Plaintiff Woodstream Corporation, a Pennsylvania corporation and long-time manufacturer and seller of animal traps, brought an action against the defendants, Herter's, Inc., a Minnesota corporation, and its president, George L. Herter, for patent infringement and unfair competition. Plaintiff, formerly the Animal Trap Company of America, is the assignee of the patents allegedly infringed. Woodstream alleges that the defendants distribute and sell animal traps embodying the improvements protected by the plaintiff's patents. The district court denied Woodstream any relief. Woodstream Corporation v. Herter's, Inc., 312 F. Supp. 369 (D.Minn.1970). Woodstream prosecutes this timely appeal.

Plaintiff alleges the infringement of two patents, Conibear, 3,010,245, issued November 28, 1961, and Lehn, 2,947,104, issued August 2, 1960.1 The defendants denied infringement, asserting the invalidity of the plaintiff's patents. The central issue on this appeal concerns the obviousness of the plaintiff's traps since the defendants concede the utility and novelty of these devices.2 The district court held the patents invalid, concluding that

[W]hile the Conibear and Lehn patents here in issue may satisfy the requirements of novelty and utility, the nature of the subject matter sought to be patented was obvious to a person having ordinary skill in the art, and for that reason the patents are invalid. [312 F.Supp. at 371]

Appellant urges a reversal of the district court's decision, contending that the court's conclusions lack any evidentiary support, or, alternatively, that the trial court applied an erroneous standard of invention in finding plaintiff's two patents invalid.

Plaintiff Woodstream's claim against the defendants for unfair competition rests upon allegations that the defendants' reproduction and cataloging of conventional leg-hold type traps were so similar to the plaintiff's products as to create confusion in the market-place between plaintiff's traps and those of the defendants. The district court treated this issue as purely a factual question, and found that there was no proof that the defendants were deceiving the public, attempting to do so, or palming off their traps for the plaintiff's traps. On this appeal, plaintiff-appellant contends that federal law required the defendants to advertise their traps as being of foreign manufacture, in this case Japanese, and, further, that the findings of fact made by the district court upon this issue are not supported by the evidence.

Having fully reviewed an extensive record on this appeal, we conclude that the district court erred in holding the Conibear patent invalid. In all other respects, we affirm the district court's decision.

We turn initially to the patent infringement claims. The structures involved in the instant case are animal traps intended for fur trapping, particularly muskrats, beaver, marten, and other fur-bearing animals. The Conibear structure is a killer type trap, a steel trap which kills the entrapped animals through suffocation or by breaking the animal's neck. The Lehn device presents a trap similar in operation and configuration, except for a modification of the trigger element. These traps, referred to as "swing frame" traps, reflect a structure consisting of a pair of substantially identical rectangular frames, with one of the frames being slightly smaller and capable of rotating within the other frame. The two rectangular frames are hinged together or pivoted at a midpoint at their ends, with the sides of the frames comprising the jaws of the traps. When hinged and placed in a vertical position, the top and bottom sides of the two rectangular frames may be placed in a position parallel with each other. The jaws of the traps are held against the tension of a coil spring or springs in an open or set position by a sliding latch mechanism made of interconnecting wires constituting the trigger of the traps.

The Conibear trap is released by a wire extending into the center of the trap which disengages the latch member from one of the rectangular frames when the trigger arm which is directly attached to the latch becomes actuated by an animal entering the trap. The latch mechanism contains a diamond shaped wire element which pivots in any direction upon movement transmitted to it by the trigger arm; this movement affords a positive and quick release of the trap. Upon release, the jaws swing rapidly in opposite directions in egg beater fashion under the influence of a powerful coil spring to catch the animal. The following photograph3 demonstrates the Conibear trap in a set position.

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE The Lehn patent, except for some minor modifications in the construction of the pivotal aspects of the frames, reflects an improvement over the Conibear patent, specifically in the trigger mechanism. In the Lehn trap, the latch consists of a key-like mechanism with a collar at one end and U-notches at the other end. The collar encircles the frame side or jaw of the trap; it freely slides and rotates thereon. The opposite end of the latch grasps the second jaw by one of the U-shaped notches fitting on a partially cut-down sleeve which also freely slides and rotates upon the second frame side or jaw. A trigger arm extends from this cut-down sleeve into the trap. Movement against the trigger arm rotates this sleeve, positively forcing the U-notched latch out of engagement with the second jaw. In this way, the mechanism affords instantaneous release action. The following photograph shows a Lehn trap as set.

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The defendants' alleged infringing traps, known as the "Hudson Bay Killer Traps," are almost identical in appearance and construction to the Lehn traps, except for a modification of the trigger arm and sleeve attachment to the second frame side or jaw of the trap. Like Lehn, a U-notched latch is rotatable and slideable on the first jaw. This latch engages the second jaw not directly upon a sleeve which is attached to the second jaw, but indirectly through a hinged second sleeve connecting to the trigger arm. This second sleeve contains a cut-down opening similar to that on the Lehn trap; it permits the notched latch mechanism to hold the two jaws in a set position. Movement against the trigger arm causes rotation and lifting of the notched latch which, in turn, produce instantaneous release action as in the Lehn device. The following photograph illustrates the slight differences between the allegedly infringing trap and Lehn. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The evidence introduced by the plaintiff establishes that the Conibear and Lehn type of swing frame trap constituted a substantial improvement over the old leg-hold trap previously used by trappers. The leg-hold trap has been the mainstay of the fur-trapping industry since first introduced in England in the Seventeenth Century. The leg-hold trap came into its modern form with the invention of the steel trap in the Nineteenth Century. The leg-hold trap, as its name implies, utilizes spring-loaded steel jaws which clamp around the leg of the animal when the animal steps into or releases the trap. These traps are deficient in several respects. They cause considerable pain and suffering for the trapped animal since the traps fail to kill the animal. They often work inefficiently because many trapped animals are able to work loose and escape. Many animals, particularly muskrats, escape by breaking the leg and then chewing it off. The leg-hold trap poses substantial danger to the trapper and, in some instances, its use reduces the quality of the animal's fur. On the other hand, the swing frame killer trap may be set faster, easier, and safer. Moreover, animals may be removed from it more quickly than from the conventional trap. One of the most significant advantages of the swing frame trap is its humane nature since in most cases it immediately kills the animal.

Despite evidence of the substantial advantages of the swing frame traps over the conventional leg-hold traps, the district court held, as to the Conibear and Lehn patents, that

Notwithstanding the greater utility, humaneness and commercial success of the Conibear type trap, factors certainly entitled to consideration when determining patentability, it appears to the Court that the features which characterize the Conibear and Lehn traps over the prior art are such as would be obvious to a mechanic skilled in the trap field at the time of the patenting of the Conibear and Lehn traps. [312 F.Supp. at 371-372].

The statutory requirement of non-obviousness for patentability rests upon 35 U.S.C. § 103, enacted in 1952, which provides:

A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.

In Graham v. John Deere Co., 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966), the Supreme Court enunciated the guidelines of inquiry into obviousness under § 103:

Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented. As indicia of obviousness or nonobviousness, these inquiries may have relevancy. [383 U.S. at 17-18, 86 S. Ct. at 694]

The Court noted that while the ultimate question of patent validity is one of law, Great A & P Tea Co. v. Supermarket Corp., 340 U.S. 147, 71 S. Ct. 127, 95 L. Ed. 162 (1950), the determination of obviousness lends itself to several basic factual inquiries. The Court recognized in Graham that the question of obviousness "is not a question upon which there is likely to be uniformity of thought in every given factual context." 383 U.S. at 18, 86 S.Ct. at 694. It advised using the same judicial approach as that employed in deciding questions of negligence and scienter. Id.

In applying the substantive standards of patentability to the claimed inventions in the instant case, we must follow established principles of patent construction. A presumption of validity attaches to patents that have survived the scrutiny of the Patent Office, and the burden of establishing patent invalidity rests upon the party asserting it.4 Mr. Justice White, writing for a unanimous Court in Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, 402 U.S. 313, 91 S. Ct. 1434, 28 L. Ed. 2d 788 (decided May 3, 1971), observed that "patentees are heavily favored as a class of litigants by the patent statute." Id. at 335, 91 S. Ct. at 1446.

In holding the Conibear and Lehn patents invalid, the district court noted with particularity that "the Eighth Circuit Court of Appeals do[es] not afford a congenial forum to the holder of a United States patent." 312 F. Supp. at 370. Although in other cases5 since Graham, supra, we have not sustained the validity of any patent claim, these decisions cannot bind us here. We have not considered any device or process similar to those submitted on this appeal. The test for obviousness must be resolved upon a case-by-case analysis. Graham, supra, 383 U.S. at 18, 86 S. Ct. 684; Agrashell, Inc. v. Hammons Products Company, 413 F.2d 89, 93 (8th Cir. 1969); National Connector Corporation v. Malco Manufacturing Company, 392 F.2d 766, 771 (8th Cir.), cert. denied, 393 U.S. 923, 89 S. Ct. 254, 21 L. Ed. 2d 259 (1968); L & A Products, Inc. v. Britt Tech Corporation, 365 F.2d 83, 86 (8th Cir. 1966). With these basic principles in mind, we turn to an examination of the obviousness of the patents in issue.

III. THE OBVIOUSNESS OF THE CONIBEAR AND LEHN PATENTS

The parties have stipulated that claim 8 of the Conibear patent and claim 2 of the Lehn patent,6 in general, represent the claims made by each of the inventors, and that the validity of other patent claims rests upon the determination of said claim 8 and claim 2. As to claim 8 of the Conibear patent, the district court held this device to be anticipated by two Mau patents, Mau (I), 2,564,811, issued August 21, 1951, Mau (II), 2,701,428, issued February 8, 1955, and Ullman, 1,436,833, issued November 28, 1922. As to claim 2 of the Lehn patent, the court ruled that it was anticipated by the same inventions as anticipated by Conibear, plus Zahm, 2,202,938, issued June 4, 1940.

The Patent Office cited three patents, Oleen, 2,459,580, issued January 18, 1949, and the two Mau patents as part of the prior art in the Conibear patent. The district court considered the Conibear patent to have been anticipated by the two Mau patents and Ullman. The expert witness called by the defendant relied upon both Mau patents and the Ullman patent in concluding that the Conibear development was obvious in light of the prior art. This expert witness, a member of the patent bar and formerly a patent examiner himself, with a technical education in chemistry and physics, testified with regard to the prior art and the obviousness of the Conibear patent, stating:

Mau has the general combination [of Conibear] and it also has the latch, either Mau does. Ullman has a trap with a latch and a release mechanism. The other thing that Mau lacks is some mechanism to lift the latch from engagement.

The witness then concluded that the Conibear patent was anticipated by either Mau patent in view of the Ullman patent. On cross-examination, the defendants' expert witness conceded that he could not find any single prior art reference that showed all of the elements of claim 8 of the Conibear patent.

Before the district court and on this appeal, plaintiff urges that the nonobvious features of the Conibear and Lehn patents were the trigger mechanism, the use of round stock, and the form or configuration of the traps, with primary emphasis being placed on the trigger mechanism device. The plaintiff's expert witness testified that this trigger mechanism and the use of round stock were the nonobvious and significant improvements over the prior art in this field. The district court rejected these claims of nonobviousness, however, concluding that "the Conibear and Lehn patents rest upon exceedingly small and quite non-technical [sic] differences in a trap which was old in the art." 312 F. Supp. at 373.

In determining whether the defendants have sustained their burden of establishing the validity of the Conibear and Lehn patents, it is necessary initially to examine the prior art in this field at the time of the patents in question. Mau II disclosed a trap comprised of two loop-shaped frames of rectangular bar stock. Mau II represented a trap very similar to the earlier Mau version, Mau I, and reflected the attempts of the inventor to correct the deficiencies of the earlier trap. The obvious deficiency of Mau I was that the trap, when set, was of such a configuration that it was possible for an animal to come in from the side or top of the trap and trigger the trap without being captured. In Mau II, the inventor attempted to solve this problem by having the coil spring of the trap bent over the top and sides of the trap and staked into position to prevent an animal from entering the trap from either side or the top and thereby triggering the trap without being caught. The Mau II trap somewhat resembled the Conibear patent as illustrated in the following diagram, and, upon release, the jaws rotated in the same egg beater fashion.

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The trigger-latch of the Mau II trap consisted of a wire fixed on one jaw and attached to the other jaw in the manner of a hook. The trap trigger was fixed and could not be adjusted along the trap frames. Movement of the trigger wire in an appropriate plane would twist the latch and spring the trap. This trigger mechanism, however, lacked the positive quick release in the Conibear trigger device. As indicated above, even the defendants' expert witness conceded that the Mau patents lack a mechanism to lift the latch from engagement and thereby rapidly spring the trap. Neither of the Mau traps were ever manufactured commercially.

The Ullman patent discloses a trap designed to catch burrowing animals. Two arms or jaws extending from a coil spring cross each other when the trap is set. The latch holding these arms or jaws in a set position through a rotating and pivoting mechanism permits the release of the trap when the burrowing animal steps on the trigger. The Ullman device represents an entirely different form and type of trap in its configuration, operation, and triggering than found in the Mau traps or in the devices here in question.

The principal differences between the Conibear patent and the prior art are the trigger mechanism which provides a positive quick release of the jaw, permitting the instantaneous killing of the animal, and a configuration of the trap which alleviates the problem encountered by the Mau patents of the animal entering the trap from the side and setting the trap off without being caught. The Conibear patent thus solved the two main problems encountered by the earlier killer-swing-frame type traps. The Conibear device solved the problem of animals entering the trap through the sides or by the top, through the use of a new configuration made of round wire stock. More importantly, the Conibear patent demonstrates the combination of a positive trigger release with a trap configuration of appropriate jaws designed to strike and hold the animal at the neck rather than the leg. These features represented a substantial improvement over the Mau patents, and over the conventional leg-hold traps.

Turning to a consideration of the level of ordinary skill in the pertinent art, the evidence establishes that a person having ordinary skill in the trapping art should have recognized the serious deficiencies in the Mau patents and the other prior art. However, while the deficiencies in the prior art were readily recognizable, it does not follow that the solution to those problems would have been obvious to one possessing ordinary skill in the art.

The situation in the instant case is greatly different from that in Graham, supra, where not only the deficiencies in the prior art, but also the solution to those deficiencies, were readily recognizable in the existing art. There, the deficiency in the prior art was overcome by simply inverting the plow shank and the hingeplate. 383 U.S. at 25, 86 S. Ct. 684. In the instant case, merely designing a trap configuration which would solve the problems encountered by the Mau traps would not alone cure the deficiencies in the prior art. Conibear not only conceived a new trap configuration, but also combined with it an effective, positive, quick release trigger mechanism which turned an otherwise impractical trap into an effective killer trap. We fail to see on this record that this solution would have been readily obvious to one having ordinary skill in the pertinent art.

To buttress its claim of patentability, plaintiff introduced evidence that the inventor of the trap, Frank Conibear, had received the only award ever given by the American Humane Association (AHA) for the development of a humane animal trap. In 1953, the AHA announced that $10,000 would be awarded for the invention of the "perfect" humane animal trap. In 1960, the AHA awarded Conibear a $600 cash prize and a certificate of merit for the invention of his swing frame killer trap. This marked the first and only recognition ever given to a trap inventor. In making the award, the AHA noted that while the trap did not completely fulfill the requirements of a perfect humane trap, it was the first trap to approach the ultimate goal.

In awarding Frank Conibear its certificate of merit and cash prize, the American Humane Association made the following statement regarding the Conibear trap and its impact on the trapping industry:

To say that a device should be equally effective for trapping muskrats in the marshes of Maryland as in trapping mink and marten under the snow and freezing conditions of Canada, may be asking entirely too much.

And yet, in an endeavor such as the AHA trap contest — which is aimed at providing more humane and effective wild animal traps — the development of the Conibear trap, which more or less revolutionizes the trapping of muskrats under many conditions, is a major step forward.

Plaintiff produced another witness, Elihu S. Abbott, a long-time professional trapper and winner of national recognition in the trapping field, who testified that the commercial version of the Conibear trap had almost completely replaced the old-style leg-hold trap, and that it operated with such effectiveness that it increased the quantity of the catch and resulted in less damage to the quality of the fur.7

Another witness, Arthur V. Harding, the publisher and editor of Fur-Fish-Game magazine for over forty years, testified as to the impact of the Conibear trap on the trapping industry. He stated:

It [the Conibear trap] was a terrific impact because before the Conibear came out, I had all kinds of letters from trappers complaining about loss of animals due to twistouts, and so forth, and Conibear just licked their problems.

Although Mr. Harding was not a trapper himself, he had been closely associated with, and very knowledgeable in, the trapping industry for most of his life.

Evidence of this nature, whether characterized as commercial success, fulfilling long-felt but unsolved needs, failure of others and similar considerations may be utilized to give light to the circumstances surrounding the origin of the patent in question, and may have relevancy as indicia of obviousness. Graham, supra, 383 U.S. at 17-18, 86 S. Ct. 684.

We deem the deposition testimony of the inventor John U. Lehn, taken and introduced by defendants, to be most significant on the issue of obviousness. His testimony demonstrates the nonobvious nature of the Conibear trap and indicates that the recognition of mechanical features of existing trap art did not foretell the Conibear invention. The seventy-five year old Lehn, an employee of the plaintiff, had spent most of his adult life as an inventor and mechanic in the animal trap field. Lehn claimed sole credit for some twenty patents and partial credit for ten others. Lehn's testimony, given with remarkable candor, discloses that even the resolution of an apparently simple problem, such as the size of frames for use in traps with particular animals, presented unusual difficulties and required actual experimentation with live animals.8 Lehn further testified as to his reaction upon reading a magazine article describing the success of the Conibear trap. His testimony on this point is most interesting and instructive concerning the issue of obviousness to one skilled in the pertinent art. He testified:

[W]hen I read that [the magazine article describing the Conibear trap], it almost knocked me off my feet, the story inside about the catches on this Conibear trap. Oh, they had caught everything. * * * Well, I couldn't take my eyes off of that picture anymore. I wanted to know how he set that trap. That's the first thing that come to my mind and it didn't come up until I had that whole thing developed. He didn't show the mechanism and that's the thing he tried to hide. I knew he tried to hide it; unless if he didn't try to hide it, the whole story in the magazine would be a fake. So I was convinced that he purposely hid the setting device of that trap, and that's what I was interested in when they sent it out to me.

After viewing an illustration of the Conibear trap in the magazine, but without seeing the actual trap or any illustration of the trigger mechanism, Lehn created his own version of this trap. He easily reproduced the frames in a similar configuration, but his efforts to construct an effective trigger mechanism proved fruitless. He experimented with wires twisted ninety degrees to hold the jaws in the set position, and later tried metal pieces notched to fit over the frames. However, it wasn't until the plaintiff company had purchased the rights to the Conibear trap and had given Lehn a model of the Conibear trap that he developed an effective trigger. Lehn testified:

[A]fter I got this [the Conibear trap] * * * I played with that awhile and I didn't fancy that [the Conibear trigger mechanism] as much. I thought I could make a simpler one.

After developing his own trigger mechanism for the trap, Lehn's employer obtained a patent thereon, but the witness conceded that, in his opinion, this patent did not "amount to a hill of beans."

On the instant appeal, the plaintiff-appellant focuses his primary attack upon the crucial testimony which lead the district court to reject the Conibear patent. The defendants' expert witness, admittedly possessing no experience in the trap art, could not state which elements should be taken from the Mau patents and which elements should be taken from the Ullman patent to "construct" this anticipatory combination. In light of the other testimony and evidence introduced, we think appellant's objection to the district court's reliance upon the defendants' expert witness is well taken.9 It is important to remember that the defendants, having raised the defense of patent in validity, assumed the burden of overcoming by "substantial evidence" the presumption of patent validity. See note 4, supra.

We entertain great doubt that defendants' evidentiary presentation warrants a conclusion that the existing state of the art suggested the solution demonstrated in the Conibear device. A conclusion of obviousness of the device, as drawn by defendants' expert, rests solely on his opinion that since other traps applied the same mechanical principle used by Conibear, that these other inventions foretold Conibear. As we have noted, a gap remains since defendants' case offers no explanation how, without hindsight, one having skill in the art would ascertain the mechanical elements he should draw from scores of existing traps, antique traps or traps disclosed in the Patent Office which would convert or create a workable killer trap adaptable to trapping fur-bearing animals out of Mau II or any other killer type trap existing in the art.

Notwithstanding Mr. Justice Clark's admonition in Graham, supra, that "[h]e who seeks to build a better mousetrap today has a long path to tread before reaching the Patent Office,"10 the plaintiff succeeds here, primarily, because the defendants have failed to rebut the § 282 presumption (35 U.S.C. § 282) attaching to the Conibear patent. Contrary to the contention of the appellees, this presumption of validity has not been weakened or destroyed in this case merely because the defendants cite Ullman as a pertinent reference not cited as relevant art in the United States Patent Office documents. Ullman bears neither physical resemblance nor utilizes the application of the same mechanics as Conibear. Although similar mechanical principles of quick trigger release may exist, this invention, as a whole, seems to bear little relevance to Conibear. Thus, the failure of the Patent Office to cite Ullman does not erode the strength of the presumption. See Shaw v. E. B. & A. C. Whiting Co., 417 F.2d 1097, 1104 (2d Cir. 1969), cert. denied, 397 U.S. 1076, 90 S. Ct. 1518, 25 L. Ed. 2d 811 (1970); Leach v. Rockwood & Co., 404 F.2d 652, 654 (7th Cir. 1968); cf. Ralston Purina Co. v. General Foods Corp., 442 F.2d 389 (8th Cir. 1971); L & A Products, Inc., supra, 365 F.2d at 86-87; American Infra-Red Radiant Co. v. Lambert Industries, Inc., 360 F.2d 977, 989 (8th Cir.), cert. denied, 385 U.S. 920, 87 S. Ct. 233 (1966).

Certainly the Patent Office cited, considered, and rejected the most pertinent prior art offered in this case, that of Mau I and Mau II. Such consideration serves to preserve the strength of the presumption as against the defendants' contrary contention. See Leach, supra, 404 F.2d at 654; Williams v. V. R. Myers Pump and Supply, Inc., 371 F.2d 192, 194 (7th Cir. 1966); Briggs v. M & J Diesel Locomotive Filter Corp., 342 F.2d 573, 576 (7th Cir.), cert. denied, 382 U.S. 801, 86 S. Ct. 11, 15 L. Ed. 2d 55 (1965); Cold Metal Process Company v. Republic Steel Corp., 233 F.2d 828, 837 (6th Cir.), cert. denied, 352 U.S. 891, 77 S. Ct. 128, 1 L. Ed. 2d 86 (1956); Mott Corporation v. Sunflower Industries, Inc., 314 F.2d 872, 877 (10th Cir. 1963).

In retrospect, many creative developments seem commonplace. The critical question under Graham is whether at the time of the invention the subject would have been obvious to one skilled in the art. The Court noted in Graham that the inquiries and tests it set out in that opinion might serve to "guard against slipping into use of hindsight * * * and to resist the temptation to read into the prior art the teachings of the invention in issue." 383 U.S. at 36, 86 S.Ct. at 703. See Shaw, supra, 417 F.2d at 1106; Columbia Broadcasting Sys. v. Sylvania Electric Prod., Inc., 415 F.2d 719, 728 (1st Cir. 1969), cert. denied, 396 U.S. 1061, 90 S. Ct. 755, 24 L. Ed. 2d 755 (1970); National Dairy Products Corporation v. Borden Company, 394 F.2d 887, 891 (7th Cir.), cert. denied, 393 U.S. 953, 89 S. Ct. 378, 21 L. Ed. 2d 364 (1968); National Connector Corp., supra, 392 F.2d at 771.

We think the remarks of Judge Learned Hand in two separate patent cases to be most appropos within the limited context of the record presented here. In B. G. Corporation v. Walter Kidde & Co., 79 F.2d 20, 22 (2d Cir. 1935), he noted:

All machines are made up of the same elements; rods, pawls, pitmans, journals, toggles, gears, cams, and the like, all acting their parts as they always do and always must. * * * But the elements are capable of an infinity of permutations, and the selection of that group which proves serviceable to a given need may require a high degree of originality. It is that act of selection which is the invention; and it must be beyond the capacity of common-place imagination.

Later, in Reiner v. I. Leon Co., 285 F.2d 501 (2d Cir. 1960), Judge Hand made the following observation concerning the role of a court in patent litigation:

The test laid down [for obviousness] is indeed misty enough. It directs us to surmise what was the range of ingenuity of a person "having ordinary skill" in an "art" with which we are totally unfamiliar; and we do not see how such a standard can be applied at all except by recourse to the earlier work in the art, and to the general history of the means available at the time. To judge on our own that this or that new assemblage of old factors was, or was not, "obvious" is to substitute our ignorance for the acquaintance with the subject of those who were familiar with it. There are indeed some sign posts: e. g. how long did the need exist; how many tried to find the way; how long did the surrounding and accessory arts disclose the means; how immediately was the invention recognized as an answer by those who used the new variant? [285 F.2d at 503-504]

Cf.Graham, supra, 383 U.S. at 36, 86 S. Ct. 684; White v. Fafnir Bearing Company, 389 F.2d 750, 754 (2d Cir. 1968).

To summarize, on review of the record in the instant case, we think that the district court's finding that the developments embodied in the Conibear patent would have been obvious to one with ordinary skill in the pertinent art must be rejected. Although through the use of hindsight the Conibear patent may have appeared obvious, we think the defendants failed to produce sufficient evidence to establish that Conibear's patent was an obvious development based upon the prior art. To invalidate a patent, the court, in evaluating the evidence, must be able to say that the differences between the prior art and the patented subject matter as a whole would have been obvious at the time of the invention to a person ordinarily skilled in the art. Graham, supra, 383 U.S. at 15, 86 S. Ct. 684. The heart of the defendants' case consisted of the testimony of one witness, admittedly completely unskilled in the trapping art, who concluded that the Conibear patent was obvious in the light of the prior art. This testimony, standing by itself, will not suffice to overcome the presumptive validity of the United States patent. This testimony stands in sharp contrast to other evidence tending to show nonobviousness of the Conibear discovery, such as inventor Lehn's expression of surprise and excitement upon learning of this new trap and other testimony showing the practical success of the trap in catching and killing fur-bearing animals. Moreover, plaintiff's testimony, in part, presents evidence of secondary considerations, such as commercial success, long-felt but unsolved needs, and failure of others, as a basis for sustaining its patents. This kind of evidence carries some weight, even though minimal in some cases. See Graham, supra, 383 U.S. at 17-18, 86 S. Ct. 684.

In Blonder-Tongue, supra, the Supreme Court observed that a defendant in a patent infringement suit must not only introduce proof to overcome the presumption of patent validity, but must also rebut whatever proof the patentee offers to bolster his claims. 402 U.S. at 335, 91 S. Ct. at 1446. The record here demonstrates that the defendants failed to sustain their burden of proof; accordingly, the validity of the Conibear patent must be sustained.

From what has been already said with regard to Lehn's development of his own invention, it is quite apparent that his trap reflected only a minor modification and improvement over the triggering device as developed by Conibear. His own testimony indicates that the Conibear invention foreshadowed his own. Accordingly, we sustain the district court's determination of the invalidity of claim 2 under the Lehn patent.

The district court rejected the plaintiff's claim for unfair competition, concluding that

Although there was evidence that some of plaintiff's informants were of the view, because of the general similarity of appearance, that plaintiff was manufacturing traps for the defendants, there was no proof that defendants were deceiving the public or attempting to do so or palming off their traps for plaintiff's traps. Defendants have not competed unfairly with the plaintiff. [312 F.Supp. at 373]

We think the evidence adequately supports this finding, and we sustain the district court's dismissal of this claim for the reasons stated in its opinion. Id.

We reverse the district court's finding of invalidity with regard to the Conibear patent, and remand for further proceedings on the issue of infringement. Within the spirit, if not the literal language, of 35 U.S.C. § 288, see Blonder-Tongue, supra, 402 U.S. at 347, 91 S. Ct. at 1452, we direct that neither party recover costs on this appeal.