This appeal questions the extent to which, without violating sections one and two of the Sherman Act, a major manufacturer of motor vehicles may subsidize its partially and wholly owned dealers (factory dealers) which compete at the retail level with independent dealers.

Plaintiff, a former independent Dodge dealership in Allegheny County, Pennsylvania, brought a private antitrust action under section four of the Clayton Act against defendants Chrysler Corporation and Chrysler Motors Corporation (collectively Chrysler) seeking treble damages resulting from alleged violations of the Sherman Act. 15 U.S.C. §§ 1, 2, 15 (1970). Plaintiff charged that Chrysler made large capital and other contributions to its partially or wholly owned Dodge dealerships (factory dealers) to offset losses incurred through extravagant expenditures for advertising, rent, and sales personnel; that Chrylser discriminated between its factory and independent dealers with respect to delivery of new model cars for display, special bonuses and rebates, special tool allowances, and bookkeeping services provided by Chrysler; that Chrysler engaged in unfair competition and deceitful advertising practices; and that Chrysler refused to permit independent dealerships to change locations. Plaintiff alleged that Chrysler committed and conspired to commit the above acts with the intent to destroy independent Dodge dealerships and that, as a result, plaintiff was compelled to surrender its franchise and cease operations.

After trial in the Western District of Pennsylvania, the jury returned the following verdict in response to special interrogatories:

1. Did the defendants engage in any combination or conspiracy in effect between July 14, 1964 and July 14, 1969 which unreasonably restrained interstate trade or commerce in Dodge vehicles at the retail level in Allegheny County?

2. Did the defendants, during the period July 14, 1964 to July 14, 1969 attempt to monopolize interstate trade or commerce in Dodge vehicles at the retail level in Allegheny County?

3. If your answer to Question 1 or 2, or either of them, is 'Yes', did the plaintiff sustain any damages by reason of any such violation of the antitrust laws?

4. If so, state the amount: $300,000.00.

The district court entered judgment awarding plaintiff treble damages in the amount of $900,000.

Chrysler timely moved for judgment notwithstanding the verdict or, in the alternative, for a new trial. It asserted, inter alia, that the record contained insufficient evidence to sustain the jury's verdict; that certain evidence, including the verdict in a similar case, should have been excluded; and that the jury was improperly instructed. The district court denied Chrysler's motions. 376 F. Supp. 546 (W.D.Pa.1974). Chrysler appealed; we vacate the judgment of the district court and remand for a new trial.

The record in this case consists of a lengthy stipulation, numerous exhibits, and oral testimony. Although later we shall refer to the evidence in more detail, we necessarily must set forth certain background material at the outset.

Prior to 1954, Chrysler followed the conventional practice of marketing its automobiles through independent franchised dealers for resale to the public. In 1954, Chrysler began to franchise dealers as well under the Dealer Enterprise (D.E.) Plan. Under this plan, Chrysler supplied up to 75 percent of the capital needed to finance a new dealership. The plan, similar to those previously adopted by General Motors and Ford, provides that the 25 percent owner becomes president and general manager of the dealership and that he ultimately may purchase Chrysler's entire interest in the dealership out of his share of profits. Until such time, however, Chrysler retains majority control at the shareholder and board of directors levels.

Plaintiff, a corporation with its principal place of business in Wilkinsburg, Pennsylvania, became a franchised Dodge dealership in 1955. Apart from its obligations under the franchise agreement, plaintiff operated independently of Chrysler. Plaintiff purchased Dodge automobiles and trucks from Chrysler under a 'Direct Dealer Agreement' which, although terminable by plaintiff on 30 days' notice, could be terminated by Chrysler only upon breach by plaintiff. Under this agreement, plaintiff agreed to sell at retail a certain quantity of automobiles within its sales area. This quantity, known as the 'minimum sales responsibility,' was calculated annually and reflected the percentage of new car registrations in plaintiff's sales area attributable to Dodge automobiles. Plaintiff stipulated that Chrysler's method of calculating the minimum sales responsibility was fair and reasonable.

Despite adoption of the D.E. Plan, Chrysler's national market penetration had dropped by 1961 from approximately 20 percent to 10.8 percent. In response to this sales decline, Chrysler in 1961 decided to encourage new franchises by establishing wholly owned dealerships operated by individuals unable to furnish even 25 percent of the necessary capital. Chrysler provided all of the initial capital for these so-called 'contractual' dealerships. The plan required the individual participants to place any available funds in an escrow account and add a portion of profits until the fund became sufficient to purchase 25 percent of Chrysler's interest. In 1964, Chrysler offered to lend the individual up to one-half the funds necessary to enable him to acquire the 25 percent and thereby become a D.E. dealer.1

Chrysler primarily supported factory dealerships in the following manner. First, Chrysler provided the initial capital infusion in the form of stock purchases and loans, thereby eliminating costs of capital for these dealerships. Second, Chrysler provided significant operating loss subsidies,2 thereby enabling factory dealerships, inter alia, to operate out of larger, more attractive showrooms3 and to spend substantially more money on advertising than did Coleman.4 Third, Chrysler provided factory dealerships with free services of key managerial employees at various times. The plaintiff, however, does not claim the existence of a significant difference in retail prices to the public between factory dealers and Coleman.

The total number of Dodge dealerships increased nationally from 2559 at the end of 1961 to 3138 in 1972, of which slightly more than 3000 were independently financed. The number of new independently financed Dodge dealerships ranged from 101 to 525 annually while the number of new Dodge factory dealerships never exceeded 83 in any year. The first factory dealership in Allegheny County, Boulevard Dodge, opened in 1961. The number of factory dealerships grew to six by 1966 when the Monroeville Dodge factory dealership was established. By 1968, however, the number of factory dealerships had suffered a decline to four. The number of independent Dodge dealers decreased from fifteen in 1961 to eleven in 1968. The aggregate number of independent Dodge dealers, both factory and independent, remained relatively constant between 1962 and 1968. After the factory dealer program was instituted, Chrysler's national market share increased to 16.25 percent in 1968. Dodge sales climbed from 3.43 percent of the national market in 1962 to 6.18 percent in 1968.

In Allegheny County, overall Chrysler sales increased from 11.28 percent of the new car market in 1962 to 18.01 percent in 1968. During the same period, local General Motors, Ford, and American Motors sales experienced a declining trend. The following table shows the respective shares of the Allegheny County new car market held by the major automobile manufacturers during the pertinent period. Overall Dodge sales in Allegheny County rose steadily from 3117 in 1962 to 6033 in 1968.

      NEW PASSENGER AUTOMOBILE REGISTRATIONS
       Gen.           Total    Only
Year  Motors  Ford   Chrysler  Dodge  A.M.C.  Other
1962  53.01   23.39   11.28    4.94    6.71    5.61
1963  51.42   22.07   13.96    6.47    6.84    5.71
1964  49.49   24.22   14.25    6.63    5.88    6.16
1965  51.86   22.15   15.26    6.56    4.32    6.41
1966  49.01   24.01   16.52    7.28    3.45    7.01
1967  50.50   20.73   17.60    7.51    2.87    8.30
1968  47.56   21.37   18.01    7.19    2.98   10.08

Plaintiff's closest factory competition came from the Boulevard Dodge factory dealership located five miles away. After Boulevard Dodge's first full year of sales activity in 1962, plaintiff's share of Dodge sales in Allegheny County dropped to 7.4 percent from 14.2 percent in 1961. When Monroeville Dodge was established in 1966, plaintiff accounted for slightly over five percent of the Dodge market. This figure declined to 3.8 percent in 1968. Boulevard Dodge, the first contractual dealership, cornered 39 percent of the local Dodge market in its first year. By 1967, all factory dealerships combined accounted for 54.2 percent of Dodge sales in Allegheny County.

However, in absolute terms, plaintiff's annual sales increased. Plaintiff sold 209 new cars in 1962, and then its sales rose approximately 50 percent to about 300 vehicles in 1963, 1964, and 1966. They then regressed to approximately 200 vehicles by 1968.

From 1955 through 1961, before the advent of Boulevard Dodge, plaintiff operated at a profit in only one year. During this period, plaintiff lost a total of approximately $74,000. During Boulevard Dodge's first year in business, plaintiff showed a loss of $16,779, but subsequently operated at a profit in 1963 ($6,350), 1964 ($2,963), and 1965 ($3,067). In 1966, the year that Monroeville Dodge was established, plaintiff lost $4,070. Plaintiff continued to lose money5 until finally it closed in 1969.

Plaintiff essentially claimed that Chrysler operated its factory dealerships in a manner designed to increase overall Dodge sales at the manufacturer level without regard for profit at the retail level. By providing large infusions of money and other benefits to its factory dealerships, Chrysler allegedly sought to recover its share of the total automobile market while slowly squeezing independent dealers out of business. This, plaintiff averred, was intended to achieve a Dodge sales structure composed primarily or entirely of several large Chrysler-owned dealerships.

Plaintiff asserted that Chrysler's conduct contravened both section one and section two of the Sherman Act.6 Under section one, plaintiff charged that Chrysler, in combination or conspiracy with its factory dealers, engaged in predatory practices which unreasonably7 restrained trade in the relevant market. Plaintiff alleged that Dodge vehicles were the relevant product market and that Allegheny County was the relevant geographical market.8 Under section two, plaintiff claimed that Chrysler unlawfully attempted to monopolize the same relevant market. The jury found violations of both sections of the Sherman Act.

Contending that its conduct was not unlawful, Chrysler advances a number of reasons on appeal why the district court should have set aside the jury's verdict. Chrysler also contends that certain trial errors require a new trial if judgment is not entered in its favor.

The jury found that Chrysler engaged in a combination or conspiracy between July 14, 1964, and July 14, 1969, which unreasonably restrained interstate trade or commerce in Dodge vehicles at the retail level in Allegheny County. Chrysler contends that the record contains insufficient evidence to support this finding.

Plaintiff did not claim in the district court, and does not contend on appeal, that Chrysler was precluded from opening factory dealerships in competition with its independent franchises. The restraint of trade, plaintiff contends, lies in the manner in which the factory dealerships were operated rather than the fact of their operation. Furthermore, plaintiff did not contend in the district court and does not urge on appeal that Chrysler's conduct constituted a per se violation of section one. Rather, plaintiff tried the case on the theory that Chrysler violated the 'rule of reason.' Standard Oil Company v. United States, 221 U.S. 1, 60, 31 S. Ct. 502, 55 L. Ed. 619 (1911).

The question posed to the jury was whether defendants combined or conspired to unreasonably restrain trade or commerce 'in Dodge vehicles at the retail level in Allegheny County.' The question we must decide is whether the facts adduced at trial support the jury's response. Admonishing ourselves that the record must be viewed in the light most favorable to the plaintiff, Andrews v. Dravo Corp., 406 F.2d 785, 789 (3d Cir. 1969), we turn to a closer examination of those facts.

It is undisputed that Chrysler financially subsidized factory dealerships. It provided the initial capital for these businesses, thereby eliminating cost of capital. Chrysler also provided significant operating loss subsidies (see n.2 supra). These funds enabled the factory dealerships to spend significantly greater amounts of money than could Coleman on larger, more attractive showrooms and on advertising. See n.3 and 4 supra.

Chrysler contends that the advertising campaign was intended to increase overall sales of Dodge vehicles in Allegheny County. It is fair to state this increase could have been achieved by spreading advertising loss subsidies over all dealers, both independent and factory. Subsidization of Boulevard Dodge's advertising losses in substantial sums cannot be justified over an eight-year period on the basis that Boulevard Dodge was a new entrant in the Dodge retail market in need of assistance in order to survive--in its first year of operation, Boulevard Dodge had cornered 39 percent of the Dodge market.

It is also notewarthy that when plaintiff went out of business, one of its close competitors, Boulevard Dodge, decreased its advertising expenditures. Between 1969 and 1971 (the first year it made a profit since 1965) Boulevard Dodge reduced its advertising expenditure from $59,403 to $29,861. Expenditures an new car advertising were reduced from $37,679 to $11,856.

There was some testimony that defendants discriminated in favor of the factory dealerships and against the private dealerships in the release of new automobiles. In 1962 when defendants for the first time marketed a full size car which had wide public acceptance, all the Dodge dealers advertised in advance about the release of the car, but only the factory dealership at Boulevard Dodge received immediate delivery of the car.

There was also evidence that Chrysler paid the salaries of key managerial employees of factory dealerships at various times; that Chrysler paid the salaries of accountants and the two Chrysler board members of factory dealerships; and that Chrysler provided free furniture, tools, and equipment to the factory dealership at Cloverleaf Dodge.

When plaintiff lost its used car lot to the redevelopment authority, it arranged to move to a new location four blocks away. Plaintiff's franchise agreement required it to seek permission from Chrysler, which was refused.

Vernon Staley, former president of Boulevard Dodge, testified that he informed Chrysler's regional manager Harris after several months of operation of his concern for the amount of losses his company was sustaining. Staley stated that Harris told him not to worry: 'Just roll those cars out. We will make our money at the front end, and, if you can get this somewhere close to a breakeven we will be satisfied.' Staley further testified that when Harris was attempting to engage him for the Dodge franchise, Harris indicated that Chrysler intended 'to reduce the number of Dodge dealers in Allegheny County from 16 to 11, under their new marketing program.'9

From this evidence, the jury could reasonably have concluded that defendant combined and conspired with managers of factory dealerships. It is well settled that, although a party cannot combine or conspire with itself, a parent corporation, such as the defendants here, can combine or conspire with its wholly owned subsidiaries. Perma Life Mufflers Inc. v. International Parts Corp., 392 U.S. 134, 141-42, 88 S. Ct. 1981, 20 L. Ed. 2d 982 (1968); Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, 340 U.S. 211, 215, 71 S. Ct. 259, 95 L. Ed. 219 (1951); Minnesota Bearing Co. v. White Motor Corp., 470 F.2d 1323, 1328--29 (8th Cir. 1973). Explicit agreement is not a necessary part of a Sherman Act conspiracy. United States v. General Motors Corp., 384 U.S. 127, 142--43, 86 S. Ct. 1321, 16 L. Ed. 2d 415 (1966). The conspiracy or agreement may be proved entirely by circumstantial evidence such as the conduct of the parties. Industrial Building Materials, Inc. v. Interchemical Corp., 437 F.2d 1336, 1343 (9th Cir. 1970); Flintkote Co. v. Lysfjord, 246 F.2d 368, 374 (9th Cir. 1957), cert. denied, 355 U.S. 835, 78 S. Ct. 54, 2 L. Ed. 2d 46 (1957). 'Where 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, the conclusion that a conspiracy is established is justified.' American Motor Inns, Inc. v. Holiday Inns, Inc., 521 F.2d 1230 at 1243 (3d Cir. 1975), quoting American Tobacco Co. v. United States, 328 U.S. 781, 810, 66 S. Ct. 1125, 90 L. Ed. 1575 (1946).

Although the evidence is weak, the jury could have found further that Chrysler's actions were unfairly competitive and that their effect was to force plaintiff out of business.10 We note that in the first years of the contractual dealership program, plaintiff's sales increased and it showed its first profits. However, from 1966 on it suffered losses until it went out of business in 1969. Plaintiff's expert, Dr. Rubin E. Slesinger, professor of Economics at Pittsburgh, testified that large infusions of funds to factory dealers, as subsidies to offset losses, and the expenditure for advertising of two to four times the amount spent by independent dealers, and other preferential treatment of factory dealers, would tend to monopolize the Dodge market in Allegheny County. The plain implication of his testimony is that factory dealers would take sales away from independent dealers by these practices.

The jury may also have determined that Chrysler deliberately took advantage of Coleman's financial situation. Plaintiff was in breach of its Direct Dealer Agreement because it had not met its minimum sales requirement between 1962 and 1969, except in 1963. Consequently, Chrysler had the right under the agreement to terminate plaintiff's franchise. Chrysler apparently chose to let plaintiff remain in existence and contribute to overall sales of Dodge vehicles. While plaintiff was losing money,11 Chrysler was subsidizing factory dealer losses. When plaintiff went out of business, Boulevard Dodge lowered its advertising expenses and, after 1971, operated at a profit.12 Thus, the jury could have concluded that Chrysler had accomplished its purposes of increasing Dodge sales and consolidating retail sales in the hands of several strong Chrysler-controlled dealerships.

The question thus posed is whether the above findings are sufficient to establish an unreasonable restraint of trade under section one of the Sherman Act. The problem is one of restraint vel non in intrabrand competition.

Although appellate courts have considered various restraints in an intrabrand context, see, e.g., United States v. Arnold, Schwinn & Co., 388 U.S. 365, 87 S. Ct. 1856, 18 L. Ed. 2d 1249 (1968) (territorial restrictions); Copper Liquor, Inc. v. Adolph Coors Co., 506 F.2d 934 (5th Cir. 1975) (territorial restrictions); Ark Dental Supply Company v. Cavitron Corporation, 461 F.2d 1093 (3d Cir. 1972) (refusal to deal); Joseph E. Seagram & Sons, Inc. v. Hawaiian Oke & Liquors, Ltd., 461 F.2d 71 (9th Cir. 1969), cert. denied, 396 U.S. 1062, 90 S. Ct. 752, 24 L. Ed. 2d 755 (1970) (refusal to deal), the precise question presented here--reduction in intrabrand competition by a tendency toward a vertically integrated distribution system--does not appear to have been addressed by any other appellate court as yet.13

Relying on Klor's Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 79 S. Ct. 705, 3 L. Ed. 2d 741 (1959), and language in Radovich v. National Football League, 352 U.S. 445, 453--54, 77 S. Ct. 390, 1 L. Ed. 2d 456 (1957); Radiant Burners, Inc. v. Peoples Gas Light & Coke Co., 364 U.S. 656, 660, 81 S. Ct. 365, 5 L. Ed. 2d 358 (1961); In re McConnell, 370 U.S. 230, 231, 82 S. Ct. 1288, 8 L. Ed. 2d 434 (1962); and Simpson v. Union Oil Co., 377 U.S. 13, 17, 84 S. Ct. 1051, 12 L. Ed. 2d 98 (1964), the plaintiff urges that no public injury is necessary to a finding of restraint of trade. Plaintiff would apparently have us hold that the mere fact that Coleman's business was damaged is sufficient to constitute a restraint of trade. This we decline to do. Insofar as plaintiff reads the above cases to stand for the proposition that one need not be a member of the consuming public to derive the protection of the antitrust laws and that an antitrust plaintiff need not prove public injury as a material element in this case, we are in full agreement. However, we do not understand the Supreme Court to have held that any damage to an individual's business can constitute a restraint of trade. That an injury to competition in the marketplace may be inferred from the evidence is a requisite to finding a restraint of trade. The main purpose of the Sherman Act is to protect the public from injury and the individual right of action is a tool to effectuate that purpose. 'Our inquiry is whether, . . . the effect upon competition in the marketplace is substantially adverse.' United States v. Arnold, Schwinn & Co., 388 U.S. 365, 375, 87 S. Ct. 1856, 1863, 18 L. Ed. 2d 1249 (1967). (Emphasis supplied.)14

Another area of competition between automobile dealers is service. If all independent dealers were eliminated, competition in price of parts and service and in quality of service could also be eliminated. Although the only evidence in the record concerning service is testimony by Coleman that he was proud of his dealership's service record, it is common knowledge that quality of service is vital to a successful automobile dealership.

To this analysis the objection might be raised that Chrysler has the right to structure its distribution system as it wishes and the law should not prevent it from achieving a totally vertically integrated distribution structure. Although that proposition may be open to question,15 it is at least subject to the qualification that a lawful end achieved by unlawful means is not protected from the antitrust laws. See Connell Co. v. Plumbers & Steamfitters, 421 U.S. 616, 625, 95 S. Ct. 1830, 44 L. Ed. 2d 418 (1975); Siegal v. Chicken Delight, Inc., 448 F.2d 43, 51 (9th Cir. 1971), cert. denied, 405 U.S. 955, 92 S. Ct. 1172, 31 L. Ed. 2d 232 (1972). While Chrysler's ability to cease doing business with franchisees may generally be restricted only by contractual provisions, the means it chose to accomplish this end here have an anticompetitive effect. If Chrysler had simply ceased doing business with Coleman, Coleman might have been able to seek a franchise from another manufacturer and become an interbrand competitor. However, the continuation of predatory practices for a number of years in the face of Coleman's substantial losses could have so severely damaged Coleman's financial ability that he could not reenter the market as an interbrand competitor. A combination of distributors, which through unfair practices eliminates a competitor and leaves it in such a condition that it lacks the ability to continue business as an interbrand competitor, has an adverse effect on the marketplace. Cf. Industrial Building Materials, Inc. v. Interchemical Corp., supra, 437 F.2d at 1342. Accordingly, Chrysler's activities might be held to be an unreasonable restraint of trade.

Chrysler argues that the effect of such a holding is to saddle it with inefficient private capital dealers who have lost the economic battle of the marketplace, and to limit its ability to compete with other automobile manufacturers. The answer is that Chrysler had permissible options available to it. If Coleman were inefficient, Chrysler had the right under the contract to terminate the franchise when Coleman failed to meet the minimum sales requirement without waiting until Coleman had lost all of its financial resources. In addition, Chrysler could have avoided the predatory practices. If Chrysler deemed additional dealerships necessary for effective competition with other manufacturers, it was free to establish them on nondiscriminatory terms.

Chrysler contends that there is insufficient evidence to support the jury's finding that Chrysler attempted to monopolize interstate commerce in Dodge vehicles at the retail level in Allegheny County. It asserts, therefore, that the district court should have granted its motion for judgment notwithstanding the verdict. We disagree.16

An essential element of a section two attempt to monopolize violation is that the actor have a specific intent to monopolize the relevant market. Times-Picayune v. United States, 345 U.S. 594, 626, 73 S. Ct. 872, 97 L. Ed. 1277 (1953); Chisholm Brothers Farm Equipment Co. v. International Harvester Co., 498 F.2d 1137, 1144 (9th Cir.), cert. denied, 419 U.S. 1023, 95 S. Ct. 500, 42 L. Ed. 2d 298 (1974); Gold Fuel Service Inc. v. Esso Standard Oil Co., 306 F.2d 61, 64 (3d Cir. 1962), cert. denied, 371 U.S. 951, 83 S. Ct. 506, 9 L. Ed. 2d 500 (1963). Another essential element of the offense is that the actor have sufficient market power to come dangerously close to success. Swift & Company v. United States, 196 U.S. 375, 396, 25 S. Ct. 276, 49 L. Ed. 518 (1905); Agrashell, Inc. v. Hammons Products Company, 479 F.2d 269, 284 (8th Cir.), cert. denied, 414 U.S. 1022, 94 S. Ct. 445, 38 L. Ed. 2d 313 (1973); Cornwell Quality Tools Co. v. C.T.S. Company, 446 F.2d 825, 832 (9th Cir. 1971), cert. denied, 404 U.S. 1049, 92 S. Ct. 716, 30 L. Ed. 2d 740 (1972).

In a case such as this one, definition of the relevant market is also critical. See Walker, Inc. v. Food Machinery, 382 U.S. 172, 177, 86 S. Ct. 347, 15 L. Ed. 2d 247 (1965); George R. Whitten, Jr., Inc. v. Paddock Pool Builders, Inc., 508 F.2d 547, 550 (1st Cir. 1974); Acme Precision Products, Inc. v. American Alloys Corp., 484 F.2d 1237, 1240 (8th Cir. 1973); Bushie v. Stenocord Corp., 460 F.2d 116, 121 (9th Cir. 1972); Bernard Food Industries, Inc. v. Dietene Co., 415 F.2d 1279, 1284 (9th Cir. 1972).17 If the relevant market is, as defendant urges, all automobiles in Allegheny County, Chrysler could not monopolize that market by driving independent dealers out of business; competition from independent and company owned dealers of other manufacturers would severely limit Chrysler's ability to determine prices. Plaintiff's expert witness, Dr. Slesinger, acknowledged as much. Moreover, there is no evidence in this record that Chrysler intended to monopolize the market of all automobiles in Allegheny County. Since Chrysler's conduct, even if it resulted in eliminating intrabrand competition from independent Dodge dealers, could not achieve a monopoly, Chrysler cannot be held liable for having attempted to monopolize. United States v. Berrigan, 482 F.2d 171, 190 (3d Cir. 1973); Hiland Dairy, Inc. v. Kroger Company, 402 F.2d 968, 971 (8th Cir. 1968), cert. denied, 395 U.S. 961, 89 S. Ct. 2096, 23 L. Ed. 2d 748 (1969).

If, however, the relevant market is, as plaintiff urges, retail trade in Dodge automobiles in Allegheny County, Chrysler could monopolize that market by driving independent Dodge dealers out of business. From the evidence of predatory practices detailed supra, the jury could infer the intent and probability of success necessary to establish a section two violation.

In this case, the trial judge first instructed the jury that the relevant market was Dodge vehicles in Allegheny County. Upon objection by the defense, he withdrew this instruction and directed the jury to determine the relevant product market.18 However, the interrogatory submitted to the jury asked only if Chrysler had attempted to monopolize retail trade in Dodge automobiles in Allegheny County. Since no objection was made to this interrogatory, it would appear that the parties waived jury trial of the relevant product market issue and that the district judge implicitly found that the relevant market was retail trade in Dodge vehicles in Allegheny County. Federal Rules of Civil Procedure 49(a).

Based as it is on such a finding, the verdict on this count cannot stand. The relevant product market is composed of products 'that have reasonable interchangeability for the purposes for which they are produced--price, use and qualities considered.' United States v. DuPont & Co., 351 U.S. 377, 404, 76 S. Ct. 994, 1012, 100 L. Ed. 1264 (1956). The parties stipulated that the line of commerce involved in this case is Dodge automobiles and trucks, but further stipulated that 'Dodge automobiles compete with all other automobiles except Cadillac, Lincoln and Chrysler Imperials.' Plaintiff's expert testified: 'You could have a relevant market for Dodges. You could also have a relevant market for automobiles.' He further testified that the 'relevant market includes those items which are comparatively reasonably (sic) substitutes for each other,' but gave no testimony on substitutability. We find his other testimony on the market issue confusing. From evidence such as this, no reasoned evaluation of the interchangeability of Dodges and other automobiles, and hence of the relevant product market, can be made.

We decline, however, to enter judgment notwithstanding the verdict on the issue of liability under section two. Since the record on this issue is confused and there is evidence from which the requisite intent and probability of success could be inferred if the product market issue were determined in plaintiff's favor, we deem this to be a case where a new trial on the issue of liability on this count would be 'just under the circumstances.' 28 U.S.C. § 2106.

At the trial of this case, plaintiff called Sam Liberto, the prior owner of Mt. Lebanon Motors, a former Allegheny County Dodge dealership. Liberto had been the plaintiff in Mt. Lebanon Motors, Inc. v. Chrysler Corp., 283 F. Supp. 453 (W.D.Pa.1968), in which the allegations were identical to those in the instant case. In Mt. Lebanon, the jury had answered six interrogatories, finding that Chrysler had not contracted, combined, or conspired to restrain trade, and had not acted in bad faith in terminating plaintiff's franchise, but had attempted and conspired to monopolize trade. However, the jury further found that the plaintiff had sustained no damages. Thus, judgment was entered for defendant; this court affirmed. Mt. Lebanon Motors, Inc. v. Chrysler Corp., 417 F.2d 622 (3d Cir. 1969).

In the instant case, to establish Liberto's bias, defendant on cross-examination asked Liberto whether he had earlier brought suit against Chrysler. When Liberto replied in the affirmative, defense counsel then asked whether the suit had not resulted in Liberto's receiving no money. Liberto again responded in the affirmative. Plaintiff did not object to this line of questioning.

On redirect examination, plaintiff's counsel asked Liberto whether in the Mt. Lebanon case 'a jury in this Court found Chrysler violated (the federal antitrust) laws.' Defendant objected, but before the court could rule, counsel re-asked the question and Liberto answered, 'They found them guilty of violation of the Sherman Anti-Trust Act.' After further questions, objections, and discussion, plaintiff's counsel was allowed to read the Mt. Lebanon interrogatories with answers to the jury. The court then explained to the jury that the testimony relating to the Mt. Lebanon Motors case was admitted solely for the purpose of evaluating Liberto's testimony.

Plaintiff's counsel was later permitted to examine a second witness concerning the Mt. Lebanon litigation and the jury's finding against Chrysler in that case. At sidebar, plaintiff's counsel stated that his purpose was to establish 'a flagrant disregard of the law on the part of Chrysler Motors Corporation. . . . I want to show that they are aware of what they are doing but it doesn't make a bit of difference to them.' Later, plaintiff's counsel made substantially this argument to the jury in his closing remarks, referring to the fact that 'in 1968 a jury in the Mount Lebanon Motorscase found that Chrysler had violated the anti-trust laws.'

Finally, the court instructed the jury that they could consider the Mt. Lebanon verdict:

in connection with, one, the believability, the credibility of Sam Liberto and, further, whether or not Chrysler carried on any acts after any kind of notice to them with respect to the whole problem here involved . . ..

Defendants have consistently objected to plaintiff's use of the Mt. Lebanon verdict and have moved for a new trial on this ground. They claim that their introduction of the verdict in the cross-examination of Liberto was proper to show bias, but that plaintiff's subsequent reference to the finding against Chrysler in redirect examination of Liberto, in examination of the second witness, and in closing argument was admitted improperly and was unduly prejudicial.

Plaintiff counters that defendant's cross-examination of Liberto created the impression that the Mt. Lebanon jury had absolved the defendant of any wrongdoing so that plaintiff was compelled to refer to the verdict on redirect examination in order to correct that impression.

We first note that these contentions do not present us with a collateral estoppel issue. It was never contended by either party that the prior litigation was conclusive of the legality of Chrysler's acts. Rather, the reference to the Mt. Lebanon case by the defendant was to show Liberto's bias. The reference by the plaintiff to the jury verdict was to show notice to Chrysler and intent.

We consider the defendant's reference to the case to be an unobjectionable attempt to show Liberto's bias. Plaintiff, on redirect examination, however, did not limit himself to an effort to rebut the implication of bias. We consider this redirect examination, the references to the prior verdict in examining another witness, and arguing the verdict to the jury to be prejudicial and to require a new trial.19 See, e.g., Garris v. McClain, 399 Pa. 261, 160 A.2d 398, 400 (1960); Annotation 15 A.L.R. 3d 1101 (1967).

Even if the jury had been adequately informed of the issues and evidence in Mt. Lebanon, we would disapprove of the introduction of that verdict. A jury is likely to give a prior verdict against the same defendant more weight than it warrants. The admission of a prior verdict creates the possibility that the jury will defer to the earlier result and thus will, effectively, decide a case on evidence not before it. We will not speculate about the role the Mt. Lebanon verdict played in this jury's deliberations, especially since there is another reason why we must grant defendants a new trial. It is sufficient that knowledge of the prior verdict could have influenced the jury's decision on either count so that the defendant was unduly prejudiced.

As this court long ago stated in a similar context:

(T)he statement made by counsel of the plaintiff in his argument to the jury as to what had been done by other juries in former trials of this case . . . are in a federal court deemed so improper as to warrant opposing counsel to request, and courts of their own motion to direct, the withdrawal of a juror and the continuance of a cause . . ..

McKibben v. Philadelphia & R. Ry. Co., 251 F. 577, 578 (3d Cir. 1918).

Plaintiff undertook to prove the amount of its damages through expert testimony based on voluminous financial records and sales data previously introduced into evidence. Two of plaintiff's experts, members of a research management consulting firm, first projected the volume of sales which plaintiff might have been expected to achieve during the damage period20 based on its market share in the years 1960 and 1961. According to one of the experts, these years were chosen because they 'were the only two years during which Coleman Motors was no longer devoting any efforts to selling Plymouths,21 and were the only two years available prior to the introduction of the first factory outlet.' The projection offered by the experts represented the percentage of Dodge vehicles actually sold in Allegheny County which plaintiff would have been expected to sell had no factory dealerships been established. In addition to a projection of 8.25 percent of the Dodge market, plaintiff's experts formulated a 'pessimistic' projection of 7 percent and an 'optimistic' projection of 9.5 percent.

The experts next developed a formula for plaintiff's annual net earnings before taxes, expressed as gross revenue less fixed costs, and less variable costs pegged to the number of units sold. This formula was derived on the basis of information contained in plaintiff's tax returns for the years 1960 to 1966. It took into account the cost of goods sold, wages and salaries,22 and sales and overhead expenses.

To arrive at the loss in net earnings before taxes for a given year, the experts multiplied the number of Dodge vehicles actually sold through all dealers in Allegheny County by plaintiff's projected market share. This product, which represented the projected number of vehicles plaintiff presumably would have sold, then was plugged into the formula previously derived for net earnings before taxes. The experts calculated the loss in net earnings before taxes by subtracting from this projected figure plaintiff's actual net earnings before taxes.

In addition to a loss in net earnings, plaintiff sought damages for the destruction of its business. Plaintiff's experts testified that the value of the ongoing business could be estimated at ten times earnings after taxes. The annual earnings before taxes previously calculated for the years 1965 to 1969 were reduced by the amount of taxes payable at 1973 rates.23 The average of the figures thus produced then was multiplied by ten. The values of plaintiff's business thus arrived at ranged from $404,000 using the 'pessimistic' sales projection to $662,000 using the 'optimistic' sales projection.24

As can be seen from the above summary, the foundation upon which all of plaintiff's general damage evidence rested was the sales projection based on the years 1960 and 1961. This projection, however, rather than representing only the sales which plaintiff would have made but for the allegedly unlawful manner in which Chrysler operated its factory dealerships, represented as well the sales which plaintiff would have made but for the mere existence of the factory dealerships. Plaintiff did not introduce any evidence tending to show the amount by which the sales projection should have been reduced to compensate for sales lost as a result of the factory dealerships' lawful competition. The experts' projection of the number of Dodge sales plaintiff would have made, had there been no factory dealers, is predicated on an expanded market in which the factory dealers played a vital role. To project plaintiff's percentage of sales in that market, the experts compared the Dodge market in Allegheny County before the advent of factory dealerships, and at a time when Dodge sales were severely depressed, with a market revitalized by the presence and sales of factory dealers. Moreover, the experts' projection ignored the deteriorating character of the neighborhood in which Coleman was located. We believe that a computation based upon such a comparison is unacceptable and inaccurate. Plaintiff's experts did point to their 'pessimistic' projection of 7 percent as a method of taking the mere existence of the factory dealers into account. They were unable, however, to offer any rational explanation for this reduction of 1.25 percent from their projection of 8.25 percent.

In considering plaintiff's damage evidence, we must heed the Supreme Court's admonition toobserve the practical limits of the burden of proof which may be demanded of a treble-damage plaintiff who seeks recovery for injuries from a partial or total exclusion from a market; damage issues in these cases are rarely susceptible of the kind of concrete, detailed proof of injury which is available in other contexts. The Court has repeatedly held that in the absence of more precise proof, the factfinder may 'conclude as a matter of just and reasonable inference from the proof of defendants' wrongful acts and their tendency to injure plaintiffs' business, and from the evidence of the decline in prices, profits and values, not shown to be attributable to other causes, that defendants' wrongful acts had caused damage to the plaintiffs.

Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123--24, 89 S. Ct. 1562, 1576, 23 L. Ed. 2d 129 (1969), quoting Bigelow v. RKO Pictures, Inc., 327 U.S. 251, 264, 66 S. Ct. 574, 579, 90 L. Ed. 652 (1946).

We do not believe that the jury in the instant case could have determined plaintiff's damages by 'just and reasonable inference' from the evidence presented. Plaintiff's own experts conceded that the damage figures they tendered were attributable at least in part to the lawful competition of Chrysler's factory dealerships. On the evidence adduced, the jury rationally could not have reduced these figures to reflect only losses due to unlawful competition. No evidence was adduced, for example, to show the effect which entry of competitors similar in size and location to Chrysler's factory dealers might be expected to have on the market share of an enterprise similar to plaintiff's.25

We of course do not suggest that an antitrust plaintiff may not rely upon projections in proving damages. Use of projections is entirely acceptable as long as the projections lead to a reasonable estimate of damages caused by the alleged violations.

The rule which precludes the recovery of uncertain damages applies to such as are not the certain result of the wrong, not to those damages which are definitely attributable to the wrong and only uncertain in respect of their amount.

Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 562, 51 S. Ct. 248, 250, 75 L. Ed. 544 (1931). In the present case, however, plaintiff's projections fail to account for significant factors bearing upon its diminished market share.26 The damage figures advanced by plaintiff's experts may be substantially attributable to lawful competition. In the absence of any guidance in the record, we cannot permit a jury to speculate concerning the amount of losses resulting from unlawful, as opposed to lawful, competition.

To summarize, we have found several basic errors which require that we vacate the judgment against Chrysler: (1) evidence of a prior verdict against Chrysler was erroneously admitted; (2) the critical market definition issue was presented to the jury in a confused manner; and (3) the testimony on damages was based on an erroneous assumption. Accordingly, we vacate the judgment and remand to the district court with directions to grant defendant's motion for a new trial.