This is an antitrust suit for treble damages, brought under 15 U.S.C. § 15 . The case was tried to a jury. Plaintiff recovered a judgment against all defendants for $65,000, trebled, plus $50,000 attorneys’ fees and costs, a total of $246,-938.54. Defendants appeal. We reverse.

While the complaint charged that the defendants had violated sections 1 and 2 of the Sherman Act ( 15 U.S.C. §§ 1 , 2), the section 2 charge was dropped. Thus the judgment rests upon a finding that the defendants had formed “a contract, combination in the form of trust or other *73 wise, or conspiracy, in restraint of trade or commerce.” ( 15 U.S.C. § 1 ).

1. The basic facts.

Plaintiff Hawaiian Oke and Liquors, Ltd. is a corporation whose business was distributing liquors at wholesale in the State of Hawaii. It claims that the defendants combined or conspired to, and did, put it out of business.

There are three sets of defendants. (1) Joseph E. Seagram & Sons, Inc. a corporation, (Seagram) is a large distiller, manufacturing alcoholic beverages. The House of Seagram, Inc., a corporation (House of Seagram), is a wholly owned subsidiary of Seagram, and distributes Seagram’s products. Within its corporate structure there are certain administrative divisions, each of which is called a “company”: Among them are Calvért Distillers Company (Calvert), Four Roses Distillers Company (Four Roses) and Frankfort Distillers Company (Frankfort). (2) McKesson & Robbins, Inc. (McKesson) is a corporation which, among other things, conducts a wholesale liquor distributing business in many parts of the United States, including Hawaii. (3) Barton Distilling Company, a corporation, (Barton) is a manufacturer of alcoholic beverages. Barton Western Distilling Co., a corporation, (Barton Western) is a wholly owned subsidiary of Barton.

In June 1965, plaintiff was the sole distributor in Hawaii of all Calvert products, of all “Four Roses” brand Four Roses products, and of two of Frankfort’s products. There were three separate written contracts between plaintiff and House of Seagram covering these products, one for each division. Each was for one year and was to expire on July 31, 1965. Plaintiff was also sole distributor of all but one of Barton’s products, under an oral agreement terminable by either party on reasonable notice. The “Kessler” brand Four Roses product, the one Frankfort product and the one Barton product not included in these arrangements were distributed by McKesson.

For reasons that are disputed, Calvert’s “president,” Murphy, proposed to Cotier, McKesson’s vice-president, that McKesson become Calvert’s distributor in Hawaii. This was in May 1965. Mc-Kesson was already distributing Seagram’s 7 Crown, the leading blended whiskey in Hawaii, and V.O., both distributed by the Seagram Company, another division of House of Seagram. Murphy therefore asked McKesson to establish a separate sales organization, a “separate house”, for the Calvert line, thinking that otherwise the Calvert line would not get adequate attention from McKesson’s salesmen. McKesson also wanted to establish a “second house” with Calvert the primary line, if it were to take the Calvert line. There was another meeting at Calvert’s New York City office on June 3, at which were present Cotier, Maloney and Kauhane of Mc-Kesson, Murphy and Fleischman of Calvert, and Yogman of Seagram. The Mc-Kesson people favored the Seagram group’s proposal.

All parties knew what appellee refers to as “an economic fact well known to all of them,” that it would be necessary for the new separate house to have additional lines. At the meeting this was mentioned, as was McKesson’s desire to get other Seagram lines!

The need for additional lines includes the fact that a distributor needs both some name brand, high profit lines, known as “Class A,” and some lower priced “Class B” lines. Calvert’s line is Class A. It was therefore necessary for McKesson, when it was approached by Calvert, to look for one or more Class B lines. Barton’s line is Class B. Frankfort and Four Roses each had both Class A and Class B lines.

McKesson was already doing considerable business with Barton on the mainland. It was interested in becoming Barton’s distributor in Hawaii. It had broached the subject with Barton in April. It again approached Barton after the June 3 meeting.

About June 7, Maloney, who was Mc-Kesson’s West Coast representative, had *74 returned to California, and he called Friedman of Barton on the telephone. Later in June Friedman and Weinstock of Barton met with Maloney of McKesson, and Maloney solicited the distributorship of Barton products in Hawaii for McKesson’s proposed new sales force. Weinstock expressed interest. On about June 15 Maloney and Cotier of McKesson decided to go ahead with the new distributorship. There is evidence from which the jury could infer that by that time Barton had agreed with McKesson to transfer its line to the new McKesson house, and that the three House of Seagram divisions had also agreed with McKesson to transfer their lines to the new McKesson house.

Plaintiff urges that the evidence supports further inferences as follows: McKesson wanted the complete deal, and would not go ahead without it. It was the prime mover in getting Barton, Four Roses and Frankfort to go along, and those three, with Calvert, all went into the new arrangement together, knowing that the deal was conditioned on the participation of each.

There were no communications between anyone representing any Seagram corporation or division and any representative of the Barton corporations. Neither at the trial nor here does plaintiff assert that the Seagram and Barton corporations made an express agreement with each other. What it does say, in substance, is that, regardless of whether one or more of the three House of Seagram divisions agreed to make McKesson its distributor before Barton did, or whether Barton was the first to agree, each went ahead knowing that, unless the •others also did so, the new house would not be established.

On June 25, Calvert told plaintiff that it would not renew its contract. On June 28, Four Roses did the same, as did Frankfort on July 2. Barton's Friedman came to Hawaii on July 5 and visited plaintiff. The same day, Friedman visited McKesson’s Kauhane. Next day, Friedman called his superior, Weinstock. Friedman then notified plaintiff that Barton was going to switch to McKesson, the change to be effective August 31. McKesson called its new distributorship and sales force “Portside.”

Each of the House of Seagram divisions and Barton knew that plaintiff was the distributor of its own line and of the others’ lines. They therefore knew that the change to Portside would deprive plaintiff of the major portion of its business. Plaintiff urges that the jury could infer (1) that they knew that the change would destroy plaintiff’s business and (2) that it did destroy plaintiff’s business.

Other pertinent facts will be stated as we consider appellant’s arguments.

2. No unreasonable restraint was proved.

In this part of this opinion, when we say “Seagram” we refer to the Seagram group. Plaintiff asserts that the agreement between Seagram, McKesson and Barton was intended to and did take the distributorship of Seagram and Barton products from plaintiff. This, it says, is a group boycott, and a group boycott is unlawful per se under section 1 of the Sherman Act. It further urges that, in such a case as this, the defendant’s business motives are immaterial. That was its position in the trial court. It asked the trial court to .so instruct the jury, and the court did so.

, Plaintiff did not assert that defendant’s primary purpose was either to put plaintiff out of business, or to compel plaintiff to conform its conduct to any anti-competitive objectives of the defendants. It offered no evidence to support either of such possible theories; it offered no instructions to support either of such possible theories.

Appellants argue that the Seagram and Barton decisions to transfer the business to McKesson were made independently, because Seagram and Barton each believed that McKesson would do a better job for it. They further argue that, even if there was an understanding between Seagram, Barton, and McKesson, that they would do what they did, such under *75 standing was not a group boycott, and so did not violate section 1. In the following discussion, we assume that such an understanding did exist, the jury being presumed to have so found under the court’s instructions.

The court instructed the jury as follows:

The essence of these instructions is that any agreement between two or more suppliers who have been selling to or through distributor A to transfer their business to distributor B, who is also *76 a party to the agreement, is a per se violation of section 1. We think that this is an overstatement of the rule.

Not every agreement is per se “in restraint of trade” within the meaning of section 1. See White Motor Co. v. United States, 1963, 372 U.S. 253, 261 , 83 S.Ct. 696 , 9 L.Ed.2d 738 . Thus, it is well settled that it is not a per se violation of the antitrust laws for a manufacturer or supplier to agree with a distributor to give him an exclusive franchise, even if this means cutting off another distributor. See United States v. Arnold, Schwinn & Co., 1967, 388 U.S. 365, 376 , 87 S.Ct. 1856 , 18 L.Ed.2d 1249 ; Lawlor v. National Screen Serv. Corp., 1957, 352 U.S. 992 , 77 S.Ct. 526 , 1 L.Ed.2d 540 [and see 3 Cir., 1959, 270 F.2d 146 ,152, and 1956, 238 F.2d 59 , 65]; United States v. Columbia Steel Co., 1948, 334 U.S. 495, 522, 524-525 , 68 S.Ct. 1107 , 92 L.Ed. 1533 ; Scanlan v. Anheuser-Busch, Inc., 9 Cir., 1968, 388 F.2d 918, 921 ; Walker Distrib. Co. v. Lucky Lager Brewing Co., 9 Cir., 1963, 323 F.2d 1, 7 ; Ace Beer Distribs., Inc. v. Kohn, Inc., 6 Cir., 1963, 318 F.2d 283 , 286-287; Packard Motor Car Co. v. Webster Motor Car Co., 1957, 100 U.S.App.D.C. 161 , 243 F.2d 418 ; Interborough News Co. v. Curtis Publishing Co., 2 Cir., 1955, 225 F.2d 289, 293 ; Naifeh v. Ronson Art Metal Works, 10 Cir., 1954, 218 F.2d 202, 206-207 ; Bascum Launder Corp. v. Telecoin Corp., 2 Cir., 1953, 204 F.2d 331, 334-335 ; Fargo Glass & Paint Co. v. Globe American Corp., 7 Cir., 1953, 201 F.2d 534, 539-540 ; Schwing Motor Co. v. Hu'dson Sales Corp., D. Md., 1956, 138 F.Supp. 899 , aff’d per curiam, 4 Cir., 1956, 239 F.2d 176 ; United States v. Bausec & Lomb Optical Co., S.D.N.Y., 1942, 45 F.Supp. 387, 398-399 , aff’d by an equally divided court, 1944, 321 U.S. 707, 719 , 64 S.Ct. 805 , 88 L.Ed. 1024 . Cf. Albrecht v. Herald Co., 1968, 390 U.S. 145, 154 , 88 S.Ct. 869 , 19 L.Ed.2d 998 , (concurring opinion). See generally Report of the Attorney General’s National Committee to Study the Antitrust Laws 27-29 (1955); Fulda, Individual Refusals to Deal: When Does Single-Firm Conduct Become Vertical Restraint?, 30 Law & Contemp.Prob. 590, 597-98 (1965); McLaren, Territorial and Customer Restrictions, Consignments, Suggested Resale Prices & Refusals to Deal, 37 Antitrust L.J. 137 (1968); Robinson, Providing for Orderly Marketing of Goods, 15 ABA Antitrust Sections 282, 286-88 (1959); Turner, The Definition of Agreement under the Sherman Act: Conscious Parallelism & Refusals to Deal, 75 Harv.L.Rev. 665, 703-05 (1962).

It is no doubt true that a manufacturer or supplier can do many things independently which he may not combine with others to accomplish. See e.g. United States v. Parke, Davis & Co., 1960, 362 U.S. 29 , 80 S.Ct. 503 , 4 L.Ed.2d 505 ; Associated Press v. United States, 1945, 326 U.S. 1, 14-15 , 65 S.Ct. 1416 , 89 L.Ed. 2013 ; United States v. Bausch & Lomb Optical Co., 1944, 321 U.S. 707, 722 , 64 S.Ct. 805 , 88 L.Ed. 1024 . But the mere fact of combination or “conspiracy” does not necessarily result in per se liability.

We turn, then, to the group boycott cases on which plaintiff relies. Such boycotts have been .held to be illegal per se under section 1 because they are “naked restraints of trade with no purpose except stifling of competition.” White Motor Co. v. United States, supra, 372 U.S. at 263 , 83 S.Ct. at 702 , (emphasis added). We find that in all of plaintiff’s cases there was a purpose either to exclude a person or group from the market, or to accomplish some other anti-competitive objective, or both.

In several of them, the objective was to put one or more so-called “discounters” or “price-cutters” out of business. United States v. General Motors Corp., 1966, 384 U.S. 127 , 86 S.Ct. 1321 , 16 L.Ed.2d 415 ; Ford Motor Co. v. Webster’s Auto Sales, Inc., 1 Cir., 1966, 361 F.2d 874 , involved a scheme similar to, but less elaborate than the General Motors scheme. Somewhat similar is Fashion Originators’ Guild of America v. FTC, 1941, 312 U.S. 457 , 61 S.Ct. 703 , 85 L.Ed. 949 , where a combination of manufac *77 turers and designers sought to suppress competition by “style-pirates” who were also price cutters.

In other eases, there was concerted action by one group to. put one or more of their competitors out of business, or to impair their ability to compete with the conspirators. See Silver v. New York Stock Exchange, 1963, 373 U.S. 341, 347 , 83 S.Ct. 1246 , 10 L.Ed.2d 389 ; Radiant Burners v. Peoples Gas Light & Coke Co., 1960, 364 U.S. 656 , 81 S.Ct. 365 , 5 L.Ed.2d 358 ; Associated Press v. United States, supra.

Another case involved the exclusion of competitors from the market by monopolistic practices violative of section 2 of the Sherman Act, together with a price-fixing conspiracy. Continental Ore Co. v. Union Carbide & Carbon Corp., 1962, 370 U.S. 690 , 82 S.Ct. 1404 , 8 L.Ed.2d 777 ; Binderup v. Pathe Exchange, Inc., 1923, 263 U.S. 291, 311 , 44 S.Ct. 96 , 68 L.Ed. 308 , is similar, although the exclusion was of a customer of some of the conspirators, rather than a competitor of the conspirators. Eastern States Retail Lumber Dealers’ Ass’n v. United States, 1914, 234 U.S. 600 , 34 S.Ct. 951 , 58 L.Ed. 1490 , involved a combination of retailers to boycott wholesalers who sold directly to consumers. See also Montague & Co. v. Lowry, 1904, 193 U.S. 38 , 24 S.Ct. 307 , 48 L.Ed. 608 . In Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 1951, 340 U.S. 211 , 71 S.Ct. 259 , 95 L.Ed. 219 , there was an agreement between sellers to refuse to sell to wholesalers who would not agree to abide by maximum resale prices fixed by the sellers. Thus the boycott of the plaintiff to that case was part of a price-fixing scheme.

In Klor’s, Inc. v. Broadway-Hale Stores, Inc., 1959, 359 U.S. 207 , 79 S.Ct. 705 , 3 L.Ed.2d 741 , on which plaintiff most heavily relies, the purpose was to put the plaintiff out of business. That was enough for the Supreme Court. And the facts show, although the Court did not rely on this, that the reason for doing so was that the plaintiff was a price-cutter. Thus the defendant’s motives were doubly anti-competitive.

There is nothing in the case before us really comparable to the facts of the foregoing cases. The proper line to be drawn, we think, is that stated by Barber in “Refusals to Deal Under the Federal Antitrust Laws,” 103 U. of Pa.L.Rev. 847 (1955), at 876-77:

Here, plaintiff presented no evidence whatever that either Seagram or Barton had any anticompetitive motive for terminating plaintiff as their distributor. There were no price fixing or other similar motives or demands or activities. There had been no anticompetitive practices imposed by either Seagram or Barton on plaintiff; none was imposed on McKesson; McKesson sought none. Of course McKesson wanted the part of the business that it got. But there is no evidence that either McKesson or any of the other defendants were primarily motivated by a desire to damage plaintiff or put it out of business.

There is also much evidence that both Seagram and Barton were dissatisfied with plaintiff as a distributor. Plaintiff presented considerable evidence that they should not have been dissatisfied, including evidence of praise of plaintiff by their agents and evidence that Portside did no better for them than plaintiff, and perhaps not as well. From this, plaintiff infers that Seagram and Barton were not dissatisfied with it. From this, it urged the jury to find that the parties conspired. It says that because it presented evidence that the expressed reason for the change may not have been valid, and hence may' not have existed, some sinister, anticompetitive motive can be inferred. We think not; the most that plaintiff’s evidence suggests is that Seagram and Barton may have been mistaken in judging the quality of plaintiff’s performance and the probable ability of McKesson to do better. Nothing in the record suggests, much less would support an inference, that their purpose was to eliminate or damage plaintiff, or to force it to comply with unlawful demands, rather than to get better distribution of Seagram and Barton products. And McKesson’s desire to get the busi less is the same as exists in every case in which a distributor persuades a supplier to give it an exclusive distributorship, whether as an original arrangement or by transfer of the business from another exclusive distributor. Surely, this is the essence of competition. Surely, liability for treble damages should not be based on such second guessing of business judgment by a jury.

We think it indisputable that a single manufacturer or seller can ordinarily stop doing business with A and transfer his business to B. All of the cases cited, supra, pp. 76-77, stand for this rule. The rule is also expressly recognized in the cases upon which appellee relies, and which we have discussed. There is language in many of those cases which, taken out of context, plaintiff construes as meaning that, if the seller agrees with a third party — a competitor of the seller, for example, or a competitor of A — to do the same thing, a per se violation of section 1 has occurred. This obviously cannot mean an agreement with B, the new distributor; he could not accept the distributorship without agreeing to do so. And the decisions cited supra, pp. 76-77, make it clear that the decision of the seller to transfer his business from A to B is valid even though B may have solicited the transfer and even though the seller and B may have agreed before the seller terminates his dealings with A. No case cited to us, and none that we have found, actually goes so far as plaintiff claims.

We confine our decisions to the facts of this case. We agree that a “combination or conspiracy” to establish a common distributor could be shown to have such an adverse purpose or effect on competition that it would violate section 1 of the Sherman Act as an unreasonable restraint of trade. See Albrecht v. Herald Co., supra; White Motor Co. v. United States, supra; United States v. Columbia Steel Co., supra, 334 U.S. at 524-525 , 68 S.Ct. 1107 ; United States v. Paramount Pictures, Inc., 1948, 334 U.S. 131, 144-147 , 68 S.Ct. 915 , 92 L.Ed. 1260 ; Walker Distrib. Co. v. Lucky Lager Brewing Co., supra, 323 F.2d at 7 .

*79 But we do not think that the group boycott per se rule should automatically apply to a situation such as the one before us.

Maple Flooring Mfrs. Ass’n v. United States, 1925, 268 U.S. 563, 579 , 45 S.Ct. 578 , 69 L.Ed. 1093 . We must always guard against “the tyranny of tags and tickets.” Cardozo, Mr. Justice Holmes, 44 Harv.L.Rev. 682, 688 (1931).

This case is similar to Instant Delivery Corp. v. City Stores Co. (Lit Bros. Div.), E.D.Pa., 1968, 284 F.Supp. 941, 947 , where four retailers who had been using a single delivery service went to two different services, but later decided to re-consolidate. The stores awarded the contract to one delivery service; the other one went to court, crying group boycott. The court denied a preliminary injunction because the evidence presented at the preliminary hearing did not show either a concerted refusal to deal or an unreasonable restraint of trade:

See also Parmelee Transp. Co. v. Keeshin, 7 Cir., 1961, 292 F.2d 794 , 804; Interborough News Co. v. Curtis Publishing Co., 2 Cir., 1955, 225 F.2d 289, 293 ; Lawlor v. National Screen Serv. Corp., 3 Cir., 1959, 270 F.2d 146 , and 1956, 238 F.2d 59 , vacated on other grounds, 1957, 352 U.S. 992 , 77 S.Ct. 526 , 1 L.Ed.2d 540 .

We recognize that the latter three cases can be distinguished from the case at bar. In Parmelee, a group of railroads had agreed to enter into a joint exclusive contract with a single “transfer service” to deliver passengers and their baggage among the various Chicago railroad terminals. The Third Circuit apparently rested its decision on the fact that “there was very strenuous competition.- This unavoidable fact undermines the plaintiff’s charges under §§ 1 and 2 of the Sherman Act.” 292 F.2d at 804. However, the court did say that the railroads, in entering into the exclusive contract, were performing duties imposed by the Interstate Commerce Act. Id. at 803. See City of Chicago v. Atchison, T. & S. F. Ry„ 1958, 357 U.S. 77, 86 , 78 S.Ct. 1063 , 2 L.Ed.2d 1174 . In Interborough News and in Lawlor, the triers of fact had found that there was no agreement at all because the suppliers had independ *80 ently decided to use the same exclusive distributor. However, it is our view that the existence of a horizontal agreement is not dispositive.

The exclusion of plaintiff was merely the incidental result of appellants’ agreement to transfer their lines to Portside. In such a case the group boycott cases are not controlling. See Barber, Refusals to Deal under the Federal Antitrust Laws, 103 U.Pa.L.Rev. 847, 872, 876-78 (1955). A manufacturer or supplier has a legitimate interest in the quality, competence, and stability of his distributors. See generally Note, Restricted Channels of Distribution under the Sherman Act, 75 Harv.L.Rev. 795 (1962). In Hawaii, liquor suppliers have usually distributed their products to retailers via exclusive wholesale representatives. Plaintiff was such a representative of appellants. As noted above, the use of exclusive distributorship is not, without more, invalid under the antitrust laws. No claim to the contrary is here made. Nor is it even suggested that there was any illegality in the fact that both Seagram and Barton had chosen plaintiff as their distributor.

A supplier who becomes dissatisfied with an existing distributor also has a legitimate interest in seeing that any new distributor to which it might turn would be viable. Manufacturers’ or suppliers’ decisions about the distribution of their products “are not made in a vacuum.” Schwing Motor Co. v. Hudson Sales Corp., supra, 138 F.Supp. at 906 . Thus Seagram and Barton each had a legitimate interest in the viability of McKesson’s proposed new Portside distributor. As we have seen, such a distributor needs a well rounded group of lines, and Seagram and Barton, therefore, each had a legitimate interest in Portside’s having such a group of lines. So, of course, did McKesson. See Interborough News Co. v. Curtis Publishing Co., supra; Instant Delivery Corp. v. City Stores Co., supra. “The antitrust laws do not require a business to cut its own throat.” Brown v. Western Mass. Theatres, Inc., 1 Cir., 1961, 288 F.2d 302, 305 .

We conclude, therefore, not only that the instructions we have quoted were erroneous, but also that there is insufficient evidence to sustain the verdict.

3. Intra-Corporate Conspiracy.

In the foregoing portion of this opinion, we have frequently referred to “Seagram” and “Barton.” We were using them respectively, as collective terms for the Seagram corporations and the House of Seagram divisions, and for the two Barton corporations. However, as the instructions that we have quoted indicate, the court took a different view of the Seagram group. In addition to what we have quoted above, it also told the jury:

Thus, to the jury, the possible conspirators were Seagram, House of Seagram, Calvert, Four Roses, Frankfort, McKesson, and Barton, or any two or more of them. The court’s theory seems to have been that there could be no actionable conspiracy between Barton and Barton Western because they were not competitors. The same reasoning applies to Seagram and House of Seagram. Evidently the court believed that, to be actionable, the charged conspiracy had to be between parties who competed with each other, or between one or more of such parties and McKesson, a competitor of appellee. We think that the court was right as to Barton and Barton Western, and wrong as to Seagram and House of Seagram. Surely a manufacturer and its national or regional dis *82 tributor, whether a subsidiary or an independent, can agree to transfer their business from one wholesaler to another without running afoul of the group boycott per se rule, in the absence of some forbidden anti-competitive or monopolistic objective. See Alpha Distr. Co. v. Jack Daniel’s Distillery, N.D.Cal., 1961, 207 F.Supp. 136 , aff’d per curiam, 9 Cir., 1962, 304 F.2d 451 . Cf. Sunkist Growers, Inc. v. Winckler & Smith Citrus Prods. Co., 1962, 370 U.S. 19, 29 , 82 S.Ct. 1130 , 8 L.Ed.2d 305 ; Syracuse Broadcasting Corp. v..Newhouse, 2 Cir., 1963, 319 F.2d 683, 687 .

The court was also in error in holding that the House of Seagram divisions should be treated as separate entities, capable of conspiring with each other and with Seagram. We consider the question because it is conceivable that it could be held that an agreement among four suppliers to terminate their business with a wholesaler is so likely to have an anticompetitive effect that it should be condemned as a per se group boycott, when a similar agreement between only two should not. We do not mean to suggest that this is what we think the law is; we merely face the possibility that it may become the law. We do not think, for reasons to be stated, that this is such a case.

a. The question is not moot.

Plaintiff urges that the question is moot because the jury returned a verdict against all defendants, thus necessarily finding that McKesson, the two Barton corporations and the two Seagram corporations all conspired together. Thus, says plaintiff, it is not now material whether the jury also found that the three divisions of House of Seagram conspired with each other. The record, however, does not support this claim.

Under the court’s instructions, the jury could have found that any two or three of the divisions first conspired among themselves, and that the other corporations later knowingly joined that conspiracy, and could have based its verdict solely on that ground. It may not have done so, but we have no way of knowing that. See Sunkist Growers, Inc. v. Winckler & Smith Citrus Prods. Co., supra, 370 U.S. at 30, 82 S.Ct. 1130 . The intra-corporate conspiracy theory was an important part of plaintiff’s case. It was emphasized in the court’s instructions ; indeed, it permeates the conspiracy instructions from beginning to end. If the instructions were erroneous, they were certainly prejudicial.

b. The instructions were erroneous.

It is now settled law that if a corporation chooses to conduct parts of its business through subsidiary or affiliated corporations^ and conspires with them to do something that independent entities cannot conspire to do under section 1 of the Sherman Act, it is no defense that the corporations are, in reality a single economic entity. The Supreme Court has said that “common ownership and control does not liberate corporations from the impact of the antitrust laws.” Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 1951, 340 U.S. 211, 215 , 71 S.Ct. 259, 261 , 95 L.Ed. 219 . See also Fortner Enterprises, Inc. v. United States Steel Corp., 1969, 394 U.S. 495, 507 , 89 S.Ct. 1252 , 22 L.Ed.2d 495 ; Perma Life Mufflers, Inc. v. International Parts Corp., 1968, 392 U.S. 134, 141-142 , 88 S.Ct. 1981 , 20 L.Ed.2d 982 ; Timken Roller Bearing Co. v. United States, 1951, 341 U.S. 593, 598 , 71 S.Ct. 971 , 95 L.Ed. 1199 ; Schine Chain Theatres, Inc. v. United States, 1948, 334 U.S. 110, 116 , 68 S.Ct. 947 , 92 L.Ed. 1245 ; United States v. Yellow Cab Co., 1947, 332 U.S. 218, 227 , 67 S.Ct. 1560 , 91 L.Ed. 2010 . The Court has never indicated what, if any, are the limits of this doctrine. Handler, Through the Anti-Trust Looking Glass — Twenty-first Annual Antitrust Review, 57 Cal.L.Rev. 182-93 (1969).

.On the other hand, it has been held that “a corporation cannot conspire with its officers or agents to violate the antitrust laws.” Chapman v. Rudd Paint & Varnish Company, 9 Cir., 1969, 409 F.2d 635 , 643 n. 9. See Poller v. CBS, *83 Inc., 1960, 109 U.S.App.D.C. 170 , 284 F.2d 599, 603 , rev’d on other grounds, 1962, 368 U.S. 464 , 82 S.Ct. 486 , 7 L.Ed.2d 458 ; Goldlawr, Inc. v. Shubert, 3 Cir., 1960, 276 F.2d 614, 617 ; Nelson Radio & Supply Co., Inc. v. Motorola, Inc., 5 Cir., 1952, 200 F.2d 911, 914 ; Kemwell Automotive Corp. v. Ford Motor Co., S.D.N.Y., 1966 Trade Cas. ¶ 71,882 ; Sperry Rand Corp. v. Nassau Research & Dev. Associates, Inc., E.D.N.Y., 1957, 152 F.Supp. 91, 93 ; Johnny Maddox Motor Co. v. Ford Motor Co., W.D.Tex., 1960, 202 F.Supp. 103, 105 ; Alexander v. Texas Co., W.D.La., 1957, 149 F.Supp. 37, 45 ; Cott Beverage Corp. v. Canada Dry Ginger Ale, Inc., S.D.N.Y., 1956, 146 F.Supp. 300 . The same rule has been applied to unincorporated divisions. See Poller v. CBS, Inc., supra; Deterjet Corp. v. United Aircraft Corp. (Hamilton Standard Div.), D.Del., 1962, 211 F.Supp. 348, 353-354 ; Johnny Maddox Motor Co. v. Ford Motor Co., supra.

The decision of the trial court is the first that has not followed these cases. Its views are set out in Hawaiian Oke & Liquors, Ltd. v. Joseph E. Seagram & Sons, Inc., D.C., 272 F.Supp. 915 . The trial court’s opinion has produced a spate of comment, pro and con. Burrus & Savarese, Developments in Anti-trust During the Past Year, 37 Antitrust L. J. 381, 384 (1968); Handler, supra, 57 Cal.L.Rev. at 185-86; Kempf, Bathtub Conspiracies, Has Seagram Distilled a More Potent Brew, 24 Bus.Law. 173 (1958); Willis & Pitofsky, Antitrust Consequences of Using Corporate Subsidiaries, 43 N.Y.U.L.Rev. 20 (1968); 6 Duquesne U.L.Rev. 157 (1968); 36 Fordham L.Rev. 607 (1968); 43 Notre Dame Law. 786 (1968); 37 U.Cinc.L. Rev. 223 (1968); 1968 U.Ill.L.F. 248; 21 Vand.L.Rev. 375 (1968).

The trial judge noted that the House of Seagram divisions had formerly been separate but related corporations, and had been found in 1951 to have conspired with each other in Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., supra, 341 U.S. 211 , 71 S.Ct. 259 . He felt that all that had happened since KieferStewart had been a formal change ■ in corporate structure, and that there had been no change in marketing technique. He concluded that, as a matter of law, the divisions were, in substance, the same entities that they were at the time of the Kiefer-Stewart decision.

We cannot agree. There is here no evidence that the “de-incorporation” of the former corporations was a sham or “shuffling of papers,” as plaintiff argues. Nor do we think that there was here a mere change in the label attached to a business entity. See United States v. Sealy, Inc., 1967, 388 U.S. 350, 352 , 87 S.Ct. 1847 , 18 L.Ed.2d 1238 . Before the 1959 reorganization, each subsidiary had its own payroll, accounting department, billing, and each had limited liability. Consolidation destroyed this limited liability, as well as certain tax advantages. The advantages of the corporate subsidiary form are spelled out in Willis & Pitofsky, supra, 43 N.Y.U.L.Rev. 20, 27-28. The trial judge relied only on the fact that the divisions had autonomous sales organizations, 272 F.Supp. at 924 , thus in effect conceding that there was no autonomy in other respects. But since sales and price decisions are not made in a vacuum, but are affected by other corporate activities, we doubt that autonomy in sales alone would ever be sufficient independence. See 43 N.Y.U.L.Rev. 172, 177 (1968).

Once the theory that “divisions” or other internal administrative units of a single corporation can “conspire” with each other is accepted, we can see no sensible basis upon which it can be decided that, in one case, there has been a conspiracy and that, in another, there has not. No corporation of any size can operate without an internal division of labor between various of its officers and agents. The larger the enterprise, the more necessary such internal units become. Moreover, sound management demands extensive delegation of authority within the organization. Yet, under the trial court’s ruling, the more delegation there is, the more danger there will be *84 that the holders of such delegated authority will be found by a court to be capable of conspiring with each other in carrying on the corporation’s business, as in this case, where the trial court so held as a matter of law. It is most unlikely that partially autonomous (here the autonomy extended to sales only) divisions of a single corporate enterprise will or can operate completely independently of each other. It is inevitable that there will be communication between them, either directly or through those persons in the corporate hierarchy to whom they report. And such communication can then be used as evidence that they arrived at understandings with each other as to what they would do. Thus, they are capable of conspiring because they are autonomous and they have conspired because they are, in fact and law, parts of a single corporation. Here, for example, plaintiff relies heavily on the fact that the top executives of the divisions “officed” in the same building, “only steps apart” in New York, and that their Western Division managers “officed” in the same building in Los Angeles. The doctrine hands to plaintiffs, on a silver platter, an automatically self-proving conspiracy.

Nor is it an answer to say, as the trial judge did, that here the conspiracy was “horizontal,” which was not the fact in Chapman and the other cases cited above ( 272 F.Supp. at 918-919 ). This is an elevation of form over substance. See Kempf, supra, 24 Bus.Law. 173 [1968]. If intra-corporate divisions are capable of conspiracy only on a horizontal plane, they could avoid the antitrust laws and still conspire by going through someone higher up in the corporate hierarchy. We do not see why the House of Seagram could order Calvert, Four Roses, and Frankfort to change their lines from plaintiff to McKesson, but Calvert, Four Roses and Frankfort cannot themselves agree to do so.

In short, we conclude that if the doctrine of intra-corporate conspiracy is accepted, there is no logical or practical way to avoid holding that all intra-corporate agreements are or may be found to be conspiracies in restraint of trade. See McQuade, Conspiracy, Multicorporate Enterprises, and Section 1 of the Sherman Act, 41 Va.L.Rev. 183, 216 (1955); Rahl, Conspiracy and the Anti-Trust Laws, 44 Ill.L.Rev. 743, 765-66 (1950); Willis & Pitofsky, supra, 43 N.Y.U.L.Rev. 20, 26.

We reject the doctrine of intra-corporate conspiracy as applied here. It follows that the only “conspiracy” shown is that described in part 2 of this opinion. And we have there held that there was no unlawful restraint of trade resulting from that conspiracy.

4. Other claimed errors.

a. Conscious parallel action.

The court gave the following instruction :

Appellants object that this instruction permitted the jury to find a conspiracy on the basis of conscious parallel action alone. It does not require such a finding, but we agree that it permits it. The court refused to tell the jury:

We think that the court erred. Conscious parallel action is indeed evidence which, with other evidence, may support a finding of conspiracy. But *85 standing alone it is not enough. See First National Bank of Ariz. v. Cities Service Co., 1968, 391 U.S. 253, 286-288 , 88 S.Ct. 1575 , 20 L.Ed.2d 569 , where the Court upheld a summary judgment for a defendant, although there was evidence of conscious parallel action by that defendant with the action of others. And see Independent Iron Works, Inc. v. United States Steel Corp., 9 Cir., 1963, 322 F.2d 656, 661 ; Delaware Valley Marine Supply Co. v. American Tobacco Co., 3 Cir., 1961, 297 F.2d 199, 202-203 , and cases there discussed. Cf. Theatre Enterprises, Inc. v. Paramount Film Dist. Corp., 346 U.S. 537, 540-541 , 71 S.Ct. 257 , 98 L.Ed. 273 . We know of no ease holding that conscious parallel action, standing alone, is sufficient to support a finding of conspiracy, in the absence of “circumstances which logically suggest joint agreement, as distinguished from individual action.” Independent Iron Works v. United States Steel Corp., supra. See Note, Conscious Parallelism —Fact or Fancy? 3 Stan.L.Rev. 679, 694 (1951).

In this case, appellants strenuously contended that Seagram’s and Barton’s decisions were arrived at independently. We agree with plaintiff that there is enough other evidence in the case to sustain a finding that they conspired (i. e., agreed, expressly or tacitly) with each other as well as separately with McKesson. But we also think that the case was close enough to make the instruction highly prejudicial.

b. Proof of damages.

Appellants attack practically all of the evidence offered by plaintiff in support of its claim for damages. Plaintiff did not claim loss of future profits. Rather, it claimed that its business had been destroyed, and sought to recover its value, less the amount received by its shareholders in liquidation, plus certain out-of-pocket expenses. Appellants do not attack this approach to damages. They do strenuously attack the admissibility and sufficiency of certain evidence received, and certain of the court’s instructions.

(1) The Caldwell charts.

A number of charts prepared by one Caldwell, a C.P.A., were received in evidence. Admittedly, he was an expert accountant. But plaintiff conceded that he was not an expert in the valuation of a business. Caldwell did not examine plaintiff’s books and records. In preparing five of plaintiff’s exhibits, he “simply followed Mr. Blecher’s [plaintiff’s counsel’s] instructions with respect to the mathematical computations.” These charts, together, were received in evidence, as projections of what plaintiff’s business would have been if Seagram and Barton had not terminated it as their distributor.

To illustrate: Plaintiff’s own financial statements, summarized in exhibit P-1, shows its earnings history as follows:

‘Year Net Profit or (Loss) from Operations Total Net Income (Loss)

1959 $(55,486.63) $(45,589.73)

1960 $(55,583.20) $(46,240.93)

1961 $(65,922.28) $(50,179.54)

1962 $(16,505.59) $ 8,060.80

1963 $(13,808.63) $ 2,640.11

1964 $(18,767.92) $ 2,497.85”

*86 However, the next chart, exhibit P-2, shows projected 1965 net income of $45,-256. And exhibit P-3 projected net profits for succeeding years as approximately :

1966 $53,000

1967 62,000

1968 72,000

1969 82,000

How was this remarkable result achieved? Appellants’ counsel summarize the process in their opening brief, and plaintiff does not attack the essential accuracy of what they say:

The record simply does not support the assumptions that were made, either by the witness or by plaintiff’s counsel. (The record is not always clear as to which we are dealing with.)

First, it was assumed that, because, in 1964, second-half sales and costs were higher while operating expenses were lower than in the first half of the year, the same would be true in 1965 and succeeding years. The result was a 1965 projected increase in sales of $230,000 and a decrease in operating expenses of $80,000. Yet in fact, in prior years other than 1964, there had been an increase in operating expenses in the second half of the year. So far as the record shows, 1964 was the exception, not the rule. And Caldwell admitted that normally operating expenses would increase as sales increase.

Second, it was assumed that the large increase in sales in the first half of 1965 would continue. The record does not support the assumption. Monthly sales in the first half of the year for the years 1962-1965 were:

1962 1963 1964 1965

January $81,796 $ 86,085 $ 89,904 $ 82,455

February 82,405 84,306 93,304 100,184

March 97,855 101,156 97,968 113,858

April 90,001 100,364 114,825 104,497

May 86,793 95,649 105,158 93,667

June 97,751 97,590 99,092 212,318”

Based upon past performance, June, 1965, appears to be an extraordinary month. And so it was. We again quote appellant’s brief:

We agree. It is no answer to say, as plaintiff does, that McKesson had more second half sales in 1965 than it did in the first half of that year. There is no evidence as to McKesson’s experience in prior years, as there is of plaintiff’s prior experience.

Third, having constructed an assumed 1965 increase in sales of 18.31%, plaintiff, using that figure, arrived at a trend in sales increase of 9.23%. Yet the actual growth rate had been, for 1963 and 1964, 4.7%. Here we have a second assumption, unwarranted in itself, and based' upon the unwarranted first assumption.

Fourth, plaintiff projected a growth rate in operating expenses of 5.8%. This was 50% of Caldwell’s projected “annual average percentage increase” of gross profits. We can find no basis for this 50% figure in the record. There being no basis shown for this assumption, the figure should have been excluded. See Lessig v. Tidewater Oil Co., 9 Cir., 1964, 327 F.2d 459, 473-474 ; Emich Motors Corp. v. General Motors Corp., 7 Cir., 1950, 181 F.2d 70, 83-84 , modified on other grounds, 1951, 340 U.S. 558 , 71 S.Ct. 408 , 95 L.Ed. 534 .

Nor is this all. Having constructed its remarkable table of projected earnings, plaintiff proceeded to arrive at an average annual profit figure of $63,349 and to capitalize this figure by multiplying it by figures ranging from 5 to 10 to arrive at a suggested fair market value of the business. Capitalization of a purely speculative profit figure at any rate may make the result look very impressive, but the result is no less speculative than its basis.

Plaintiff’s charts are “an array of figures conveying a delusive impression of exactness in an area where a jury’s common sense is less available than usual to protect it.” (Friendly, J. in Herman Schwabe, Inc. v. United Shoe Machinery Corp., 2 Cir., 1962, . 297 F.2d 906, 912 .) They should not have been admitted.

We are fully aware that “the plaintiff is not required to prove with mathematical certainty the amount of its damage resulting from a defendant’s violation of the antitrust laws.” Richfield Oil Corp. v. Karseal Corp., 9 Cir., 1959, 271 F.2d 709, 714 . But this does not mean that the door is open to present to a jury the kind of rampant speculation that went to the jury in this case. See also Emich Motors Corp. v. General Motors Corp., supra; Volasco Prods. Co. v. Lloyd A. Fry Roofing Co., 6 Cir., 1962, 308 F.2d 383 , 390-393.

*88 If we were not reversing on substantive grounds, we would feel compelled to require a new trial on the damage issue because of the foregoing errors.

(2) McKesson (Portside) as a yardstick.

Appellee used Portside’s performance as a “yardstick” to measure what its performance would have been if it had not been terminated by Seagram and Barton. We think that it was entitled to do this. Exact comparability is neither possible nor necessary. See Bigelow v. RKO Radio Pictures, Inc., 1946, 327 U.S. 251 , 66 S.Ct. 574 , 90 L.Ed. 652 ; Timberlake, The Legal Injury Requirements and Proof of Damages in Treble Damage Actions under the Antitrust Laws, 30 Geo.Wash.L.Rev. 231, 269-72 (1961).

There was, however, one serious weakness in this proof. McKesson’s average ratio of total income to sales was 7% times as great as plaintiff’s yet Caldwell assumed that plaintiff would have had as good a rate of return as McKesson. This was arguably a matter for cross-examination and argument, and can be said to go to the weight, not the admissibility of the evidence as a whole. But it is essentially speculative, and such evidence could be very misleading to the average juror. We think that, unless there was some showing of justification for Caldwell’s assumption, it should have been excluded, and that at the very least, the trial judge should have told the jury to examine such an assumption with care, and to disregard conclusions or opinions based upon it unless the jury found that the assumption was justified.

(3) Expressions of interest in purchasing plaintiff’s business.

Plaintiff’s Wong was permitted to testify that various persons had expressed an interest in buying plaintiff’s business. No figures were mentioned; no negotiations occurred. Gonzalez, of Seagram, testified that he told his brother-in-law that he was thinking of going into the wholesale liquor business in Hawaii, that he thought it would take $360,000 to do it, and that he approached two dealers, one of them plaintiff. No figure for plaintiff’s business was mentioned by him.

We think that this testimony should not have been admitted; no case has been cited to us, and we know of none, that would permit it.

(4) Out-of-pocket losses.

Plaintiff showed that in 1965 it lost $35,000 and in 1966 $14,000, before it was liquidated. But plaintiff made no showing whatever as to what the losses were. Thus there was no way in which the jury could decide what losses, if any, were caused by the termination of the Seagram and Barton distributorships. We have already noted that plaintiff suffered losses in prior years. Thus, we think that something more than gross figures was required to show the fact of damage, i. e., that the loss was caused by the termination. See Talon, Inc. v. Union Slide Fastener, Inc., 9 Cir., 1959, 266 F.2d 731, 736 ; Flintkote Co. v. Lysfjord, 9 Cir., 1957, 246 F.2d 368, 392 .

(5) Instructions relating to damages.

The court told the jury that it could consider, in fixing damages, “the interest of third persons in purchasing the plaintiff’s business.” For reasons already stated, that was error.

The court also told the jury that “plaintiff may also recover * * * for out-of-pocket operating losses occasioned as a result of any defendant’s conduct. Thus you may consider the out-of-pocket losses alleged by plaintiff to have been occasioned in 1965 and 1966 as a result of the liquidation of its business.” This instruction can be argued to imply, but does not state, that only such losses as were caused by defendants could be recovered. The jury should have been plainly told that it must first find causation, and that only those losses that it found to have been caused by defendants are recoverable.

The court declined to instruct that Caldwell’s exhibits P-2, P-3, P-4 and P- *89 5 were not to be treated as expert testimony. For reasons' already stated, we think that this was error. It was compounded by an instruction that the court did give, that Caldwell was an expert in accounting, and that the jury “should consider each expert opinion received in evidence in this ease, and give it such weight as you think it deserves.”

In sum, we are- convinced that, at the least, there should be a new trial, both as to liability and on the issue of damages. We are also convinced that a conspiracy in restraint, of trade was not proved.

The judgment is reversed, with directions to enter a judgment dismissing the action.