Concurring with most of Chief Judge Lumbard’s opinion, Judge Medina and I do not consider that a remand as to the claim with respect to the “tied” purchases of flavoring, toppings and cones is either required or warranted in the light of the pre-trial order to which plaintiffs agreed.
The relevant portion of this order is as follows:
“4. That plaintiffs agree that they are relying solely upon alleged per se violations of the antitrust laws and make no claim other than that their evidence will establish per se violations of the antitrust laws.
“5. Plaintiffs’ evidence at the trial on the issue of liability will be limited to (a) written agreements entered into between the Carvel de- . fendants and the plaintiffs, and other documents supplementary thereto or explanatory thereof, (b) written agreements between the Carvel defendants and the supplier defendants, and other documents supplementary thereto and explanatory thereof, and (c) testimony heretofore taken of a Mr. Vettel and a Mr. Bittner.”
'It could hardly be plainer that plaintiffs were confining themselves to claims of violation of the antitrust laws which, as they thought, could be sustained without other market data than the meagre amount contained in the documents of- ' f ered.
In agreeing to the pre-trial order, plaintiffs may have been relying on the. inclusion of tying arrangements in the list of per se violations in Northern Pacific Ry. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514 (1958). But, as that opinion makes clear, not every tying arrangement is illegal per se. Tying arrangements "are unreasonable in and of themselves whenever a party has sufficient economic power with respect to the tying product to appreciably restrain free competition in the market for the tied product and a ‘not insubstantial’ amount of interstate commerce is affected.” 356 U.S. at 6, 78 S.Ct. at 518. See also Times-Picayune Publishing Co. v. United States, 345 U.S. 594, 608-609, 73 S.Ct. 872 (1953). More recently the Court has said that tying arrangements “may fall” in the category of per se violations, “though not necessarily so.” White Motor Co. v. United States, 372 U.S. 253, 262, 83 S.Ct. 696, 9 L.Ed.2d 738 .(1963).
Here the facts to which plaintiffs were limited by the pre-trial order showed neither that Cárvel had “sufficient economic power with respect to the tying product to appreciably restrain free competition in the market for the tied product” nor that “a not insubstantial amount of commerce is affected.” Indeed, such figures as exist would prove the contrary. In 1960 there were in New York, Connecticut and Massachusetts, 250 Carvel dealers out of a total of 125,000 outlets where ice cream cones could be purchased — amounting to one fifth of one per cent of the outlets and apparently doing about one per cent of the business.1 The other large Carvel states are New Jersey and Pennsylvania; the balance of a 1960 total of 400 Carvel dealers were scattered at one time over as many as 8 or 9 states from Maine to Florida and as far west as Wisconsin. These dealers competed not only with similar chains — Dairy Queen, Tastee Freez, Dari-Delite, King Kone, Dari-Isle, and others, but also with chains and independents utilizing mobile units, with chain stores and operations such as Howard Johnson, and with the ubiquitous *519corner drug-store. Although Carvel’s aggregate sales are “not insubstantial,” the totals shed no light on the amount of the “tied” sales here complained of, which common sense tells us must be a minor part.2 The figures in footnotes 1 and 2 give some indication of the “insubstantiality” of the commerce affected even under the rather narrow test indicated by dictum in Brown Shoe Co. v. United States, 370 U.S. 294, 330, 82 S.Ct. 1502, 8 L.Ed.2d 510 (1962), especially with reference to the cones, whose tie-in was hardest to justify in terms of quality control. Not only was the amount of commerce in these not consequential, but any damage to the plaintiffs was even less so. . Eagle Cone’s billings to Carvel averaged $460 per dealer per year, and Carvel’s mark-up was slightly over 5% or about $25 per dealer; 3 whether the dealers suffered even that much damage is questionable since the price Carvel charged them was less than they could have obtained if they bought in smaller quantities than Carvel, see fn. 7 to Chief Judge Lumbard’s opinion. And, of course, it remained open to competing suppliers to bid for the Carvel business by soliciting that company — in itself an important contrast with eases where the tied item is produced by the seller.
Our brother Lumbard thinks the first of the deficiencies in proof as to economic power was remedied by Carvel’s license of a package of trade-marks, design patents relating to the shape of the building etc., and a patented freezing and dispensing machine. We cannot agree. In the first place, the patented items cannot realistically be considered the ^"tying product”- or the focus of the arrangement. Whatever has been said about the evils of “ties” to patented or copyrighted items is meaningful only in the situation where the desirability of the patented item is what motivates the purchaser to make further commitments or to give up some liberty of choice as to other products. See, e. g., United States v. Loew’s Inc., 371 U.S. 38, 45,' 83 S.Ct. 97 (1962). In this case, the patented items appear to have been virtually without motivating significance in' bringing about the agreement. The true tyingLitem was_ rather the Carvel trademark, whose growing repute was intended to help the little band of Carvel dealers swim a bit faster than their numerous rivals up the highly competitive stream. There may, of course cases where a trade-mark has acquired such prominence that the coupling of some further item to its. license would--constitute a peí — violation; but such a trademark would satisfy the market dominance test of Times-Picayune and Northern Pacific. The figures show thal Carvel is not such a mark.
Tying arrangements differ from other per se violations, such as price-fixing, United States v. Trenton Potteries Co., 273 U.S. 392, 47 S.Ct. 377, 71 L.Ed. 700 (1927), in that they can be justified on occasion, as by proof that “the protection of goodwill may necessitate” their use “where specifications for a substitute would be so detailed that they could not practicably be supplied,” Standard Oil Co. of Calif. and Standard Stations v. United States, 337 U.S. 293, 306, 69 S.Ct. 1051, 1058 (1949). Since the value of a trade-mark depends solely on the public image it conveys, its holder must exercise controls to assure himself that the mark is not shown in a derogatory light. The record affords no sufficient basis for upsetting the finding of the District Judge that “To require Carvel to limit itself to advance specifica*520tions of standards for all the various types of accessory products used in connection with the mix would impose an impractical and unreasonable burden of formulation * * * ” Although instances of impossibility of control through specification may indeed be rare in cases involving the proper functioning of mechanical elements of a machine, see International Business Machines Corp. v. United States, 298 U.S. 131, 139, 56 S.Ct. 701 (1936); International Salt Co. v. United States, 332 U.S. 392, 397-398, 68 S.Ct. 12, 92 L.Ed. 20 (1947); Turner, The Validity of Tying Arrangements under the Antitrust Laws, 72 Harv.L.Rev. 50 (1958); but see Dehydrating Process r Co. v. A. O. Smith Corp., 292 F.2d 653 (1 Cir.), cert. denied, 368 U.S. 931, 82 S.Ct. 368, 7 L.Ed.2d 194 (1961); United States v. Jerrold Electronics Corp., 187 F.Supp. 545 (E.D.Pa.1960), aff’d per curiam, 365 U.S. 567, 81 S.Ct. 755, 5 L.Ed.2d 806 (1961), such cases are scarcely relevant to the problem of controlling something so insusceptible of precise verbalization as the desired texture and taste of an ice cream cone or sundae; that Carvel was able to specify this to its source of supply, whose product it regularly checked, does not show that administration could be confided to 400 dealers. Furthermore in most states Carvel would risk liability to anyone injured by a foreign substance in the frozen mix, see ALI, Restatement (Second), Torts 402A (Tent. Draft No. 7, 1962); although it might not be so with respect to ingredients purchased from suppliers of the dealer’s own choosing, i the difficulties of proof are such that it is entitled to insist on products in which it has complete confidence. Vettel’s testimony that the success of the enterprise required distinctive flavorings of uniform quality and that he considered it impracticable to handle the problem by specification and policing was sufficiently reasonable and persuasive that the judge was entitled to credit it, whether or not one of us would agree.4 We see little force in the fact that a competitor, which licenses only freezing machines and the trade-mark and does not sell a mix made by a secret formula, is satisfied with less exacting provisions. See Engbrecht v. Dairy Queen Co., 203 F.Supp. 714 (D. Kan.1962). Moreover, as Vettel also explained, by packing the flavoring in tins or bags that contain exactly the amount required for a 10-gallon can of mix, Carvel encouraged the dealers to use an adequate quantity and provided an automatic control. Finally there is a limit to the amount of inspecting feasible when each dealer purchases an average of less than $12,000 a year of Carvel products (including equipment) and sells in 100 and 200 units.
Even were we to consider the patented machines and design patents to be a relevant part of the tying arrangement, the Northern Pacific opinion ruled out the idea, thought by some to have been implicit in earlier opinions, that proof of the license of a patented device sufficed semper et ubique to show the market dominance required to render a tying arrangement a per se violation.5 In doing so it appropriately recognized the facts of business life. The society of patents is not egalitarian. As had been well said three years earlier, although with some dissent, in the Report off the Attorney General’s Committee to Study the Antitrust Laws (1955), 238:
“The patent may be broad and basic, in which event the economic power incident to the patent makes the tying clause illegal. On the other *521hand, the patent may be narrow and unimportant, in which event it may confer virtually no real market power. Accordingly, where the tying produet is patented, the patentee should be permitted to show that in the entire factual setting, including the scope of the patent in relation to other patented or unpatented products, the patent does not create the market power requisite to illegality of the tying clause.”
Indeed, the statement in Northern Pacific had been anticipated by the remark in Standard Oil Co. of Calif. and Standard Stations v. United States, supra, 337 U.S. at 307, 69 S.Ct. 1051, that a patent is “prima, facie evidence” of market control — evidence which was clearly rebutted here. Although there is lanaguage in United States v. Loew’s Inc., supra, 371 U.S. at 44-48, 83 S.Ct. 97, which lends support to a theory that the mere existence of a patent of any description is not merely prima facie but irrebutable evidence of market control, we find it hard to believe that the Court would thus have obliquely reversed the position taken in Northern Pacific, a position underscored by Mr. Justice Harlan’s remarks in dissent, 356 U.S. at 17-19, 78 S.Ct. 514; the language rather must be read in the context of the Court’s previous proscription of block-booking of motion picture films in United States v. Paramount Pictures, Inc., 334 U.S. 131, 156-159, 68 S. Ct. 915, 92 L.Ed. 1260 (1948), see 371 U. S. at 48, 83 S.Ct. 97. There is scant analogy between unique hit movies, or the tabulating and computing machines in International Business Machines Corp. v. United States, supra, as to which the defendant had but one smaller competitor, or the salt-processing machines in International Salt Co. v. United States, 332 U. S; 392, 68 S.Ct. 12 (1947), and Carvel s dispensing machine by which, along with the rest of its package of distinctive devices, it had gained only 1% of the ice cream cone market in the area of its heaviest concentration, see fn. 1.
The triviality of whatever financial hardships the plaintiffs may have suffered as a result of these allegedly illegal contracts explains why they were willing to limit themselves to per se violations rather than incur the expense of assembling market data to gain what at best would be a negligible recovery. They have had two trials, one in which they were defeated in their claim of fraud, with a transcript of nearly 4,000 pages; we see no sufficient reason to order a third,
. There are figures which indicate that 1960 sales of all “ingredients” other than mix to all dealers amounted to $1,354,599 —or an average of less than $3,400 per dealer per year; the record does not make it dear whether all these “ingredients” were tied. Two-thirds of the figure of $3,965,923 in Chief Judge Lumbard’s opinion constitute sales of the Carvel secret formula mix which surely could legally be tied to the Carvel trademark if that mark was to retain its significance, as he concedes.
. There was evidence that one of the plaintiffs had purchased $621 worth of cones in the big year, 1959; Carvel’s mark-up on these would have been around $30.
. Vettel testified that one of tlie specifications to the manufacturer was that certain fruits be purchased at crop site; we are at a loss to understand how this could be policed by inspection on the dealer’s premises.
. “In arriving at its decision in International Salt the Court placed no reliance on the fact that a patent was involved nor did it give the slightest intimation that the outcome would have been any different if that had not been the case. If anything, the Court held the challenged tying arrangements unlawful despite the fact that the tying item was patented, not because of it.” 35G U.S. at 9, 78 S.Ct. at 520.