Full opinion text
MAJOR, Circuit Judge. These appeals are from a final judgment in favor of plaintiff, entered August 26, 1959, after accounting, for infringement of claims 18, 20, 22 and 23 of patent No. 2,043,960 on a welding flux (sometimes referred to as Jones or the patent in suit). Judgment was rendered in the amount of $3,010,077.16, apportioned as follows: Award of damages based on reasonable royalty.. $1,361,891.25 Interest at 6% per annum from June 30, 1950, the date of the last infringement ................ 748,185.91 Additional damages..... 900,000.00 Plaintiff, Union Carbide Corporation (hereinafter referred to as Union Carbide) appeals on the basis that the judgment is grossly inadequate (appeal No. 12797). Defendants, Graver Tank & Mfg. Co., Inc. (hereinafter referred to as Graver) and The Lincoln Electric Company (hereinafter referred to as Lincoln) appeal on the basis that it is excessive (appeal No. 12798). This action was commenced October 1, 1945, by Linde Air Products Company (hereinafter referred to as Linde) against the present defendants. Linde at that time was a subsidiary of Union Carbide and was subsequently merged into that corporation, which thereupon became the plaintiff. Lincoln (hereinafter called the defendant) since 1907 has been engaged in the manufacture of electric arc welding equipment and supplies and from the beginning has conducted the defense on behalf of both defendants. Graver has a manufacturing plant at East Chicago, Indiana, where defendant’s 660 flux, the infringing flux, was used for welding operations during the period of infringement. (The patented flux was called Unionmelt; the infringing flux, Lincolnweld.) The litigation has been before one court or another from the time it was commenced in 1945, to the present time. Linde in its original action sued on sixteen claims to a process of electric welding and on seven claims to a composition or flux for use in the process. The case was heard by Honorable Luther M. Swygert, who held claims 18, 20, 22 and 23 valid and infringed and composition claims 24, 26 and 27 invalid. He also held all process claims invalid. Linde Air Products Co. v. Graver Tank & Mfg. Co., D.C., 86 F.Supp. 191. Upon appeal by both parties, this Court affirmed as to claims 18, 20, 22 and 23, but held valid and reversed the District Court as to all process claims and composition claims 24, 26 and 27. 167 F.2d 531. The Supreme Court agreed with Judge Swygert in toto, both as to claims 18, 20, 22 and 23, which he had held valid and infringed, and as to the other composition and process claims which he had held invalid. 336 U.S. 271, 69 S.Ct. 535, 93 L.Ed. 672. The valid composition claims were held infringed by Lincoln in the manufacture and sale of its 660 flux. The Supreme Court, upon petition of Graver, granted a rehearing on the issue of infringement, limited to a consideration of the doctrine of equivalents. 337 U.S. 910, 69 S.Ct. 1046, 93 L.Ed. 1722. The majority of the Court (two Justices dissenting) held that although Lincoln did not utilize the material called for by the claims, the 660 flux infringed by application of that doctrine. 339 U.S. 605, 70 S. Ct. 854, 94 L.Ed. 1097. After the 660 flux was adjudged an infringement by the District Court, Lincoln made and marketed four new fluxes, the 760, 770, 780 and 840 (known as the 700-series fluxes). Union Carbide, on the premise that these new fluxes infringed and, therefore, violated the District Court’s decree, brought an action to have defendants adjudged in contempt. This proceeding was heard by Honorable Charles A. Dewey, D.C., 106 F.Supp. 389, sitting by special assignment. He held that the new fluxes infringed and that defendants were in contempt. This Court, on the basis of the doctrine of “File Wrapper Estoppel,” held that Lincoln’s new fluxes did not infringe and reversed the decision of Judge Dewey. 196 F.2d 103. The Supreme Court denied certiorari, 343 U.S. 967, 72 S.Ct. 1059, 96 L.Ed. 1363, and petition for rehearing, 344 U.S. 849, 73 S.Ct. 6, 97 L.Ed. 660. We approach the issues argued on these appeals with knowledge of the careful and extensive consideration which has been accorded to the parties and their respective contentions in the Court below. The accounting proceeding was referred to Honorable Henry R. Sackett, as Special Master, by order dated February 10, 1953. The pre-trial hearing and proceedings extended over a period of three and one-half years. The hearing before the Master occupied a full month, September 25 to October 25, 1956. Twenty witnesses were examined and cross-examined, and the stenographic transcript of the trial came to 3,045 pages. Seven hundred and twelve documentary exhibits were put in evidence. Motion pictures, welding demonstrations and other forms of evidence were presented. Records and exhibits from the previous hearings were incorporated by reference. After the trial, both sides submitted printed briefs to the Master, together with proposed findings of fact and conclusions of law. Oral argument before the Master on these proposed findings occupied four days. The Master, after having the case under consideration for almost a year, submitted a draft report to counsel. Numerous objections were interposed to the Master’s proposed findings and conclusions. Union Carbide in the main acquiesced in the proposed findings but interposed objections to certain of the Master’s conclusions. On the other hand, Graver objected to many of the proposed findings, as well as to some of the conclusions. After a hearing on the objections to the proposed report, the Master made a number of changes in his proposed findings and conclusions of law. On June 4, 1958, the report was signed in printed form and filed with the District Court. Objections to the report were argued before Judge Swygert for two days, at the conclusion of which the Court adopted the Master’s findings of fact but modified his conclusions of law in certain respects. Union Carbide in its brief states, “It is not likely that any other matter has had more thoughtful and thorough study and attention than the Master’s findings in this record.” With this statement we agree. In addition to the Master’s outstanding work, we are not unmindful of the fact that Judge Swygert, who heard and passed upon objections to the Master’s report, has been with this case from its inception in 1945, and has demonstrated a clear understanding of the many difficult and complicated issues which have arisen during the course of the litigation. Thus, we are possessed of an acute awareness of the force and effect imposed upon a reviewing court by Rule 52 (a) of the Rules of Civil Procedure, 28 U.S.C.A. Both parties disclaim any intent to attack directly the findings of the District Court; in fact, Union Carbide expressly accepts them. In its brief Lincoln states, “The evidentiary facts in this case are undisputed. The factual issues in this appeal relate to conclusions —i. e,, inferences of ultimate fact— drawn by the District Court from underlying facts which of themselves are not in controversy.” Did Lincoln Copy and Thereby Become a Conscious and Wilful Infringer? Prior to a consideration of the issues argued by the parties in their separate appeals, it appears appropriate to consider the Master’s finding, approved by the Court, that Lincoln’s 660 flux “was copied from the teachings of the patent.” Based on this finding, the Court characterized Lincoln as a conscious and wilful infringer, notwithstanding that the Master had refused on request to do so. We shall consider this issue in the beginning for the reason that it permeates numerous of the issues raised on both appeals, particularly by Lincoln in appeal No. 12798; in fact, it is controlling as to some of such issues. In treating the issue we must start, of course, with the premise that Lincoln by the manufacture and sale of its 660 flux infringed the four patent claims which remained in suit. That fact, as already shown, has been adjudicated by the courts up to and including the Supreme Court. The question, however, as to whether Lincoln copied its infringing flux from the teachings of the patent and thereby became a conscious and wilful infringer has not been adjudicated; in fact, it has not been previously in issue. This distinction must be kept in mind in appraising the relevancy of what has heretofore been shown and decided. Another factor is that the issue must be determined in the light of circumstances in existence at the time Lincoln commenced the manufacture and sale of its infringing flux rather than from those after issues have been litigated and determined. After much thought, we find ourselves in the dark as to what it was that Lincoln copied. The Supreme Court on the second appeal, 339 U.S. 605, 610, 70 S.Ct. 854, 857, stated: “The patent under which Union-melt is made claims essentially a combination of alkaline earth metal silicate and calcium fluoride * and Lincolnweld’s composition “substitutes silicates of calcium and manganese — the latter not an alkaline earth metal — for silicates of calcium and magnesium.” Thus, the claims did not call for the use of magnesium, while the infringing flux consisted of more than 90% of manganese silicate, which is not an alkaline earth metal silicate specified in the claims. Certainly under such circumstances no one would contend that the composition of the infringing flux was copied from the claims in suit. The Supreme Court in the last cited case so recognized. It stated (339 U.S. at page 607, 70 S.Ct. at page 855): “If accused matter falls clearly within the claim, infringement is made out and that is the end of it.” The Court denied the application of that theory but concluded that the methods by which the compositions were employe were “identical in operation and produce the same kind and quality of weld.” It therefore evoked the doctrine of equivalents and held that Lincoln’s flux infringed. In doing so, it appears that the Court relied upon what was disclosed by the specifications rather than that specified in the claims. This was the view held by two Justices who separately dissented. Justice Douglas in his dissent stated (339 U.S. at page 618, 70 S.Ct. at page 861): “The claims of the patent are limited to a flux ‘containing a major proportion of alkaline' earth metal silicate.’ Manganese silicate, the flux which is held to infringe, is not an alkaline earth metal silicate. It was disclosed in the application and then excluded from the claims. It therefore became public property. See Mahn v. Harwood, 112 U.S. 354, 361 [5 S.Ct. 174, 178, 6 S.Ct. 451, 28 L.Ed. 665]. It was, to be sure, mentioned in the specifications. But the measure of the grant is to be found in the claims, not in the specifications. Milcor Steel Co. v. [George A.] Fuller Co., 316 U.S. 143, 145, 146 [62 S.Ct. 969, 970, 971, 86 L.Ed. 1332]. The specifications can be used to limit but never to expand the claim. See McClain v. Ortmayer, 141 U.S. 419, 424 [12 S.Ct. 76, 77, 35 L.Ed. 800].” If Lincoln had copied its flux from the patent, that, as the Supreme Court said, would have been “the end of it,” and there would have been no occasion to resort to the doctrine of equivalents. That doctrine rests upon the premise that the infringer made something different. It strikes us as an anomaly to find that Lincoln copied from the patent but that infringement was found only by application of the doctrine of equivalents. The Court, in Graham v. Cockshutt Farm Equipment, Inc., 5 Cir., 256 F.2d 358, 359, stated: “Infringement, exists only if the patentee’s arrangement is copied, or if, by use of equivalents, the same arrangement is used except that for some element of it equivalents are substituted.” In our view, the word “copied” as used to describe Lincoln’s conduct is a misnomer. It is a conclusion arrived at by process of reasoning with which we do not agree. Moreover, other findings of the Master, subsequently shown, militate strongly against the finding that Lincoln copied. In any event, we do not think it of vital importance whether the finding is approved or rejected because, if approved, it does not under the circumstances support a conclusion that Lincoln was a conscious and wilful infringer. As stated, the Master refused to so conclude and we think, for reasons subsequently shown, the Court erred in holding that it was. Lincoln in developing its infringing flux acted upon the opinion of its patent counsel, Honorable John F. Oberlin, admitted on all sides as an experienced and highly competent lawyer in the field of patent law. Oberlin investigated many process, equipment and flux patents in connection with a “dual-arc process” then being developed by Lincoln. He advised Lincoln that pre-fusing the ingredients of a flux was an old step and that Lincoln could pre-fuse any flux without infringing, so long as it did not contain a major proportion of alkaline earth metal silicate as demanded by the claims. Long before the 660 was first made, he advised Lincoln extensively as to what it could and could not do in respect to welding fluxes. After the 660 had been tested at Graver and before it was placed on the market, Lincoln submitted it to Oberlin to obtain his specific opinion as to whether it could be sold without infringing the patent. By a written opinion dated December 8, 1942, after careful study of plaintiff’s patent and the prior art, Oberlin advised Lincoln that plaintiff’s asserted patent monopoly over the automatic submerged arc welding process was void, that the broad flux claims were void and that Lincoln’s 660 flux did not infringe any of the remaining claims. Oberlin’s opinion was vindicated by the courts in all respects except that pertaining to the four claims in suit, which the courts held valid and infringed on the doctrine of equivalents. Even as to these claims his opinion that they were not infringed was agreed to by two dissenting Justices of the Supreme Court. The above sketchy statement as to Lincoln’s relation with its patent counsel and its reliance upon his opinion is fully recognized by the Master in his report. Among other things, he found: “After careful study of the plaintiff's patents and the prior art, counsel advised Lincoln that plaintiff’s asserted patent monopoly over the submerged arc automatic welding process was void, that the broad composition claims 24 and 26 of the patent were void, and that the 660 flux did not infringe the remaining claims of the patent. Counsel expressed the opinion that Lincoln 660 did not infringe the patent in suit because it was not an alkaline earth metal silicate flux and that the valid claims of the patent in suit encompassed only alkaline earth metal silicate fluxes. Lincoln 660 flux is a manganese silicate flux and is not an alkaline earth metal silicate flux. Counsel’s advice to Lincoln with respect to the invalidity of plaintiff’s asserted monopoly over the submerged arc automatic welding process and with respect to the invalidity of the broad composition claims 24 and 26 was correct and was sustained by the Courts. His advice with respect to the 660 flux proved to be in error by virtue of the Courts applying the doctrine of equivalents and holding the 660 flux did infringe claims 18, 20, 22 and 23 of the patent in suit.” The Master further found: “The defendant Lincoln relied on the opinion of Mr. Oberlin that the plaintiff’s composition claims were either void or not infringed by the Lincoln 660 by virtue of its being a manganese silicate flux and not an alkaline earth metal silicate flux in major proportion as encompassed by the claims that Mr. Oberlin considered to be valid and in so relying on said opinion thought lack of infringement by the Lincoln 660 would most likely be established by virtue of the Lincoln 660 being outside the scope of any valid claims of plaintiff’s patent.” (Italics supplied.) That the issue of infringement was highly doubtful and controversial was also recognized by the Master. He found: “The question of infringement was sufficiently doubtful that the Supreme Court granted a rehearing on that issue, and two justices dissented from the final judgment. Plaintiff ‘ considered the question whether a manganese silicate flux was the equivalent of an alkaline earth metal silicate flux so doubtful that even after the District Court, the Court of Appeals, and the Supreme Court’s first decision had held it was an equivalent, plaintiff nevertheless withheld payment of royalties to the inventors on the Union-melt 50, a manganese silicate flux, until after the second decision of the Supreme Court.” Union Carbide attempts to escape what ordinarily might be expected to follow from Lincoln’s activity and conduct based upon the good faith advice of its counsel, through the loophole that Lincoln failed to advise its counsel that its flux was copied from the teachings of the patent. This theory appears to have been embraced by the Master and the District Court. The Master, after finding that Oberlin was a competent and conscientious lawyer and that his opinion was rendered in good faith, found: “The defendant Lincoln did not supply Mr. Oberlin with information as to the work done by the Lincoln engineers in creating and compounding the Lincoln 660 flux or the fact that the Lincoln 660 was not the result of independent research or experimentation but was copied from the teachings of the patent. * * * Had Lincoln disclosed to its counsel that the 660 ñux was copied from the teachings of plaintiff’s patent the opinion could, and probably would have, contained an admonition that a copyist was more likely to be found an infringer by the Courts than one infringing as a result of independent research.” The Court went further and stated in effect that Lincoln’s failure to advise counsel that it copied showed that it was aware of possible infringement because “otherwise it would not have sought its patent counsel’s advice about that possibility.” We have difficulty in comprehending this reasoning. Lincoln denies and has always denied that its flux was copied from the patent. Notwithstanding, it appears to be the theory that Lincoln should have advised its counsel that it had done so. It is true that Lincoln not only procured a copy of Union Carbide’s patent but also a quantity of its flux for the purpose of analysis. This, so we think, it had a right to do. As stated by this Court in Atkins et al. v. Gordon, 86 F.2d 595, 596: “One may legitimately study the patent and microscopically examine the language of the claim in order to make a product which will serve the same purpose and yet avoid infringement of the patent. Such a right is the logical, beneficial result which was sought through the adoption of the comprehensive patent system of our government.” The boundaries of the monopoly which Union Carbide acquired by its patent were defined and limited by its claims. All outside that boundary was in the public domain. We think it does not detract from the good faith of a competitor for him to procure and examine a patent and its claims, as well as the subject matter thereof, in order to ascertain what is open to the public. Moreover, no inference of bad faith can be drawn from the fact that Lincoln procured an opinion from a competent patent counsel. If anything, it shows that Lincoln was making a good faith effort to avoid infringement. The Master in a rather mild fashion appears to so recognize. Referring to Lincoln’s preparation of its 660 flux, he found: “By making the substitution disclosed in plaintiff’s patent of an ingredient (manganese silicate) concerning which Lincoln had independent knowledge of its usefulness in welding fluxes and then proceeding to prepare the 660 flux in accordance with the teachings of the patent, Lincoln sought to stay outside the claims of the patent and to avoid literal infringement of the patent and hoped and thought it most likely, according to Mr. Oberlin’s opinion, that in so doing the 660 flux would be found non-infringing.” (Italics supplied.) Furthermore, it should be kept in mind that the legal opinion upon which Lincoln acted was rendered by its counsel on a study of the patent (also other patents in the prior art) and an analysis of Lincoln’s flux, subsequently held to infringe. It was upon this basis that he rendered the opinion that there was no infringement. Neither the Master nor the Court indicates that the opinion of Oberlin would have been different on the question of infringement had he been advised by Lincoln that which it has always denied, that is, that it copied its composition from the teachings of the patent. As already noted, the Master does find that if Oberlin had been notified that Lincoln’s composition was copied, he probably would have included in his letter an admonition “that a copyist was more likely to be found an infringer by the Courts than one infringing as a result of independent research.” Considering the fact that Oberlin had all the necessary information to render an opinion, it is purely a matter of speculation as to what he might have added had he been informed by Lincoln that its composition was copied. Certainly there is no reason to believe that his opinion on the question of infringement would have been different. The admittedly good faith opinion by Oberlin, and more particularly Lincoln’s good faith in acting in accordance therewith, is strongly buttressed by evidence which was before the Court but not introduced at the hearing before the Master. It consists of representations made to the Patent Office in applications for patents by Edwin A. Clapp and W. D. Miller, which resulted in the issuance of No. 2,200,737 to Clapp and No. 2,308,194 to Miller. Both of these applications were prepared by the same solicitors who prepared the application for the patent in suit, and both patents are now owned by Union Carbide. Both the Clapp and Miller patents issued on the premise that they disclosed patentable improvement over the prior art. The patent in suit was cited as such and material representations were made under oath to the Patent Office in an effort to avoid its teachings. In Clapp’s application he, who was a welding engineer in charge of Union Carbide’s research laboratories, stated, “While this reference [referring to Jones] broadly suggests that manganese oxide may be present, there is clearly no teaching for what purpose or in what amounts it is to be added.” There was submitted in support of Clapp’s application an affidavit by James M. Keir, who had been engaged for fifteen years in the design, testing and supervision of welding equipment. He stated that he was familiar with the progress and status of the art, that Clapp’s application described “the use of an alkaline earth metal silicate welding composition containing manganese oxide within the range of 4% to 16% of the composition,” and that the “Clapp invention contributes a novel and unobvious advance in the welding art not apparent from the teaching of U. S. patent No. 2,043,960 [Jones].” The Miller patent relates to a process of electric welding by means of a prefused welding composition containing from 30% to 65% of manganese oxide, 9% to 40% of alumina, and the remainder principally silicate. An affidavit by Clapp was submitted in support of the Miller application, in which he stated that when he filed his application which matured into patent No. 2,200,737, he was of the opinion that in welding sulfur-banded steel by a process of the type disclosed in patent No. 2,043,960 (Jones), using a manganese-containing welding medium, “the welding conditions were adversely affected if the manganese percentage was above 16% calculated as MnO, and it is surprising to me in view of my own work in connection with the problems encountered in welding sulfur-banded steel that the use of the welding medium defined in the above identified application produces entirely satisfactory results even though the manganese content is far higher than I had thought permissible.” He further stated, “It is my opinion based upon my experience in the type of welding disclosed in patent No. 2,043,960 [Jones] that the result obtained by the use of the welding medium disclosed in the above identified application could not have been predicted from the disclosures of the cited references and that the welding composition described possesses advantages over other compositions which render its use commercially attractive.” Thus, in résumé, the skilled representatives of Union Carbide told the Patent Office, referring to the Jones patent, that “there is clearly no teaching for what purpose or in what amounts it [i. e., manganese] is to be added.” In fact, Clapp found out only “after a great deal of experimentation” that it would be possible to use as much as 4% to 16%, and this, he and his colleagues said, was “surprising, new and unobvious,” and was “not apparent from the teaching of U.S. patent No. 2,043,960 [Jones].” In view of this appraisement by Union Carbide’s experts, it is paradoxical that Lincoln’s infringing flux with its 95% manganese silicate has been or could on any logical basis be held to have been copied from such teachings. It is suggested that these representations made to the Patent Office are immaterial as they bear only upon the issue of infringement which has been adjudicated adversely to Lincoln. True, that issue is no longer open but, even so, it is an interesting, though futile, matter for speculation as to what the courts would have decided on the issue of infringement if they had been favored with the appraisement which Clapp and Miller later placed upon the teachings of the Jones patent. The representations thus made, however, are relevant because they not only fortify the soundness of the opinion which Lincoln obtained from its counsel but they come close to demolishing any basis for a finding that Lincoln copied its infringing flux from the teachings of the patent and thereby became a conscious and wilful infringer. Another facet of the argument relied upon in support of the thesis that Lincoln’s infringing flux was copied from the teachings of the patent arises from the Master’s finding “that the 660 flux was not the result of independent research or experimentation.” The Supreme Court in its second Graver opinion, 336 U.S. 271, 276, 69 S.Ct. 535, 538, stated: “The petitioners introduced no evidence to show that their accused flux was derived either from the prior art, by independent experiment or from any source other than the teachings of the patent in suit.” Lincoln in the instant proceeding, in response to this observation so we assume, introduced a large volume of testimony, not to disprove infringement but to show that its infringing flux was not copied from the teachings of the patent and that it therefore was not a conscious and wilful infringer. In this connection the Master found: “The defendant Lincoln had knowledge independent of plaintiff’s patent that manganese silicate was a useful ingredient for welding fluxes. * * * The defendant Lincoln in making its 660 flux substituted manganese silicate for the alkaline earth metal silicates described in the claims of the patent, and with this substitution then proceeded to prepare the 660 flux in accordance with the teachings of the patent.” Thus, it is recognized that Lincoln had knowledge that manganese silicate, of which its flux was largely composed, was a useful ingredient and substituted it for the alkaline earth metal silicates described in the claims. This, however, according to the opinion of its counsel, Lincoln had a right to do. The fact that such substitution was later held by the courts to be an equivalent and therefore an infringement does not reflect unfavorably on the good faith opinion rendered by counsel; neither does it justify branding Lincoln as a conscious and wilful infringer. Even the Master appears to have equivocated on the issue of good faith. He did not find that Lincoln acted in bad faith or that good faith was lacking ; instead he found that Lincoln “did not act in wholly good faith belief that it was acting within its legal rights as to the 660 flux.” (Italics supplied.) The argument presented here is permeated with the idea that because “a parade of judges” has adjudicated Lincoln an infringer, it follows that such infringement was conscious and wilful. Such thought is beside the issue as to whether Lincoln honestly and in good faith believed that its 660 flux was not an infringement. We repeat the finding of the Master: “Lincoln sought to stay outside the claims of the patent and to avoid literal infringement of the patent and hoped and thought it most likely, according to Mr. Oberlin’s opinion, that in so doing the 660 flux would be found non-infringing.” Bad faith should not be attributed to Lincoln because it contested and lost on the issue of infringement. Many a legal battle has been fought honestly and in good faith, but lost. Union Carbide has shared in that experience in the instant prolonged litigation. It attempted all the way to the Supreme Court to sustain the validity of sixteen process and seven composition claims. It lost as to all of the former and three of the latter, attempted without success to obtain in adjudication that Lincoln was an infringer because it made and marketed four new fluxes, known as the 700-series. In these matters, as well as in others, it no doubt acted under the advice and guidance of its experienced and competent counsel. But the fact that it lost does not, in our view, raise an inference that it or its attorneys acted in bad faith. Neither do we think on this record that Lincoln can properly, be charged with bad faith or characterized as a conscious and wilful infringer when it acted under the advice of equally experienced and competent counsel. It Another thought in connection with the good faith issue arises from Judge Swygert’s observation that Lincoln was “aware of possible infringement.” We think the observation is valid but we also think it could be made in all instances where there is a question as to what a patent encompasses. In fact, we suppose awareness of the possibility of infringement is always present, or should be, when the manufacture and sale of a new article or product is commenced. It is for that reason that competent patent counsel is called upon for advice and, as shown, that was the course pursued by Lincoln. Union Carbide in its brief states, “To say at this stage that infringement was doubtful is really ‘whistling by the graveyard.’ ” This sage observation obviously is true at the present time; however, we are now concerned with Lincoln’s situation as it was some eighteen years ago. If its foresight had been as good as its hindsight, it no doubt would have acted differently. Unfortunately ,such was not the case. If Lincoln, and Union Carbide as well, had then known what they know now this prolonged and ■expensive litigation would have been avoided. That the rights of the parties in 1942 were uncertain and doubtful is admitted on all sides. As previously shown, the Master so found and the Court approved. Union Carbide thought the issue was doubtful. The Master found that even after the Supreme Court’s first decision it “withheld payments of royalties to the inventors on the Unionmelt 50, a manganese silicate flux, until after the second decision of the Supreme Court.” At that time Union Carbide wrote to the inventor, “There is a question as to whether Lincoln’s 660 infringes the valid composition claims of the patent.” As Mr. Justice Black said'in his dissenting opinion, 339 U.S. 605, 617, 70 S.Ct. 854, 860, “I think petitioners had a right to act on the belief that this Court would follow the plain mandates of Congress that a patent’s precise claims mark its monopoly boundaries, and that expansion of those claims to include manganese could be obtained only in a statutory reissue proceeding.” It is, therefore, our conclusion and we so hold that there is no substantial support for the finding, if such it may be characterized, that Lincoln copied its infringing flux from the teachings of the patent and, in any event, there is no basis for a conclusion that Lincoln in the manufacture and use of its 660 flux was a conscious and wilful infringer. Appeal No. 12797. Union Carbide appeals on the ground that the judgment in its favor is grossly inadequate. Based upon accounting statements introduced in evidence, Union Carbide in its brief states: “The proof established that pláintiff’s Unionmelt Division had suffered a substantial shrinkage in its margin of profit during the period of infringement. The total loss as computed on this basis came to $8,521,234.” This shrinkage was predicated upon a comparison of the period prior to infringement, 1937-1942, with the period of infringement, 1943-1950. We think it not essential to relate the complicated process by which Union Carbide calculated the amount of its loss. The Master found that its computed loss was as follows: “Loss suffered on sales of patented composition and royalties ............. $5,356,462 Loss suffered on other Unionmelt operations .. 3,164,711 Total loss suffered on Unionmelt operations .. $8,521,234” As to the infringing flux, the Master found: “The Lincoln Company produced 40,998,305 pounds of the 660 flux. The net sales thereof during the period of infringement, 1942-1950, totaled $4,353,777.12 and Lincoln’s profit thereon before Federal income taxes was $940,275.60. More than 99% of all the 660 flux was sold to purchasers who were already customers of Lincoln for other products.” Thus, Union Carbide claims that it was entitled to a judgment of $8,521,234, when Lincoln’s net sales of the infringing composition totaled only $4,353,-777.12, on which it realized a profit of $940,275.60. (This makes no allowance for Federal income taxes.) The mere statement of these facts presents a serious challenge to the validity of plaintiff’s contention. The loss thus computed by Union Carbide was rejected by the Master and the Court as a basis for an ascertainment of its damages. This was for the reason that the assumptions upon which the calculation rested were contrary to fact. In this connection, the Master found that the computations treated income realized upon the license by which Union Carbide exploited its process (until the process claims were declared invalid) “as if it were income from flux sales”, whereas, said the Master, “this is not the fact.” The computation thus “ignores the fact that the plaintiff, when the royalties were paid, was illegally claiming a patent monopoly over the submerged arc welding process. Only a part of the royalty paid by the licensees was for the use of the patented flux.” The Master found: “The computations * * * assume that the decline in plaintiff’s profit to expense ratio * * * was the direct and sole result of the infringement on the part of Lincoln. This is not the fact.” The Master further found: “The computations * * * assume that the plaintiff would have sold all of said flux if Lincoln had not infringed. This is not the fact. A substantial part of such sales would have been made by plaintiff but not all of them.” Union Carbide’s calculations were based upon a comparison of its profits during the five year non-infringing period (1937-1942) with the infringing period (1943-1950). Such comparison was recognized by the Master and the Court as proper, providing the figures employed were relevant to the issue for decision, that is, Union Carbide’s loss by reason of Lincoln’s sale of the 660 flux. However, Union Carbide made no such comparison. Instead, its comparison was based upon profits derived from all of its arc welding business, including not only its flux but also process license royalties, welding machines, welding equipment and welding rods. All of these items other than its flux were outside the scope of its patent monopoly. We need not dwell at length on the accounting method thus employed by Union Carbide as proof of lost profits. It is illustrative and sufficient, we think, to refer to that portion of its calculations relative to royalties received from its licensees. The licenses covered numerous items other than its flux, including its process claims held invalid and other patents owned by Union Carbide. A reading of these license agreements strongly suggests that the process disclosed by the patent was an important consideration for the royalties paid by licensees. This computation of “loss suffered on sales of patent composition and royalties” amounts to $5,356,462. Union Carbide in its brief states that Peter Pirnie, Jr. was chief accounting officer in charge of its books and records and supervised the preparation of the accounting statements. Lincoln asserts in its brief that the technique employed by Union Carbide in combining the figures on flux with royalties on the invalid process was “devised specifically for this case.” There is support for this accusation. Pirnie testified that he mixed flux with other products even though it would have taken less time to give the figures on flux alone. He testified that he did this because “that is the way I was asked to develop the situation, and that is the way we set it up,” and “I am not setting the policy in this particular case.” Pirnie was asked, “Then there is nothing in your former treatment of the breakdown of this composite Unionmelt enterprise that indicates any reason for combining the flux sales with the license income as one category, and the other two items, Unionmelt equipment and Unionmelt rod as another category ?” and responded, “I don’t know. This was done according to the way people at that time in charge wanted it done, I suppose, and that is how we did it. * * * I did what I did per — in accordance with the request of our counsel.” Union Carbide’s method of computing its loss as a basis for the award of damages was rejected by the Master and confirmed by the Court. The Master’s conclusion of law in this respect, which Union Carbide challenges in this Court, is as follows: “Plaintiff has been unable to and therefore has failed to prove with sufficient accuracy for judicial determination the following: “(a) The amount of its damage from loss of profits it would have made on such additional sales of the patented composition it would have made but for the infringement. “(b) The amount of its damage from loss of additional profit on its sales which would have resulted from higher prices it could, and, but for the infringement, would have charged for the patented composition sold by plaintiff.” Judge Swygert, in overruling Union Carbide’s objection to this conclusion, stated: “I think the Master properly rejected the profits-to-expense ratio method in this case in recommending the amount of damages. The license agreements included a royalty on both the process and the composition claims of the patent. What plaintiff would have charged as a royalty on the composition claims alone from the starting period of infringement to February 28, 1949 is purely a matter of speculation. * * * The scope of the invention is measured by what has been determined by the Court, not by the claims granted by the Patent Office. ‘ Since the invention is now limited to four composition claims, plaintiff’s royalty payments based on the invalid process claims cannot be included in a proper base for determining the amount plaintiff lost by reason of the infringement.” Union Carbide contends that the fact of loss having been determined as a consequence of Lincoln’s infringement and the total amount of such loss having been determined, it then became the burden of Lincoln to prove the portion of such loss not due to Lincoln’s infringement. Three cases are cited in support of this contention. Westinghouse Electric & Mfg. Co. v. Wagner Electric & Mfg. Co., 225 U.S. 604, 32 S.Ct. 691, 56 L.Ed. 1222; Bigelow et al. v. RKO Radio Pictures, Inc., 327 U.S. 251, 66 S.Ct. 574, 90 L.Ed. 652, and Anderson et al. v. Mt. Clemens Pottery Co., 328 U.S. 680, 66 S.Ct. 1187, 90 L.Ed. 1515. As Union Carbide suggests in its brief, Westinghouse is the only case involving patent infringement. Both parties in their briefs have discussed Westinghouse at length, and we have given it careful study. We think a brief statement of the facts is sufficient to disclose the issue which the Court decided. The patent was on a transformer used in the transmission of electric power. Infringing transformers sold by Wagner had features disclosed by the patent, together with others developed by Wagner. In other words, Wagner sold a transformer which by the addition of certain features made it an improvement over the patent in suit. The Master, on accounting, found Wagner’s total profit on the sale of transformers and that all of the commercial value was due to the use of the patent claims. The Court of Appeals for the Eighth Circuit (173 F. 361, 367-368) held in effect that the improvements made by Wagner were substantial and produced the desired result more certainly and efficiently. That Court also held that the burden of apportionment was upon the patentee, and that it having failed to allocate the profits as between the patent features and other features was entitled to a decree only for nominal damages. The Supreme Court, after discussing the general rule which places the burden upon the patentee to prove its loss of profits, came to the issue as to which party had the burden in a situation where it was impossible for the patentee to prove the portion of the infringer’s profits resulting from use of the patent and that from use of features outside the patent. The Court pointed out that plaintiff had called witnesses employed by defendant to keep the books, purchase the material, superintend the construction and fix the price of the transformers, none of whom was able to show that profits had been made and consequently was not able to show what part of the profits was attributable to the patent and what to the non-infringing improvements. The gist of the Court’s decision and its reason therefor is contained in the following statement (225 U.S. at page 618, 32 S.Ct. at page 696): “But when a case of confusion does appear, — when it is impossible to make a mathematical or approximate apportionment, — then, from the very necessity of the case, one party or the other must secure the entire fund. It must be kept by the infringer or it must be awarded by law to the patentee. On established principles of equity, and on the plainest principles of justice, the guilty trustee cannot take advantage'of his own wrong. The fact that he may lose something of his own is a misfortune which he has-brought upon himself; and if, as argued, the fund may have been made by the use of other patents also, for which he may be liable in another, case, it is again a misfortune which he has brought upon himself and an instance of a double wrong causing double liability.” It is apparent that the facts in Westinghouse are in marked contrast to those here. In that case, profits resulted from the sale of transformers, portions of which were covered by the patent and others which were not. A portion of the profits belonged to the patentee and a portion to the infringer. They were commingled by the infringer in such a manner that apportionment by the patentee was impossible. If the infringer desired to extricate himself from a situation of his own creation, the Court said the burden was on him to do so. In contrast, we have a situation here where the Master specifically found, as previously noted, the total amount of infringing flux sold by Lincoln, the amount received therefor and the profit derived therefrom. The confusion which Union Carbide visualizes resulted from its own system of bookkeeping and its manner of doing business. It granted licenses and received royalties not only on its flux but on its process claims and other patents, all of which it commingled. We agree that Lincoln took the risk incident to its infringement, but it did not take the risk incident to the program of Union Carbide by which it extracted royalties based in part upon a process which was held invalid. And neither did Lincoln take the risk inherent in the situation created by Union Carbide whereby its profits from all its business were lumped together. Certainly there was no burden on Lincoln to extricate Union Carbide from this situation of its own creation. Our interpretation of Westinghouse finds support in Hamilton-Brown Shoe Co. v. Wolf Brothers & Co., 240 U.S. 251, 36 S.Ct. 269, 60 L.Ed. 629, a suit for trademark infringement, which the Court compared with the analogous situation in suits for patent infringement. Following the citation of numerous cases, the Court stated the general rule as follows (240 U.S. at page 260, 36 S.Ct. at page 272): “In such case, where the invention is used in combination with other elements of value not covered by the patent, so that plaintiff’s patent creates only a part of the profits, he is entitled to recover only that part, and must give evidence tending to apportion the profits between the patented and unpatented features.” The Court stated: “But, as pointed out in the Westinghouse Case (p. 618), there is a recognized exception where the plaintiff carries the burden of proof to the extent of showing the entire profits, but is unable to apportion them, either because of the action of the wrongdoer in confusing his own gains with those which belong to plaintiff, or because of the inherent impossibility' of making an approximate apportionment. There, ‘on established principles of equity, and on the plainest principles of justice, the guilty trustee cannot take advantage of his own wrong.’ ” Neither Bigelow (327 U.S. 251, 66 S. Ct. 574) nor Anderson (328 U.S. 680, 66 S.Ct. 1187) supports the contention that Lincoln carried the burden of proof as to the apportionment of profits. While neither of these cases was an action for infringement, the Court in Bigelow discussed the rule in such cases and cited Westinghouse and Hamilton in support of the following statement (327 U.S. at page 265, 66 S.Ct. at page 580): “And in cases where a wrongdoer has incorporated the subject of a plaintiff’s patent or trade-mark in a single product to which the defendant has contributed other elements of value or utility, and has derived profits from the sale of the product, this Court has sustained recovery of the full amount of defendant’s profits where his own wrongful action has made it impossible for the plaintiff to show in what proportions he and the defendant have contributed to the profits.” Anderson was an action by employees against an employer to recover sums claimed to be due under the Fair Labor Standards Act, 29 U.S.C.A. § 201 et seq. The Court placed the burden on the employer because (328 U.S. at page 687, 66 S.Ct. 1187, 1192): “ * * * it is the employer who has the duty under § 11(c) of the Act to keep proper records of wages, hours and other conditions and practices of employment and who is in position to know and to produce the most probative facts concerning the nature and amount of work performed.” In the instant situation, however, it was Union Carbide, not Lincoln, who had the duty and who was in a position to know if it was damaged and, if so, the extent thereof. Union Carbide argues that in any event the process and composition claims stated a single invention and that the invalidation of the former had no appreciable effect upon the royalties paid by its licensees. It states in its brief, “any diminution attributable to loss of the ‘process’ claims was de minimis.” This is a strange argument following Union Carbide’s effort, asserted through all the courts, to sustain the validity of its process claims. Particularly is it so to this Court, which was persuaded to the view that such claims were of great merit and sustained their validity. 167 F.2d 531. If this concession had been made in the beginning, much extensive litigation and labor on the part of counsel and the courts might have been avoided, or at least greatly reduced. However, in the brief before this Court we were told by Union Carbide, among other things, that the Trial Court “was unwarrantedly harsh” when it “deprived the inventors of protection for their valuable process * * * ”; “These are claims to a process. They are drawn to a process of electric welding. They are not claims to a composition of matter or chemical compound,” and “In the patent in suit the inventors had not only to devise a new process but also to invent an entirely new agent for use in that process.” Union Carbide professes to discern some relevancy in the Master’s “copying” finding and the Court’s characterization of Lincoln as a “conscious and wilful” infringer. We have already discussed and expressed our disagreement with that finding and the conclusion drawn therefrom. However, even if the finding and conclusion were accepted, we think it would be of no consequence to the issue under consideration, that is, that the showing made by Union Carbide as to its claimed damages cast the burden of proof upon Lincoln. We agree with the conclusion of law under attack and the rejection by the Master, approved by the District Court, of the method employed by Union Carbide in its attempt to prove damages. Appeal No. 12798. On this appeal Lincoln attacks the judgment as excessive in numerous respects. The Master and the District Court, as shown in appeal No. 12797, rejected the method proposed by Union Carbide for the ascertainment of its damages. In the absence of an established royalty, the Master resorted to the statute which provides for an award of damages “adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer” (Title 35 U.S.C.A. § 284). The Court approved the action of the Master in this respect and no question is raised here as to this method of computing plaintiff’s damages. The Master found the total poundage of the infringing flux manufactured and sold by Lincoln. The poundage thus found was divided between that sold before and after the filing of the complaint (October 1, 1945). A royalty of 50 per pound was calculated by the Master only on the latter, which amounted to $1,212,-545.86. The Court increased this amount to $1,361,891.25, by applying the royalty rate determined by the Master to the flux sold by Lincoln prior to the filing of the complaint. The Master awarded as compensatory damages the sum of $600,-000, which was increased by the Court to $900,000. The Master refused the allowance of interest for the period prior to the time damages were determined' on the coming in of the Master’s report. The Court overruled the Master in this respect and included in the judgment interest in the sum of $748,185.91, calculated from June 30, 1950, when infringement ceased, to August 26, 1959, when the judgment was entered. The contested issues emerge from Lincoln’s contention that the Court and the Master applied erroneous standards in determining a reasonable royalty; that the Court erred in increasing the total amount of the royalty, in awarding increased damages and in allowing interest prior to the time damages were liquidated. There is the further contention that there was error in the Master’s denial, affirmed by the Court, of Lincoln’s request for an allowance of attorney fees incurred in the defense of the action for contempt instituted by Union Carbide. As previously noted, defendant was adjudged in contempt by the District Court, which ruling was reversed by this Court (196 F.2d 103). In that proceeding, this Court referred to the District Court Lincoln’s application for the allowance of attorney fees. The Master made numerous findings which portray a situation from which he was required to determine a reasonable royalty. A review of his findings in detail would serve no good purpose. The critical finding is as follows: “Applying the reasonable royalty method of computing damages to the evidence and to the findings of fact herein set forth and with due consideration of the evidence and the findings herein concerning plaintiff’s loss of sales of flux and loss of profits on sales of flux and welding equipment and supplies during the infringement period resulting from Lincoln’s infringing activities, the minimum royalty that the plaintiff should be willing to accept and the defendants should be willing to pay for the right to make, use and sell the patented composition and being a reasonable royalty, was the sum of 5 cents per pound of rod deposited.” The Court, on Lincoln’s objection to the Master’s finding, stated: “Unless it can be determined that the Master’s finding in fixing a reasonable royalty at 5 cents per pound of rod deposited is clearly erroneous, Rule 52(a) of the Federal Rules of Civil Procedure prevents the setting aside of the finding. The Master heard a lengthy trial. He also had the benefit of extensive briefs and arguments. Although these factors, themselves, do not insure against error, they do support the view that the issue in regard to the amount of a reasonable royalty was presented before the Master in a manner to afford full and deliberate consideration of the evidence. The evidence, as well as the Master’s related findings of fact, sustain the conclusion that on this point the Master was not in error.” Lincoln argues that the Court failed to apply three recognized rules in the determination of a reasonable royalty: (1) that it failed to apply the willing seller and willing buyer rule; (2) that it failed to measure the margin of superiority of the patent flux over non-infringing alternatives, and (3) that it failed to confine the royalty to the use made of the invention and to exclude all else. The willing seller and buyer rule is aptly stated in Horvath v. McCord Radiator & Mfg. Co., 6 Cir., 100 F.2d 326, 335, as follows: “In fixing damages on a royalty basis against an infringer, the sum allowed should be reasonable and that which would be accepted by a prudent licensee who wished to obtain a license but was not so compelled and a prudent patentee, who wished to grant a license but was not so compelled.” We think this Court is in no position to say that the rule was not considered and applied by the Master and the Court; in fact, the Master in his finding fixing a royalty rate of 5^ per pound stated that it was “the minimum royalty that the plaintiff should be willing to accept and the defendants should be willing to pay.” (See finding lastly above quoted.) Lincoln’s contention that the Court failed to take into consideration that during all the infringement period it had available alternative fluxes which it could have used had there been reason for supposing its counsel was wrong in advising that the 660 flux did not infringe, presents a more serious question. It is argued that the findings of the Master, approved by the Court, disclose that Lincoln had available for its use 620 and 650 fluxes during the period from October 8, 1942, when infringement started, until July 1, 1948, and from the latter date until the end of infringement it had available its 700-series fluxes. On this premise it is contended that under the decision of this Court in Columbia Wire Co. v. Kokomo Steel & Wire Co., 194 F. .108, the measure of recovery could only be the difference between the infringing flux and those which Lincoln had available as alternatives. The Master found: “The use of the Lincoln 620 and 650 fluxes * * * did not produce wholly satisfactory welding results while the Lincolnweld 660 flux produced excellent results. Lincoln at the time it introduced its 660 flux did not have any substitute or alternative fluxes which would produce welding results equal to those produced by the use of Lincolnweld 660 or Unionmelt.” This finding is supported, in fact amplified, by the testimony of Lincoln’s witness Smith, who admitted he encountered difficulties with porosity when he welded with the 620 flux. As to the 650 flux, he conceded that he was “by no means satisfied with what had been accomplished” and that it was “not ready to turn over to the customer.” As far as we are aware, there was no proof that •either of these fluxes was salable or that Lincoln ever attempted to place either upon the market. In response to Lincoln’s contention that in fixing a reasonable royalty the Master did not consider the available alternative non-infringing fluxes, the Court stated: “In support of their position they cite Columbia Wire Co. v. Kokomo Steel & Wire Co., (C.C.A.7, 1911) 194 F. 108. I think this case states the law as to the effect to be given to noninfringing articles available to the public which may be used as alternatives for the patented article. * * * “The difficulty with defendants’ position is that until Lincoln developed the 700-series of fluxes there were no noninfringing fluxes available that were as good in performance as the patented composition. The Master’s finding that ‘ * * * the use of Lincoln 620 and 650 fluxes at the Graver plant did not produce wholly satisfactory welding results while the Lincolnweld 660 flux produced excellent results’ is supported by the evidence.” Union Carbide, like the Court, recognizes the validity of the Columbia Wire rule and also, like the Court, reasons that it is without application to the facts as found. With this reasoning we agree as it relates to the 620 and 650 fluxes. Their utility was near zero and they were never sold or offered for sale to the public. Under such circumstances, there is little, if any, room for the belief that they could have been used by Lincoln as a substitute or alternative for the patent flux. As to the 700-series fluxes, a different situation is presented. As to these the Master found: “Near the end of the period of infringement, after adverse decisions of the Court of Appeals and the United States Supreme Court, the Lincoln engineers were able to develop and field test non-infringing fluxes and to substitute them commercially for the infringing 660, as follows: “The 770 flux was first field tested in July, 1948, and was first shipped commercially in September, 1948.” This 770 flux was superseded by the 780 and 760 fluxes, the first of which, according to the findings, was “shipped commercially in March, 1949” and the latter, “shipped commercially in June, 1950.” The Master also found: “These fluxes are made from the same raw material as 660 and cost less than 660 to manufacture.” The main contention of Union Carbide on this point is that the 700-series fluxes were not available to Lincoln during the infringing period. (The second period of infringement extended from July 1, 1948 to June 17, 1950.) The findings as shown are to the contrary. Moreover, there is testimony to the effect that these 700 fluxes were as good as the patent flux, that they were interchangeable with Lincoln’s infringing flux and that more than eleven million pounds of these fluxes were sold and commercially used during the period of infringement. That they were as good in performance as the patent flux was recognized by the Court, as previously shown. The Court did not deny the application of the Columbia Wire rule to the 700-series fluxes because they were not as good as the patent flux or because they were not available as a substitute or alternative. Rather, the Court stated: “As for the 700-series fluxes, the answer to defendants’ argument is that even though Lincoln developed this noninfringing flux in 1948, it continued to produce and sell the infringing flux until the issuance of the Supreme Court’s mandate in June, 1950.” In our view, this statement does not meet the problem. It is only where the infringer continues to produce and sell the infring