Plaintiffs, a group of manufacturers of knitted goods, instituted this declaratory judgment action seeking a declaration of the invalidity of patent No. 2,-839,909 issued for a knitted fabric, method and process of manufacture to the defendant John E. Morgan on June 24, 1958. Morgan, who had made the application on May 16, 1957, assigned his patent rights to the corporate defendant, John E. Morgan Patents, Inc.
The defendants denied invalidity and asserted a counterclaim, in Count I, for damages for infringement by all but one of the plaintiffs, and, in Count II, for treble damages by reason of a group boycott by all plaintiffs in violation of the Sherman Act, 15 U.S.C. § 1.
The district court held the patent invalid. 244 F.Supp. 219 (E.D.Pa. 1965). It found that the plaintiffs had engaged in an illegal group boycott, but concluded that since the patent was invalid, the patentee suffered no damage. 244 F.Supp. 235 (E.D.Pa.1965).
This appeal followed.
On review, we are of the opinion that the patent involved is valid, except claims 14-18 insofar as they relate to a process for finishing the knitted fabric. As to those claims, it is admitted that the process described was already known in the prior art.
The patent involved, called the “Morgan Patent”, is for a knitted fabric which because of its light weight and warmth is particularly suited for use in the production of such winter garments as “thermal underwear”. It is for a broad rib fabric knitted on “weft” (circular) knitting machines as opposed to *453“warp” knitting machines and is identifiable by the multitude of air-entrapping cells which are on both sides of the fabric. It is claimed that these cells keep their shape despite prolonged wear. By providing dead air spaces, it is claimed that the fabric achieves an enhanced insulating quality.
The sides of the air-entrapping cells are formed by the longitudinal ribs of the fabric; multiple tuck stitches form the bottom and top walls of these cells. These tuck stitches are utilized to break up the valleys and ribs of the broad rib fabric into individual air-entrapping cells.1 In the words of the patent claim “when these tuck stitches are ‘knitted off’ the loops so formed are stretched between adjacent longitudinal ribs and function to draw these ribs together at spaced apart points along the length thereof. This serves to accentuate the cellular construction which is produced. At the same time, the tuck strands underlie the ribs and function to brace and maintain the height of these ribs when the fabric is stretched or subjected to prolonged wear.” Additionally, the patent teaches that by knitting a tight fabric with heavy yam this will aid the cells in keeping their three-dimensional character.
Prior to the Morgan patent the only fabric to gain consumer acceptance as possessing the “thermal” qualities of light weight and warmth in cold weather was the “Raschel” fabric earlier developed by the Navy. This fabric, however, could only be produced on warp knitting machines. Consequently, knitters such as Morgan who had weft knitting machines which made circular knit fabrics had to buy their requirements from others. Morgan thus undertook the experimentation that resulted m his patent with the hope of developing an imitation of the Raschel fabric that could be made on a circular knit machine.
Insofar-as the law of patentability is concerned, we only note the decisions of the Supreme Court in Graham v. John Deere Co., 381 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545, and United States v. Adams, 383 U.S. 39, 86 S.Ct. 708, 15 L.Ed.2d 572, both decided February 21, 1966, for exposition and application of currently effective principles. These, because of their recency and completeness, need not be reiterated here.
We turn, then, to the determination of the validity of the Morgan Patent.
Indefiniteness of the Claims
The district court found the patent to be invalid because of the indefiniteness of the claims made by the patent. The court reached this conclusion because it found that the term “heat-insulating” as used in the patent specifications and claims was a relative and comparative term which did not meet the statutory requirement.2
Regardless of the fact that the term “heat-insulating” may be one of degree, there is abundant evidence on this record taken as a whole that when one skilled in the art reads the Morgan Patent, he would be capable of producing the patented fabric and of determining the limits of the patent. As we interpret the statutory standard, this is all the inventor need show. 35 U.S.C.A. § 112.
As succinctly stated by the Second Circuit in Georgia-Pacific Corporation v. United States Plywood Corporation, 258 F.2d 124 (2 Cir. 1958), cert. den. 358 U.S. 884, 79 S.Ct. 124, 3 L.Ed.2d *454112, the requirement of 35 U.S.C.A. § 112 serves two primary purposes:
“ * * * those skilled in the art must be able to understand and apply the teachings of the invention and enterprise and experimentation must not be discouraged by the creation of an area of uncertainty as to the scope of the invention. On the other hand, the policy of the patent statute contemplates granting protection to valid inventions, and this policy would be defeated if protection were to be accorded only to those patents which were capable of precise definition. The judicial function requires a balancing of these competing considerations in the individual case.” (258 F.2d, at 136).
In performing this balancing, Georgia-Pacific teaches that:
“ * * * patentable inventions cannot always be described in terms of exact measurements, symbols and formulae, and the applicant necessarily must use the meager tools provided by language, tools which admittedly lack exactitude and precision. If the claims, read in the light of the specifications, reasonably apprise those skilled in the art both of the utilization and scope of the invention, and if the language is as precise as the subject matter permits the courts can demand no more.” (258 F.2d, at 136)
The cases cited by the plaintiffs, General Electric Co. v. Wabash Appliance Corp., 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402 (1938); United Carbon Co. v. Binney & Smith Co., 317 U.S. 228, 63 S.Ct. 165, 87 L.Ed. 232 (1942); Fisher & Porter Co. v. Brooks Rotameter Co., 107 F.Supp. 1010 (E.D.Pa. 1952), aff’d 211 F.2d 400 (3 Cir. 1954), cert. den. 348 U.S. 871, are not inapposite to the teachings of Georgia-Pacific. It is clear that each case presents a different question, and the rule is not a technical one, but one made to protect the public.
Plaintiffs’ witness Carroll Anderson testified that he believed the fabric put out by his company, Munsingwear, would infringe the Morgan Patent if the patent were upheld. We can assume that Anderson’s willingness to make such a statement is based on his ability to determine the limits of the Morgan Patent. Plaintiffs’ expert witness, Professor Shinn, also demonstrated ability to recognize the Morgan fabric. So also defendants’ expert, Robert H. Lawson, was able to clearly identify an infringing fabric.
The testimony of Professor Shinn on the issue of invention also supports a finding that one skilled in the art can read the patent so as to know that in addition to the stitch structure a knitter would utilize a tight knit and a heavy yam.3
Anderson testified that he had no difficulty in understanding the Morgan Patent and that he could identify the Morgan fabric as well as infringements.
The testimony of Mr. Lawson also illustrated that one skilled in the art can read the teachings of the patent without any reliance on the term, “heat-insulating”.
These references suffice for us to state, in summary, our conviction that the evidence shows that the patent teaches with sufficient conciseness how the invention might be put into use and also gives sufficient warning to the would-be infringer of the patent’s limits. The fact that the patent also uses a term capable of only relative definition is not fatal. See Eibel Process Company v. Minnesota & Ontario Paper Company, 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523 (1923); Chicago Pneumatic Tool Co. v. Hughes Tool Co., 97 F.2d 945 (10 Cir. 1938), cert. den. 305 U.S. 643, 59 S.Ct. 242, 83 L.Ed. 436.
*455The Prior Art
The district court concluded that the Morgan Patent was anticipated in the prior art primarily on the basis of a fabric claimed to have been made by Carroll Anderson in 1933. But assuming similarity, more is required to constitute prior art under the statute.4 The Reviser’s Notes to Section 102, 35 U.S.C.A., make clear that the interpretation of “known” by the courts assimilated it to “publicly known”. See Pennock v. Dialogue, 27 U.S. (2 Pet.) 1, 7 L.Ed. 327 (1829).
The district court concluded that there was such a public knowledge and use because garments made from this Anderson fabric had allegedly hung on display racks in the Munsingwear Designs Division and that these racks were accessible to the public. Assuming that such a use is sufficient to meet the requirements of “publicly known,” we conclude that the evidence is insufficient to support a finding of such use.
The defense of prior use is one that an alleged infringer, although free to raise, has the burden of persuasion. This burden is a heavy one, having been described by this Court as necessitating evidence that “must be so strong as to remove all reasonable doubt thereof.” Smith v. Carter Carburetor Corporation, 130 F.2d 555, 560 (3 Cir., 1942). See also American Machine & Foundry Company v. Liggett & Myers Tobacco Co., 172 F.Supp. 12, 16 (D.N.J.1959), aff’d per curiam, 272 F.2d 451 (3 Cir.). Similarly, the Supreme Court has stated that every reasonable doubt should be resolved against the one who avers this defense. Washburn & Moen Manufacturing Company v. Beat ’Em All Barbed Wire Company (“The Barbed Wire Patent Case”), 143 U.S. 275, 284, 12 S.Ct. 443, 36 L.Ed. 154 (1892); Coffin v. Ogden, 85 U.S. (18 Wall.) 120, 21 L.Ed. 821 (1874). In a more recent reappraisal, the Supreme Court, in Radio Corporation of America v. Radio Engineering Laboratories, Inc., 293 U.S. 1, 7, 8, 55 S.Ct. 928, 931, 79 L.Ed. 163 (1934) expressed itself thus:
“A patent regularly issued, * * * is presumed to be valid until the presumption has been overcome by convincing evidence of error. * * * [It is the] common core of thought and truth, that one otherwise an infringer who assails the validity of a patent fair upon its face bears a heavy burden of persuasion, and fails unless his evidence has more than a dubious preponderance.”
Plaintiffs’ attempt to meet this burden, by proving display of the Anderson fabric, rests upon the testimony of Anderson himself, and that of Ruth K. Friedman, Jenny M. Lindh, and Mary Przybylka, who were employees of Munsingwear at the time these garments were allegedly displayed. The district court in its opinion recognized that the disposition testimony of the three employees of Munsingwear was entitled to no probative weight (244 F.Supp., at 224).
This leads us to recognize that the asserted “public use” rests solely upon the oral testimony of Anderson. The fact that plaintiffs seek to carry their burden solely on the basis of oral testimony is not necessarily fatal to their case. However, because of the nature of this testimony, this court is under a duty to scrutinize it very carefully. The rationale behind such close scrutiny was cogently stated in “The Barbed Wire Patent Case”, supra, 143 U.S. at 284-285, 12 S.Ct. at 447:
“We have now to deal with certain unpatented devices, claimed to be complete anticipations of this patent, the existence and use of which are proven only by oral testimony. In view of the unsatisfactory character of such testimony, arising from the forgetfulness *456of witnesses, their liability to mistakes, their proneness to recollect things as the party calling them would have them recollect them, aside from the temptation to actual perjury, courts have not only imposed upon defendants [in infringement suits] the burden of proving such devices, but have required that the proof shall be clear, satisfactory, and beyond a reasonable doubt. Witnesses whose memories are prodded by the eagerness of interested parties to elicit testimony favorable to themselves are not usually to be depended upon for accurate information. The very fact, which courts as well as the public have not failed to recognize, that almost every important patent, from the cotton gin of Whitney to the one under consideration, has been attacked by the testimony of witnesses who imagined they had made similar discoveries long before the patentee had claimed to have invented his device, has tended to throw a certain amount of discredit upon all that class of evidence, and to demand that it be subjected to the closest scrutiny. Indeed, the frequency with which testimony is tortured, or fabricated outright, to build up the defense of a prior use of the thing patented, goes far to justify the popular impression that the inventor may be treated as the lawful prey of the infringer.”
Viewing Anderson’s testimony in this light, we cannot say that plaintiffs’ evidence sustains their burden. Three factors must be noted in support of this conclusion: (1) the length of time that has elapsed between the event and the date of trial: Anderson was testifying to a recollection that is, from the date of his depositions, approximately twenty-five years old and concerns a matter which was not then directly related to his particular line of work, (2) there is a lack of any corroborating evidence: swatches of the fabric with a 1933 date were introduced into evidence, but this is not proof that the material was made into garments or displayed. No company records were produced nor was any witness whose employment with Munsingwear at that time involved the making and displaying of garments. In other instances the latter factor alone has been sufficient for a finding of failure to carry the burden of proof. Jepson v. Egley, 231 F.2d 947, 953, 43 CCPA 853 (1956); American Machine & Foundry Company v. Liggett & Myers Tobacco Co., supra; Radio Corporation of America v. Philco Corporation, 201 F.Supp. 135, 149 (E.D.Pa.1961), aff’d per curiam, 309 F.2d 397 (3 Cir. 1962). Moreover, Anderson cannot be considered á disinterested witness. He admitted that it was his belief, as well as that of Munsingwear officials, that the thermal wear fabric they were then producing infringed the Morgan Patent, that they were interested in nullifying the Morgan Patent, and that for these reasons they were assisting plaintiffs in this proceeding.
Further, Anderson’s testimony as to the manufacture and display was not, as we view it, so clear that he was testifying from actual recollection on this score rather than from a reasoned belief.
In these circumstances, it cannot be said with sufficient conviction to justify negation of a patent that the evidence sustains the defense of prior anticipation. Cf. General Radio Company v. The Superior Electric Company, 321 F.2d 857 (3 Cir. 1963), cert. denied, 376 U.S. 938, 84 S.Ct. 793, 11 L.Ed.2d 659, wherein we specifically found corroboration of oral testimony, and cited Smith, Ex’r v. Hall, 301 U.S. 216, 57 S.Ct. 711, 81 L.Ed. 1049 (1937), indicating the decisive significance of corroboration.
Lack op Invention
The essence of the Morgan invention is the air-entrapping cells, which keep their shape even after prolonged wear, on both sides of a broad rib circular knit fabric. This is achieved through multiple tucking, utilizing heavy yarn and tight stitching. Such cells are present in the “Raschel” fabric, but Morgan claims to be the first to produce this effect in a circular knit fabric. The district court concluded this result would have been ob*457vious at the time to one of ordinary skill in the art.
The standard of invention was codified in 1952 and is set out in 35 U.S.C.A. § 103.5 This section was not meant to establish a new standard but merely to codify prior case law. This is the conclusion reached by the Supreme Court in Graham v. John Deere Co., supra. The court there reasoned that Congress had hoped by codifying the standard to emphasize “non-obviousness” as the operative test of the section, rather than the less definite “invention” language of Hotchkiss v. Greenwood, 11 How. 248, 13 L.Ed. 683 (1850).
Graham also makes clear that the “flash of genius” theory has no place in the requirement of invention. At footnote 7 of the opinion, the Court indicated that the theory originated from a misunderstanding of its opinion in Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 62 S.Ct. 37, 86 L.Ed. 58 (1941), and that Cuno merely required that the device, and not the method, display invention. This is the view we entertained in R. M. Palmer Company v. Luden’s Inc., 236 F.2d 496 (3 Cir. 1956). There we said that patentability shall not be negatived by the manner in which the invention was made but that rather the court must look at the end product to see whether a step beyond the prior art has been taken.
Admittedly, Morgan was not a skilled knitter, and many of the steps and experiments he performed were necessitated by this lack of skill. Regardless of the nature of his struggles, it is the end product that matters, and if this product advances the art a significant step, there is invention so long as that step is not obvious to those of ordinary skill in the art. The district court ignored the end product to concentrate on the inventor’s technique. This can be observed in the court’s own language: “It is the means by which the idea is reduced to practice which must contain the elements of inventiveness as distinguished from mechanical skill and know how.” 244 F.Supp. at 225. This strikes too closely to the “flash of genius” theory.
Examining the prior art, the district court concluded that the Morgan Patent was obvious at the time to one of ordinary skill in the art of knitting. The state of the prior art as found by the district court (Findings No. 42 and No. 43) consisted of the Raschel fabric, a “2 X 2 rib knit fabric with at least two tuck stitches on the front and no tuck stitches on the back” and “a 2 X 2 rib knit fabric with pockets or cells formed on one side by plural tuck stitches and with one tuck stitch on the other side.” The court then concluded that to one of ordinary skill, given this prior state of the art, it would be obvious how to make the Morgan Patent by utilizing multiple tucking in such a way as to form cell pockets on both sides of the rib knit fabric. We cannot agree with this conclusion and find it to be error insofar as it is based on the testimony of Carroll Anderson and Professor Shinn.
While Anderson testified that given the prior art the Morgan Patent is obvious, his own experiments in 1952 and 1957 were unsuccessful and his own company decided to copy the Morgan fabric. There is also his deposition testimony, which was, however, denied by Anderson at the trial, that when he claimed the Morgan idea became obvious to him, he was already familiar with the Morgan Patent.
Professor Shinn, a recognized expert in the knitting field, with considerably more' *458skill than the ordinary knitter, was unable to offer one constructive suggestion to inquiries concerning the manufacture of a Raschel type of fabric on a circular machine.
Here are two skilled knitters testifying to obviousness where one of them failed to produce the result through his own experimentation and the other did not offer constructive suggestions toward a solution. Where before the disclosure of a patent, two experts 6 are unable to foresee the solution provided by the patent, an inference of nonobviousness is not without justification. See Honolulu Oil Corporation v. Shelby Poultry Company, 293 F.2d 127, 131 (4 Cir. 1961). Cf. United States v. Adams, supra.
The evidence indicates that given the state of the prior art as found by the district court the Morgan invention is not obvious to one of ordinary skill in the knitting art, and that the invention represents a significant advance in the knitting field.
We do not rest this conclusion merely on the negative inferences to be drawn from the testimony of plaintiffs’ experts, but also note the commercial success of the Morgan fabric, the long felt need for the invention, and the availability of the tools to produce the invention and the failure of others to achieve the result. The relevancy of such considerations is acknowledged in the statement of the Supreme Court in Graham v. John Deere Co., supra:
“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. See Note, Subtests of ‘Nonobviousness,’ 112 U.Pa.L.Rev. 1169 (1964).”
Such consideration failed to tip the scales in favor of patentability in a companion case to Graham (86 S.Ct. 684, 15 L.Ed.2d, at 561) (plastic finger sprayer), but in United States v. Adams, supra, these factors were used to support a finding of nonobviousness (86 S.Ct. 708, 15 L.Ed.2d, at 580-581). See also, Hartford National Bank & Trust Company v. E. F. Drew & Co., 237 F.2d 594, 596 (3 Cir. 1956); Reiner v. I. Leon Co., 285 F.2d 501, 504 (2 Cir. 1960), cert. den. 366 U.S. 929, 81 S.Ct. 1649, 6 L.Ed.2d 388 (1961); Allen v. Standard Crankshaft & Hydraulic Company, 323 F.2d 29, 34 (4 Cir. 1963); Copease Manufacturing Co. v. American Photocopy Equipment Co. & Poray, Inc., 298 F.2d 772, 781 (7 Cir. 1961).
There is abundant and uncontradicted evidence, much of it supplied by plaintiffs’ own witnesses, that while the tools existed to produce the Morgan fabric for a number of years and there was a real need for such a fabric no one came up with a commercially successful product until Morgan. We have already considered the evidence relating to the state of the prior art and the failure of two skilled knitters to arrive at or suggest in any manner a solution. That there was a need for a solution is well documented. Circular knit underwear mills were forced to buy their requirements from Raschel manufacturers and let their own machines stand idle. Thus there was great interest in the 1950’s to develop a circular knit imitation of the Raschel fabric. Finally, there is the overriding commercial success of the Morgan fabric. The fabric was desired by the underwear salesmen and copied by competing manufacturers. In addition, the Navy’s proposed specifications for circular knit waffle type fabric is based on the Morgan fabric. All of this is very strong evidence of invention entitling the defendants to a valid patent in the face of a lack of any convincing testimony, on the part of those skilled in the art, of obviousness.
Our conclusion as to the validity of the Morgan Patent will require the district court, on remand, to dispose of the issue *459of infringement raised in defendants’ counterclaim, which it did not reach.
The Gkoup Boycott
Upon examination of the record, we agree with the fact-finding of the district court, 244 F.Supp. 235, 238, 239, that the twelve member plaintiff group did more than merely enter into an agreement to challenge the validity of the Morgan Patent; that the “group was formed not only for purposes of bringing suit, but also for purposes of refusing to negotiate with Morgan for licenses”, and that under the terms of the agreement “[t]he freedom of each plaintiff to deal freely with Morgan was restrained by the requirement of giving notice” of any dealing with Morgan to all the members of the group. We further agree with the district court’s ultimate finding of fact that the plaintiffs engaged in a “group boycott”, and its conclusion of law “that plaintiffs have restrained trade in violation of § 1 of the Sherman Act.” Radiant Burners, Inc. v. Peoples Gas Light & Coke Co., 364 U.S. 656, 659-660, 81 S.Ct. 365, 5 L.Ed.2d 358 (1961); Klor’s Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 212, 79 S.Ct. 705, 3 L.Ed.2d 741 (1959); Fashion Originators’ Guild of America, Inc. v. Federal Trade Commission, 312 U.S. 457, 61 S.Ct. 703, 85 L.Ed. 949 (1941).
Since the district court based its denial of damages on its ruling that the patent is invalid, our holding that the patent is valid will require it, on remand, to reach the issue of damages with respect to this phase of the litigation.
For the reasons stated, the Final Decree of the district court will be reversed, and the cause remanded with directions to proceed in accordance with this Opinion.
. In this respect, Fig. 3 of the drawings of the Morgan Patent is incorrect as it shows no tuck stitches as described or claimed. Pre-trial Order, Admitted Pact No. 11.
. The statute, as here relevant, reads:
“The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.” 35 U.S.C.A. Sec. 112.
. “Q. What are the factors which govern whether it is a tight cell or a loose cell?
“A. Tightness of the knit and the size of the yarn are factors.
“Q. Would you say in your judgment that that would be something that any knitter would know, if be wanted to tighten up the tuck stitch he could do it by knitting tighter or using heavier yarn?
“A. Well, I believe so. I am confident that is true.”
. “§ 102 Conditions for patentability; novelty and loss of right to patent “A person shall he entitled to a patent unless—
“(a) the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for patent, if if if f>
. Ҥ 103. Conditions for patentability, non-obvious subject matter
“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. July 19, 1952, c. 950, § 1, 66 Stat. 798.”
. There was also evidence that a third skilled knitter, Mr. Paul, was asked to find a solution, bnt it is not clear exactly how far he had proceeded in his work.