure of the spring bearing directly upon it; and this fuse is placed directly in front of the refractory disk, where it will get the full benefit of the flame, not eighteen inches away as in the Kemp patent. Even if complainant's device be regarded as a mere change in the device of the old Kemp patent, it constituted a valuable improvement in the art and was clearly patentable under the rule laid down by the Supreme Court in Seymour v. Osborne, 11 Wall. 516, 548, 20 L. Ed. 33, where the court speaking through Mr. Justice Clifford, said: "Particular changes may be made in the construction and operation of an old machine so as to adapt it to a new and valuable use not known before, and to which the old machine had not been, and could not be, applied without those changes, and, under those circumstances, if the machine, as changed and modified, produces a new and useful result, it may be patented, and the patent will be upheld under existing laws. Such a change in an old machine may consist merely of a new and useful combination of the several parts of which the old machine is composed, or it may consist of a material alteration or modification of one or more of the several devices which entered into its construction, and whether it be the one or the other, if the change of construction and operation actually adapts the machine to a new and valuable use not known before, and it actually produces a new and useful result, then a patent may be granted for the same, and it will be upheld as a patentable improvement."

In Webster Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L. Ed. 1177, the patent in suit dealt with old elements in a way to greatly improve the efficiency of an old machine, and the rule applicable in such cases was well stated by Mr. Justice Bradley as follows: "It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention. It was certainly a new and useful result to make a loom produce fifty yards a day when it never before had produced more than forty; and we think that the combination of elements by which this was effected, even if those elements were separately known before, was invention sufficient to form the basis of a patent."

And the rule is thus stated by the late Judge Sanborn in Luten v. Kansas City Bridge Co. (C. C. A. 8th) 285 F. 840, 845: "A new combination of old elements, whereby an old result is obtained in a more facile, economical, and efficient way, or whereby a new and useful result is secured, may be protected by patent as securely as a new machine or a new composition *919 of matter. [Citing cases.] A new combination of old elements, in which, by a different location of one or more of the elements, a new and useful result is attained, or an old result is produced in a better way, is patentable. [Citing cases.] Some of the combinations claimed and patented in this case appear from the present record to have been new combinations of old mechanical elements, whereby the desired result was obtained in an easier, more economical, more efficient, and less dangerous way."

And in Star Brass Works v. General Electric Co. (C. C. A. 6th) 111 F. 398, 400, the rule is thus stated by Judge (later Mr. Justice) Day: "It is too well settled to need extended citation of authorities to support the proposition that a new combination of elements old in themselves, producing a new and useful result, entitles the inventor to the protection of a patent. Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L. Ed. 1177. It may be true that Anderson has only taken the familiar contact spring or brush, and placed it in a protected position, but this change seems to have made the difference between a defective mechanism and a practical method of attaining the desired end."

And for other statements and applications of the rule see Eibel Process Co. v. Minnesota, etc., Co., 261 U.S. 45, 43 S. Ct. 322, 67 L. Ed. 523; Richmond Screw Anchor Co. v. U. S., 275 U.S. 331, 48 S. Ct. 194, 72 L. Ed. 303; Black & Decker Mfg. Co. v. Baltimore Truck Tire Service Corp. (C. C. A. 4th) 40 F.(2d) 910; In re Kamrath (Cust. & Pat. App.) 67 F.(2d) 928, 930; Star Can Opener Co. v. Owen Dyneto Co. (C. C. A. 2d) 16 F.(2d) 353; Mead-Morrison Mfg. Co. v. Exeter Mach. Works (C. C. A. 3d) 225 F. 489; Toledo Computing Scale Co. v. Computing Scale Co. (C. C. A. 7th) 208 F. 410; Dowagiac Mfg. Co. v. Superior Drill Co. (C. C. A. 6th) 115 F. 886; Edison Electric Light Co. v. U. S. Electric Lighting Co. (C. C. A. 2d) 52 F. 300.

It is urged, however, that nothing more than the skill of a mechanic was shown in adopting from old devices the device of the patent. In this we cannot concur. Defendant seems to have been working without success for a number of years in an attempt to evolve a device of this character; and we are inclined to think that the problem seems simple now only because complainant's device has taught us how it may be solved. As said by Mr. Justice McKenna in the oft-quoted passage from Diamond Rubber Co. v. Consol. Tire Co., 220 U.S. 428, 435, 31 S. Ct. 444, 447, 55 L. Ed. 527: "Knowledge after the event is always easy, and problems once solved present no difficulties, indeed, may be represented as never having had any, and expert witnesses may be brought forward to show that the new thing which seemed to have eluded the search of the world was always ready at hand and easy to be seen by a merely skillful attention. But the law has other tests of the invention than subtle conjectures of what might have been seen and yet was not. It regards a change as evidence of novelty, the acceptance and utility of change as a further evidence, even as demonstration."

We think there can be no doubt as to the patentability of complainant's device; but, if there were doubt, there can be no question but that this doubt should be resolved in favor of the validity of the patent. The ordinary presumption of novelty arising from the grant of the patent is greatly strengthened because of the contest in the Patent Office. Hildreth v. Mastoras, 257 U.S. 27, 32, 42 S. Ct. 20, 66 L. Ed. 112; Gairing Tool Co. v. Eclipse Interchangeable Counterbore Co. (C. C. A. 6th) 48 F.(2d) 73; Smokador Mfg. Co. v. Tubular Products Co. (C. C. A. 2d) 31 F. (2d) 255; Ensign Carburetor Co. v. Zenith-Detroit Corporation (C. C. A. 2d) 36 F. (2d) 684, 686. The invention filled a want in the industry and entered into immediate use when placed on the market by the defendant. Temco Co. v. Apco Co., 275 U.S. 319, 48 S. Ct. 170, 72 L. Ed. 298; Diamond Rubber Co. v. Consolidated Rubber Tire Co., supra; Pangborn Corporation v. W. W. Sly Mfg. Co. (C. C. A. 4th) 284 F. 217. And in addition to this we have the presumption arising from the imitation of the patented article by the manufacturers of the alleged infringing device. As to this, we agree with what was said by Judge Hough, speaking for the Circuit Court of Appeals of the Second Circuit in Kurtz v. Belle Hat Lining Co., 280 F. 277, 281: "The imitation of a thing patented by a defendant, who denies invention, has often been regarded, perhaps especially in this circuit, as conclusive evidence of what the defendant thinks of the patent, and persuasive of what the rest of the world ought to think." See, also, Black & Decker Mfg. Co. v. Baltimore Truck Tire Service Corp., supra, 40 F.(2d) 910, at page 914.

*920 And on the question of infringement, we have no doubt, either as to the first device manufactured by defendant, or as to the later device which it began to manufacture about the time it abandoned its contest in the Patent Office. The first device copied not only the principle of complainant's device, but also its form. The use of the wire fuse for the pin was so clearly the use of an obvious equivalent as not to warrant discussion. The device which complainant began to make in 1931, while different in form, accomplishes the same result in the same way by the use of equivalents. There is the same sort of spring valve held open by a lever, which is held in tension by a pin. It is so arranged, however, that the lever is placed in front of the refractory disk; and, instead of the pin being constructed in two parts so as to separate when subjected to heat, the end of the lever is so constructed. Its construction appears from the following drawing; and its operation is thus described in defendant's brief:

"The casing is divided into upper and lower halves by a substantially horizontal partition 123. The lower half is itself divided by a substantially vertical partition 104 so as to provide two chambers in that lower half. The gas and air flows into one of these chambers in the lower half and through an opening or valve port in the partition 104 to the other chamber. The opening or port serves as a seat for a spring-pressed valve 105 having a stem 108 guided in appropriate bores in the casing. From the second chamber in the lower half of the casing, and after passing by the valve, the gas flows upwardly through an enlarged opening in the substantially horizontal partition to the upper half of the casing. In the opening in the latter partition, there is disposed a short nipple packed with the usual wire gauze or other flame arresting material so that the gas flows through the same but flame cannot return. From the upper half of the casing, the gas flows outwardly and forwardly to the burner. The horizontal partition which divides the casing into upper and lower halves is provided with a vertically disposed bore which serves as a bearing for a short rock shaft. The lower end of the rock shaft disposed in the lower half of the casing is provided with a forked arm 110, which engages a collar 109 on the valve stem 108. The upper end of the rock shaft, in the upper half of the casing, is provided with an elongated arm 112 having a slotted free end 117. In the upper half of the casing adjacent the free end of the arm 112, there is disposed a fixed stud 115 which serves as an anchor for a fuse link 113. The latter device comprises a flat body plate having a head 116 soldered thereto. When the short rock shaft is moved in a clockwise direction (Figure 5) by moving the arm 112 toward the pin 115, the forked arm on the other end of that shaft will move the valve to open position against the force of spring 107. The parts are retained in this open position by engagement of the fuse link 113 and the arm 112.

"In the normal operation of fire checks of this type, the gas flows through the casing in a direction from the left to the right. In the event of a flash back, the flame will travel rearwardly in the opposite direction, from the right to the left. The flame will be checked in its rearward travel by the customary flame arresting wire gauze 119 disposed in the nipple carried by the horizontal dividing partition. When the flame is thus checked, it will burn on the surface of the gauze and will immediately raise the temperature of the fuse to a point where the solder will melt so that the retaining head 116 will become *921 dislodged. When this occurs, the arm 112 will be free to swing under the action of the spring pressed valve, and the valve will move to its seat to cut off the flow of gas."

It will be noted that this latest device of defendant provides for the closing of the valve by a spring, just as does the device of complainant. The valve is held open by a lever in tension, just as is complainant's device. The release of the lever is accomplished by the melting of a fusible device. And this fusible device is placed in front of the refractory disk so that the flame of the backfire will play upon it. It contains all of the elements of complainant's device, and all operate the same way to produce the same result. Infringement is not avoided because a housing is provided for the lever, nor because the fusible element is attached to the lever, instead of to a pin attached to the lever, nor by the slightly different location of the members of the combination. The lever held in tension by the fusible head is clearly the equivalent of the lever held in tension by the jointed pin of complainant's patent. And while it may be patentable as an improvement over complainant's device, it is well settled that infringement is not avoided by improvement. Temco Electric Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 328, 48 S. Ct. 170, 72 L. Ed. 298.

The learned District Judge was of opinion that complainant was not entitled to invoke the doctrine of equivalents and must be confined to the exact device disclosed by his patent. In this we think there was error. Whether complainant's patent be treated as a pioneer or basic patent or not, he unquestionably made a valuable contribution to the art; and it is well settled that in such case his patent is entitled to liberal treatment. As said by Chief Justice Taft in Eibel Process Co. v. Minnesota, etc., Co., 261 U.S. 45, 63, 43 S. Ct. 322, 328, 67 L. Ed. 523: "In the case before us, for the reasons we have already reviewed, we think that 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."

The rule applicable is thus stated in the leading case of Winans v. Denmead, 15 How. 330, 342, 14 L. Ed. 717:

"Patentable improvements in machinery are almost always made by changing some one or more forms of one or more parts, and thereby introducing some mechanical principle or mode of action not previously existing in the machine, and so securing a new or improved result. And, in the numerous cases in which it has been held, that to copy the patentee's mode of operation was an infringement, the infringer had got forms and proportions not described, and not in terms claimed. If it were not so, no question of infringement could arise. If the machine complained of were a copy, in form, of the machine described in the specification, of course it would be at once seen to be an infringement. It could be nothing else. It is only ingenious diversities of form and proportion, presenting the appearance of something unlike the thing patented, which give rise to questions; and the property of inventors would be valueless, if it were enough for the defendant to say, your improvement consisted in a change of form; you describe and claim but one form; I have not taken that, and so have not infringed.

"The answer is, my improvement did not consist in a change of form, but in the new employment of principles or powers, in a new mode of operation, embodied in a form by means of which a new or better result is produced; it was this which constituted my invention; this you have copied, changing only the form; and that answer is justly applicable to this patent."

And the rule was applied and stated with great clarity by Mr. Justice Clifford in Union Paper Bag Machine Co. v. Murphy, 97 U.S. 120, 125, 24 L. Ed. 935, from which we quote as follows:

"Except where form is of the essence of the invention, it has but little weight in the decision of such an issue, the correct rule being that, in determining the question of infringement, the court or jury, as the case may be, are not to judge about similarities or differences by the names of *922 things but are to look at the machines or their several devices or elements in the light of what they do, or what office or function they perform, and how they perform it, and to find that one thing is substantially the same as another, if it performs substantially the same function in substantially the same way to obtain the same result, always bearing in mind that devices in a patented machine are different in the sense of the patent law when they perform different functions or in a different way, or produce a substantially different result.

"Nor is it safe to give much heed to the fact that the corresponding device in two machines organized to accomplish the same result is different in shape or form the one from the other, as it is necessary in every such investigation to look at the mode of operation or the way the device works, and at the result, as well as at the means by which the result is attained.

"Inquiries of this kind are often attended with difficulty; but if special attention is given to such portions of a given device as really does the work, so as not to give undue importance to other parts of the same which are only used as a convenient mode of constructing the entire device, the difficulty attending the investigation will be greatly diminished, if not entirely overcome. Cahoon v. Ring [Fed. Cas. No. 2,292] 1 Cliff. [592] 620.

"Authorities concur that the substantial equivalent of a thing, in the sense of the patent law, is the same as the thing itself; so that if two devices do the same work in substantially the same way, and accomplish substantially the same result, they are the same, even though they differ in name, form, or shape. Curtis, Patents (4th Ed.), sect. 310."

In the case of Crown Cork & Seal Co. v. Aluminum Stopper Co. (C. C. A. 4th) 108 F. 845, 866, this court, speaking through Judge Brawley, said: "Infringement is not avoided by mere change of form, or renewals of parts, or reductions of dimensions, or the substitution of mechanical equivalents, or the studious avoidance of the literal definition of specifications and claims, or the superadding of some improvement. The court will look through the disguises, however ingenious, to see whether the inventive idea of the original patentee has been appropriated, and whether the defendants' device contains the material features of the patent in suit, and will declare infringement even when those features have been supplemented and modified to such an extent that the defendant may be entitled to a patent for the improvement."

And what we said in Frick Co. v. Lindsay (C. C. A. 4th) 27 F.(2d) 59, 62, is particularly applicable here. We said: "While the patent of complainant is not a basic or pioneer patent, this does not mean that it is not entitled to invoke the doctrine of equivalents. The range of equivalents, it is true, depends upon the extent and nature of the invention, and if the invention is broad and primary in character, the range of equivalents will be correspondingly broad; but, where the patent is for a mere improvement, a narrower range of equivalents will be applied. Paper Bag Patent Case [Continental Paper Bag Co. v. Eastern Paper Bag Co.] 210 U.S. 405, 415, 28 S. Ct. 748, 52 L. Ed. 1122. Any patent, however, has some range of equivalents, unless form is made the indispensable thing. And the rule is especially applicable where the infringer takes the whole gist of the invention, as in this case. U. S. Slicing Mach. Co. v. Wolf, Sayer & Heller (D. C.) 249 F. 245, 247 [Wolf, Sayer & Heller v. U. S. Slicing Mach. Co.], Id. (C. C. A. 7th) 261 F. 195."

See, also, Imhaeuser v. Buerk, 101 U.S. 647, 656, 25 L. Ed. 945; Morley Machine Co. v. Lancaster, 129 U.S. 263, 9 S. Ct. 299, 32 L. Ed. 715; Leader Plow Co. v. Bridgewater Plow Co. (C. C. A. 4th) 237 F. 376; Yancey v. Enright (C. C. A. 5th) 230 F. 641; Baldwin v. Abercrombie & Fitch Co. (C. C. A. 2d) 228 F. 895; Parker v. Automatic Mach. Co. (D. C.) 227 F. 449; United States Slicing Machine Co. v. G. S. Blakeslee & Co. (C. C. A. 7th) 227 F. 442; Peter T. Coffield & Son v. Spears & Riddle (C. C.) 169 F. 641; Lourie Implement Co. v. Lenhart (C. C. A. 8th) 130 F. 122; Bresnahan v. Tripp Giant Leveller Co. (C. C. A. 1st) 72 F. 920; Herrick v. Tripp Giant Leveller Co. (C. C. A. 1st) 60 F. 80; 20 Rawle C. L. 1155, 1156.

We come then to the question as to whether complainant is entitled to relief on account of the manufacture and sale by defendant of devices which embody complainant's invention prior to the grant of his patent; and, as heretofore indicated, we think that this, also, must be answered in the affirmative. The general rule, of course, is that the monopoly of a patent which entitles a patentee to damages for *923 infringement commences only when the patent is granted; but where, in advance of the granting of a patent, an invention is disclosed to one who, in breach of the confidence thus reposed, manufactures and sells articles embodying the invention, such person should be held liable for the profits and damages resulting therefrom, not under the patent statutes, but upon the principle that equity will not permit one to unjustly enrich himself at the expense of another. The question was before the Circuit Court of Appeals of the Seventh Circuit in the recent case of Booth v. Stutz Motor Car Co. of America (C. C. A. 7th) 56 F.(2d) 962, and we think that the decision there rendered is eminently sound and just. It would be a reproach to any system of jurisprudence to permit one who has received a disclosure in confidence to thus appropriate the ideas of another without liability for the wrong.

It is argued that there was no confidential relationship existing between complainant and defendant with respect to the disclosure of complainant's invention; but this contention is groundless. Complainant offered to disclose his invention to defendant with a view of selling it to defendant, and so stated in his letter. Defendant was interested in the proposition and invited the disclosure, otherwise it would not have seen complainant's specification and drawings until the patent was granted. While there was no express agreement that defendant was to hold the information so disclosed as a confidential matter and to make no use of it unless it should purchase the invention, we think that in equity and good conscience such an agreement was implied; and having obtained the disclosure under such circumstances, defendant ought not be heard to say that there was no obligation to respect the confidence thus reposed in it.

Defendant argues, however, that in the manufacture and sale of automatic fire checks prior to the grant of the patent, it was not using the ideas of complainant, but ideas which had been evolved in its own plant prior to the disclosure; and it introduces evidence tending to show that as far back as 1918 it was working on the idea, that in 1926 it had drawings and patterns made for the development of an automatic fire check, and that early in 1929, before it received the specification and drawings of complainant's patent, it had developed in its own plant the fire check with the wire fuse which we have held to be an infringement. Defendant concedes, however, that it is not entitled to claim priority over complainant; and we think that the same rules of evidence which would preclude it from claiming to be the prior inventor of the device constitute a complete answer to the contention that in manufacturing the device it was using its own ideas and not ideas gained from complainant. It is well settled that where an unpatented device, the existence and use of which are proven only by oral testimony, is set up as a complete anticipation of a patent, the proof sustaining it must be clear, satisfactory, and beyond a reasonable doubt. The Barbed Wire Patent, 143 U.S. 275, 284, 12 S. Ct. 450, 36 L. Ed. 154; Adamson v. Gilliland, 242 U.S. 350, 37 S. Ct. 169, 61 L. Ed. 356. And we think the same rule should be applied against one who admittedly receives a disclosure from an inventor, proceeds thereafter to manufacture articles of similar character, and, when called to account, makes answer that he was using his own ideas and not the ideas imparted to him.

But without applying this rule to defendant, we think that the same result follows. We must weigh the testimony of the witnesses in the light of defendant's conduct; and, when we do so, we cannot credit the testimony as to the development of the fire check by defendant's employees. The fact that these employees were working on a fire check in 1918 and 1926 shows that the problem of producing one had not at that time been solved. If it had been solved in April, 1929, when complainant sent defendant his specification and drawings, it is inconceivable that defendant should have failed to mention that fact in its letter of May 15th and should have contented itself with comparing complainant's device with the old Kemp fire arrester. Sawyer v. Strauss, 57 Ohio App. D. C. 118, 18 F.(2d) 160, 162. If upon receipt of complainant's letter defendant had immediately written him that it had already worked out the device, a different complexion would be placed on the matter; but, instead of doing this, it delayed its answer more than two weeks, and in its letter said nothing about its recent invention of the fire check, stated that it was not interested in complainant's invention because it was not an improvement over the old backfire preventer, and that the fire check incorporated no new features over that old device, was *924 considered by the writer as inoperative and contained one very objectionable feature which would prevent its acceptance by insurance authorities. Then, after writing this letter, which was manifestly calculated and intended to discourage complainant, defendant proceeded, without saying anything further to complainant, to manufacture articles embodying his design and to have two of its employees apply for a patent covering it. The similarity of defendant's device to that of complainant is strong proof that one was copied from the other; for it is hardly probable that different persons should independently of each other invent devices so nearly similar at so nearly the same time. Adamson v. Gilliland, supra; Orange-Crush Co. v. American Ornamental Bottle Corp. (C. C. A. 4th) 60 F.(2d) 518, 520. But this is not all. When the patent office had declared the interference and complainant had learned for the first time of defendant's effort to obtain a patent, defendant attempted to purchase his rights, and, failing in this, moved to dissolve the interference on the ground that the patent was void, instead of fighting the matter out on the issue of priority. After this course of conduct, the defendant can hardly expect that its testimony as to prior invention should be accepted. What was said by the Supreme Court in Atlantic Works v. Brady, 107 U.S. 192, 203, 2 S. Ct. 225, 234, 27 L. Ed. 438, when dealing with the contention of one Brady that he was the inventor of a device which had been disclosed to him by General McAlester, is applicable here to the testimony of the employees of defendant, viz.: "Interested as he is in the result of the suit, his own testimony cannot be allowed to prevail against a course of conduct so utterly at variance with it. It may be true; but we cannot give it effect against what he himself did, and did not do, without disregarding the ordinary laws that govern human conduct."

For the reasons stated, the decree appealed from will be reversed, and the case will be remanded for further proceedings not inconsistent herewith.

Reversed.

SOPER, Circuit Judge (dissenting).

When Hoeltke, the patentee, first informed the representatives of the Kemp Manufacturing Company that he had made an invention that would be of value in the manufacture of gas appliances, they requested him to apply for a patent before he divulged to them the nature of his discovery. It is difficult to reconcile this conduct with the theory that these men were prone to dishonesty, and willing to appropriate to their own use the property of another. Their subsequent conduct in the interference proceeding is even more significant. They then knew not only the nature of the Hoeltke device, but also that he had made it between April 24, 1928, and December 4, 1928. With this knowledge in their possession they asserted that they had first embodied their invention on April 1, 1929. This circumstance alone seems well-nigh sufficient to dispel the charge of fraud; for if they were then engaged in a conspiracy to purloin the Hoeltke invention, it is hard to understand why they should have chosen a false date later than Hoeltke's discovery, as we are asked to believe, rather than an earlier one. They were not inexperienced in patent matters, and we cannot suppose that they were devoid of average intelligence, for during two decades they had successfully conducted a business founded sixty years ago, and had supplied the industry with gas appliances of such efficiency as to command a price higher than their competitors.

A third important circumstance, found as a fact by the District Judge, and abundantly established by uncontradicted evidence, demonstrates that long before Hoeltke made his disclosure to the defendant, its employees had conceived the idea of an automatic fire check. Pattern makers, connected with an independent business, refreshing their recollection from their books, showed that in May, 1926, they had made patterns from drawings of an automatic fire check designed by Van Horn, an employee of the business for fifty years. A model thereof was made during the trial and was said by Hoeltke's attorney in argument in this court to be an infringement of the Hoeltke patent. It is not suggested that priority over Hoeltke is established by these facts, for the delay of the defendant before and after 1926 must be considered. See Automatic Weighing Machine Co. v. Pneumatic Scale Corp. (C. C. A.) 166 F. 288, 300. But when due weight is accorded to these important circumstances, it is more reasonable to believe that the parties to the cause independently conceived and completed automatic fire checks rather than to find that reputable men had been guilty of dishonesty and false swearing. It is true that *925 when a patent is attacked on the ground that it was anticipated by an unpatented device, whose existence and use are proved only by oral testimony, the proof must be clear, satisfactory, and beyond a reasonable doubt. But it is not the rule that when a patentee charges that an infringer has secured possession of the patented device by fraud prior to the issuance of the patent, the person accused must prove his innocence beyond a reasonable doubt. On the contrary, the law never presumes fraud, and the burden devolves upon him who alleges it to show it by clear and convincing proof. Lalone v. United States, 164 U.S. 255, 257, 17 S. Ct. 74, 41 L. Ed. 425; U. S. v. American Bell Tel. Co., 167 U.S. 224, 241, 17 S. Ct. 809, 42 L. Ed. 144. Even if the rule were otherwise, it would still be unreasonable in the pending case to conclude that the first idea of the invention came to the defendant from Hoeltke, in view of the evidence that its employees were in possession of the idea years before. One who is familiar with the numerous interference proceedings in the United States Patent Office is not surprised when the evidence shows that two men, working independently of one another upon a mechanical problem, have found the same solution.

The delay on the part of the defendant in putting its idea into commercial practice is satisfactorily explained. While there had been backfires in the defendant's apparatus and in the appliances of other manufacturers, they had been neither so numerous nor so destructive as to be of great importance. Since 1901, the Kemp devices, equipped with the backfire preventer, had been successfully used, and since 1911, appliances equipped with backfire preventers and with hand-operated fire checks had not only been successfully and safely used, but had received the approval of the Fire Underwriters. The fire checks were so constructed that a flash back would be retarded from entering the gas pipe for a considerable interval of time, so that with the use of odorous paint, the danger could be discovered by the attending workmen; and if the fire succeeded in making its way back to the fire preventer, its automatic arrangement was such that the bottom of the device, purposely made weak and unsubstantial, would give way and prevent the access of the fire to the gas supply. Under these circumstances, the need for an automatic fire check was not considered imperative, and the improvement was delayed until the defendant began to manufacture gas ovens, in which it was necessary to install fire checks that were not accessible to the workmen. It was then that efforts were made to develop a practical and efficient automatic fire check which culminated in the device conceived and manufactured in the defendant's plant.

Emphasis is placed upon the fact that when the defendant received for the first time the Hoeltke invention on April 27, 1929, it sent a letter of May 15, 1929, in reply, in which, without disclosing the work which it had done along the same lines in its own factory, it expressed the opinion that the Hoeltke device was not patentable, since it merely incorporated the principles of the old Kemp fire preventer. It is suggested that fraud must be inferred from this concealment. The fact that a well-established concern did not choose to divulge what it was doing in its own factory to an outsider seems a slender basis for an inference of crime. Moreover, the evidence clearly shows that the defendant did not conceal that it was making an automatic device, for beginning in the month of September, 1929, it caused advertisements of its automatic fire check to be inserted in the trade journals and offered them for sale. It is also suggested, as evidence of criminality, that the defendant first offered $3,500 and then $5,000 to settle the controversy. The great delay and expense attendant upon patent litigation are well known to all experienced persons; and heretofore it has not been supposed that an offer to compromise a disputed claim was evidence of liability. It has never been received as evidence of fraud. Hoeltke on his part demanded $75,000 at one time, and $100,000 at another.

The filing of the application for the Van Horn patent, the subsequent participation in the interference proceedings, the failure to follow them up, and the abandonment of the application for the patent were done upon the advice of counsel. Unless one starts with a presumption of fraud, these actions have no tendency to show fraud in view of the circumstances already discussed; for they merely show that the applicants for the Van Horn patent adopted the usual way to raise the question of patentability of the Hoeltke device in the Patent Office. The fact that the defendant, through its employee, made application for a patent on a device which may involve the same principles as that of *926 the Hoeltke structure, does not preclude the defendant from alleging now that the Hoeltke device lacks invention. The Supreme Court has recently reiterated the rule that there is no estoppel under such circumstances. See Paramount Publix Corp. v. American Tri-Ergon Corp., 55 S. Ct. 449, 455, 79 L. Ed. 997, where it is said: "However inconsistent this early attempt to procure a patent may be with petitioner's present contention of its invalidity for want of invention, this Court has long recognized that such inconsistency affords no basis for an estoppel, nor precludes the court from relieving the alleged infringer and the public from the asserted monopoly when there is no invention. Haughey v. Lee, 151 U.S. 282, 285, 14 S. Ct. 331, 38 L. Ed. 162."

It may be added that Hoeltke laid his charges against the defendant and its representatives before the Department of Justice of the United States and endeavored to secure a prosecution for crime. Government agents visited the defendant's factory and made an examination, and, as the result, the idea of criminal prosecution was abandoned.

A finding of fraud cannot be made in this case unless the conclusion is reached that the defendant's witnesses deliberately gave false testimony. Hence it is of the utmost importance to observe the rule repeatedly laid down by this and other courts that the findings of the trial judge in an equity case, who had the opportunity of seeing the witnesses, hearing their story, judging their appearance, manner, and credibility on the question of fact, is entitled to great weight, and will not be set aside unless clearly wrong. Standard Transportation Co. v. Wood Towing Corp. (C. C. A.) 64 F.(2d) 282. The District Judge heard all of the defendant's important witnesses in person, except one. The record shows that he proceeded with great care and deliberation during the taking of the testimony and afterwards, hearing argument on more than one occasion, and receiving a number of written briefs. His conclusions, even if the issue should be considered doubtful, should not be disregarded.

The question still remains as to whether Hoeltke's device involved a patentable invention. It was of narrow scope, for all that he did, as he said in a letter to his lawyer was to take the well-known fire check, consisting of a perforated refractory metal disc, a mat of copper wire gauze and a hand-operated valve, and make it automatic. He accomplished this very readily, as soon as the idea of an automatic valve occurred to him, by placing a fusible pin in front of the metal disc and connecting it by a lever to a spring valve so that when the fuse melted, the lever was released and the valve closed. The District Judge, taking into account that the same principle was involved in the Kemp fire preventer, properly held that it required merely the skill of a mechanic thus to alter the fire check by the use of familiar devices. It is a mistake to suppose that Hoeltke found that for which the art had been searching in vain, for the uncontradicted evidence shows that an equally serviceable structure was independently made in the defendant's factory as soon as the need for it in a commercial sense was felt, and this evidence may reasonably be believed, since it merely proves that defendant's experienced employees were also able to do that which Hoeltke so easily accomplished. The conclusion of the District Judge that the Hoeltke patent is void for lack of invention should be affirmed.

On Rehearing.

PARKER, Circuit Judge.

This case has been reheard and the questions involved have been carefully considered in the light of the briefs filed and the arguments made on the rehearing. On the question of patentability and infringement nothing need be added to what was said in our original opinion.

Upon the reargument it was most strenuously insisted that we ought not find that, in manufacturing and selling its automatic fire check between the time of plaintiff's disclosure and the granting of plaintiff's patent, the defendant appropriated the idea of plaintiff in breach of the confidence reposed in it; and that, at any rate, we ought not disturb the findings of the District Judge with respect to this matter. We are thoroughly familiar with the salutary rule that in equity cases the findings of the District Judge on questions of fact will not be disturbed unless in our opinion such findings are clearly wrong. But as we recently had occasion to point out in Standard Acc. Ins. Co. v. Simpson (C.C.A.4th) 64 F.(2d) 583, 588, "in an appeal in equity we review the facts as well as the law, and the review *927 is a real review and not a perfunctory approval." The finding of the judge below is persuasive, not controlling; otherwise review on the facts would be a delusion and a snare. In the case before us, we think that facts as to which there is no controversy establish so conclusively the liability of the defendant that the verbal testimony of officers and employees of defendant, unsupported as it is by anything in writing or of record, to the effect that the device of defendant was perfected prior to plaintiff's disclosure, is not sufficient for defendant's exoneration.

The following facts are undisputed: The defendant had been working without success upon the development of an automatic fire check. Plaintiff on December 4, 1928, wrote defendant calling attention to the fact that a gas explosion had occurred in the factory in which he was working "because an operator was not around just then to shut off the gas" and that he had made a device that would overcome that trouble. On December 10, 1928, defendant wrote plaintiff that it was interested in the device, and later, not hearing from plaintiff, sent one of its employees to advise him to patent his device and submit it for defendant's consideration. On April 27, 1929, plaintiff wrote defendant sending copies of the drawings and specification of his patent. On May 15, 1929, defendant replied to this letter stating that it was not interested in his device because it was a fire check instead of an improvement on a back-fire preventer, that plaintiff's invention comprised nothing but the old features of the Kemp back-fire preventer, patented in 1901, that there was no invention in the device and no claims of any consequence could be obtained on it, and that the device was considered inoperative and had one very objectionable feature which would prevent its acceptance by insurance authorities. Defendant said nothing in this letter indicating that it had perfected a fire check of similar character, or that it was interested in a fire check, but on the contrary stated that it was disappointed in the invention because it was for a fire check instead of for improvement on back-fire preventer. Nevertheless, defendant had its sales manager call on plaintiff in June and talk with him about his invention, plaintiff contends with a view of purchasing it.

It is further established, by undisputed testimony, that in August 1929, defendant caused two of its employees to file an application for a patent on a device functioning in the same way as plaintiff's device and similar thereto, except that, for the soldered pin which holds the lever in tension in plaintiff's patent, a wire fuse in a sleeve was substituted with a soldered joint in the flame of the back-fire, and about the same time defendant began making and selling these devices, all without any notice to plaintiff. About this time, August 25, 1929, plaintiff wrote defendant, directing the letter to the sales manager who had talked with him in June, and asking whether it had been decided whether defendant could use his invention. On August 28, 1929, defendant wrote that one Ponder, the employee who had called on plaintiff in January, would call and discuss the matter with him. Ponder did call on plaintiff and told him that if he had a proposition he should submit it to defendants, whereupon plaintiff, on September 23, 1929, wrote defendant offering to sell his invention for $75,000. Plaintiff heard nothing further from defendant and did not learn of its manufacture of the device covered by his patent application, or of the attempt of defendant through its employees, to obtain a patent thereon, until he learned of it in connection with proceedings in the Patent Office. After interference had been declared in the Patent Office between the plaintiff's application and that of employees of defendant, defendant offered plaintiff first $3,500 and then $5,000 for his patent rights. When plaintiff declined these offers, defendant moved to dissolve the interference on the ground that the invention was not patentable.

If there were no explanation offered of these undisputed facts, the conclusion that defendant had appropriated the ideas disclosed to it by plaintiff would be inescapable, and we do not think that the matters offered in explanation are sufficient to justify us in reaching any other conclusion. Defendant's contention is that it had been having its employees work on this fire check for years, that the basic idea had been worked out by Van Horn in 1918, that drawings embodying the idea, with the exception of the fuse, had been made for it in 1926, and that the fuse was developed by its employee Hunt on or about April 1, 1929. The only written evidence offered in support of any part of this contention is the evidence of the drawings which were made in 1926; but these do not support the contention. They *928 cover a mechanism of a complicated character, which was never found to be workable and which does not suggest the simple mechanism of plaintiff's patent and defendant's manufacture. Instead of helping, they hurt the defendant's case; for they show beyond question that as early as 1926 defendant was attempting without success to find a solution for the problem which plaintiff solved.

Defendant's case, on this point, in its final analysis rests upon the testimony of its president W. W. Kemp, its sales manager Dunkak, and its two employees Van Horn and Hunt to the effect that about April 1, 1929, Hunt perfected the device which defendant manufactured and attempted to have patented, and that they saw it in operation in defendant's plant before the receipt of plaintiff's letter of April 27th. If this were true, Hunt perfected the mechanism after plaintiff had perfected his invention and after defendant knew that this was so; for at that time it had plaintiff's letter of December 4, 1928, telling of his invention and had answered it under date of December 10th inviting him to make disclosure. When the letter of April 27th came, showing that the device which plaintiff had invented was of the same character as that worked out by Hunt, defendant, instead of trying to get a patent on the device for itself, or proceeding with the manufacture and sale of an article which at the most was a mere improvement on what plaintiff had invented, should certainly, in the exercise of good faith, have notified plaintiff of Hunt's achievement and, if there was any superiority in what Hunt had done, should have sought a patent on the improvement only, and not on plaintiff's idea. Instead of doing this, it proceeded to make and sell and attempted to patent a device which it knew at the time was substantially the device which plaintiff had invented some time before.

But we cannot accept this unsupported verbal testimony of interested officers and employees of defendant as to Hunt's independent development of the idea. One who invites the disclosure of an invention, and thereafter begins to manufacture articles embodying the principle of the disclosure, labors under a heavy burden when he seeks to justify his action on the ground of independent invention, and he ought to offer something of greater weight than the verbal testimony of interested witnesses. Here there is no writing or other record to sustain the contention, and no one, except Kemp, the two employees who acted under his direction in attempting to obtain the patent, and the sales manager who negotiated with plaintiff, claims to have seen the Hunt device prior to plaintiff's disclosure. Kemp's testimony is discredited by the fact that he failed to mention the Hunt device in his letter of May 15th and took the position in that letter that he was disappointed in the fire check because it was not an improvement on the back-fire preventer, and also by the fact that he had Hunt and Van Horn make application for a patent on the Hunt device when he knew that it was in substance the device of plaintiff's invention and admits that he thought it not patentable over the old Kemp backfire preventer. Hunt and Van Horn's testimony is discredited by the fact that they made the application for the patent on the Hunt device, claiming to be the original inventors thereof, when they knew of plaintiff's invention and knew that it embodied the same principle of operation.

There can be no question but that defendant's sales of infringing devices prior to the grant of plaintiff's patent were made after plaintiff had made his invention and had disclosed it to defendant upon its invitation; and certainly the device of defendant was so nearly like that of plaintiff as to lead irresistibly to the conclusion that one was copied from the other, once the disclosure to defendant prior to its manufacture and sale is established. The defendant's explanation that its device was independently developed is one which we cannot accept in the light of its conduct, particularly in failing to mention the development of the device in its letter of May 15th, and in seeking to obtain a patent on a device embodying the idea of plaintiff's invention which had been disclosed to it, and which it knew that plaintiff had perfected prior to the perfection of its device. Under such circumstances, we think that defendant should account to plaintiff for the profits realized from its sale of the infringing device prior to the grant of plaintiff's patent, not under the patent statutes, but under the long-recognized principle of equity that no man ought to be allowed to enrich himself unjustly at the expense of another.

For the reasons herein stated, as well as for the reasons stated in the original opinion, *929 the decree appealed from will be reversed, and the cause will be remanded for further proceedings.

Reversed.

SOPER, Circuit Judge.

I adhere to my dissent previously filed.