Apropos of most patent litigation, the instant claims of infringement- countered by the invalidity defenses of anticipation, obviousness and a failure of complete disclosure to the patent office (coupled with a denial of infringing use, and a claim for anti-trust relief)- present this court with problems which are difficult to resolve. Our task is further complicated by the subject matter of the patent, which is a drug useful for human medication composed of a complex organic molecule bearing the convenient, "brief" name, propoxyphene hydrochloride. As though this were not enough, the Gordian knot we must judicially sever is completed by intertwined objections to the procedures followed by the lower court, which defendants-appellants claim moved the preliminary hearing along too erratically and then erroneously resulted in the entry of a final judgment in derogation of their demands for a trial by jury, after a hearing they thought was restricted to consideration of preliminary injunctive relief.

The chemical compound here involved is widely distributed by the present patent holder, Eli Lilly & Company, Inc. (Lilly) under the registered trade name Darvon. It is a synthetic analgesic of the morphine type which acts on the central nervous system and is claimed to be the most efficacious non-addictive pain reliever. A patent for this compound was applied for by Dr. Albert Pohland, on December 3, 1952; and on December 27, 1955, Patent No. 2728779 was granted to Lilly's parent corporation as the assignee of Dr. Pohland, its employee. The drug proved to be a prompt and sustained commercial success. It has been classified as the most often prescribed drug dispensed in retail pharmacies in the United States, and enjoys a current annual rate of sale exceeding 65,000,000 dollars. No manufacturer other than Lilly has ever produced or sold the drug in the United States. Consent judgments have been entered against two companies who previously proposed to import propoxyphene hydrochloride for sale in this country.

The present litigation was instituted by Lilly's complaint seeking preliminary and final injunctive and declaratory relief, damages and attorney's fees against the appealing drug companies and one of their top officials. These defendants not only contested Lilly's right to any relief sought, but counterclaimed for injunctive relief in their favor as well as monopoly damages. In their response the defendants demanded a jury trial on all issues arising from the complaint and the counter-claims. After a lengthy evidentiary hearing on the motion for preliminary injunction, the district judge entered detailed and highly technical findings of fact and conclusions of law, D.C.Fla.1971, 324 F. Supp. 715. These were followed by the entry of a preliminary injunction, from which the defendants took an initial appeal to this court. Shortly after that appeal was taken, the district judge, acting sua sponte, set the matter down for hearing on final judgment and permanent injunction. Subsequent to a chambers conference with counsel for all parties, whereat Lilly renounced its claim for damages, the court concluded that no triable issue of fact remained in the action and thereupon permanently enjoined the manufacture and sale operations contemplated by defendants, and dismissed the counterclaim.

The burden is ordinarily on the party seeking preliminary injunctive protection in a patent infringement suit to demonstrate beyond question that the patent he sues on is valid and infringed, as well as showing that other equitable grounds are present. Hoeme v. Jeoffrey, 100 F.2d 225 (5th Cir. 1938). However, this rule is ameliorated where the patent at issue has been the subject of long acquiescence or has been adjudicated to be valid. Rosenberg v. Groov-Pin Corp., 81 F.2d 46 (2nd Cir. 1936); See also Carter-Wallace, Inc. v. Davis-Edwards Pharmacal Corp., 443 F.2d 867 (2nd Cir. 1971).

The trial court's determination to flex the injunctive strong-arm of equity is a matter committed to the discretion of that court. The exercise of this discretion can only be reversed upon an appellate determination that such discretion has been abused. The test is one that must be applied in terms of the district court's right to make the determination it did, and not by any subtle substitution of this court's discretion for that of the tribunal which weighed the equities and made the choice. See Bayless v. Martine, 430 F.2d 873 (5th Cir. 1970) and Detroit Football Company v. Robinson, 283 F.2d 657 (5th Cir. 1960).

Both sides submitted proposed findings to the district court which adopted verbatim the findings of fact suggested by the plaintiff. While acceptance of a party's findings after a decision has been reached (a condition not present here) is the subject of judicial criticism, all findings of fact and conclusions of law upon which a district court bases its actions are to be reviewed on an identical standard regardless of whether such findings are court-originated or party-suggested. That standard is: Are such findings clearly erroneous? We have recently analyzed this exact point in a patent context. See Part II of Railex Corp. v. Speed Check Co., 457 F.2d 1040 (5th Cir. 1972).

From a detailed examination of the record, we conclude that the critical findings of fact by the district court were not clearly erroneous. Cf. Graver Tank & Mfg. Co. v. Linde Air Products Co., 336 U.S. 271, 274, 69 S. Ct. 535, 93 L. Ed. 672 (1949), modified on other grounds 339 U.S. 605, 70 S. Ct. 854, 94 L. Ed. 1097 (1950). In view of the age of the patent and the wide, commercially successful distribution enjoyed by the patented product, these findings are legally sufficient to answer the basic factual inquiries necessary to demonstrate beyond question that the patent was both valid and infringed. If the legal question of patent validity may be answered in the affirmative, then it was within the ambit of the federal chancellor's discretion to grant preliminary injunctive relief. Graham v. John Deere Co., 383 U.S. 1, 17, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966).

The district court's findings, D.C.Fla., 1971, 324 F. Supp. 715, make it unnecessary for us to treat extensively with the background details of the inventor's claimed process of discovery or with the factual issues involved in the disputes between the parties, in reasoning our decision whether, as a matter of law, the preliminary order must be affirmed. However, before launching into a discussion of the chemical controversy about which the merits and procedural issues revolve, a foreword of an unusual sort seems appropriate. Since most readers will be unfamiliar with organic chemistry, the following lexicon-drawn largely from the stipulation of the parties and tailored to this particular controversy-should facilitate the understanding of both the district court's findings and our reasoning here.

analogue-one of a series of slightly varied molecules which all have a similar basic structure of atoms.

homologue-one of a series of similar molecules which vary by a regular difference in atomic structure, especially the members of a series which vary by only one carbon and two hydrogen atoms.

stereoisomerism-a long known and predictable condition of certain chemical compounds in which two identical groups of atoms are linked in the same order, but differ in their arrangement in space. Simple analogies may be drawn to the left and right hand gloves from a single pair or to an object and its image in a mirror. The stereoisomers pertinent here are opticial isomers, which means that one of the separate atomic groups, or isomers, will rotate a straight beam of light to the left and the other will rotate it to the right. The right rotating isomer is called the dextro or disomer, and its opposite is designated the levo or l-isomer.

diastereoisomer-a compound containing more than one set of stereoisomers. In the case at bar we are concerned with a compound in which two such sets are involved. The distinction between the sets of isomers or the stereoisomers in such a compound is based upon a difference in the solubility of each set. The less soluble pair here is designated by the Greek letter alpha, which is written a; and the more soluble pair by the letter beta, written B.

racemate-a name coined by Pasteur for a compound consisting of equal parts of a dextro and a levo isomer, also referred to as a racemic mixture.

ester-a product which results when an organic acid is mixed with an alcohol. More importantly to the problems at hand, esters fall into one of two classes, normal or reverse, according to the manner of their chemical bonding.

Typical of drug patents, the patent in suit consists of specifications describing the uses of the chemical, a detailed explanation of the method of preparation, and the claim which consists solely of the full technical name of the compound and a schematic diagram of its formula. The 779 patent, which claimed six compounds, specified that each of them were analgesics "characterized by their ability to produce analgesia without toxic side effects, such as respiratory depression." The specifications further stated: "The a-dl diastereoisomers are the preferred compounds of the invention since they possess marked analgesic activity in contrast to the B-dl diastereoisomers, which are substantially inactive." Only Claim 2, covering a-dl propoxyphene hydrochloride, is in suit.

The almost reflexive counterattack defenses to a patent infringement claim-charges of invalidity of the patent due to anticipation and obviousness -as well as the less frequently encountered claim that relevant prior art was withheld from the patent examiner, are rendered meritless in fact by the trial court's findings and conclusions which we here uphold. Those findings and conclusions likewise dispose of defendants' contention that the proposed use would not constitute an infringement of the patent even if it were conceded arguendo to be valid. Nevertheless, we may not rest our decision regarding the preliminary injunction solely upon such a fact review and approval basis. Graham requires a review of the enmeshed legal issues as well.

The patent in suit is not a newcomer to the drug art. It has been publicly disclosed for well over 19 years and was formally patented and thereafter widely sold in the United States by Lilly alone for over 15 years-almost the entirety of its statutory lifetime of 17 years. See 35 U.S.C.A. Sec. 154. The only prior inconsistent claims were promptly quashed. Such long continued public acquiescence in the monopoly created by a patent claim furnishes a valid basis of reinforcement for the statutory presumption of the patent's validity for purposes of a preliminary injunction. 35 U.S.C.A. Sec. 282; Rosenberg v. Groov-Pin Corp., supra; Norwich Pharmacal Co. v. Veterinary Corp. of America, 296 F. Supp. 937 (M.D.Ga.1968).

Chemical compounds of the type of propoxyphene which are intended for human dosage, must be developed empirically. The district court accepted expert opinion:

. . . that minor changes in the substituents of a compound can profoundly affect its analgesic potency; and that reliable prediction of either analgesic potency or addiction liability on the basis of structure cannot be made. Although in hindsight, structure-to-function generalizations can sometimes be made in chemistry, such relationships cannot be predicted in the field of analgesic compounds but can be ascertained only retrospectively after experimentation and testing.

Analogical reasoning is necessarily restricted in many chemical patent cases because of the necessity for physiological experimentation before any use can be determined. In the case of propoxyphene's development the process of chemical synthesis constituted only a preliminary portion of the invention process, which consisted not only of originating the compound but also of establishing through extensive biological testing that the newly-synthesized compound was an effective medicine. The fact that no reliable way of predicting utility based upon chemical formulas alone had been developed in the field of analgesic research when Dr. Pohland worked, bars our use of the more ordinary analysis processes which determine invention in mechanical devices. In fact, such lack of predictability of useful result from the making of even the slightest variation in the atomic structure or spatial arrangement of a complex molecule not only deprives the instant claims of obviousness and anticipation of most of their vitality, but also largely renders impotent the claim that Dr. Pohland dealt unfairly with the Patent Office in not noting that the existence of a chemically closely related compound was disclosed in published literature.

These claims of obviousness and anticipation by the prior art principally, and the claim of bad faith non-disclosure solely, relate to a chemical compound described by Dr. Pohland's coworker, Dr. K. K. Chen, in a paper published by the New York Academy of Sciences approximately four years before the patent application. In the course of this article, Dr. Chen described the experimentation being done by the parent Lilly company in the synthetic analgesic drug field and gave the formulae for a number of compounds that had been produced and tested. A majority of the compounds described had been prepared by Dr. Pohland. When Dr. Pohland communicated the technical data relative to his discovery to the company's patent attorney, he described one of the mixtures which Dr. Chen had disclosed, together with two other published compounds, all three of which were chemically similar reverse esters that had been synthesized in the search for better analgesic drugs. This memorandum described the Chen compound as the parent of the series of reverse esters which included the proposed invention. For some reason which is really not satisfactorily explained, no compound disclosed by Dr. Chen was brought to the attention of the patent office. But in no sense can this non-disclosure be said to approach concealment or unfair dealing of the type discussed and condemned in Beckman Instruments, Inc. v. Chemtronics, Inc., 439 F.2d 1369 (5th Cir. 1970).

Dr. Pohland was not claiming to be the originator of a series of reverse ester analgesics or any other general type of synthetic pain reliever. Man-made analgesics were well known both in public and patented forms in this country and abroad. The raison d'etre for the research was not simply to produce another analgesic compound, but was rather to produce one which would exert this therapeutic value with minimal toxic side effects. Most importantly, Dr. Pohland's application disclosed the existence of a compound which, in its chemical composition, was an adjacent homologue to propoxyphene and thus had chemical characteristics more nearly identical to propoxyphene than the closest Chen compound. The district court found, on what can only be read as confusing evidence, that the closest compound disclosed by Dr. Chen was clinically determined to have no significant analgesic activity at all. In addition to the resolved dispute as to this fact, it is undisputed that the Chen-described drug had no patent interest to Lilly as an analgesic because testing showed it caused the appearance of toxic signs at the dosage level where its analgesic activity began. Also, the evidence failed to disclose that the Chen compound, or indeed any compound of this reverse ester series other than propoxyphene, has ever been patented or marketed. In view of our recent collation of authorities and analysis in Part VIII of Hobbs v. United States, 451 F.2d 849 (5th Cir. 1971), we pretermit further legal elaboration on the anticipation and non-obviousness issues.

Bearing in mind that the patent office is not equipped to do more than examine the formulae and the stated results of trial and error testing of drug compounds, the disclosures made here were fair and reasonable. The trial court found that Dr. Pohland did not consider the Chen compound a relevant prior disclosure. The record is devoid of any proof which would taint the good faith of this determination on his part. Since, if he had called it to the examiner's attention he would have made the undisputable assertion that it lacked the essential value of propoxyphene, we cannot say its nondisclosure violated any basic concept of doing equity. Cf. Part IV of Carter-Wallace, Inc. v. Davis-Edwards Pharmacal Corp., supra, 443 F.2d 867 at pp. 881, 882. Without in the least undermining Beckman's mandate to leave to the patent office such decisions as it is able to make concerning the effect of prior art, we hold that while this rule requires no less than full, frank disclosure, it does not mean an applicant must list out the full spectrum of his knowledge to establish bona fides.

In deciding that it was sufficient for Dr. Pohland to provide a full, frank disclosure of that which he honestly considered to be the relevant prior art when that disclosure brought to the patent examiner's attention closer compounds than the one known but not disclosed, we apply the same reasoning as that which would refuse to overturn patent validity because unknown prior art exists which is not as pertinent as that which was considered by the patent office. Arthur J. Schmitt Foundation v. Stockham Valves & Fittings, Inc., D.C., 292 F. Supp. 893, aff. 404 F.2d 13 (5th Cir. 1968), cert. denied 398 U.S. 965, 90 S. Ct. 2177, 26 L. Ed. 2d 549 (1970). The public function of the patent system remains equally protected in either case as long as the proposed invention is judged by work as close or closer to it than work not shown.

In the case at bar, we necessarily decide the issue which we did not have to reach in Monsanto Co. v. Dawson Chemical Co., 312 F. Supp. 452 (S.D.Tex. 1970), rev'd. on other grounds 443 F.2d 1035 (5th Cir. 1971). In that case a chemical compound was discovered to exhibit novel, useful, herbicidal activity. An adjacent homologue was disclosed in the prior art, but recognized to be of value only in making pigment. That appeal was controlled by the application of collateral estoppel and we did not decide the merits. In today's holding, we follow the rationale which underlies the decision of the Court of Customs and Patent Appeals in Application of Stemniski, 444 F.2d 581 (CCPA 1971); and Commissioner of Patents v. Deutsche Goldund-Silber-Scheideanstalt Vormals Roessler, 130 U.S.App.D.C. 95, 397 F.2d 656 (1968). See Note: Patentability of Chemical Compounds, 50 Tex.L.Rev. 566.

In the field of drug patents today therapeutic value, not chemical composition, is the substance of all incentive to invent. Except where the state of the medical art and the state of the chemical art have been advanced and coordinated to the point that it is possible for the mind to conceive or predict with some minimal reliability a correlation between chemical analogues, homologues or isomers and their therapeutic value, reason compels us to agree that novelty, usefulness and non-obviousness inhere in the true discovery that a chemical compound exhibits a new needed medicinal capability, even though it be closely related in structure to a known or patented drug. When such a fresh, efficacious, undisclosed use is identified, its inventor deserves the full ambit of statutory protection. A limitation to "use" or process patentability, based solely on the existence of prior chemical formulations, would not accord with the basic constitutional power being exercised by the Congress to promote science and the useful arts. Such a niggardly patent reward for costly and painstaking research would discourage both the inspiration-perspiration process of the laboratory and the incentive to publicly disclose products of value to mankind.

However, where a court finds the alleged inventor's work in a field filled with formidable prior art to be no more novel or non-obvious than the conducting of a biological or physiological testing program among catalogued compounds or an easily formulated series of homologues or analogues that logically or predictably should disclose helpful uses, the grant or validation of a patent on the product would be out of keeping with the letter or spirit of the law. This conclusion would be even more compelled where the use claimed by such a tester in his application turns out to be a minor value of the drug which the patent would monopolize. See Judge Dooling's opinion on remand in Carter-Wallace, Inc. v. Davis-Edwards Pharmacal Corp., 341 F. Supp. 1303 (E.D.N.Y.1972).

Contrary to expressed opinions that the patent laws must be amended or that drug patenting must be largely limited to "use" or process claims, Eggert: Uses, New Uses and Chemical Patents -A Proposal, 1968 Wisc.L.Rev. 901, 51 J.Pat.Off.Soc'y 768, we view 35 U.S.C.A. Secs. 101-103 as presently providing sufficient authority to permit courts to adjudicate whether or not chemical invention has occurred. The hazard that a compound patent will preempt an after-discovered therapeutic value of a drug during its statutory term exists whether the drug is chemically novel or not. Never-theless, this possibility must not make us timid in assaying for novelty, usefulness and non-obviousness on the evidence developed in each case, unbound by a stultifying per se rule of chemical similarity, albeit requiring that a close eye be kept on the state of the art in the area where and when the alleged invention took place.

The alternative claim of non-infringement asserts two major premises. The first is that the racemate a-dl compound claimed in the patent should not be allowed to monopolize the use of the a-d isomer which defendants-appellants propose to distribute. A substitutent of this argument is that if the racemic a-dl mixture includes the a-d isomer, it also includes the a-l isomer which is not analgesic, hence making the claim invalid by the inclusion of an inoperable species. The second premise is that the doctrine of equivalents cannot be applied. There are three dependent branches to this latter argument: (a) propoxyphene is not a pioneer invention, (b) Lilly cannot assert the doctrine of equivalents because of its inequitable failure to disclose the Chen compound to the patent office, and (c) the a-d isomer is not an equivalent of the a-dl racemic mixture, that is to say, it does not perform substantially the same function in substantially the same way to obtain the same result.

While chemical compounds contain many a riddle wrapped in a mystery inside an enigma, some things are knowable and predictable. Long before propoxyphene-indeed, long before Dr. Pohland-racemates and their component isomers were known to the drug art. The synthetic analgesics prior to propoxyphene were known to have optical isomers and the parties stipulated that this property could be reliably predicted by an inspection of the structure of the compound. The parties further stipulated that the analgesic activity was known to exist exclusively or predominantly with one of the isomers of such racemates. Dr. Pohland did not allow his claim to outstrip his invention when he applied for a patent on the a-dl racemate, but clearly stated in the specification that he only sought patent protection for the property of propoxyphene which was characterized by an ability to produce analgesia without toxic side effects such as respiratory depression. It was thereafter immaterial whether it was discovered that the claimed therapeutic value lay in the -l or the -d isomer. It was known and predictable that this value would lie either in the racemate or in only one of its two isomers. In this state of affairs, the subsequent development of the art would serve to limit the claim if it were required to be limited. We therefore reject the contention that the patent on the a-dl racemic mixture here did not protect the use of the a-d isomer for analgesic purposes once it was later established that the isomer was the repository of the useful and novel invention claimed. Whether it is conceptualized as a claim limited by a specification, United States v. Adams, 383 U.S. 39, 86 S. Ct. 708, 15 L. Ed. 2d 572 (1966); or as an alleged infringement reached by the doctrine of equivalents, Laitram Corp. v. Deepsouth Packing Co., 443 F.2d 928 (5th Cir. 1971); the result is the same-the optical isomer in which the asserted quality lay cannot be utilized without infringing the patent on the racemic mixture. See also Hobbs v. United States, supra, Part X, and the cases there cited. Due to exactly the same state-of-the-art considerations, the fact that the a-l isomer proved to be inoperable as an analgesic did not make this claim invalid. We expressly pretermit any ruling on whether a patentee acquires the breadth of protection accorded by a compound patent in such an isomer, i. e. whether a patent on a-dl propoxyphene as an analgesic would protect the a-d isomer as an insect repellant. That issue is not before us. It is unnecessary for us to test the validity of the district court's conclusion that 977 is a landmark or pioneer invention.

The original complaint in this action was filed on October 27, 1969. On February 16, 1970, a consent order was entered which provided that the defendants-appellants would not make, use or sell a-d propoxyphene hydrochloride in the United States, its possessions or territories until 45 days after they had received a required predistribution approval from the Federal Food & Drug Administration. On November 17, 1970, the trial court was advised that notice of this approval had been received. The court was also informed that the defendants Ellencee and Carbide intended to manufacture, produce and sell approximately 3 to 4 million capsules of propoxyphene per month in almost identical form and packaging beginning 45 days after receipt of such approval.

When the notice that FDA's new drug approval was given, Lilly moved for an immediate trial on the merits or on its motion for preliminary injunction. On December 1, 1970, the court ordered that hearings in the cause be commenced on December 8, 1970. During the course of the hearing the court took two five-day recesses, and on three days heard testimony for less than a full day. One of the longer recesses and two of the partial trial days occurred during defendants' case; the other lengthy recess and the other partial trial day occurred during the presentation of plaintiffs' rebuttal evidence.

On November 20 and again on November 23, 1970, the defendants-appellants noticed the taking of a number of depositions. They were unable to complete all of these depositions before the hearing was completed. At the conclusion of the hearing the court stated it hoped to reach a decision "within about twelve days"; however, it was over 70 days before its decision was rendered.

The scheduling of trial procedures is a matter uniquely committed to the sound discretion of the nisi prius court. This is particularly true in a case such as the one at bar where the proceedings are being tried to the court without a jury. An appellate court will not interfere with the trial court's exercise of its discretion to control its docket and dispatch its business in such cases, except upon the clearest showing that the procedures have resulted in actual and substantial prejudice to the complaining litigant. The record now before this court wholly fails to make any such demonstration. Under the circumstances that confronted the trial judge in the case at bar, his exercise of discretion cannot be faulted-either as to bringing the matter on for trial as promptly as he did or for keeping the matter pending despite the necessity for recesses from time to time during its course.

Likewise, we decline to find error in the court's determination to commence this hearing on a day over 13 months after the filing of the litigation despite the fact that one of the parties had recently noticed the taking of several depositions. Every litigant has a duty to his client and to the court to proceed with discovery with reasonable promptitude. Reversible error arising from curtailment of discovery procedures must be premised on a demonstration that the court's action made it impossible to obtain crucial evidence, and implicit in such a showing is proof that more diligent discovery was impossible. The proof here does not rise to that level.

Defendants-appellants assert that the court's expressed hope for a quick decision misled them into voluntarily refraining from beginning the manufacture of propoxyphene. Given the eventual decision of the trial court, the contention approaches the frivolous. The court's overly optimistic prediction of the time within which it could reach a decision not only did not harm those complaining, but, if it did indeed keep them from acting, actually prevented the losses they would have sustained as a result of the manufacturing start-up which would have been shut off by the court's preliminary injunction.

Two contentions are raised by the defendants-appellants concerning the court's sua sponte action in entering a final order in this cause after an appeal had been taken from the grant of the preliminary injunction. First, it is asserted that the court failed to apprise the parties, as it was required to do under Fed.R.Civ.P. 65(a) (2), that the action on its merits would be advanced and consolidated with the hearing on the application for preliminary injunction, and that this deficit of notice prevented them from developing before the court a record suitable for making a final determination relative to patent invalidity-in-fringement, as well as the Sherman Act anti-trust claims. Second, it is argued that the act of consolidation and the issuance of a final judgment, caused them to lose their jury trial rights to the legal issues inherent in the case.

The first of these contentions requires determinations which cannot be made at the appellate level. In Puerto Rican Farm Workers ex rel. Vidal v. Eatmon, 427 F.2d 210 (5th Cir. 1970) and in Nationwide Amusement, Inc. v. Nattin, 452 F.2d 651 (5th Cir. 1971), this court expressed its concern over injunctive evidentiary proceedings which resulted in final dismissal after hearings which were nominally related to preliminary relief alone. These authorities make it clear that while this circuit does not interpret Rule 65(a) (2) to require a formal written "order", it does require some form of notice to the parties that their final day in court has come. However, as Nationwide similarly makes clear, surprise alone is not a sufficient basis for appellate reversal; appellant must also show that the procedures followed resulted in prejudice, i. e., that the lack of notice caused the complaining party to withhold certain proof which would show his entitlement to relief on the merits. There is some indication in the record before us that the appellants were indeed informed that a Rule 65(a) (2) consolidation had occurred. For example, at several points during the course of the hearing on the preliminary injunction, the court advised the parties that the issues being presented on preliminary injunction were sufficiently broad to resolve almost everything except damages. The court also urged the parties not to withhold evidence material to the issues raised merely because the proceedings had been formally titled as preliminary. In fact, the brief originally filed by the defendants-appellants on the appeal from the preliminary injunction complained that the court had treated the preliminary hearing, throughout, as one that would be dispositive of the case on its merits. On the other hand, there was some equivocation by the court concerning its specific intentions on this issue. At one point, the appellants inquired whether or not they were attending a full trial on the merits. No unmistakable reply was forthcoming. Moreover, the appellants now argue before this court that there were several matters they would have presented, and would wish now to present, at a final hearing on these proceedings.

At length, we conclude that we cannot accurately determine from the record now before us either the lack of fair notice or the resulting prejudice issues. On remand, we contemplate further meaningful adversary proceedings by the parties on these issues, and the necessary factual and legal determinations thereon by the lower court. We leave the matter completely open for that court to decide: (1) whether a fair opportunity was given to the defendantsappellants to present their entire case; and (2) if all parties were fairly on notice to make full proof, whether the evidence shows clearly that Lilly was entitled to prevail on the merits as to permanent injunctive relief regardless of any additional proof which the defendants-appellants might be able to adduce. If after such a hearing the district judge should conclude that this is a case wherein no material fact issues remain and there is indeed nothing left to be tried, he should not hesitate to afford Lilly final relief. Standard Oil Company of Texas v. Lopeno Gas Company, 240 F.2d 504, 509 (5th Cir. 1957). Contrariwise, if the rights of the defendants-appellants to a day in court on the merits has not been accorded, they must be given this right regardless of the quantum of proof they developed (or failed to develop) on the preliminary hearing.

However, before the district judge can reach any such decision regarding the issuance of a permanent injunction, the defendants-appellants must be afforded the opportunity, if they still choose to proceed thereon, to try to a jury their counterclaims against Lilly for anti-trust violations. Admittedly, the trifurcation of this lawsuit into a preliminary hearing before a judge, followed by a trial before a jury on a single segment of the action, then concluded thereafter by yet another hearing before a judge, wreaks havoc with our strong policies against piecemeal litigation. But, since we are convinced that such a separate trial of the issues may be had without injustice resulting, Swofford v. B & W Incorporated, 336 F.2d 406, 415 (5th Cir. 1964), in the circumstances of this particular case no other procedure would be permissible. The Supreme Court's decision in Beacon Theaters, Inc. v. Westover, 359 U.S. 500, 79 S. Ct. 948, 3 L. Ed. 2d 988 (1959) and Dairy Queen, Inc. v. Wood, 369 U.S. 469, 82 S. Ct. 894, 8 L. Ed. 2d 44 (1962), and our own in Swofford, supra, and Thermo-Stitch, Inc. v. Chemi-Cord Processing Corp., 294 F.2d 486 (5th Cir. 1961) mandate in the strongest of terms a strict observance of the right to trial by jury. Thus, where certain fact issues are common to both the legal and equitable claims in a lawsuit, only under the most imperative circumstances should a judge proceed first with the equitable claims in such a fashion that he, rather than the jury the parties have demanded, decide those fact issues. Fitzgerald v. United States Lines Co., 374 U.S. 16, 83 S. Ct. 1646, 10 L. Ed. 2d 720 (1963); Beacon, 359 U.S. at 510-511, 79 S. Ct. 948; Thermo-Stitch, 294 F.2d at 490-491. That is precisely what has happened here. The court, not a jury, has made a final determination of the facts crucial to defendants-appellants antitrust claim, to wit, whether Lilly acted fraudulently or inequitably in its dealings with the patent office. The question before us now is whether the circumstances of the case were so imperative as to require that the court make such a final determination. The answer is, they were not.

The Supreme Court indicated in Beacon, 359 U.S. at 510-511, 79 S.Ct. at 956, that if there should be a case where the joinder in a single lawsuit of both legal and equitable causes:

would not in all respects protect the plaintiff seeking equitable relief from irreparable harm while affording a jury trial in the legal cause, the trial court will necessarily have to use its discretion in deciding whether the legal or equitable cause should be tried first.

The Court continued by saying that this discretion is "very narrowly limited" and indicated that it could not even anticipate circumstances sufficiently imperative wherein the right to a jury trial of legal issues could be lost through a prior determination of equitable claims. We are satisfied, however, that this stringent test was met insofar as the judge's decision to proceed with the hearing on the preliminary injunction is concerned. He had at hand the defendants-appellants announced intention to forthwith begin substantial manufacture and distribution of the allegedly infringing product. We cannot say he abused his discretion in concluding that the threat of irreparable injury to the plaintiffs due to sales losses and confusion in the market place necessitated an expedited disposition of the prayer for immediate injunctive relief. Such a conclusion required him to make preliminary fact findings concerning the validity of the Lilly patent which would govern the rights of the parties pendente lite, this necessarily included findings on the question of fraudulent dealings with the patent office. To this point, then, we are in complete agreement with the procedural as well as the legal actions of the lower court.

However, at the moment the preliminary injunction issued: (1) the previous threat of irreparable injury, (2) the once imperative circumstance, and (3) the judge's erstwhile discretion to make final fact determination of issues common to legal and equitable causes- all vanished from this litigation. With the status quo preserved, there is no reason left which would justify the subversion of defendants-appellants' right to have a jury decide whether Dr. Pohland met his duty to disclose to the Patent Office. We conclude that the judge abused his discretion in not safeguarding this right to defendants-appellants after granting the preliminary injunction. We remand this matter for him to give them that opportunity now. We further take careful note that should they avail themselves of such an opportunity the judge will be bound at the subsequent hearing on the permanent injunction by any factual determination that such a jury might make.

The order granting the preliminary injunction is affirmed; the order granting the permanent injunction is vacated and the cause is remanded for further proceedings not inconsistent with this opinion.

Affirmed in part; and in part vacated and remanded.