Antitrust litigation seldom partakes of elegant simplicity. Its elaborations find expression both in the intricacies of finely woven legal theory and in the expansiveness of painstakingly detailed factual submissions. In the face of such complexity, considerable value attaches to efforts by the trial judge to confine the task of the jury to manageable dimensions by the judicious use of directed verdicts. Nonetheless, the Seventh Amendment guarantees a party the right to commit the task of determining facts in civil proceedings to a jury. And the difficulties of deciphering the import of extensive and conflicting evidence may not be allowed to disturb the exercise of that right.

In the matter before us, the resolution of charges of antitrust violations in the market for aluminum culvert pipe turns on disputed factual issues of conspiracy, motive, and market definition. The trial judge, however, deemed it appropriate to enter a directed verdict at the close of the plaintiff's case. We are now called upon to determine whether the extensive trial record permitted such an action.

This proceeding arose out of the interaction of five entities: Columbia Metal Culvert Company, Inc. (the plaintiff), Kaiser Aluminum & Sales, Inc. (KACSI), Kaiser Aluminum Chemical Corporation (KACC), Robert Kennedy, and Kennedy Culvert & Supply Company.

The plaintiff in this case originated in 1959, when Joseph Bonjourno founded Columbia Steel, a corporation that was to fabricate metal culvert pipes used in highway construction. Three years later, after some financial difficulties, Columbia was reconstituted as Columbia Metal Culvert, the current plaintiff. Columbia Metal's plant in Vineland, New Jersey, specialized in the manufacture of aluminum culvert pipes, which the company then marketed.

KACC and its wholly owned subsidiary, KACSI, constitute the second alignment in the case. KACC manufactures aluminum sheet and coil, from which aluminum culvert pipe, such as Columbia's product is constructed; it then conveys these pipes to KACSI, at a "transfer price" which is not necessarily reflective of either costs or market price. KACSI in turn sells the sheet and coil to pipe fabricators such as Columbia. In addition, KACSI fabricates and sells aluminum culvert pipe itself in competition with the fabricators to whom it supplies sheet and coil. KACSI sells 80% Of the aluminum culvert pipe in the United States, although only 79% Of its pipe output is aluminum (the rest being steel). Neither of the other large aluminum producers manufactures aluminum culvert.

The third set of parties before us is comprised of Kennedy Culvert & Supply Company, a distributor of KACSI culvert pipe in Southern New Jersey, and its owner, Robert Kennedy, a former salesman for Columbia Metal.

In earlier times, the parties had enjoyed a close business relationship. Columbia's decision in 1962 to begin producing aluminum culvert was encouraged by KACC and KACSI, who supplied technical assistance, as well as all of Columbia's raw materials. Columbia was one of the few strictly aluminum culvert manufacturers in the country and KACSI manifested an interest in using Columbia as a spearhead to break into the culvert market a market which had previously been dominated by steel and concrete products. In these amicable years, Robert Kennedy was one of the most successful of Columbia's salesmen.

The core of Columbia's antitrust claim is that in 1971, when Columbia decided to place some of its aluminum orders with Reynolds Aluminum, KACSI determined to put Columbia out of business in retaliation. After learning of the errant orders, the evidence reveals, KACSI refused to sell aluminum to Columbia.1 In addition, Columbia alleges that in an act of vengeance for Columbia's infidelity, KACSI decided, with KACC's approval, to locate a new culvert manufacturing plant in New Castle, Delaware, within fifty miles of Columbia's plant in Vineland, New Jersey, despite the fact that KACSI's new production unit had earlier been slated for Virginia. Because of the high cost of transporting culvert, there is evidence that the Delaware location meant that Columbia would become subject to severe new competition from KACSI culvert.

At the same time, Columbia maintains, KACSI conspired with Robert Kennedy, still a Columbia salesman, to set Kennedy up in business as an independent distributor of KACSI products in the Vineland area, and thereby to complete the destruction of Columbia's market. Columbia charges that Kennedy used confidential knowledge he had gained from his work with Columbia to underbid Columbia, and that Kennedy and KACSI conspired to avoid competing with one another.

Finally, Columbia claims that it was the victim of a price squeeze consciously engineered by KACSI and KACC to drive independent fabricators such as Columbia out of business. Sheet and coil were transferred from KACC to KACSI below cost, enabling KACSI to make a "profit" on sales of aluminum pipe at prices too low to be matched by independents who had to buy their sheet and coil at market prices. Simultaneously, it is claimed, KACSI, with KACC's concurrence, contrived to raise the price of raw materials confronting Columbia.

As a result of these maneuvers, Columbia asserts, it was driven out of business.

In 1974, Columbia brought the present law suit, which charges Kennedy, KACSI, and KACC with violations of §§ 1 and 2 of the Sherman Act. In addition, the complaint alleges that KACSI is guilty of an infraction of § 3 of the Clayton Act by its attempt to insist that Columbia deal exclusively with KACSI for coil and sheet.

After the jury had heard the twelve days of testimony comprising the plaintiff's case, the trial judge directed a verdict for the defendants on all of Columbia's claims. Judge Cahn held:

1. The evidence "tended to establish" that the relevant market in the case was not, as Columbia contended, a market for aluminum culvert, of which KACSI controlled 80%, but a market for both steel and aluminum culvert, where KACSI sales represented less than 10%.2

2. As a result, KACC/KACSI could not be guilty of monopolization in violation of § 2 of the Sherman Act since they possessed no monopoly power in the relevant market.3. Neither could KACC/KACSI be guilty of attempted monopolization since the limited market position precluded a "dangerous probability of success."

4. KACC, KACSI and Kennedy could not be liable for conspiracy to monopolize, on the ground that insufficient evidence had been adduced for a jury to find either specific intent to monopolize or a conspiracy to monopolize in the relevant product market.

5. A jury could not find KACC and KACSI guilty of a conspiracy or combination in restraint of trade under § 1 of the Sherman Act, since not enough evidence had been presented to allow a reasonable inference of conspiracy between KACSI and KACC, and because the market share was too small to allow a finding of restraint of trade.

6. Kennedy and KACSI could not be held liable for conspiracy in restraint of trade because there was inadequate evidence to allow a jury to find a conspiracy between Kennedy and KACSI, a substantial restraint of trade, or an intent to put Columbia out of business.

7. The failure to present evidence on the proper definition of the market for the Clayton Act claim precluded a finding in plaintiff's favor on that issue.

Columbia has appealed each aspect of the directed verdict.

As the trial judge recognized, in evaluating defendants' motion for a directed verdict, the question is whether sufficient evidence has been introduced, when viewed in the light most favorable to the plaintiff and allowing all reasonable inferences in its behalf, to allow a jury to find that relief is warranted.3

The badinage between the plaintiff and defendants in this case over whether "some" evidence, "substantial" evidence, "any" evidence or a " scintilla" of evidence is necessary in order to defeat a motion for directed verdict impels us to reiterate the Supreme Court's comments of 35 years ago in Galloway v. United States:4

Nor is the matter greatly aided by substituting one general formula for another. It hardly affords help to assist upon "substantial evidence" rather than "some evidence" or "any evidence" or vice versa. The matter is essentially one to be worked out in particular situations and for particular types of cases. Whatever may be the general formulation, the essential requirement is that mere speculation not be allowed to do duty for probative facts, after making due allowance for all reasonably possible inferences favoring the party whose case is attacked.

Where the issues are ones for resolution by the jury, the judge may deprive the jury of its role at trial only where such action is necessary to guard against a verdict founded solely on "mere speculation."

The fulcrum of the district court's opinion was its determination that no jury could reasonably find the relevant market in this case to be composed of any spectrum of products more restricted than "aluminum culvert and steel culvert." This determination reduced the market position of the defendants, as proved at trial, from over 80% Of the aluminum market to less than 10% Of the aluminum and steel market.5 Such a reduction in market share allowed the trial judge to proceed to draw the conclusions that the case contains no adequate evidence to allow the jury to find monopolization, dangerous probability of monopolization, or restraint of trade. The correctness of the district judge's market determination, therefore, must frame the analysis of the charges brought by Columbia.

Well-settled principles govern the delineation of the relevant product market in an antitrust case, even if the outcome of their application to individual fact patterns is often less than obvious.

The "monopoly" condemned by § 2 of the Sherman Act6 inheres in the "the power to control prices or exclude competition."7 To the extent that competition from related products limits the market power of an entity with a dominant position in one product, such an entity is less likely to be found to hold "monopoly" power forbidden by law. This is so because the ongoing competition from other products guards against the ability of the dominant entity to increase prices and makes exclusionary tactics by such a party fruitless, impossible or unbearably expensive. Thus, in resolving the proportion of the "market" controlled as a prelude to an examination of the extent of a firm's power to control prices or exclude competition, the courts look to the range of "commodities reasonably interchangeable by consumers for the same purposes."8

Similarly, in the context of § 1's prohibition of conspiracies in restraint of trade, except where practices fall under a judicially crafted Per se ban, a finding of illegality presupposes a determination in any given case that the "effect upon competition in the marketplace is substantially adverse."9 This inquiry, in turn, necessitates an examination of the boundaries of real competition. As the Supreme Court stated in Times-Picayune Publishing Co. v. United States "our inquiry to determine reasonableness under § 1 must focus on 'the percentage of business controlled, the strength of the remaining competition (and) whether the action springs from business requirements or purpose to monopolize.' "10

Therefore, under either § 1 or § 2 of the Sherman Act, judges must attempt to ascertain the flow of commercial interactions. And in exploring this pattern of competition, courts are adjured to follow the well-trodden trail illuminated by the late Chief Justice Warren in Brown Shoe Co., Inc. v. United States :11

The outer boundaries of a product market are determined by the reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it. However, within this broad market, well-defined submarkets may exist for antitrust purposes. The boundaries of such a submarket may be determined by examining such practical indicia as industry or public recognition of the submarket as a separate economic entity, the product's peculiar characteristics or uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.

That path, however, has led to a variety of destinations. It is true, as the defendants here point out, that the Supreme Court in DuPont, a § 2 controversy, refused to treat "cellophane" as a product market distinct from other flexible wrapping paper.12 Yet, in subsequent cases, the Court has countenanced the separation of "accredited central station alarm systems" from non-accredited alarms, noncentral station alarms, and watchman services;13 the distinction between markets for insulated aluminum and insulated copper wire, despite the fact that the same customers bought each for similar purposes;14 and the differentiation between championship boxing matches and all other boxing matches.15 Thus, the breadth of the spectrum of products which have been held to constitute "markets" for antitrust purposes bears out the Supreme Court's observation in United States v. Continental Can that:(T)he (legal) guidelines offer no precise formula for judgment and they necessitate, rather than avoid, careful consideration based on the entire record.16

As this Court recently indicated, a pronouncement as to market definition is not one of law, but of fact, and as such, a party in a private action may allocate it to the jury.17 Meticulous "color-matching"18 of precedent to determine how similar a particular product is to others which have or have not constituted markets in other cases is thus not nearly so important as a detailed examination of the record developed in the trial court.19

According to the evidence here, culvert pipe is used primarily in large construction projects, such as housing developments and highways, to provide channels for draining surface water. Three types of culvert pipe are manufactured: steel, concrete, and aluminum. There was testimony from a number of witnesses that since all three materials have different physical properties, with rare exceptions specifications for construction projects call for only one type of culvert.20 As a result, on most occasions once the engineering specifications for the project have been fixed the contractor who actually purchases culvert pipe may choose only among different manufacturers of the same material. From the contractor's point of view, there could reasonably be said to exist distinct markets for aluminum, for steel and for concrete pipes. Consequently, if a monopolist in aluminum culvert avoided exceeding the price necessary to induce engineers not to specify aluminum, he might well be able to raise prices without fearing encroachment on his sales from other products.

Nor would a jury have acted unreasonably in finding, on the basis of the record here, that the patterns of action on the part of engineers who specify the type of culvert to be used establish aluminum as a separate market. For there was testimony that price was not the crucial variable in the choice by engineers among the materials used to drain surface water. Roy Elam, a former city engineer who qualified as an expert witness, testified that in his experience "cost is checked out at the end of the design," and "budget numbers are not influenced much" by the cost differentials between concrete, aluminum and steel.21 He also indicated that the choice among the materials was made primarily on the basis of their physical properties, rather than price.22

Credible testimony was produced to show that there is considerable difference among the physical properties of concrete, aluminum and steel: Concrete is stronger than steel, which in turn is stronger than aluminum. Concrete culverts, being smooth, conduct water better than steel or aluminum culverts, which are made out of corrugated metal. But steel and aluminum are easier to work with.23 Further, steel generally is stronger than aluminum. Because of this, and because it has been in the market longer, steel is more widely used than aluminum. The State of New Jersey and the Federal government on occasion specify steel, but forbid aluminum.24 There was also testimony, however, that aluminum resists corrosion better than steel and is more durable, particularly in salt-water regions. As a result, Elam asserted, aluminum is used to the exclusion of steel "along the shore."25 Aluminum is also lighter than steel, and thus is preferred in marshy areas, where access for heavy equipment would be difficult.26

In sum, Elam stated that, as to engineering properties, there was "no way you could say these three (types of culvert) are interchangeable,"27 and that the materials "were not in competition from an engineering standpoint."28

Thus, there is evidence from which the jury could reasonably have concluded, as did Mr. Elam, that aluminum is recognized by customers as a "specialty material used only in special applications."29 Aluminum culvert could therefore have been found by the jury to constitute a separate submarket.30

Defendants point to testimony by producers of aluminum culvert pipe, including the plaintiff's president, which makes reference to "competition" between aluminum and steel manufacturers in attempting to have their respective products specified by engineers. They also refer to statements by KACSI and Reynolds employees that prices of aluminum culvert are set with reference to steel.31

Although such testimony is surely not irrelevant to the definition of a market, neither is it determinative in the context of this case. First, it is not the perceptions of manufacturers but those of consumers which are most salient in the determination of market boundaries.32 More important, the existence of competition between two product lines does not alone preclude market power within each line, if each product has a cadre of customers in which it enjoys a decisive advantage. For example, in United States v. Grinnell,33 the Supreme Court acknowledged competition between the defendants' "central station alarm" system and other protective devices. Indeed said the Court:

We recognize that (as the district court found) (the defendants) do not have unfettered power to control the price of their services . . . due to the fringe competition of other alarm or watchman services.34

Nonetheless, the Court held that the central station services constituted a separate market, since "some customers will . . . be unwilling to consider anything but central station protection."35 From the evidence presented, it would not have been unreasonable for the jury here to have made a similar finding. Accordingly, we must reject the trial judge's conclusion that the evidence "establishes that aluminum culvert pipe is not a separate submarket from steel culvert pipe."36 Inasmuch as there was sufficient evidence to go to the jury on this central question, we treat the market, at least for purposes of this appeal, as composed of aluminum culvert pipe.

The directed verdict against the plaintiffs on the claims that the defendants monopolized, attempted to monopolize, and conspired to monopolize in violation of § 2 was based on the trial judge's prior conclusion that steel and aluminum, rather than aluminum culvert pipe alone, constituted the relevant market. Based on the small percentage of the steel and aluminum market which the defendants represented, the trial judge held that there was insufficient evidence to allow a jury to find monopolization, the probability of success necessary for attempted monopolization,37 or the specific intent prerequisite to liability on a conspiracy to monopolize.

Since we hold that a conclusion that aluminum culvert pipe constituted a relevant market was not unreasonable, the foundation of the directed verdicts on the § 2 counts collapses.

There was sufficient evidence from which the jury could have inferred that KACSI attempted to drive Columbia out of business. Holmes Collins, the manager of KACSI's highway division, was heard to say that if Columbia bought metal from Reynolds he, Collins, would locate an aluminum culvert plant in Columbia's "backyard."38 This was relevant because transportation of culvert is a significant item of expense for the product in question, and a local manufacturer of culvert has an advantage over one whose plant is at a distance from the place of delivery.39 In fact, Collins did so locate a plant, despite the fact that there was evidence that his original plans called for a Virginia situs.40

Moreover, the jury could have found that Collins extended credit to Kennedy, a competitor of Columbia, against the recommendations of the KACSI credit department,41 thus showing that KACSI engaged in conduct against its business interest in order to damage the plaintiff.

More generally, there was evidence that KACC transferred sheet and coil to KACSI at accounting prices that were less than production cost and well below market price, enabling KACSI to sell pipe more cheaply than independent fabricators could manage.42 While this may not be a classic predatory pricing endeavor, it effectively allowed KACSI to drive independents out of the market. Finally, the prices at which the Delaware KACSI plant sold culvert were lower than those KACSI charged at certain other plants, despite the fact that costs in Delaware were no lower giving rise to a permissible inference that prices were depressed in an effort to harm Columbia.43

From the testimony and from KACSI's actions against its business interests, the jury could have inferred that KACSI intended specifically to punish Columbia for buying from Reynolds. And given the formerly close relations between KACSI and Columbia, the jury might well have inferred that KACSI knew of the drastic impact its maneuvers would have on Columbia's prospects for survival. Such activities, engaged in by an entity controlling more than 80% Of the market, would be sufficient to allow the jury to find both the anticompetitive effects and intent requisite to liability for monopolization or attempted monopolization.

Accordingly we reverse the directed verdict as to defendants KACSI and KACC on the § 2 counts.44

By its terms, § 1 of the Sherman Act proscribes "every combination . . . or conspiracy in restraint of trade." Liability under that section obviously requires a finding of a combination or conspiracy. In addition, under well-worn judicial interpretation, a finding of a § 1 violation entails a determination that such conspiracy or combination Unreasonably restrains trade.45 The trial judge here supported his decision to grant a directed verdict on the § 1 charge both on the ground that there was insufficient evidence from which a jury could have found a combination or conspiracy, and on the basis that the evidence could not as a matter of law support a finding that whatever conspiracy allegedly existed unreasonably restrained trade.

The second rationale is untenable in light of our conclusion that the jury could have found that aluminum culvert is a relevant market. If the jury ascertained that KACSI, which controlled 80% Of the aluminum culvert market, conspired to drive one of its few competitors (Columbia) out of business, such a conspiracy could have been found by the jury to violate § 1. Since, as we have noted previously,46 the presence of independent producers in an oligopolistic market serves to limit the market power of the dominant firms, the departure of an independent producer in such a situation adversely affects competition. Where such an elimination is consciously brought about by concerted action to which a dominant entity is a party, at least in the absence of a compelling business justification for such actions, an unreasonable restraint of trade may be found to exist.47

Thus the issue as to the § 1 violation is whether the trial judge's alternative ground for decision is viable: that there was inadequate evidence to allow the jury to find a conspiracy involving KACSI to put Columbia out of business.

As observed above, there was evidence which could have permitted a jury to infer that KACSI acted with an intent to drive its competitor, Columbia, out of business. The propriety of the directed verdict thus turns on whether, with regard to KACSI and any other defendant,

. . . the circumstances are such as to warrant (the trier of fact) in finding that the (alleged) conspirators had a unity of purpose or a common design and understanding or a meeting of minds in an unlawful arrangement.48

Consequently, we now turn to examine separately the two groupings which, according to Columbia, harbor such common unlawful purposes.

The first combination alleged to violate § 1 arises out of the joint activities of KACC and KACSI. KACSI is a wholly-owned subsidiary of KACC, sharing joint management on several levels of operation. Because of this, KACSI and KACC suggest that they are incapable of conspiring together. Though legally distinct for most purposes, defendants argue, the reality of the corporate structure of KACC and KACSI negatives the plurality of actors legally necessary to form an illegal conspiracy.

This argument is not well-founded in law, for the prerequisites of a conspiracy under § 1 are fulfilled by the presence of two or more legally distinct corporations. Indeed, we recently had occasion to reject a contention of legal inability to conspire with regard to subsidiaries which were "part of an integrated Chrysler enterprise engaged in the manufacturing and marketing of automobiles." The claim that such corporation could not form a § 1 conspiracy, we held, was foreclosed by a line of cases in this Court reaffirming the viability of the "intra-enterprise conspiracy" theory as applied to separately incorporated entities.49

Judge Cahn's holding that there was insufficient evidence to go to the jury on the existence of a KACC/KACSI conspiracy was grounded on a different rationale. The trial judge pursued the opposite approach from the defendants' argument of "functional identity;" while he commented that he harbored doubts as to the capacity of KACSI and KACC to conspire, the trial judge held that rather than being too closely aligned to constitute a conspiracy, KACSI and KACC were not closely enough related. He determined that there was inadequate evidence to allow a jury to find that the two entities shared the requisite unity of purpose. We disagree with this evaluation.

Some of the efforts which allegedly engineered Columbia's elimination were concededly undertaken by KACSI alone. However, the jury could have found that the combined efforts of KACC and KACSI were utilized (1) to impose the price squeeze and (2) to open the New Castle plant.

In finding that prices for sheet and pipe were arrived at by Holmes Collins, KACSI's division manager, independently of KACC, and thus that KACC was not implicated by the price squeeze, the trial judge relied on the testimony of Collins as evoked by the defendants on cross-examination. Were this a finding of fact, and a judgment of credibility, such reliance could not be reversed. However, the defendants' motion was one for directed verdict, and Collins' testimony had been contradicted.

Collins himself stated elsewhere in his testimony that he had no part in setting the "specification price" of coil.50 This casts doubt on his later assertion that the responsibility for setting all prices was solely his. Moreover, in a memorandum to his subordinates, which was admitted into evidence, Collins noted that he "had been chastised for not getting our prices up high enough."51 Since there was testimony that Collins, as manager of the Highway Products Division of KACSI, reported to the Sheet and Plate Division of KACC,52 the jury could have inferred that the pricing policies of KACSI were not independent as Collins claimed at trial, but rather arrived at jointly with KACC. Likewise, another KACSI employee, David Thomas, stated that "someone higher up than Mr. Collins" was responsible for a crucial price increase.53 Again, the jury could reasonably have concluded from such testimony that pricing responsibility was shared with KACC.

Finally, and quite significantly, the keystone of the alleged price squeeze was the price at which KACC transferred coil and sheet to KACSI. So long as the transfer occurred at a price below cost, the predatory effect could be obtained if KACSI kept a sufficient and constant profit margin. Since KACC set the transfer price at a low level, apparently against its business interests, the jury could reasonably have found participation by KACC in the price squeeze.54

In regard to the placement of the New Castle plant, it appears that KACC's approval of Collins' "Request for Investment" was a precondition to his constructing the plant. From this involvement, along with the pricing policy, the close relationship between the companies, and the incentive for both KACC and KACSI to keep control of the market, it would not have been unreasonable for the jury to infer a conspiracy between KACC and KACSI to drive Columbia out of business.

In contrast to the broad-ranging activities of KACC and KACSI, Kennedy's role in the scenario of Columbia's demise was a limited one. Kennedy left Columbia and opened his own business in competition with his former employer.55 On its face, such conduct seems pro-competitive, although perhaps demonstrating less than punctilious faithfulness to Columbia, for it created a new competitive entity. And Bonjourno, president of Columbia, testified that two months before Kennedy left, Bonjourno, in the course of a dispute over Kennedy's salary, told Kennedy that Kennedy could "look around" in Southern New Jersey to see whether he could improve his remuneration.56 Moreover, Bonjourno reiterated that there were no contractual impediments to Kennedy's leaving Columbia for other employment.57

Certainly Kennedy's actions could not be characterized as being against his economic interest. In the four years he was employed by Columbia, Kennedy's sales of pipe rose to $300,000 per year, making Kennedy Columbia's best salesman.58 Nonetheless, Kennedy's salary increased only from $200 to $325 per week.59 Nor is there any testimony that Kennedy was attempting to destroy Columbia. A finding that Kennedy consciously shared KACSI's anti-competitive purposes would thus be mere conjecture.

Likewise, the relationship between Kennedy and KACSI was not such as to give rise to a reasonable inference that Kennedy was aware of KACSI's strategy and knowingly associated himself with it. While it is true, as plaintiff maintains, that certain testimony can be taken to imply that KACSI gave Kennedy favored treatment, the record is devoid of evidence that Kennedy was privy to KACSI's broader machinations. And a jury could not have reasonably found that Kennedy knew that the probable effect of KACSI's actions would be to eliminate Columbia. The record is barren of testimony that Kennedy was aware either of Columbia's production costs, or of the price at which it bought metal. Absent this information, Kennedy could not know that the competitive pressure his entrance into the market evoked would prove fatal to Columbia.60

Inasmuch as we agree with the district court that there was insufficient evidence from which the jury could find a "unity of purpose, design or understanding," between Kennedy and the other defendants, we sustain the directed verdict as to Kennedy.

The final count of Columbia's complaint charges KACSI with a violation of § 3 of the Clayton Act, 15 U.S.C. § 14, which makes it unlawful for:

. . . any person engaged in commerce, in the course of such commerce, to lease or make a sale or contract for sale of goods . . . or other commodities . . . on the condition, agreement, or understanding that the lessee or purchaser thereof shall not use or deal in the goods . . . or other commodities of a competitor or competitors of the lessor or seller, where the effect of such lease, sale, or contract for sale . . . may be to substantially lessen competition or tend to create a monopoly in any line of commerce.

Columbia contends that KACSI imposed upon it an exclusive dealing arrangement which precluded it from purchasing aluminum sheet and coil from other manufacturers so long as it bought KACSI materials. Such arrangement, Columbia maintains, substantially lessened competition in the aluminum coil and sheet market. This was true, Columbia alleges, because its purchases, which were foreclosed to other sellers, represented at least 8% Of the aluminum coil and sheet market where KACSI held 50% Of the total sales. Such foreclosure, Columbia contends, violates the Clayton Act.61 The district court, however, entered a directed verdict on this question because of the failure of Columbia adequately to prove that aluminum coil and sheet used in culvert fabrication was a "line of commerce" within the meaning of the Clayton Act.

In determining the share of the market preempted as a result of an exclusive dealing arrangement imposed by a producer, a court must look to the alternative customers available to purchase the output of other producers. Here, no evidence was adduced to show whether the output of the facilities which produced the aluminum sheet and coil for the culvert industry could be or were used to manufacture material sold only to fabricators of aluminum culvert pipe. Since this is so, there was insufficient evidence to demonstrate whether aluminum sold to culvert manufacturers, or some much larger market constituted the relevant line of commerce. Thus, the jury would be left without adequate guidance in attempting to delineate the share of the market foreclosed by the alleged exclusive dealing arrangement, or its probable market impact.

Consequently, we do not disturb the directed verdict on the Clayton Act claim.

(1) There was sufficient evidence to allow a jury reasonably to conclude that a relevant market for Sherman Act purposes was composed of aluminum culvert.

(2) There was sufficient evidence to go to the jury on the charge that KACSI and KACC violated § 2 of the Sherman Act by monopolizing or attempting to monopolize the aluminum culvert market.

(3) There was sufficient evidence to go to the jury on the charge that KACC and KACSI conspired in restraint of trade in violation of § 1 of the Sherman Act.

(4) There was insufficient evidence to allow the jury reasonably to conclude that Robert Kennedy and Kennedy Culvert conspired with KACSI either in violation of § 1 or § 2 of the Sherman Act.

(5) There was insufficient evidence to allow the jury reasonably to define the "line of commerce" restrained by any exclusive dealing arrangement, and, thus, there was insufficient evidence to go to the jury on the claimed violation of § 3 of the Clayton Act.

The judgment of the district court will therefore be affirmed as to the defendants Kennedy Culvert and Robert Kennedy. As to KACC and KACSI, the judgment will be affirmed insofar as it granted a directed verdict against the plaintiffs on the Clayton Act claims; the remainder of the judgment will be reversed and the case remanded for action consistent with this opinion.