In Re Coordinated Pretrial Proceedings in Antibiotic Antitrust Actions. Pfizer, Inc. v. International Rectifier Corp.

GIBSON, Chief Judge. This patent infringement case was initially filed by Pfizer, Inc. seeking damages and declaratory and injunctive relief in the Central District of California. The defendants, International Rectifier Corp. (IRC) 1 and USV Pharmaceutical Corp. (USV), answered, pleading that Pfizer’s United States Patent No. 3,200,149 2 is invalid and unenforceable for failure to meet statutory requirements of patentability and for fraud and misconduct before the Patent Office. Both defendants at first admitted infringement but subsequently amended their answer to deny infringement and assert unfair competition and antitrust counterclaims. Upon defendants’ motion, the case was transferred to the United States District Court for the District of Minnesota and assigned to District Judge Miles W. Lord for coordinated pretrial proceedings with IRC’s antitrust case and other antibiotic antitrust litigation, by order of the Judicial Panel on Multidistrict Litigation dated March 12, 1973. The case is scheduled to be returned to the Central District of California for trial after completion of coordinated pretrial proceedings. While the case was pending for pretrial processing, defendants filed successive motions for partial summary judgment maintaining that Pfizer’s conduct in prosecuting the patent application before the Patent Office, from the time the application was filed in May, 1961, until the patent was issued in August, 1965, constituted fraud, inequitable conduct and unclean hands, and that Pfizer’s conduct before the District Court from 1973 until 1975 was also fraudulent and inequitable, separately justifying refusal to enforce the patent. On July 16, 1975, in an extensive memorandum opinion, the District Court granted partial summary judgment and declared Pfizer’s doxycycline patent invalid and unenforceable. Pfizer, Inc. v. International Rectifier Corp., 186 U.S.P.Q. 511 (D.Minn. 1975). The court set forth more than 150 “uncontroverted” facts from the record of documents 3 and testimony taken in eleven days of hearings on the summary judgment and related discovery motions conducted in *183 termittently from September, 1974, through April, 1975. On these 150 facts, deemed uncontroverted, Pfizer was found guilty of five acts of inequitable conduct before the Patent Office, each of which, standing alone, the court considered sufficient to bar enforcement of the patent. They are: (1) Pfizer’s alleged failure to disclose both the existence of, and its inability to distinguish, a prior Belgian patent of a similar compound, in order to avoid Patent Office rejection for anticipation (lack of “novelty”) under 35 U.S.C. § 102 (1964) 4 and “obviousness” under 35 U.S.C. § 103 (1964); 5 (2) Pfizer’s argument of an allegedly false and misleading analogy between doxycycline and certain prior art epimers, 6 similarly to avoid rejection for anticipation and obviousness; (3) Pfizer’s alleged concealment of test data concerning the antibacterial activity of doxycycline, with similar motivation; (4) Pfizer’s alleged concealment of its scientists’ beliefs that doxycycline may have been co-produced in prior art processes, similarly motivated; and (5) Pfizer’s alleged concealment of repeated experimental failures of processes described and claimed in the doxycycline patent with the motive of obtaining claims broader than permissible under 35 U.S.C. § 112 (1964). 7 Additionally, Pfizer’s entire course of conduct before the Patent Office was characterized as fraudulent or “at the very least a calculated recklessness about the truth” for the purpose of obtaining as broad a patent as possible. Finally, Pfizer was summarily found guilty of concealing critical facts from the court, amounting to inequitable conduct and fraud on the court, that independently justified refusal to enforce the doxycycline patent. On this appeal, Pfizer challenges the District Court’s rulings as violative of the cardinal principle that summary judgment is permitted only if there is “no genuine issue as to any material fact,” Fed.R.Civ.P. 56(e), and also the court’s interpretation of the principles of law governing the patent infringement defense of unclean hands resulting from fraud or inequitable conduct by the patentee before the Patent Office and the District Court. Pfizer contends that *184 the court resolved bitterly disputed issues of material fact, drew inferences unfavorable to the nonmovant and determined credibility without affording Pfizer a trial by jury in the Central District of California as demanded in its pleadings. 8 We think the court erred in granting partial summary judgment, as genuine issues of material fact remain as to each of Pfizer’s five alleged acts of misconduct before the Patent Office, and the record does not reveal sufficient misconduct by Pfizer before the District Court in the pretrial proceedings to warrant a finding of fraud on the court and the imposition of a sanction of such severity. We express no opinion as to Pfizer’s conduct before the Patent Office nor on the merits of the legal and factual issues of patent validity. The scientific issues at stake are of such moment that all parties should be permitted to present their evidence fully in a plenary trial not restricted to a trial by affidavit. 9 I. PROPRIETY OF SUMMARY JUDGMENT. The standards to be applied in ruling upon a motion for summary judgment pursuant to Fed.R.Civ.P. 56(c) have been clearly stated by the United States Supreme Court: Summary judgment should be entered only when the pleadings, depositions, affidavits, and admissions filed in the case “show that * * * there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” Rule 56(c), Fed.Rules Civ.Proc. Poller v. Columbia Broadcasting System, Inc., 368 U.S. 464, 467, 82 S.Ct. 486, 488, 7 L.Ed.2d 458 (1962). To obtain a summary judgment, the movant must demonstrate the absence of any genuine issue of material fact, and the evidence submitted to the court “must be viewed in the light most favorable to the opposing party.” Adickes v. S. H. Kress & Co., 398 U.S. 144,157,90 S.Ct. 1598,1608, 26 L.Ed.2d 142 (1970). Movant must show “his right to a judgment with such clarity as to leave no room for controversy and that the other party is not entitled to recover under any discernible circumstances.” Ozark Milling Co. v. Allied Mills, Inc., 480 F.2d 1014, 1015 (8th Cir. 1973); accord, Cervantes v. Time, Inc., 464 F.2d 986, 993 (8th Cir. 1972), cert. denied, 409 U.S. 1125, 93 S.Ct. 939, 35 L.Ed.2d 257 (1973). In determining whether a genuine issue of material fact exists, the court must give the nonmoving party the benefit of all reasonable factual inferences. Adickes v. S. H. Kress & Co., supra, 398 U.S. at 158-59, 90 S.Ct. 1598; McSpadden v. Mullins, 456 F.2d 428 (8th Cir. 1972), and must do so without assessing credibility. United States v. United Marketing Association, 291 F.2d 851, 853-54 (8th Cir. 1961). Summary judgment is to be used not as a substitute for trial, but only when “it is quite clear what the truth is [and] that no genuine issue remains for trial.” Sartor v. Arkansas Natural Gas Corp., 321 *185 U.S. 620, 627, 64 S.Ct. 724, 728, 88 L.Ed. 967 (1944). Summary judgment is notoriously inappropriate for determination of claims in which issues of intent, good faith and other subjective feelings play dominant roles. White Motor Co. v. United States, 372 U.S. 253, 259, 83 S.Ct. 696, 9 L.Ed.2d 738 (1963); Louis Schlesinger Co. v. Kresge Foundation, 388 F.2d 208, 212 (3d Cir.), cert. denied, 391 U.S. 934, 88 S.Ct. 1847, 20 L.Ed.2d 854 (1968); Severson v. Fleck, 251 F.2d 920, 924 (8th Cir. 1958); 6 J. Moore, Federal Practice H 56.17 [41.-1] (2d ed. 1976). Here matters of intent, good faith and credibility have been resolved by the court in a summary proceeding. The defendants are unable to cite any Court of Appeals decision approving a summary judgment invalidating or refusing to enforce a patent on grounds of fraud or inequitable conduct before the Patent Office. 10 This court has long held the view that summary judgment is ordinarily inappropriate for disposition of patent cases in which issues of fact predominate and the requirements of Rule 56(c) are rarely met. Long v. Arkansas Foundry Co., 247 F.2d 366, 369 (8th Cir. 1957). The impropriety of summary judgment in the instant case is compounded for the reason that forfeiture of the patent is sought on the basis of the patentee’s personal misconduct in procuring the patent as well as technical and scientific fact. II. THE DEFENSE OF MISCONDUCT BEFORE THE PATENT OFFICE. The principle that a defendant in a patent infringement action may interpose as a complete defense the patentee’s failure to deal candidly with the Patent Office is a corollary of the equitable doctrine of unclean hands. Precision Instrument Manufacturing Co. v. Automotive Maintenance Machinery Co., 324 U.S. 806, 814, 65 S.Ct. 993, 89 L.Ed. 1381 (1945). The Supreme Court has set forth the duty of candor owed by a patent applicant as follows: Those who have applications pending with the Patent Office or who are parties to Patent Office proceedings have an uncompromising duty to report to it all facts concerning possible fraud or inequitableness underlying the applications in issue. * * * Public interest demands that all facts relevant to such matters be submitted formally or informally to the Patent Office, which can then pass upon the sufficiency of the evidence. Only in this way can the agency act to safeguard the public in the first instance against fraudulent patent monopolies. Precision Instrument Manufacturing Co. v. Automotive Maintenance Machinery Co., supra at 818, 65 S.Ct. at 999. (Emphasis added.) The equitable origins of this doctrine, combined with recognition of the growing administrative burden facing the Patent Office, have led to expansion of the defense in recent years to encompass also a wide variety of inequitable conduct short of common law fraud or deceit. 11 See, e. g., Monsanto Co. v. Rohm & Haas Co., 456 F.2d 592, 599 (3d Cir.), cert. denied, 407 U.S. 934, 92 S.Ct. 2463, 32 L.Ed.2d 817 (1972). In the instant case, the District Court adopted a far-reaching interpretation of the doctrine, emanating from what the court described as an obligation on Pfizer’s part to disclose to the Patent Office any fact that “may be relevant to an issue of patent- *186 ability.” Further, the court held that the defendants in proving Pfizer’s breach of this obligation were not required to prove that Pfizer intended to deceive the patent examiners, and that Pfizer’s claims of good faith are immaterial and do not create genuine issues of fact. We believe this interpretation imposes an unworkable standard of conduct upon the patent applicant and expands the inequitable conduct defense beyond legitimate limits. 12 It would be unwise to attempt to formulate a standard of conduct setting forth all elements of the defense embracing all misconduct before the Patent Office that might justify refusal to enforce a patent. However, we note that the standard is not one of strict liability for innocent or even negligent omissions or misstatements before the Patent Office. Schnadig Corp. v. Gaines Manufacturing Co., 494 F.2d 383, 393 (6th Cir. 1974). Rather, to result in refusal to enforce a patent, the misconduct must be accompanied by “some element of wrongfulness, willfulness, or bad faith.” Parker v. Motorola, Inc., 524 F.2d 518, 535 (5th Cir. 1975); see Precision Instrument Manufacturing Co. v. Automotive Maintenance Machinery Co., 324 U.S. 806, 814-815, 65 S.Ct. 993, 997-998, 89 L.Ed. 1381 (1945) (a “willful act * * * which rightfully can be said to transgress equitable standards of conduct”). This requirement of proof has been uniformly applied in infringement actions by a majority of the circuits to claims of both fraud 13 and lesser inequitable conduct. 14 Moreover, proof of *187 misconduct under either theory must be established by “clear, unequivocal and convincing” evidence. United States v. American Bell Telephone Co., 167 U.S. 224, 251,17 S.Ct. 809, 42 L.Ed. 144 (1897); accord, Schnadig Corp. v. Gaines Manufacturing Co., supra at 392; Monsanto Co. v. Rohm & Haas Co., supra at 601 n.14. Before summarizing the unresolved issues' of fact relating to Pfizer’s alleged misconduct before the Patent Office, we briefly review some undisputed background to facilitate understanding of the issues. A. Terminology and Background. Doxycycline is derived by synthesizing a fermentation produced drug called oxytetracycline (Pfizer tradename Terramycin) with a process called hydrogenolysis. A significant concern to Pfizer in patenting doxycycline was the prior existence of another compound of nearly identical chemical structure. This prior discovery, referred to as the McCormick compound, was produced in the 1950’s by Dr. J. R. D. McCormick for American Cyanamid, and independently by Drs. Stephens and Conover, working for Pfizer, using the same process of hydrogenating oxytetracycline. It was patented by Dr. McCormick in Belgium in 1959. All three chemical compounds are molecules consisting of component groups of atoms, called substituents, in a configuration unique to each compound. Chemists are able to describe three dimensionally the molecular structure of each compound using the terminology of stereochemistry. The spacial arrangement of a molecule’s substituents is known as its stereochemical configuration or orientation. By convention, the various positions (asymmetric centers) on the basic molecule common to the compounds of the tetracycline family have been assigned numbers. (See Fig. 1.) One can distinguish a particular compound by referring to its stereochemical configuration at one or more individually numbered positions on the molecule. Figure 1. The complete doxycycline molecule. If two compounds are identical at all but a single position at which one compound has a configuration opposite that of the other, the two compounds are known as “epimers” of one another. Doxycycline and the McCormick compound are 6-position epimers of each other. That is, the chemical structure of doxycycline is identical to that of the McCormick compound except at the 6-position, where their structures are opposite: McCormick compound Figure 2. Comparison of the 6-positions of Doxycycline and the McCormick compound. In the above diagrams, the substituents shown with dotted lines are on one side of *188 the molecule (referred to as “down”), and those shown with solid lines are on the other side (referred to as “up”). This close structural similarity between any two compounds can pose a problem in patent prosecution. Under the Patent Act, 35 U.S.C. § 103, the existence of the second discovered epimer is presumed to be “obvious,” and hence unpatentable, absent a showing that it has unexpected beneficial properties. See generally Note, Standards of Obviousness & the Patentability of Chemical Compounds, 87 Harv.L.Rev. 607 (1974). Further problems can result if the evidence suggests that the second-discovered compound was previously coproduced (as a byproduct) or otherwise disclosed in the prior art so as to defeat its “novelty” within the meaning of 35 U.S.C. § 102. B. Pfizer’s Alleged Misconduct Before the Patent Office. 1. The Belgian Patent. Pfizer is accused of fraud and inequitable conduct in connection with its failure in May, 1961, upon filing the doxycycline application, to cite to the Patent Office the prior art Belgian patent. In the application, Pfizer referred to the compounds of the prior art as the “known 6-de-oxy-tetracyclines,” without citing Dr. McCormick’s Belgian patent. Pfizer’s motivation for doing so, according to the defendants, was to mislead the Patent Office in order to avoid two issues of patentability: “anticipation” of Pfizer’s compound by McCormick under 35 U.S.C. § 102, and the possible limitation of the doxycycline patent to “product-by-process” claims under 35 U.S.C. § 112. 15 The Belgian patent contains a diagram that purports to show the stereochemical configuration common to compounds such as McCormick’s. According to the defendants, Pfizer suspected that the diagram in the Belgian patent did not correctly depict the McCormick compound. Defendants concede that regardless of the inaccuracy of the diagram’s depiction of the McCormick compound, Pfizer was confident that its new compound, doxycycline, had a 6-posi-tion configuration opposite that of the McCormick compound. 16 However, defendants maintain that this would not necessarily excuse Pfizer’s failure to cite the Belgian patent for the reason that according to the Von Bramer doctrine, then in effect in the Patent Office, but later overruled, 17 the *189 Belgian diagram alone could have served as a complete anticipation of doxycycline even if its illustration of doxycycline was only accidental. Thus, defendants maintain that, in view of Pfizer’s concession that upon filing the application it believed the Belgian diagram correctly, albeit accidentally, depicts doxycycline at the 6-position, Pfizer knew it needed another distinction between doxycycline and the diagram to avoid the problem of anticipation. Pfizer maintains it then believed, and has since confirmed, that the diagram differs from doxycycline at the 5-position despite their identity at the 6-position. Thus, Pfizer contends, in 1961 it believed the Belgian patent did not accurately depict either the McCormick compound (different at the 5- and 6-positions) or doxycycline (different at the 5-position). Pfizer insists the Belgian diagram was not a true Von Bramer anticipation and need not have been cited to the patent examiners; however, Pfizer contends that if it erred in failing to cite the Belgian patent initially as prior art, its mistake was later cured by full disclosure to the patent examiners in oral interviews during the prosecution — well after the demise of the Von Bramer doctrine rendered the entire matter immaterial to patentabilityThe District Court held that these later unrecorded oral disclosures were inadequate to cure Pfizer’s breach of candor to the Patent Office and concluded that regardless of the later demise of the Von Bramer doctrine, rendering the Belgian diagram immaterial, Pfizer’s initial failure to disclose its uncertainty concerning the difference between the Belgian diagram and doxycycline estops it from relying upon later disclosures in defense of its conduct. 18 The court concluded that Pfizer’s deliberate initial failure to disclose the Belgian patent and delay thereafter in doing so were inequitable, and alternatively that even without the estoppel, Pfizer’s deliberate delay alone of almost one year was sufficient to justify forfeiture of the patent. However, despite the court’s characterization of facts supporting these conclusions as “uncontroverted,” we believe genuine issues of material fact remain as to the states of mind of Pfizer’s scientists and patent counsel, precluding summary judgment, and that additional issues of technical fact are present that are unsuited for summary resolution without the benefit of live expert testimony and cross-examination in a plenary trial. 2. The False Analogy to Prior Art Epimers. Pfizer also is accused of misconduct in connection with its response to Patent Office rejection of doxycycline as an “obvious” routine development in the art unpatentable because closely similar to the McCormick comppund and lacking any unexpected beneficial properties. In its application, Pfizer described doxycycline as an epimer of the prior art McCormick compound, differing only in that the latter had an opposite configuration at the 6-position. Then, in response to Patent Office rejection, according to the defendants, Pfizer argued that the epimer doxycycline was not obvious because, inter alia, the superiority of its antibacterial properties, compared with the properties of the McCormick compound, was an unexpected departure from the norm of prior art epimers of tetracycline antibiotics that revealed lower activity than their epimer counterparts. This analogy, the defendants contend, was falsely drawn to deliberately confuse and mislead the patent examiners. The basis for this allegation is that Pfizer knew doxycycline had the same 6-position structure as the conventional, or “normal” form of tetracyclines *190 (such as oxytetracycline), not the epimeric form, and thus Pfizer should have expected it to perform better. 19 The McCormick compound, on the other hand, had an opposite configuration at the 6-position and, defendants contend, might be expected to compare unfavorably with doxycycline. 20 Pfizer’s response, however, is that its label “conventional” when applied to a chemical structure was intended to refer only to a member of an epimer pair as first-discovered, and the prefix “epi” or “epimer” to the member discovered second. Pfizer contends it did not intend the word to imply that doxycycline had a “normal” or “natural” configuration in an absolute sense or to compare it with the “natural” process of making oxytetracycline, from which the McCormick and doxycycline compounds are derived “synthetically.” 21 The District Court concluded that Pfizer’s responses to rejections by the patent examiners, who eventually allowed the doxycyline patent, were deliberate misrepresentations using intentionally confusing language to draw the false analogy and conceal the fact that tetracycline has a 6-position structure similar to the “natural” fermentation produced tetracyclines such as oxytetracycline. Pfizer denies any intent to mislead or confuse the examiner. Again, we believe genuine issues of material fact are present, concerning but not limited to, Pfizer’s willfulness and the examiner’s understanding of technical facts critical to evaluating Pfizer’s conduct. 3. Misrepresentations of Antibacterial Superiority of Doxycycline. Pfizer is accused of misconduct in connection with a second argument that it advanced to overcome the examiners’ rejection of doxycycline as unpatentable because obvious. Pfizer submitted scientists’ affidavits to show that the antibacterial properties of doxycycline were not only superior to those of the McCormick compound but so dramatically superior that they were unexpected to persons skilled in the art. At the same time, however, defendants contend that Pfizer withheld other less favorable or unfavorable test results, and in so doing, Pfizer was guilty of putting its best foot forward, as an advocate, in violation of its duty of candor to the Patent Office. Pfizer responds that doxycycline is vastly superior to the McCormick compound in antibacterial activity against most organisms both in vitro (in test tubes) and in vivo (in experiments on animals). However, in *191 Pfizer’s experiments doxycycline proved to be inactive in vivo against an organism known as Staph 400. Pfizer did not report this shortcoming of doxycycline against Staph 400 in vivo to the Patent Office, but did report favorable data for doxycycline against Staph 400 in vitro. Pfizer’s explanation for doing so is that no comparable data revealing the McCormick compound’s activity against Staph 400 in vivo were available and it contends that, in granting the patent, the examiner did not assume that doxycycline had activity against Staph 400 in vivo. Pfizer also has an explanation for a second alleged misrepresentation relating to antibacterial activity defendants accuse it of perpetrating — Pfizer’s failure to report its 1964 test results that, defendants contend, reveal a less dramatic contrast between doxycycline and the McCormick compound than that indicated by the 1962 test results Pfizer previously submitted to the examiners. Pfizer contends that the 1964 test results need not have been submitted because they were not inconsistent with the 1962 results, but were within the range of variation to be expected from tests in different laboratory conditions. Again, we believe the issues of fact joined by these contrasting contentions are genuine and material to defendants’ claim of fraud and inequitable conduct. The District Court was unjustified in determining summarily the facts necessary to conclude that Pfizer’s argument by analogy and nondisclosure of test data amounted to “deliberate concealment, * * * withholding * * * and misrepresentation” to invoke the unclean hands defense and refuse to enforce the patent. 4. Failure to Disciose Beliefs Concerning Coproduction of Doxycycline. Pfizer is also accused of misconduct in delaying disclosure of its early belief that doxycycline had been coproduced in trace amounts by Drs. Stephens and McCormick independently in their work on the McCormick compound during the late 1950’s. 22 This nondisclosure, according to defendants, deprived the Patent Office for three years of information bearing upon the patent issues of “novelty” under 35 U.S.C. § 103 and erroneous claims of inventorship that may have justified striking the application under the Patent Office rules. 23 Pfizer responds that although it originally believed trace amounts of doxycycline were coproduced in prior research in the 1950’s, overwhelming evidence developed since April, 1962, reveals that no coproduction occurred in the prior processes. During prosecution of the doxycycline application, Pfizer investigated the possibility of prior coproduction and, it contends, in July, 1964, and February, 1965, presented the results to the examiners and summarized the matter in the file wrapper. The examiners thereafter declined Pfizer’s offer to produce more records and Pfizer employees for examination, as the examiners were of the view that the absence of prior art suggestion of coproduction rendered the matter immaterial to patentability. Pfizer’s patent *192 counsel, Mr. Ogelsby, contends that before he drafted the doxycycline application he concluded that earlier beliefs held by Pfizer scientists that trace coproduction had occurred were of no legal significance for the reason that the four named inventors were the men whose research had led to isolation and identification of doxycycline and determination of its proper utility. Pfizer contends, therefore, that Dr. Stephens executed the inventor’s oath in good faith and that, in light of its later explanations to the examiners, omission of the suggestions of coproduction from the 1961 application was not fraudulent or inequitable. The District Court concluded that Pfizer had deliberately withheld Dr. Stephens’ belief and unfavorable test data relating to coproduction and that Pfizer’s offers of additional information were insufficient to remedy its prior inequitable conduct. Again, however, we are of the view that these conclusions resolve genuine issues of material fact against Pfizer, not appropriate by summary judgment. At the least, Pfizer’s willfulness must be determined after full presentation of evidence by all parties in a plenary trial. 5. Concealment of Unsuccessful Efforts to Carry Out Processes Claimed in the Patent. The final acts of misconduct before the Patent Office charged against Pfizer are its alleged misrepresentations concealing unsuccessful efforts to carry out two of the many chemical processes claimed in the patent. The first is the statement in Claim 1 of the patent that the noble metal Ruthenium is a “preferred” and “entirely opera-five” catalyst for producing doxycycline by hydrogenolysis. The second is Pfizer’s implicit representation that the process described in Example 35 of the patent, for making a doxycycline-related compound called 7-chloro-doxycycIine, is operative. Defendants contend that until the issuance of the patent in 1965, Pfizer had failed in every experimental attempt to use the processes, concealing its failures from the Patent Office, and that but for the concealment, the patents on these two processes would not have issued. 24 The District Court concluded that Pfizer was unable to carry out the claimed processes and that its failure to disclose its inability was inequitable. Pfizer responds that the documents upon which the District Court relied in so concluding actually report Pfizer’s successful use of Ruthenium and that throughout the prosecution Pfizer had evidence that both processes were operable. Thus, Pfizer insists, technical issues are present requiring expert interpretation of scientific records. Further, according to Examiner Adams’ 1975 affidavits, he was not interested in Pfizer’s unsuccessful experiments as long as Pfizer believed the processes were operable. Adams stated he would have allowed the patent to issue unless he were shown conclusive scientific evidence of inoperability and no such evidence was presented. 25 Our review of the record reveals unresolved technical issues concerning Pfizer’s success with the chemical processes and genuine issues of material fact as to Pfizer’s willfulness in allegedly concealing lack of success. The District Court erred in *193 summarily finding Pfizer guilty of inequitable conduct before the Patent Office barring enforcement of the patent. We reiterate that summary judgment is not a proper procedure for final determination of factual issues. III. FRAUD ON THE COURT AND OTHER INEQUITABLE CONDUCT. The District Court also found Pfizer guilty of making false statements and recklessly disregarding the truth in pretrial proceedings before it, independently justifying refusal to enforce the patent. Pfizer flatly denies any misconduct and complains of the summary manner in which it was found guilty and punished. We have independently reviewed the record and reverse that judgment on the merits. The record of proceedings in the District Court does not reveal sufficient misconduct by Pfizer’s agents or counsel to constitute fraud on the court, reckless disregard for the truth or unclean hands, and the court’s findings thereof are clearly erroneous. Fed.R.Civ.P. 52(a). Imposition of the severe sanction was not proper, the factual situation being inadequate for a final disposition. The facts allegedly concealed were contained in memoranda prepared by the patent counsel who prosecuted Pfizer’s doxycyeline application. The memoranda concern Pfizer’s conduct in connection with disclosure of the Belgian patent to the patent examiners. 26 In the present litigation defendants allege that Pfizer withheld the Belgian patent from the patent examiners to conceal its uncertainty of doxycycline’s 5-position stereochemistry,. but Pfizer insists that all such information was later disclosed to the examiners in oral interviews not recorded in the file wrapper. 27 In *194 support, Pfizer relies upon internal memoranda relating to the interviews contained in the patent prosecution files of its patent counsel. However, due to a special “ground rule” of discovery that evolved in the case in connection with the masters who were appointed to rule on claims of privilege and work product, these memoranda were not produced by Pfizer in the discovery process. Under the rule, internal documents remaining in the files of the law firm that created them need not be produced to the masters for determination of privilege claims so long as the documents were attorney work product and were never disclosed to anyone, including the client, outside the firm. 28 When Pfizer relied upon the undisclosed memoranda in September, 1974, to corroborate its claim that it orally disclosed the Belgian patent and uncertainty of 5-posi-tion stereochemistry to the examiners during prosecution, defendant USV moved to compel Pfizer to produce the memoranda and accused it of unlawfully withholding them. Pfizer maintained, as it had previously, that the memoranda were protected under the discovery ground rule as attorney work product that had never been disclosed outside the law firm. However, during January and February, 1975, in hearings on USV’s motion, Pfizer’s counsel disclosed that the files of its patent counsel containing the memoranda had been loaned to Pfizer early in uhe litigation. Thus, Pfizer conceded that the memoranda were not covered by the discovery ground rule and the District Court characterized Pfizer’s conduct as fraudulent and reckless. Pfizer responds that its continuing misstatement that the files had not been disclosed outside the firm stemmed from an innocent error of counsel, promptly corrected when discovered on January 28, 1975. Moreover, it contends that by the time the error was discovered it was no longer significant because all facts pertaining to Pfizer’s earlier uncertainty of 5-position stereo-chemistry had been disclosed to the court and explored by the defendants in other discovery. In fact, copies of the memoranda had been voluntarily submitted to the overburdened masters on October 18, 1974; but as of February, 1975, the masters had not yet screened documents submitted as early as May, 1974. Thus, Pfizer contends, defendants could not have been prejudiced by the delayed disclosure resulting from the error. Moreover, Pfizer contends, the memoranda are favorable to its position in the litigation in that they confirm that the Belgian patent and Pfizer’s uncertainty of 5-posi-tion stereochemistry were disclosed to the examiners during prosecution. 29 Thus, Pfizer points out that it had no motive to conceal the memoranda, and on February 3, 1975, it voluntarily waived its privilege and work product claims and delivered them directly to the defendants. *195 Fraud on the court, though not easily defined, can be characterized as a scheme to interfere with the judicial machinery performing the task of impartial adjudication, as by preventing the opposing party from fairly presenting his case or defense. Kupferman v. Consolidated Research & Manufacturing Corp., 459 F.2d 1072, 1078 (2d Cir. 1972); England v. Doyle, 281 F.2d 304, 309 (9th Cir. 1960); see Annot., 19 A.L.R.Fed. 761 (1974). A finding of fraud on the court is justified only by the most egregious misconduct directed to the court itself, such as bribery of a judge or jury or fabrication of evidence by counsel, United States v. International Telephone & Telegraph Corp., 349 F.Supp. 22, 29 (D.Conn.1972), aff’d sub nom. Nader v. United States, 410 U.S. 919, 93 S.Ct. 1363, 35 L.Ed.2d 582 (1973); 7 J. Moore, Federal Practice H 60.33, at 515 (2d ed. 1976), and must be supported by clear, unequivocal and convincing evidence. Barr Rubber Products Co. v. Sun Rubber Co., 425 F.2d 1114, 1120 (2d Cir.), cert. denied, 400 U.S. 878, 91 S.Ct. 118, 27 L.Ed.2d 115 (1970). We have reviewed the record of proceedings in the District Court and we believe the conduct of the accused individuals did not evidence a scheme to improperly influence or impair the court’s function, and was not egregious misconduct characterizable as fraud on the court. Kupferman v. Consolidated Research Manufacturing Corp., supra at 1081; cf. Hazel-Atlas Glass Co. v. Hartford-Empire Co., 322 U.S. 238, 245-46, 64 S.Ct. 997, 88 L.Ed. 1250 (1944) (scheme by patentee to defraud the court by relying upon an article supposedly written by a disinterested author, but actually written and published by patentee); Keystone Drill er Co. v. General Excavator Co., 290 U.S. 240, 243, 54 S.Ct. 146, 78 L.Ed. 293 (1933) (scheme by patentee to suppress evidence by paying witness to remain silent); Mas v. Coca-Cola Co., 163 F.2d 505 (4th Cir. 1947) (scheme by patentee to establish priority of invention in court with forged documents). Nor does the record support the District Court’s alternative finding that Pfizer’s conduct in the pretrial proceedings, even if not a fraud on the court, amounted to unclean hands, justifying refusal to enforce the patent. Such extreme sanction, predicated upon the unclean hands doctrine, also requires a showing of fault, willfulness or bad faith more substantial than presented here. See Societe Internationale v. Rogers, 357 U.S. 197, 212, 78 S.Ct. 1087, 2 L.Ed.2d 1255 (1958); General Dynamics Corp. v. Selb Manufacturing Co., 481 F.2d 1204,1211,1214 (8th Cir. 1973), cert. denied, 414 U.S. 1162, 94 S.Ct. 926, 39 L.Ed.2d 116 (1974). The root of confusion that triggered the court’s drastic action was but a single misrepresentation that the memoranda had not been released by the law firm, repeated by many individuals on Pfizer’s behalf until discovered by attorney Ogelsby to be erroneous on January 28, 1975. 30 Thereafter, on January 31, 1975, counsel for Pfizer promptly brought the error to the court’s attention. Pfizer’s conduct had the effect of impeding discovery and prolonging this already protracted litigation and is not to be viewed lightly, but it did not justify the severe sanction imposed. Though Pfizer’s failure to investigate the handling of the internal memoranda at an earlier date was a serious error in judgment, the conduct of *196 the accused individuals acting on its behalf was not fraudulent or reckless, nor indicative of a scheme to conceal material facts. The facts so concealed were basically supportive of Pfizer’s contentions. The basic issues in this case are the validity and enforceability of the patent alleged to have been infringed. An infringement defendant in complex litigation should not be permitted to sidestep these main issues by nit-picking the patent file in every minute respect with the effect of trying the patentee personally, rather than the patent. A patentee’s oversights are easily magnified out of proportion by one accused of infringement seeking to escape the reach of the patent by hostilely combing the inventor’s files in liberal pretrial discovery proceedings. Unjustified damage to professional and social reputations can result, as here, without fostering any corresponding public benefit in the form of inhibiting future improvident grants of patent monopolies. IV. RELIEF. Finally, Pfizer requests that we remand this action to the Central District of California for completion of pretrial proceedings and trial. It argues that the District Judge herein has made factual determinations that he may not be able to retract, that IRC’s antitrust action against Pfizer has now been settled, leaving no other litigation involving IRC and USV before this District Judge, and that no other judge in the District of Minnesota is familiar with the instant case. We deny Pfizer’s request. The case was placed by the Judicial Panel on Multidistrict Litigation in the hands of the District Court for pretrial proceedings. When ready for trial, the case will take its normal course under the direction of the Panel without direction from this court. 31 In conclusion, it should be understood that we are not attempting to resolve disputed issues of fact or pass upon the serious charges made by the defendants in regard to the manner in which the patent application was prosecuted by Pfizer before the Patent Office. Nor do we attempt any resolution of the statutory issues of validity or equitable issues of enforceability of the patent raised by the defendants. All of these issues remain yet to be resolved in a plenary trial. We, however, do hold that according to the record before us, no fraud was practiced on the District Court in these pretrial proceedings. The judgment of the District Court is reversed and the case remanded for completion of pretrial proceedings consistent with this opinion, and a plenary trial in a district to be selected by the Judicial Panel on Multidistrict Litigation. Costs are assessed against defendants. Footnotes [1] International Rectifier Corp. and its four subsidiaries, Rachelle Laboratories Italia, S.p.A., Rachelle Laboratories, Inc., Rachelle Pharmaceuticals International, S.A., and Rachelle Laboratories (Philippines), Inc., are hereafter referred to jointly as IRC. [2] U.S. Patent No. 3,200,149 was issued on August 10, 1965, for a chemical compound, and the process of producing the compound, called doxycycline (Pfizer tradename Vibramycin), a member of the tetracycline family of broad spectrum antibiotic drugs. [3] Pfizer opposed the motion with various memoranda, 54 exhibits and 24 affidavits. The affidavits were prepared by Patent Examiner Adams, one of three Examiners (Adams, Berg and Modance) who examined and allowed the doxycycline patent, scientists employed by Pfizer, Pfizer managers, outside patent counsel and outside legal counsel. The defendants supported their motion with memoranda and one affidavit of Examiner Adams, otherwise relying upon Pfizer’s documents, deposition testimony of Pfizer’s representatives, and the official prosecution record of the patent in the Patent Office, called the “file wrapper.” [4] 35 U.S.C. § 102 states in relevant part as follows: Conditions for patentability; novelty and loss of right to patent A person shall be 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, or * * sis * * * (f) he did not himself invent the subject matter sought to be patented, or (g) before the applicant’s invention thereof the invention was made in this country by another who had not abandoned, suppressed, or concealed it. [5] 35 U.S.C. § 103 states in relevant part as follows: 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. [6] The term “epimer” is defined in part IIA., infra. [7] 35 U.S.C. § 112 states in relevant part as follows: Specification 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. * * * * * * An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. [8] We express no view concerning Pfizer’s asserted entitlement to a jury trial on its claim of infringement or on the defendants’ assertions of patent invalidity and nonenforceability on equitable grounds. See Beacon Theatres, Inc. v. Westover, 359 U.S. 500, 510-11, 79 S.Ct. 948, 3 L.Ed.2d 988 (1959); Annot., 18 A.L.R. Fed. 690 (1974). The sole issues raised in this appeal are the propriety of determining the enforceability of the doxycycline patent summarily prior to any plenary trial, and whether Pfizer’s conduct before the District Court merited the severe sanction. [9] We express no view of the facts sought to be established in Pfizer’s 170 requests for admission by the defendants filed pursuant to Fed.R. Civ.P. 36 on March 17 and 19, 1975. The defendants did not file a response to the requests within 30 days. On April 21, 1975, the District Court ordered them answered in an additional 45 days. Thereafter, on the due date, the defendants filed an objection to all the requests as “burdensome and harassing” and suggested that partial summary judgment in their favor would moot the need for any response. Ultimately, the District Court held the requested admissions “irrelevant” and granted summary judgment for defendants declaring the patent invalid and unenforceable. As all the facts will have to be established on remand, it will also be necessary for the District Court to reexamine its position concerning Pfizer’s requests for admissions. [10] The lower court cases cited by the defendants, Interlego, A.G. v. F.A.O. Schwartz, Inc., 187 U.S.P.Q. 580 (N.D.Ga.1975), and Farmer Brothers Co. v. Coca-Cola Co., 384 F.Supp. 595 (C.D.Cal.1974), are distinguishable and not controlling. [11] The cases have distinguished between fraud, which invalidates the patent, and inequitable conduct which is said to render the patent unenforceable rather than invalid. However, in an infringement action involving misconduct in patent procurement the distinction carries little practical significance as under either theory the patent monopoly is terminated if sufficient misconduct is proved. Timely Products Corp. v. Arron, 523 F.2d 288, 297-98 (2d Cir. 1975). It has been suggested that a stronger showing of materiality of the information undisclosed or misstated is required to prove fraud than to prove inequitable conduct. Id. We express no view of the standards of materiality applicable under either theory. [12] The reasons for rejecting such a broad standard were well stated by District Judge Mansfield in Xerox Corp. v. Dennison Mfg. Co., 322 F.Supp. 963, 968 (S.D.N.Y.1971): [T]o deny enforcement as a matter of law merely because of an innocent or good faith non-disclosure would go beyond what is necessary to protect the public against the improvident granting of a monopoly. Such a standard could also have the harmful effect of forcing a patent solicitor to flood the Patent Office in each case with a mass of data of doubtful materiality rather than take the risk that an inventor might later be denied the fruits of his monopoly because of failure to reveal some fact later magnified out of proportion by an infringer seeking to escape the reach of the patent by combing the inventor’s files under our liberal pretrial discovery procedures and dredging up new found “facts.” See also Becton Dickinson & Co. v. Sherwood Medical Industries, Inc., 516 F.2d 514, 524 (5th Cir. 1975); Schnadig Corp. v. Gaines Mfg. Co., 494 F.2d 383, 393 (6th Cir. 1974). In correcting the District Court’s interpretation of the law governing this defense we express no view of Pfizer’s conduct before the Patent Office or the merits of the defendants’ claims. [13] Monsanto Co. v. Rohm & Haas Co., 456 F.2d 592, 598, 601 n.14 (3d Cir.), cert. denied, 407 U.S. 934, 92 S.Ct. 2463, 32 L.Ed.2d 817 (1972); Kolene Corp. v. Motor City Metal Treating, Inc., 440 F.2d 77, 83 (6th Cir.), cert. denied, 404 U.S. 886, 92 S.Ct. 203, 30 L.Ed.2d 169 (1971); Scott Paper Co. v. Fort Howard Paper Co., 432 F.2d 1198, 1204-05 (7th Cir. 1970), cert. denied, 401 U.S. 913, 91 S.Ct. 882, 27 L.Ed.2d 812 (1971); Norton v. Curtiss, 433 F.2d 779, 793-95, 57 CCPA 1384 (1970) (interference proceeding to challenge priority of invention and strike application for fraud); see Acme Precision Products, Inc. v. American Alloys Corp., 484 F.2d 1237, 1239 — 10 (8th Cir. 1973) (infringement claim dismissed due to deliberate misrepresentations and patentee’s knowledge of fraudulent procurement); Cataphote Corp. v. DeSoto Chemical Coatings, Inc., 450 F.2d 769, 772 (9th Cir. 1971), cert. denied, 408 U.S. 929, 92 S.Ct. 2497, 33 L.Ed.2d 341 (1972) (antitrust counterclaim alleging fraud); Nashu Corp. v. RCA Corp., 431 F.2d 220, 227 (1st Cir. 1970) (claim for refund of royalties due to fraud in procurement); Triumph Hosiery Mills, Inc. v. Alamance Industries, Inc., 299 F.2d 793, 796 (4th Cir.), cert. denied, 370 U.S. 924, 82 S.Ct. 1566, 8 L.Ed.2d 504 (1962); Edward Valves, Inc. v. Cameron Iron Works, Inc., 286 F.2d 933, 947, modified on other grounds, 289 F.2d 355 (5th Cir.), cert. denied, 368 U.S. 833, 82 S.Ct. 55, 7 L.Ed.2d 34 (1961); Haloro, Inc. v. Owens-Coming Fibreglas Corp., 105 U.S.App.D.C. 320, 266 F.2d 918 (1959). [14] Parker v. Motorola, Inc., supra at 535; Schnadig Corp. v. Gaines Mfg. Co., 494 F.2d 383, 393 (6th Cir. 1974); Monsanto Co. v. Rohm & Haas Co., 456 F.2d 592, 598, 601 n.14 (3d Cir.), cert. denied, 407 U.S. 934, 92 S.Ct. 2463, 32 L.Ed.2d 817 (1972); Carter-Wallace, Inc. v. Davis-Edwards Pharmacal Corp., 443 F.2d 867, 882 (2d Cir. 1971); Scott Paper Co. v. Fort Howard Paper Co., 432 F.2d 1198, 1204-05 (7th Cir. 1970), cert. denied, 401 U.S. 913, 91 S.Ct. 882, 27 L.Ed.2d 812 (1971); see Haloro, Inc. v. Owens-Corning Fibreglas Corp., 105 U.S.App.D.C. 320, 266 F.2d 918 (1959); Norton v. Curtiss, 433 F.2d 779, 795-96, 57 CCPA 1384 (1970). See also Iron Ore Co. v. Dow Chemical Co., 500 F.2d 189, 195 (10th Cir. 1974); Carney, Misrepresentations Before the Patent Office: Antitrust and Other Legal Effects, 12 B.C.Ind. & Com.L.Rev. 1005 (1971). *187 Defendants’ contention that willfulness or a similar state of mind need not be proved to render misconduct sufficiently inequitable to vitiate a patent is not supported by Monsanto Co. v. Rohm & Haas Co., supra; Beckman Instruments, Inc. v. Chemtronics, Inc., 428 F.2d 555, duplicate report at 439 F.2d 1369 (5th Cir.), cert. denied, 400 U.S. 956, 91 S.Ct. 353, 27 L.Ed.2d 264 (1970); and Monolith Portland Midwest Co. v. Kaiser Aluminum & Chemical Corp., 407 F.2d 288 (9th Cir. 1969), upon which they rely. In Monsanto, the court found that knowingly false statements were made to the Patent Office and material facts deliberately concealed that would otherwise have resulted in denial of the patent. 456 F.2d at 595, 601 n.14. Similarly, in Beckman Instruments, the court found that knowledge of the prior art was deliberately withheld and would have resulted in denial of the patent. 428 F.2d at 560, 564-65. And in Monolith Portland, a deliberate withholding of crucial information that would also have resulted in denial of the patent was recognized to justify an award of attorneys fees under the “exceptional cases” rule of 35 U.S.C. § 285. 407 F.2d at 294-97. [15] A “product-by-process” claim is a form of product patent for a compound that cannot be described other than by reference to the process of making it. P. Rosenberg, Patent Law Fundamentals 76 (1975). It is substantially narrower and less desirable than the broad product claims, covering hundreds of compounds, in the doxycycline patent. Defendants contend that because Pfizer admittedly did not know the structure of doxycycline when it filed the application, Pfizer should have referred to the prior art as the “known 6-deoxy-tetracyclines, the structure of which has not been fully established.” Pfizer, they contend, experienced the same problem in its prior tetracycline patent application and should have known that it could overcome patent office rejection, as it had before, by truthful and complete disclosure. Instead, the defendants contend, Pfizer intentionally suppressed the prior art to avoid disclosing its ignorance of the structure of doxycycline and resultant inability to distinguish the prior art. [16] Pfizer’s diagrams compare the compounds in part as follows: [17] Under In re Von Bramer, 127 F.2d 149, 151, 29 CCPA 1018 (1942), before being modified in Application of LeGrice, 301 F.2d 929, 942, 49 CCPA 1124 (1962), and sharply limited in Application of Brown, 329 F.2d 1006, 1009-11, 51 CCPA 1254 (1964), a prior art diagram depict *189 ing the structure of a later-produced compound, even accidentally, could serve as a complete anticipation precluding patentability under 35 U.S.C. § 102(a). In the instant case, the District Court concluded that the patent examiners did not believe until June, 1964, that Von Bramer had been effectively overruled. Pfizer disputes this finding, and maintains that the correct date is May, 1962. [18] According to Pfizer, such disclosures took place in March, 1962, March, 1963, April, 1964, and July, 1964. [19] Thus, according to defendants, Pfizer knew that the fact that prior art epimers showed a pattern of lower anti-bacterial activity than their conventional counterparts did not support the argument that doxycycline’s higher activity was unexpected. For, although Pfizer referred to doxycycline as having an “epimeric” configuration, defendants contend that Pfizer knew doxycycline was in no way analogous to prior art epimers since doxycycline has a configuration comparable to, rather than opposite, that of the conventional tetracycline antibiotics having higher activity. [20] Pfizer’s diagrams illustrate the so-called “normal” 6-position form of oxytetracycline and doxycycline by a dotted line indicating the methyl group (CH3) as pointing “down.” The epimeric or opposite form of the McCormick compound is shown by a solid line indicating CH3 in the “up” position: [21] At one point in the patent prosecution, the examiner referred to the 6-methyl group of doxycycline as having an orientation opposite that of the fermentation produced compounds. Pfizer contends that it immediately corrected the examiner by explaining that doxycycline has an orientation the same as that of the fermentation produced tetracyclines. The examiner accepted Pfizer’s correction and has filed affidavits explaining his understanding of this and other issues in the instant proceeding. [22] In 1958 and 1959, Drs. Stephens and Con-over of Pfizer filed two patent applications describing their own hydrogenolysis process stating that the process produced another compound “thought to be” what is now called doxycycline. Defendants contend that Dr. Stephens further believed that Dr. McCormick’s Belgian patented process would also have co-produced doxycycline. However, in its 1961 doxycycline patent application, Pfizer did not disclose these beliefs of possible coproduction but claimed that doxycycline had been invented jointly by Drs. Blackwood, Stephens, Rennhard and Beereboom with a process different from that used before by Drs. Stephens, Conover and McCormick. [23] Patent Office Rule 56, 37 C.F.R. § 1.56 (1961), provides that any application filed fraudulently “or in connection with which any fraud is practiced or attempted” may be stricken from the files. In their oath of invention, dated May 2, 1961, Dr. Stephens and three other inventors swear that they “believe” they “are the original, first and joint inventors” of the claimed tetracycline derivatives and processes and that they “do not know and do not believe [doxycycline] was ever known or used before [their] invention thereof or patented or described in any printed publication in any country before.” [24] Affidavits were submitted by Pfizer scientists claiming that the processes are operative, but the defendants discount them for failure to cite actual test results. Defendants also challenge Pfizer’s offer of later-developed successful test data because not raised below and not relevant to Pfizer’s inequitable concealment from the Patent Office years before. Defendants also note that Examiner Adams, in his affidavit of February, 1975, stated that he would have disallowed patents on the processes had he been informed of conclusive scientific evidence that they are inoperable. [25] In his second supplemental affidavit, dated February 24, 1975, Examiner Adams states: I was not interested in any and every unsuccessful experiment carried out by Pfizer’s research scientists. * * * * * * I would have been interested, therefore, only in conclusive scientific evidence that showed that the hydrogenation could not be conducted. if; s(c if: * if: if: I do not regard failure to obtain the desired result in several exploratory experiments as conclusive scientific evidence of inoperability. [26] As noted above, Pfizer maintains that when it filed the doxycycline application in May, 1961, it believed the diagrams in the Belgian patent did not depict doxycycline and hence ■ were not Von Bramer references required to be cited as prior art. During prosecution, however, in 1963, a Pfizer scientist expressed doubt as to the difference between the 5-position structure of doxycycline and the prior art compounds, and Pfizer resumed experimentation, finally establishing in 1964 that its original belief was correct. [27] For example, Pfizer contends that in March, ' 1962, its patent counsel, Mr. Ogelsby, brought the Belgian patent to the examiners’ attention and stated that Pfizer believed the diagrams contained therein were in error at the 5- and 6-positions. Although the defendants contend that under Patent Office Rules 2 and 133, 37 C.F.R. §§ 1.2, 1.133 (1961), all information must be submitted to the Patent Office in writing, it appears that the practice is not so restrictive and that many items are discussed with the patent examiners orally rather than in writing. One commentator has recognized that “the personal appearance of an applicant and/or his * * * agent before the examiner frequently facilitates the disposition of his case.” P. Rosenberg, Patent Law Fundamentals 221 (1975). The U. S. Patent Office examiner’s manual explains the practice under Rules 2 and 133 and distinguishes situations in which an oral agreement is reached between the attorney and the examiner from those in which an interview is had but no agreement is reached. In the former situation, a formal memorandum of the interview, signed by both, must later be filed by .the applicant; but in the latter situations the exam- = iner is expected to place in the file an informal memorandum that should later be “removed from the file if and when the case is passed to issue.” U. S. Patent Office, Manual of Patent Examining Procedure § 713.04 (3d ed. 1961), as amended (1970), reprinted in, A. Deller, Walker on Patents ch. XVI, appendix at 383 (2d ed. 1972). The Supreme Court has recognized that “all facts concerning * * * [possible fraud or inequitableness should] be submitted formally or informally to the Patent Office” to safeguard the public interest in proper patent administration. Precision Instrument Mfg. Co. v. Automotive Maint. Mach. Co., supra, 342 U.S. at 818, 65 S.Ct. at 999. (Emphasis added.) We will not attempt in this opinion to resolve the impact of the above Patent Office rules and procedures on evidentiary rulings in district court infringement litigation. The District Court made no precise ruling on that issue and if the court’s decision rests upon enforcement of the Patent Office rules, we would be confronted with an additional issue which we do not perceive. We express no view of Pfizer’s alleged noncompliance with required procedures in the Patent Office or of the effect of such noncompliance, if any, upon Pfizer’s right to rely on past oral interviews in its proof with reference to statutory issues of patentability. But quite aside from the admissibility of the evidence as it relates to such statutory issues, we believe evidence of matters brought orally or informally to the attention of the Patent Office is relevant to the issues of fraud and *194 inequitable conduct by the patentee and should be admitted in the instant litigation as to these issues, if sufficiently reliable, regardless of the patentee’s alleged noncompliance with Patent Office rules and procedures governing proceedings determining statutory patentability. The District Court should resolve these separate issues on remand and reconsider its prior estoppel of Pfizer from relying upon evidence of oral Patent Office interviews. [28] The opposing party seeking such undisclosed documents is required to proceed instead by subpoena duces tecum directed to the law firm in the forum where the firm is located. The rule thus enables the firm to defend the sanctity of its own files. All other documents not covered by the ground rule were to be produced to the masters for consideration of claims of privilege. Both parties could then dispute the masters’ rulings as to discoverability of- individual documents. [29] For example, a March 12, 1963, memorandum of Dr. Knuth, Associate Director of Patents for Pfizer, produced to defendants on May 13, 1974, states that the Belgian patent and uncertainty of 5-position stereochemistry (raised by a publication of a Pfizer consultant, Prof. Muxfeldt, in December, 1962) were discussed with Examiner Berg in an interview in the Patent Office on March 5, 1963, and summarizes Pfizer’s presentation. Also, Pfizer had produced its patent files to the masters in 1973, including Pfizer’s March 7, 1961, letter to Ogelsby discussing the Belgian patent and revealing Pfizer’s preapplication beliefs concerning 5- and 6-position stereochemistry. In June, 1973, the masters ruled the document privileged. [30] We have reviewed the record relating to four additional items asserted to have been part of Pfizer’s scheme and hold them to be inadequate even in combination to justify a finding of fraud on the court or inequitable conduct to trigger the severe sanction. These are: (1) an argument by Pfizer’s counsel, Mr. Cooper, that 5-position (as opposed to 6-posi-tion) stereochemistry was always irrelevant to patentability of doxycycline, found by the court to be a deliberate misrepresentation; (2) an affidavit of Mr. Ogelsby, dated July 30, 1974, found by the court to have contained the same misrepresentation; (3) Mr. Ogelsby’s preparation for and testimony during his March 5, 1974, deposition, found by the court to have been deliberately incomplete and misleading; and (4) a notation on a draft patent application contained in Ogelsby’s files, found by the court to indicate that some of Ogelsby’s files were provided to Pfizer in February, 1974, apparently suggesting that Ogelsby had knowledge thereof prior to his March, 1974, deposition. Moreover, in our view, Pfizer has a reasonable explanation relating to each of these four allegations of misconduct. [31] Orders of the Panel are reviewable only by extraordinary writ. 28 U.S.C. § 1407(e). Pfizer, of course, is free to request an order of remand from the Panel itself. Rules of Procedure of the Judicial Panel on Multidist. Lit. 11(c) and (d) (1975).