Full opinion text
GOLDBERG, Circuit Judge: In this clash between two giants of the chrysanthemum business, we confront a myriad of antitrust and plant patent issues. Yoder Brothers (Yoder), plaintiff in the district court, sued, alleging infringement of twenty-one chrysanthemum plant patents by California-Florida Plant Corp. (CFPC) and California-Florida Plant Corp. of Florida (CFPCF) (sometimes referred to collectively as Cal-Florida). CFPC and CFPCF denied the infringement and filed antitrust counterclaims under sections 1 and 2 of the Sherman Act. As to seven of the chrysanthemum plant patents, the lower court directed verdicts for Yoder that the patents were valid and infringed and awarded treble damages. The court also ruled for Yoder on Cal-Florida’s section 2 claim. CFPC and CFPCF, however, prevailed in their antitrust counterclaim under section 1 and received treble damages for Yoder’s derelictions. Because many of the issues in this case turn on the particular nature of the ornamental plant industry and the specific characteristics of chrysanthemums, we shall describe the background facts in some detail before discussing the many complex legal issues presented on this appeal. Following our description of the facts, we shall briefly sketch the procedural history of the case. Finally, we shall consider the antitrust claims and the issues relating to the plant patent law. I. General Background A. The Chrysanthemum Industry Chrysanthemums, in their natural state, blossom only during the fall. This is because they are photoperiodic in nature, meaning that their growth is affected by the relative lengths of lightness and darkness in the day. When the days are long, the chrysanthemum plant remains in a vegetative state. As the nights become longer, the initiation process of the chrysanthemum bud begins. Thus, in early August, when the nights achieve a duration of nine and one-half continuous dark hours, the chrysanthemum plant in its natural state will begin the process of developing a flower. During the fall and early winter months, the mature flower appears. Yoder began doing business in the 1930’s as a simple greenhouse operator, specializing in tomatoes. Soon thereafter, because the fall tomato crop was less profitable than the spring crop, it decided to replace the fall crop with chrysanthemums. In 1939 or 1940, Yoder employees began research into out-of-season flowering of chrysanthemums. By applying black cloth shades over the chrysanthemums when dark hours were needed and applying artificial light when light hours were needed, it became possible to flower chrysanthemums on a year-round basis. Yet this breakthrough was not without its problems. For example, the use of black cloth shades resulted in an abnormally high temperature build-up around the plants, which in turn retarded bud initiation. Similarly, when the finishing temperatures were too warm, the chrysanthemums would not hold their color. In an effort to adjust for these conditions and to improve the quality of the chrysanthemum generally, Yoder initiated a breeding program in the early 1940’s. One of the most important goals of the breeding program was the development of new varieties for consumers. Although the ornamental plant industry encompasses many different kinds of flowers, including azaleas, carnations, roses, african violets, geraniums, snapdragons, and others, chrysanthemums are one of the most popular of the genre. According to the United States Department of Agriculture, in 1971 approximately 2,134 growers in twenty-three states sold nearly 145 million blooms from about 129 million standard variety chrysanthemum plants, 34.5 million blooms from 136 million pompon chrysanthemum plants, and 17.5 million potted chrysanthemum plants. At the time of the trial there were over 475 different varieties of chrysanthemums available. The total wholesale value of growers’ sales in the twenty-three states that year was approximately 83.5 million dollars. Chrysanthemums have been subject to intensive breeding efforts over the past thirty years; each individual specimen is a genetically unique complex organism. Several definitions of the term “variety” of chrysanthemum were offered at trial. Mr. Duffett, Yoder’s head breeder, defined a variety as a group of individual plants which, on the basis of observation by skilled floriculturists and according to reasonable commercial tolerances, display identical characteristics under similar environments. Cal-Florida defined variety in its complaint as “a subspecies or class of chrysanthemums distinguishable from other subspecies or classes of chrysanthemums by distinct characteristics, such as color, hue, shape and size of petal or blossom or any of them.” New varieties of chrysanthemums are developed in two major ways: by sexual reproduction and by mutagenic techniques. Sexual reproduction, the result of self or cross pollination, produces a genetically unique seedling, the characteristics of which are impossible to predict. Mutagenic techniques simply accelerate the natural rate of mutation in the chrysanthemum plant itself. A mutation was defined by Mr. Duffett as “a change in the number of chromosomes or a change in the chromosome position or a specific change in the genes within those chromosomes.” Technically, only those mutations that first express themselves as bud variations are properly called “sports”; however, the word is used loosely in the industry as a general synonym for mutation, and we will so use it. Two types of sports can appear: spontaneous sports and radiation sports. The cells of all living things occasionally mutate, and spontaneous sports are simply the result of that process. Radiation sports, on the other hand, are induced artificially, through exposure to such things as gamma radiation from radioactive cobalt and X-rays. These techniques do nothing that could not occur in nature apart from speeding up the natural mutation process. Although most of the mutations induced by radiation are not commercially usable plants, a skilled breeder will select for further development those that display such desirable characteristics as fast response time, temperature tolerance, durability, size, and vigor. After a breeder has successfully isolated a new variety, the only way he can preserve his creation is by means of asexual reproduction. In the case of chrysanthemums, the most common technique of asexual reproduction is the taking of cuttings from a stock plant. Cuttings, as defined in the Cal-Florida complaint, are “sections or parts of chrysanthemum plants which may' be grown into mature plants for sale as cut flowers and/or potted plants or from which additional cuttings may be harvested.” According to Yoder’s suggested definition, cuttings are simply immature chrysanthemum plants. Since a cutting is genetically ■identical to the parent plant, it will develop into a plant whose characteristics match the parent’s exactly, so long as the same environmental conditions obtain. A central fact of life in the chrysanthemum industry is the ease with which cuttings can be taken from parent plants: from one chrysanthemum, it is theoretically possible to develop an infinitely large stock, by taking cuttings, maturing some into flowered plants, taking more cuttings, and so on. Over the years since Yoder first entered the chrysanthemum business, the industry has become internally specialized. At the first functional level are the breeders, who create new varieties of chrysanthemums. Breeding is an expensive, complex procedure. The breeder must possess the skill and discrimination to spot potential new varieties and recognize whether they possess desirable traits; facilities for elaborate testing and development must be available. Because chrysanthemums mutate rapidly, a breeder must always be on the lookout for new changes. At the next level in the industry are the propagator-distributors. The propagator-distributors build up mother stock from sources such as breeders, retail florists, or their existing flowers, and reproduce cuttings from that mother stock. In a sense they are simply mass producers of cuttings. They do not develop cuttings to the mature flower stage (except for purposes of their own testing). Next are the growers, who develop cuttings purchased from propagator-distributors into mature plants either for cut flowers or potted plants. Combining the function of propagator-distributors and growers are the self-propagators. Cal-Florida defined a “self-propagator” as “a person who either buys or establishes stock and takes cuttings for the sole purpose of producing cut flowers and/or potted plants for resale or own use.” In other words, the self-propagators are vertically integrated into one step. Finally, the growers (or self-propagators) sell their products to retail florists, who in turn sell to ultimate consumers. B. The Parties During the times relevant to this litigation, Yoder operated on two levels in the business: as a substantial (if not the largest) breeder of new varieties of chrysanthemums, and as a large propagator-distributor. In addition to chrysanthemums, Yoder dealt with carnation cuttings, azalea liners (baby azaleas), and snapdragon seeds. Yoder is an Ohio corporation, and it sells its products nationwide. CFPC, which is incorporated in California, and which sells primarily in the western part of the United States, was a propagator-distributor. CFPCF, a wholly owned subsidiary of CFPC, was also a propagator-distributor. CFPCF is incorporated in Florida and it sells in the eastern United States. Both CFPC and CFPCF specialize in chrysanthemums. They entered the market in 1957, at a time when Yoder was clearly the largest of the propagator-distributors. During the period in question, Yoder and the two Cal-Florida companies competed horizontally as propagator-distributors— they did not compete as breeders, although Cal-Florida did make a minor foray into breeding during the 1960’s. C. The Plant Protection Programs The issues in this litigation arose out of Yoder’s breeding operations and its desire to secure a fair return from those efforts. Theoretically, once the first plant of a new variety is sold, it is impossible for a breeder ever again to be compensated for his efforts in developing it. As indicated above, anyone can take a cutting from that new plant, propagate a number of cuttings from the first cutting, and obtain an infinite supply of the plant. Even as a practical matter, the evidence at the trial suggested that it was relatively easy to obtain plant material of new varieties without the consent of the breeder. Yoder’s first effort to obtain compensation for its breeders took the form of a program entitled the Yoder Grower Agreement, or YGA, instituted around 1958. In return for access to new varieties developed by Yoder, growers were required to sign an agreement that prohibited purchasers of Yoder cuttings from selling, loaning, or otherwise disposing of purchased cuttings. Specifically, growers were prohibited from selling Yoder cuttings to self-propagators or to propagator-distributors. The agreement also contained a “sport return clause,” which required purchasers of Yoder cuttings to return to Yoder any mutations which appeared either directly or indirectly on Yoder cuttings. Yoder enforced the YGA program by refusing to ship covered varieties to persons who did not sign a YGA agreement. The most significant aspect of the YGA program was the fact that a royalty was charged on all Yoder cuttings propagated or used. In the early 1960’s, the YGA program was replaced by a new system that took its name from the Breeder-Grower Agreement that was its central reason for being. A corporation called BGA, International [BGA] was created to administer the program. Any breeder could be a member of BGA. According to the members’ regulations, voting strength was proportional to the amount of expenses the member bore. Expenses, in turn, were assessed in proportion to the amount of royalties collected on the breeder’s new varieties. The practical effect of these provisions was to secure control of BGA in Yoder’s hands. The bylaws and articles of incorporation of BGA indicated that its primary purpose was to insure a measure of remuneration to the breeders. A breeder would list his new variety of ornamental plant with BGA, and BGA would make plant material of that variety freely available to propagator-distributors. The breeder members of BGA agreed on the amount of royalty to be charged. Significantly, during most of the time that the BGA program was in existence, it was administered within Yoder’s offices. Three kinds of agreements were used in administering the BGA program. The first was the Propagator-Distributor Agreement, which permitted the signatory to make any desired commercial use of purchased cuttings or cuttings harvested from the stock plants. Participating propagator-distributors had an obligation to send a grower or grower license agreement to customers who wanted to purchase a BGA variety. For each cutting sold, the propagator-distributor had a contractual obligation to pay BGA a $.006 royalty. He also was required to report the number of cuttings sold quarterly, not to give cuttings to non-signatories, to exercise reasonable care to keep others from getting cuttings, and to allow the breeder to inspect and inventory his plantings at all reasonable times. The propagator-distributor agreement also contained a provision whereby the propagator-distributor was entitled to full credit from BGA if he was unable to collect the $.006 royalty from his customers. The agreement required the propagator-distributor to return all mutations and sports to the breeder, who retained all rights to them. Grower License Agreements were signed by self-propagators. These agreements conferred the right to grow and to propagate plants to sell as cut flowers or potted plants. The restrictions and conditions in the agreement were essentially the same as those in the Propagator-Distributor Agreement, except that the royalty payment was to go to the propagator-distributor who had furnished the cutting, instead of to BGA. Finally, the growers signed a Grower Agreement. The Grower Agreement covered growers who purchased cuttings from propagator-distributors for the purpose of selling flowers or potted plants. All propagation rights were again reserved to BGA, and the grower agreed not to propagate without BGA’s consent, not to give BGA varieties to others for the purpose of propagation, to allow reasonable inspections, and to return sports. Typically, the BGA program operated as follows: A propagator-distributor would propagate a large number of cuttings of a BGA new variety. For each cutting he sold to a grower or a self-propagator, he would pay BGA' $.006. On his invoices to his customers, a base price for the cutting would appear, and separately stated would be the amount of BGA royalty due. Evidence at the trial indicated that the industry was generally aware of the ■ existence of the BGA royalties and understood that these royalties were in essence compensation to the breeders of the new variety. The customer would therefore pay the base price plus royalty to the propagator-distributor, and the latter would in turn remit the full royalty amount to BGA. Thus, the role of the propagator-distributor was that of a BGA administrator; his cooperation was essential in the process of collecting royalties from those who sold or used the protected varieties and channeling the monies to the appropriate breeder. The degree of enforcement of the BGA program was the subject of some dispute at the trial. If Yoder knew that a grower or a self-propagator had not signed a BGA agreement, it would not ship the requested BGA variety. Instead, a substitute variety would be sent. On the other hand, the testimony indicated that a substantial number of complaints were voiced about the lack of enforcement of the BGA program against non-signatories. No lawsuits were ever filed. If a grower or self-propagator went out and purchased his mother stock from a retail florist, for example, there was nothing that Yoder could or would do about the fact that he had obtained a protected variety without signing a contract. Yoder explained its lack of enforcement by the need to maintain good will in the industry. If a grower member informed BGA that someone had access to BGA varieties who had not signed the contract, a BGA representative would check with the alleged pirate and try to persuade him to become a member. Yoder’s representative testified that in almost every case, once‘the purpose of the BGA system was explained to a non-participant, the grower or self-propagator would usually agree to sign a contract and to pay the royalty to BGA. From Cal-Florida’s perspective, Yoder’s tactics were tantamount to strong-arming. Both parties agreed that new varieties were helpful to everyone in the industry. It was Yoder’s position that BGA, by providing a means for breeder compensation, was helping in the development of new varieties of chrysanthemums. The GRA program [Grower Rights Agreement], developed by Yoder in 1968 to supplement BGA, was similar to the latter program in many ways. When a grower or propagator-distributor or self-propagator discovered a mutation on plant material that was not covered by a BGA agreement (in other words, any free plant), he could send the mutation to Yoder Brothers for evaluation of its commercial possibilities. After extensive testing, if Yoder decided that the new variety could profitably be introduced, the grower who discovered the variety would be entitled to 50% of the royalty return. The agreements used to administer GRA followed the BGA pattern — a Propagator-Distributor Agreement, a Grower License Agreement, and a Grower Agreement. Unlike BGA, under GRA the royalties collected were returned directly to Yoder. The amount of the royalty under GRA was again $.006 per cutting. Cal-Florida participated in the BGA and GRA programs only as a propagator-distributor. Although it did conduct a breeding program of its own during the 1960’s, it never registered any new varieties with BGA. Instead, it developed the CFPC program. The CFPC program used the same three kinds of agreements as the BGA program — -the propagator-distributor contract, the grower-propagator license, and the grower agreement. In one aspect, however, the CFPC program was more restrictive than the BGA program: sales to self-propagators and other propagator-distributors were prohibited. The royalty rate was the same $.006 per cutting. Like the BGA program from which it was copied, the CFPC program’s basic purpose was to obtain remuneration for the company’s breeding efforts. Sports discovered by participants in the CFPC program were required to be returned to Cal-Florida, the breeder. As a propagator-distributor participant in the BGA program, Cal-Florida of course paid royalties to BGA. During the relevant period, CFPC and CFPCF combined paid $229,805.12 in BGA royalties and $27,941.18 in GRA royalties — a total of $257,746.30. The evidence showed that over the years, more and more of Cal-Florida’s sales were of varieties controlled by Yoder under either BGA or GRA. In 1963, 0.19% of their cutting sales were BGA or GRA varieties; by 1969, the number had grown to 17.59% of total sales, and by 1971, to 41.22%. . BGA and GRA royalties were always separately stated on Cal-Florida’s invoices to its customers. In addition, the following explanation was to be found in its catalogs: BGA (BREEDER GROWER AGREEMENT) VARIETIES. The California-Florida Plant Corp. is licensed by BGA International to propagate and distribute BGA varieties. The terms of our Propagator-Distributor Contract call for the customer to sign a BGA Agreement prior to the shipment of any BGA variety. BGA varieties are subject to all discounts of Volume, Advance Order and Prompt Payment. The current BGA Royalty is $0.60 per 100 cuttings, rooted or unrooted, and is in addition to the listed base price. BGA Royalties are not subject to Discount or Adjustment of any kind and the total amount of BGA Royalty collected by us is returned to BGA International. GRA (GROWER RIGHTS AGREEMENT) VARIETIES. The California-Florida Plant Corp. is licensed to grow, propagate and distribute GRA varieties. The terms of our Propagator-Distributor contract call for the customer to sign a GRA Agreement prior to us shipping any GRA varieties. GRA varieties are subject to all discounts of Volume, Advance Order and Prompt Payment. The current GRA Royalty is $0.60 per 100 cuttings, rooted or unrooted, and is in addition to the base price. GRA Royalties are not subject to Discount or Adjustment of any kind and the total amount of GRA Royalty collected by us is returned to the developer. ROYALTY CHARGES. All Royalty Charges (CFPC, BGA, and GRA) will be billed separately and included in the monthly statement. Thus, CFPC clearly segregated the BGA and GRA royalty charges from the prices charged for the cuttings it sold. D. Government Intervention On April 20, 1970, the United States brought an action against Yoder, alleging that the BGA and GRA programs violated sections 1 and 2 of the Sherman Act,, 15 U.S.C. §§ 1, 2. The Government’s suit ended in a consent judgment entered on March 15, 1972 in the Northern District of Ohio. The consent decree abolished BGA and GRA and prohibited further collection of royalties and further enforcement of the sport return clauses; it also required Yoder to take certain affirmative actions to inform the former participants of the changed status quo. The judgment expressly stated that it did not apply to any rights that Yoder had obtained under the patent laws of the United States or of any foreign country. Cal-Florida had moved to intervene in the Government’s case on February 16, 1972. On March 15, 1972, the same day as the consent decree was finally approved, the court denied its motion. E. Post BGA: Plant Patents After BGA ended, around the end of 1971, Yoder started patenting some of its new varieties under the Plant Patent Act, 35 U.S.C. § 161 et seq. Several salient differences existed between the rights conferred by a plant patent and the rights secured under the old BGA and GRA agreements. For example, under a plant patent, sports of the patented plant are not covered by the original patent. See Part IV, infra. Second, the royalty- event for a -patented plant is the asexual reproduction of the plant, instead of its use or sale. Even so, the Plant Patent Act and the BGA/GRA programs were quite similar. Under both, licenses for propagation by others could be issued, and royalties could be charged for the use of the plant. These similarities have led Cal-Florida to allege that Yoder’s new use of the plant patent laws is simply a continuation of its old and illegal BGA program. Since BGA and GRA ended, Yoder has secured plant patents on all new varieties it has introduced to the trade. Shortly after the Government suit was terminated, and after extensive unsuccessful negotiations with Cal-Florida, Yoder filed its complaint commencing this litigation. We thus arrive at last at the beginning — the procedural history of the case before us. II. Summary of Proceedings Below On March 6, 1973, Yoder filed its complaint in the United States District Court for the Southern District of Florida, alleging infringement of twenty-one chrysanthemum plant patents by CFPC and CFPCF. CFPCF answered on April 12,1973, denying the infringement and setting forth antitrust and trade disparagement counterclaims. On the same day, CFPC moved to dismiss for improper venue under Rule 12(b)(3), Federal Rules of Civil Procedure. In addition, CFPC filed suit in the Northern District of California on June 5, 1973, for a declaratory judgment on the validity of Yoder’s patents and for trade disparagement damages. On December 26, 1973, the California action was ordered transferred to the Southern District of Florida pursuant to 28 U.S.C. § 1404(a) (transfer in the interests of justice to district where suit might have been brought). In the first pretrial order, filed January 16, 1974, the district court denied CFPC’s motion to dismiss on venue grounds. The two cases were consolidated by an order entered March 4, 1974. The trial before a jury began on April 23, -1974. The issues presented for trial were stipulated by the parties in their joint Pretrial Stipulation and involved both patent and antitrust claims. Yoder claimed infringement and contributory infringement of twenty different United States plant patents by either CFPC or CFPCF or both. The two Cal-Florida companies asserted the invalidity of twenty-two United States plant patents. In its counterclaim, Cal-Florida asserted that Yoder, by its participation in the BGA and GRA programs, combined to restrain trade in violation of Sherman Act § 1,15 U.S.C. § 1, and further alleged that Yoder had committed acts of monopolization of the trade in chrysanthemum cuttings, in violation of Sherman Act § 2, 15 U.S.C. § 2. At the close of the evidence, the district court directed verdicts on several critical issues. On June 12,1974, it ruled for Yoder on the issues of patent validity and infringement of all twenty patents, subject to Cal-Florida’s claims of contract rights and prior commercial exploitation; additionally, it ruled in Yoder’s favor on all issues of inventorship, newness and distinctness, asexual reproduction, and adequacy of description. On June 13, 1974, the court granted a verdict in favor of Yoder on Cal-Florida’s monopoly counterclaims. On Cal-Florida’s side, the court directed a verdict that Yoder, through the BGA and GRA programs, had participated in a group boycott which constituted a per se violation of Sherman Act § 1. The court denied Yoder’s motions for directed verdicts claiming that there was insufficient evidence to sustain a verdict: (1) that BGA and GRA were illegal; (2) that CFPC and CFPCF were injured as a result of the BGA and GRA programs (i. e. lack of “fact of damage”); (3) that CFPC and CFPCF were in the “target area” of the BGA and GRA programs (i. e. that they had standing to sue under the antitrust laws); and (4) that any damages were supportable, since the damage proof and theories were legally improper, factually unsupported, and speculative. The patent claims were submitted to the jury on special interrogatories. The antitrust claims, in contrast, were submitted under a general verdict form. Two theories of antitrust damages were submitted to the jury: the royalty payments theory, and the price differential theory. In connection with the royalty payments theory, the jury was told that unless the BGA and GRA systems were analogous to a “pre-existing cost plus contract,” they would not be permitted to consider whether Cal-Florida had “passed on” the incidence of the royalties paid under those programs to its customers. This theory was submitted over Yoder’s strenuous objection. The price differential theory, based on evidence showing the comparative prices charged by CFPC and CFPCF (taken together) and Yoder, permitted the jury to find that the Cal-Florida companies were damaged to the extent that their prices were lower than Yoder’s. Again Yoder voiced its objection both to the introduction of this evidence and to the jury’s consideration of that theory. On June 21, 1974, after eight days of deliberation, the jury returned its verdict, finding a violation of section 1 of the Sherman Act, causation and damages in the amount of $64,500 for CFPC and $64,500 for CFPCF. With regard to the patent issues, the jury found that CFPC and CFPCF had established a contractual right to use eight of the patented plants. For another seven patented plants, the court had instructed the jury that the patents were valid and infringed as a matter of law; as to these, the jury awarded compensatory damages. The jury also awarded damages on one patent, covering the “Deep Conquest” chrysanthemum, which it found valid and infringed. It was unable to reach a verdict on the rest of the patents. On this appeal, we are concerned only with the seven patents that the district court declared valid and infringed as a matter of law and with Deep Conquest. For those patents, the court trebled the amount of the damages found by the jury, giving Yoder a total recovery of $66,917.64. Following the jury verdict, Yoder moved for judgment notwithstanding the verdict as to the award of damages to CFPC and CFPCF and, in the alternative, moved for a new trial as to the award to CFPCF only, for reasons related to the amount of the damage award. In orders dated January 16, 1975, and March 4, 1975, the court denied Yoder’s post-trial motions. Thus, on April 3, 1975, Yoder noticed its appeal; on April 14, 1975, CFPC and CFPCF responded with notices of their cross appeal. III. Antitrust Appeal On appeal from the antitrust judgment rendered against it, Yoder raises three main issues: (1) that judgment should be entered for it because appellees lack standing to sue under section 4 of the Clayton Act, 15 U.S.C. § 15; (2) that the lower court’s ruling that the BGA and GRA programs were per se illegal group boycotts was incorrect; and (3) that neither of the two possible theories of damages — the “price differential” theory or the “royalty payments” theory — could provide a basis for the jury’s award. Cal-Florida raises both antitrust and patent claims on its cross appeal. On the antitrust issues, it argues that the district court erroneously granted Yoder’s motion for a directed verdict on the Sherman Act § 2 claim of monopolization and attempted monopolization. Its other antitrust point asserts that the district court erred in denying it the benefit of the tolling provision of 15 U.S.C. § 16(i), the Clayton Act statute of limitations. We have decided that the district court correctly ruled that Cal-Florida had standing to challenge the BGA and GRA programs and that those programs were per se violations of section 1 of the Sherman Act. Furthermore, we find that the court’s ruling that Cal-Florida had failed to show a section 2 claim of monopolization or attempted monopolization was justifiable. Finally, although the lower court correctly ruled that Cal-Florida was not entitled to the tolling provisions contained in 15 U.S.C. § 16(i), it erroneously admitted evidence comparing Yoder’s and Cal-Florida’s prices. Since the jury may have relied on the price differential theory in its award of damages, we must remand the antitrust claims to the trial court for further proceedings. In this connection, we note that the district court should have disallowed Yoder’s “passing on” defense as a matter of law, rather than submitting it to the jury. In the interest of an orderly discussion of the issues before us, we have decided to organize them as follows: (1) whether Cal-Florida had standing to sue under section 4 of the Clayton Act, 15 U.S.C. § 15; (2) whether the Government’s suit was still “pending” at the time Cal-Florida filed its antitrust claims, thereby entitling Cal-Florida to the benefit of the tolling provision of the statute of limitations under Clayton Act section 4B, 15 U.S.C. §§ 15b, 16(i); (3) whether the BGA and GRA programs were illegal per se under section 1 of the Sherman Act, 15 U.S.C. § 1; (4) whether Cal-Florida failed to prove a relevant market as a matter of law for purposes of Sherman Act § 2,15 U.S.C. § 2, and whether it failed to show dangerous probability of success in such market; and (5) whether Cal-Florida successfully proved fact of damage and causation under its price differential and royalty payments theories of damages. A. Standing to Sue Section 4 of the Clayton Act provides in pertinent part: Any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court of the United States . 15 U.S.C. § 15. Focusing on the “by reason of” language, Yoder argues that private enforcement of the antitrust laws should be granted only to those plaintiffs who suffer sufficiently direct injury that it is proper that they act as private attorneys general. Additionally, they note this Court’s language in Jeffrey v. Southwestern Bell, 5 Cir. 1975, 518 F.2d 1129, 1131: The “target area” test [for standing] arose as a means of limiting the class of potential treble-damage plaintiffs to those persons who could most adequately vindicate the purposes of the antitrust laws. Those purposes include, according to Yoder, the twin goals of easing the burden on the courts of meritless antitrust claims while at the same time furthering the deterrent impact of the laws on anticompetitive business behavior. Yoder directs our attention to the Supreme Court’s comment in Hawaii v. Stan dard Oil Co., 1972, 405 U.S. 251, 92 S.Ct. 885, 891-92 n. 14, 31 L.Ed.2d 184, noting the virtual unanimity of the lower courts in concluding that Congress did not intend the antitrust laws to provide a remedy in damages for all injuries that might conceivably be traced to an antitrust violation. The question of standing, Yoder argues, is a legal question for the court to resolve. Starting with the proposition that the Fifth Circuit has adopted the “target area” test for standing, Yoder suggests that the purpose and specifically intended impact of the violative conduct must be to injure the alleged victim. Furthermore, the area of the economy wherein the impact of the violation is felt must be specifically foreseeable. Applying its proposed test, Yoder concludes that propagator-distributors were not the targets of the BGA and GRA programs, because only growers and self-propagators suffered out-of-pocket expenses due to the royalties. Propagator-distributors merely collected royalty payments from growers and self-propagators and transmitted those monies to BGA or Yoder. Neither program was designed to hurt CFPC and CFPCF, nor was Yoder in its capacity as propagator-distributor helped vis a vis CFPC and CFPCF by BGA and GRA. From these premises, Yoder concludes that neither CFPC nor CFPCF was injured in its business or property by reason of a violation of the antitrust law and that this Court should dismiss their action for lack of standing. In response to Yoder’s arguments, Cal-Florida argues that a “direct injury” test should apply. It suggests that the courts have followed a twofold traditional tort analysis: (1) was there causation in fact; and (2) was the violation a proximate cause of the victim’s injuries. Both of these questions are questions of fact which should be submitted to the jury. Finally, Cal-Florida asserts that the Fifth Circuit’s “target area” test, expressed in Battle v. Liberty National Life Ins. Co., 5 Cir. 1974, 493 F.2d 39, cert. denied, 1975, 419 U.S. 1110, 95 S.Ct. 784, 42 L.Ed.2d 807, is the equivalent of the tort analysis that it proposes. Applying its test, Cal-Florida first asserts that the royalty payments that it made to BGA and to Yoder were directly caused by Yoder’s act — i. e. Yoder’s requirement that it sign Propagator-Distributor Agreements. Second, it argues that the illegal BGA program was a factor in depressing Cal-Florida’s overall prices for chrysanthemum cuttings. Third, it alleges that it lost sales because the royalty imposed on BGA varieties caused prices to be so high that some growers turned to self-propagation. Finally, it maintains that all three elements of damage were foreseeable. From this, it concludes that the evidence was adequate for the jury to consider the issue of causation and, in accordance with the court’s instructions, to render a special verdict against Yoder. We begin our analysis of this problem by noting that standing to sue is a preliminary matter, to be evaluated upon the allegations of the complaint. See Malamud v. Sinclair Oil Corp., 6 Cir. 1975, 521 F.2d 1142, 1150. Cf. Battle v. Liberty National Life Ins. Co., supra, 493 F.2d at 48. This Circuit’s test for standing was recently expressed in Jeffrey v. Southwestern Bell, supra, as follows: To attain standing a person (whether corporation or individual) must be one against whom the conspiracy is aimed. Or, put in plutonomic terms, the complainant must show that he is within that section of the economy which is endangered by a breakdown of competitive conditions in a particular industry. 518 F.2d at 1131. See Tugboat, Inc. v. Mobile Towing Co., 5 Cir. 1976, 534 F.2d 1172; Battle v. Liberty National Life Ins. Co., supra, 493 F.2d at 49. See also Southern Concrete Co. v. United States Steel Corp., 5 Cir. 1976, 535 F.2d 313; Buckley Towers Condominium, Inc. v. Buchwald, 5 Cir. 1976, 533 F.2d 934. One might win the battle of standing to sue but still lose the war on another issue, such as the violation issue, see E. A. McQuade Tours, Inc. v. Consolidated Air Tour Manual Comm., 5 Cir. 1972, 467 F.2d 178, cert. de nied, 1973, 409 U.S. 1109, 93 S.Ct. 912, 34 L.Ed.2d 690, or the damages issue. A test focusing on the sector of the economy is more easily stated than applied. The Ninth Circuit suggested one approach in In Re Multidistrict Vehicle Air Pollution M.D.L. No. 31, 9 Cir., 481 F.2d 122, 129, cert. denied sub nom. Morgan v. Automobile Mfr’s Ass’n, 1973, 414 U.S. 1045, 94 S.Ct. 551, 38 L.Ed.2d 336: A proper application of “by reason of” focuses on whether the anti-competitive conduct directed against an area of the economy injured business operations conducted by the claimant in that sector of the economy. The resulting two-step approach first requires identification of the affected area of the economy and then the ascertainment of whether the claimed injury occurred within that area. Examining Cal-Florida’s complaint, we find first the undisputed fact that both CFPC and CFPCF were involved as propagator-distributors in the chrysanthemum production industry. Next, we find that the two companies alleged that purchasers of BGA and GRA cuttings were required to pay a royalty, to the purchasers’ injury, as a result of the illegal programs. Propagator-distributors, including CFPC and CFPCF, were among those who purchased cuttings under BGA and GRA. Furthermore, they pointed out other restrictions which were also imposed on propagator-distributors, pursuant to the programs, including the obligation to return sports of BGA and GRA cuttings, the fixing of the royalty amount, and restrictions on which of their customers could receive BGA and GRA cuttings. These claims were summarized in the amended answer and counterclaims of both defendants in paragraph 16(h) as follows: CFPC and CFPCF were and continue to be damaged in their business through, inter alia, over-payment for cuttings by the payment of royalties, loss of past profits, loss of future profits, loss of sales, and the prospective destruction of its business. From these allegations, we see that one affected area of the economy against which the anti-competitive conduct was directed was the propagator-distributor level of chrysanthemum production. CFPC and CFPCF clearly fall within this area. The claimed injuries included the payment of royalty to BGA and to Yoder. In fact, propagator-distributors literally sent royalty checks drawn on their own accounts to BGA and Yoder. Propagator-distributors were necessary participants in the BGA and GRA schemes, even if their role was simply that of an administrator of the program or a conduit for the funds. Unless propagator-distributors cooperated with the restrictions on access to protected varieties for non-signatories, the programs could not operate. The restrictive effect of the BGA combination on the amount of royalty charged by different breeders could have been felt directly by á propagator-distributor. The administrative burdens of the BGA and GRA programs might have dissuaded propagator-distributors from participation or imposed some illegal economic burden on them. Finally, the effect of the two programs would almost certainly have been different for Yoder than for other propagator-distributor participants, because of Yoder’s vertical integration of the breeder function and the propagator-distributor function. Yoder qua propagator-distributor would have been more than happy to return sports of BGA varieties to Yoder qua breeder, whereas another propagator-distributor might have preferred to keep the sport for himself. Cal-Florida was therefore clearly disadvantaged competitively by the existence of the BGA and GRA programs. All of the foregoing considerations convince us that Cal-Florida’s claimed injuries did occur within a sector of the economy which was endangered by the BGA and GRA programs. Moreover, because Yoder certainly did intend to operate the BGA program, and because it indisputably did intend to secure compensation for breeders in the industry and to get sports back to the breeders, we believe that it must have intended the necessary consequences of its acts, thus satisfying whatever purpose and intention requirement might exist for standing under section 4. One need not be sitting on the bull’s-eye in order to be within the target area of an antitrust conspiracy. We therefore find that Cal-Florida did have standing to seek treble damages under section 4 of the Clayton Act. B. Statute of Limitations Section 4B of the Clayton Act, 15 U.S.C. § 15b, provides a four year statute of limitations for actions brought under section 4 of the Clayton Act, 15 U.S.C. § 15. If the private action has been preceded by a Government proceeding and if it is based in whole or in part on any matter complained of in the Government’s action, then under Section 5(b) of the Clayton Act, 15 U.S.C. § 16(i), the statute of limitations is tolled during the pendency of the suit by the United States and for one year thereafter. The tolling provision of the statute has the effect of giving antitrust plaintiffs a period of four years plus the length of the Government suit for which they can recover. The United States instituted its suit against Yoder Brothers and BGA on April 20, 1970. On January 26, 1972, Yoder and the United States filed a stipulation containing a proposed consent decree which would become effective thirty days following the filing date. On February 16, 1972, CFPC and CFPCF moved to intervene in the consent proceeding. The district court denied both motions on March 15, 1972, and entered the consent decree on that date. Over a year later, on April 12, 1973, CFPCF’s counterclaim alleging antitrust violations was filed. Even later, on June 5, 1973, CFPC’s counterclaim under the antitrust laws was filed. CFPC and CFPCF seek to escape the consequences of their tardy filing by arguing that the Government’s action did not cease to “pend” until the statutory periods provided for appeal elapsed. They assert that they were entitled to the full sixty days provided under the Expediting Act, 15 U.S.C. § 29, to appeal from the denial of their motions to intervene. Additionally, since the parties to the lawsuit might have appealed from the provisions of the consent decree, they assert that the Government’s suit continued to pend until May 15, 1972, thus making CFPCF’s filing date within the one year period following the end of the Government’s suit. CFPC’s claim, since it was filed on June 5, would still have to relate back to the Florida filing date in order to be timely. Yoder argues that the Government’s suit ceased to pend on the date that the consent decree was entered and relies in the first instance on three Supreme Court cases in which the Court assumed that the date on which the judgment or decree is entered governed. American Pipe & Constr. Co. v. Utah, 1974, 414 U.S. 538, 94 S.Ct. 756, 38 L.Ed.2d 713, 731; Zenith Radio Corp. v. Hazeltine Research, Inc., 1971, 401 U.S. 321, 91 S.Ct. 795, 803-04 n. 5, 28 L.Ed.2d 77; Minnesota Mining & Mfg. Co. v. New Jersey Wood Finishing Co., 1965, 381 U.S. 311, 85 S.Ct. 1473, 1475, 14 L.Ed.2d 405. Because the question of when government proceedings cease to “pend” for purposes of Section 5(b) was not squarely before the Court in any of those cases, we do not regard them as dispositive of the question. Rather, we examine the problem in the light of the precedents more directly on point to decide whether the date that the judgment or decree was entered or the date when the time for appeal expired governs. The Ninth Circuit considered a question analogous to the one before us in Marine Firemen’s Union v. Owens-Corning Fiberglass Corp., 9 Cir. 1974, 503 F.2d 246. The Marine Firemen’s Union had filed a private antitrust action against the Owens-Corning Company alleging violations of section 1 of the Sherman Act, 15 U.S.C. § 1. Marine’s action followed a criminal proceeding that had been brought by the United States on December 28,1964, which had also alleged a combination and conspiracy in violation of Sherman Act § 1. On February 6,1969, the last of the defendants entered a plea of nolo contendere in those proceedings. On February 17, 1969, the district court orally pronounced its sentence. However, the judgment of conviction on those pleas was not signed by the sentencing judge until February 19, 1969; the clerk entered the judgment on February 20, 1969. Marine’s action was instituted on February 18, 1970, less than one year after entry of the judgment, but more than one year after entry of the plea and oral pronouncement of the sentence. In the subsequent private action, the district court ruled that the “pendency” of the Government’s criminal proceedings terminated on February 17, 1969, when the defendants were orally sentenced. Reversing, the Ninth Circuit held that the pendency of the Government’s case “continues for a period of one year from and after the date of entry of judgment of conviction against the last remaining defendant in the related criminal proceeding.” 503 F.2d at 249. (Emphasis in original.) The court made the following comment on the judicial administration aspect of the problem: Case law chaos results if different procedural points and dates thereof are to be for one reason or the other selected as the commencement date of time limits on the myriad of post judgment remedies open to any given party, such as appeal, motions for new trial, reduction of sentence or filing of costs bills, to name a few. Orderly procedural necessity dictates uniformity. We sense no logical reason to differentiate the procedural point or date of final adjudication and termination of the “pendency” of a given criminal proceedings [sic] for the purpose of computing statute of limitations time from that of calculating appeal time for any given party. Accordingly, we hold that the “pendency” of the related criminal proceeding referred to in § 16(b) [§ 16(i)] terminates at the procedural point and date of the clerk’s notation in the “criminal docket” for the case of the entry of the judge’s signed written judgment of conviction of the last remaining defendant in the criminal proceeding. 503 F.2d at 250. The court did not reach Marine’s contention that the “pendency” of the criminal proceeding continued through appeal time. Although Marine Firemen’s Union arose in the case of a private action following a criminal proceeding, and upheld the litigant’s right to the tolling period, we believe that its reasoning is equally applicable here. Rule 58 of the Federal Rules of Civil Procedure provides that “a judgment is effective only . . . when entered as provided in Rule 79(a).” Rule 79(a) directs the clerk to enter all papers chronologically in the civil docket and specifically provides that “the entry of an order or judgment shall show the date the entry is made.” Rule 4 of the Federal Rules of Appellate Procedure relies on the date of the entry of the judgment or order appealed from for determining the time for appeal. Unless either the district court or the appellate court grants a stay pending appeal, execution can normally take place on a final judgment of a district court. Fed.R.Civ.P. 62, Fed.R.App.P. 8. Particularly with respect to consent decrees, it makes sense to look to the date of the entry of the decree. Since the scope of review of consent decrees is extremely narrow, the outcome of the lawsuit is practically certain as of the time the decree is entered. Generally, the only matters that can be raised on appeal are lack of jurisdiction over the subject matter or facts which would vitiate the consent. Martin Marietta Corp. v. FTC, 7 Cir., 376 F.2d 430, 434, cert. denied, 1967, 389 U.S. 923, 88 S.Ct. 237, 19 L.Ed.2d 265. See Fuller v. Branch County Road Comm’n, 6 Cir. 1975, 520 F.2d 307. Thus, cases involving consent decrees may be distinguishable from cases in which the Government action terminated in a final judgment after full litigation. Deciding a case in the latter category, the Second Circuit decided that the “pendency” of a Government enforcement action continues until the expiration of the time to appeal from the final decree. Russ Togs, Inc. v. Grinnell Corp., 2 Cir., 426 F.2d 850, 857, cert. denied, 1970, 400 U.S. 878, 91 S.Ct. 119, 27 L.Ed.2d 115. The court rested its decision on the ground that [a] judgment or decree in a government enforcement action becomes “final” only when the government and the defendants are satisfied with the result and determined not to appeal. Therefore, only after the time to appeal has expired can private litigants rely on the irrevocability of determinations made in the government action. 426 F.2d at 857. Even the Russ Togs court did not require absolute irrevocability of determination, however, for the court expressly refused to hold that pendency included the period subsequent to a final decree during which a court exercises continuing jurisdiction for purposes of modification and enforcement. 426 F.2d at 856 n. 8. Rather, the court’s concern appeared to be with the ability of private litigants to rely upon the finality of the provisions in the judgment or decree. When the Government litigation terminates in a consent decree, parties will rarely, if ever, be injured by reliance on the decree’s provisions. Thus, even taking into account the concerns of the Russ Togs court, We see no reason to include the time for appeal within the “pendency” of the Government’s suit in the circumstances sub judice. Here, since the Government proceedings terminated in a consent decree, we hold that the Government’s action ceased to “pend” for purposes of section 5(b), 15 U.S.C. § 16(i), on March 15, 1972, the date on which the consent decree was entered. Therefore, since neither CFPC nor CFPCF filed its claim within one year of that date, neither company is entitled to take advantage of the tolling provision of section 5(b). Each is relegated to the four year statute of limitations contained in section 4B, 15 U.S.C. § 15b. Yoder also attacked the district court’s ruling that the CFPC claim filed on June 5, 1973, could relate back for statute of limitations purposes to the filing date of CFPCF’s counterclaim, on April 11, 1973. If Yoder is correct, then the CFPC claim would cover only the period from June 5, 1969, onward. We have decided not to disturb the lower court’s ruling on this point. In light of the close parent-subsidiary relationship between the two companies, the identity of their business, and the identity of their claims, we think the court properly held that all claims were to be measured from April 11, 1969. C. Per se Illegality of BGA and GRA Section 1 of the Sherman Act prohibits “[e]very contract, combination . ., or conspiracy, in restraint of trade . .” 15 U.S.C. § 1. Although generally this section prohibits only “unreasonable” restraints on competition, see Standard Oil Co. v. United States, 1911, 221 U.S. 1, 31 S.Ct. 502, 55 L.Ed. 619, the Supreme Court has written that: [T]here are certain agreements or practices which because of their pernicious effect on competition and lack of any redeeming virtue are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use. This principle of per se unreasonableness not only makes the type of restraints which are proscribed by the Sherman Act more certain to the benefit of everyone concerned, but it also avoids the necessity for an incredibly complicated and prolonged economic investigation into the entire history of the industry involved, as well as related industries, in an effort to determine at large whether a particular restraint has been unreasonable — an inquiry so often wholly fruitless when undertaken. Among the practices which the courts have heretofore deemed to be unlawful in and of themselves are price fixing, United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 210, [60 S.Ct. 811, 883, 84 L.Ed. 1129]; division of markets, United States v. Addyston Pipe & Steel Co., [6 Cir.], 85 F. 271, aff’d, 175 U.S. 211, [20 S.Ct. 96, 44 L.Ed. 136]; group boycotts, Fashion Originators’ Guild v. Federal Trade Comm’n, 312 U.S. 457, [61 S.Ct. 703, 85 L.Ed. 949]; and tying arrangements, International Salt Co. v. United States, 332 U.S. 392, [68 S.Ct. 12, 92 L.Ed. 20]. Northern Pacific Ry. v. United States, 1958, 356 U.S. 1, 5, 78 S.Ct. 514, 518, 2 L.Ed.2d 545. In United States v. General Motors Corp., 1966, 384 U.S. 127, 86 S.Ct. 1321, 16 L.Ed.2d 415, the Court reaffirmed its position that group boycotts are among those classes of restraints that are illegal per se. To label an arrangement a group boycott, however, is merely to state a conclusion. In order to determine whether or not the label fits, it is necessary to ascertain whether the presence of exclusionary or coercive conduct warrants the view that the arrangement in question is a “naked restraint of trade.” E. A. McQuade Tours, Inc. v. Consolidated Air Tour Manual Comm., 5 Cir. 1972, 467 F.2d 178, 187, cert. denied, 1973, 409 U.S. 1109, 93 S.Ct. 912, 34 L.Ed.2d 690. See Sulmeyer v. Coca Cola Co., 5 Cir. 1975, 515 F.2d 835, cert. denied, 1976, 424 U.S. 934, 96 S.Ct. 1148, 47 L.Ed.2d 341. The McQuade Tours court discerned three categories of eases in which collective refusals to deal were condemned as per se violations of section 1. In our opinion, the third category comes closest to describing the fact situation before us. In that group of cases, the combinations were designed to influence coercively the trade practices of boycott victims, rather than those of direct competitors. The leading case of this type is Fashion Originators’ Guild of America, Inc. v. FTC, 1941, 312 U.S. 457, 61 S.Ct. 703, 85 L.Ed. 949 (hereinafter referred to as FOGA). In FOGA, a combination of women’s garment designers, textile manufacturers, and textile dyers, in an effort to stamp out “style piracy,” refused to sell their products to retailers who purchased and sold garments that were surreptitiously copied from the FOGA designers. Retailers were required to sign agreements that assured their cooperation with the FOGA boycott; those who did not sign would not receive original designs from FOGA participants. The FOGA system was enforced by anonymous visits to retail outlets, audits of members’ books, and a variety of prohibitions on members’ business practices. The Supreme Court held that FOGA’s refusal to deal with retailers who patronized style pirates was illegal per se. Because the purpose and object of the combination was to destroy one type of manufacture and sale that competed with FOGA members— i.e. imitation — the Court found irrelevant FOGA’s proffered evidence tending to show that the program was reasonable since it protected all persons in the manufacturing chain from the evils of style piracy. Like the ill-fated FOGA system, BGA required all persons who wanted access to new varieties developed by its breeder members to sign an appropriate BGA agreement. Propagator-distributors were forbidden to sell, loan, or in any way to place BGA cuttings in the hands of nonsignatories. Like FOGA, BGA had the right to audit members’ books. Like FOGA, BGA contained other restrictions on members’ business practices — most importantly, the requirement that all sports found on BGA varieties be returned to BGA. Like FOGA, if a potential customer refused to sign an agreement, the requested BGA variety would not be sent to him. Finally, unlike FOGA but like General Motors, supra, an indispensible part of the BGA program was the assessment and collection of a royalty whose amount was fixed in advance by the breeder members of BGA. See 384 U.S. at 147, 86 S.Ct. at 1331, 16 L.Ed.2d at 427. We believe that these factors present the kind of exclusionary or coercive conduct characteristic of a “naked restraint of trade,” see McQuade Tours, supra. The differences in degree of enforcement between BGA and FOGA, to the extent they existed, do not warrant a contrary conclusion. The central purpose of the BGA program was to ensure that a set royalty was paid on every BGA cutting sold, and that BGA cuttings were not released to those who would not pay. A secondary and equally exclusionary purpose was to retain control over the sports that appeared on BGA plants. In light of all these factors, we hold that the district court correctly ruled that the BGA and GRA programs were per se violations of section 1 of the Sherman Act. D. Monopolization and Attempted Monopolization On this phase of the appeal, Cal-Florida asserts that the district court erred in granting Yoder’s motion for a directed verdict on the section 2 claims of monopolization and attempted monopolization. Without specifying what the relevant market was, the court held that Yoder did not have a sufficient share of the market to permit an inference of monopoly power, citing United States v. Grinnell Corp., 1966, 384 U.S. 563, 86 S.Ct. 1698, 16 L.Ed.2d 778, and Cliff Food Stores, Inc. v. Kroger, Inc., 5 Cir. 1969, 417 F.2d 203. Additionally, the court found that the evidence tending to show dangerous probability of success in an attempt to monopolize was insufficient to go to the jury. Cal-Florida presents two arguments in support of its position: first, that the district court erroneously believed that a defendant had to have something more than 50 percent of the market before a monopolization or an attempt offense could be made out, misinterpreting this Court’s Cliff Food Stores decision; and second, that the facts were sufficient to allow the jury to find even a 50 percent market share, depending on how the market is defined. It offers four possibilities for the product market: (1) ornamental plants; (2) chrysanthemums grown; (3) chrysanthemum cuttings sold; and (4) new varieties of chrysanthemum cuttings sold. Yoder responds that Cal-Florida failed to prove a relevant market and mounts various attacks on Cal-Florida’s computations attempting to show a market limited to chrysanthemum cuttings. 1. Monopolization.—In United States v. Grinnell Corp., 1966, 384 U.S. 563, 570-71, 86 S.Ct. 1698, 1703-1704, 16 L.Ed.2d 778, 785-786, the Supreme Court summarized the section 2 monopolization offense as follows: The offense of monopoly under § 2 of the Sherman Act has two elements; (1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident. Monopoly power, defined as “the power to control price or exclude competition,” is measured with reference to a relevant market. United States v. E. I. du Pont de Nemours & Co., 1956, 351 U.S. 377, 391, 76 S.Ct. 994, 1005, 100 L.Ed. 1264 (sometimes referred to as the Cellophane case). Because the definition of relevant market is essentially a fact question, see Sulmeyer v. Coca Cola Co., supra, 515 F.2d 835, 849, the precise issue we