concurring in part and dissenting in part.
In addition to requiring divestiture of Autolite, the District Court made ancillary injunctive provisions that go far beyond any that have been cited to the Court. Ford is forbidden to manufacture spark plugs for 10 years; Ford is ordered to purchase one-half of its total annual requirement of spark plugs from the divested company under the “Autolite” name, and Ford is forbidden for the same period to use its own trade name on any spark plugs. These provisions are directed to prevent Ford from making an independent entry into the spark plug market and, in effect, to require it to subsidize Autolite for a period of time. Despite the Draconian quality of this restriction on Ford, I can find no justification in the District Court’s findings for this *583remedy. I dissent from the broad sweep of the District Court's remedial decree. I would remand for further consideration of the remedial aspects of this case.
An understanding of the District Court’s findings as to the spark plug market shows three reasons why it was in error in requiring Ford to support Autolite. First, the court did not find that the weakness of an independent Autolite’s competitive position resulted from Ford’s acquisition. Rather, a reading of its findings makes apparent that the precariousness of Autolite’s expected post-divestment position results from pre-existing forces in the market. Therefore, the drastic measures employed to strengthen Autolite’s position at Ford’s expense cannot be justified as a remedy for any wrong done by Ford. Second, the remedy will perpetuate for a time the very evils upon which the District Court based a finding of an antitrust violation. Third, the court’s own findings indicate that the remedy is not likely to secure Autolite’s competitive position beyond the termination of the restrictions. Therefore, there is no assurance that the judicial remedy will have the desired impact on long-run competition in the spark plug market.
The Court makes two critical errors in order to avoid the effect of this reasoning. It rejects the factfinding by the District Court in order to uphold its remedial order; and it repeats that court’s error by discussing the assistance necessary to restore Autolite to the status quo ante without ever delineating that prior state of affairs or indicating how Ford, by acquiring Autolite and holding it for a number of years, had undermined its ability to reassume its former independent competitive position.
The District Court made extensive findings on the nature of the spark plug market. Some of these findings appear in the Court’s opinion, but some factors that *584seem crucial to me are either omitted or not adequately set forth. Therefore I will sketch these findings at some risk of repetition.
Beyond doubt, the spark plug market has been overwhelmingly dominated by three manufacturers for a long period: AC, owned by General Motors, which had about 30% of the market in 1961; Champion, which had supplied Ford since 1910 and had approximately 50% of the market in 1961; and Autolite, which had supplied Chrysler since 1941 and had 15% of the market in 1961. Together these three companies had over 95% of the total market in 1961.
The reason for the continued domination of the market by the three big plug manufacturers is the pervasive feature of the plug market known as the “OE (original equipment) tie.” This denominates the phenomenon that mechanics who replace spark plugs in a car engine have tended, almost exclusively, to use the brand of plug installed by the auto builder as original equipment. Though not required by spark plug technology, mechanics have followed this practice because of a strong desire to avoid any chance of injuring an engine by putting a mismatched plug into it. Further, because plugs are low-profit items, those who install them tend to carry an inventory of a small number of brands. Most carry only two and some carry three brands, and they choose the brands installed by the' big auto manufacturers as original equipment. Thus, it takes a position as supplier to a large auto maker to gain recognition in the spark plug replacement market. The Government conceded in the District Court, for instance, that American Motors, with 5% of the auto market, would not be able to create market acceptance for an independent brand of plug by installing it as original equipment in its cars.
Because of the competitive importance of having their plugs installed as original equipment by one of the three *585auto companies, plug manufacturers have over a long period been willing to sell OE plugs for initial installation by auto manufacturers at a price below their production cost. The longstanding price for OE plugs, about 6 cents, is now approximately one-third of the cost of producing these plugs. Such below-cost selling is profitable for the plug companies because of the foothold it gives them in competing for the normal five or six sets of replacement plugs necessary in the lifespan of an automobile. This pricing policy has been partially responsible for the semipermanent relations between the plug manufacturers and the auto manufacturers: it is only those plug companies that profit from the OE tie over the long run that can afford this below-cost sale to the auto companies.
The strength of the OE tie is demonstrated by the inability of well-known auto supply manufacturers to gain a significant share of the spark plug market in the absence of an OE tie. As the District Court found, no company without the OE tie
“ever surpassed the 2% level. Several have come and gone. Firestone Tire and Rubber Company merchandised ‘Firestone’ replacements for 35 years before it gave up in 1964. Although it owned some 800 accessory stores and successfully wholesaled other items to more than 50,000 shops and filling stations, it could not surmount the patent discrimination against brands not blessed with Detroit’s approbation. Goodyear Tire and Rubber Company quit in only three years. Globe Union, a fabricator which had barely 1% of the nation’s shipments, withdrew in 1960.” 286 F. Supp. 407, 434-435.
Two small manufacturers survive, producing plugs for private-label brands. Thus “Atlas” plugs, sponsored by *586the Standard Oil companies, has 1.4% of the replacement market; “Prestolite” and Sears, Roebuck’s “Allstate” each have 1.2%; and Montgomery Ward’s “Riverside” label has 0.6% of the replacement market.
An independent entry into the plug market by Ford, with the expected substitution of its own plugs as original equipment in its cars, would have necessarily deprived one of the two significant independent plug producers of its OE status. The District Court found that, because of the importance of the OE tie, the plug producer deprived of this support would most likely have lost any significant position in the market.1 Autolite, with only 15% of the market before the acquisition, would certainly have lost any significant position in the market if an independent entry by Ford had led Chrysler to shift its patronage from Autolite to Champion. The District Court asserted that a Champion without OE status would have had some chance of maintaining a significant market position because of its size, although it gave no reason for thinking Champion’s size immunized it from dependence on OE status. Before 1961, Champion had just under 50% of the market. As a result of Champion’s move to Chrysler in 1961, its position in the market dropped to 33% by 1966. The District Court found no basis for predicting which of the two big independents would have won such a competition for continued OE status.
Thus, an independent entry by Ford would not likely have increased the number of significant competitors in the spark plug market. Rather, it would simply have substituted Ford for one of the two significant independent manufacturers. The result of this expectation *587is that the District Court did not base its finding of illegality on the ground typically present when a potential entrant enters an oligopolistic market by acquisition rather than internal expansion, i. e., that such a move has deprived the market of the pro-competitive effect of an increase in the number of competitors. Here an independent entry would not have increased the number of competitors but simply would have exchanged one competitor for another. In noting this paradoxical fact, the District Court concluded that “Ford may well have been more useful as a potential than it would have been as a real producer, regardless how it began fabrication.” 2 286 F. Supp., at 441.
Not finding that Ford’s entry by acquisition had deprived the spark plug market of any pro-competitive effect of an independent entry, the District Court relied on two other grounds for finding a violation of the antitrust laws. First, it concluded that as a potential entrant on the edge of the market which was also a major purchaser in the market, Ford exercised a “moderating” influence on the market; the second basis for determining the acquisition illegal was the finding that the acqui*588sition “foreclosed” other companies from competing for the business of supplying Ford with spark plugs.
With respect to Autolite itself, the District Court made several relevant findings. First, it found that Autolite is a fixed-production plant. In other words, it can be profitable only turning out approximately the number of plugs it now manufactures. It could not, for instance, reduce its production by half and sell that at a profit. Second, it made extensive findings with respect to Autolite’s distribution system:
“Ford received six regional offices, personnel and a list of Electric Autolite’s warehousers and jobbers. All of these have been and still are at liberty to deal with anyone they wish. Each old direct account had to be visited individually and, if it consented, be re-signed by defendant [Ford]. Within a few months, 52 did enter into new ignition contracts. However, 50 of these for the previous year had also been . . . [distributors of other Ford products]. By mid-1966, direct accounts totaled 156, of which 104 in 1960 had been pledged to neither Ford nor Autolite. The same bloc of 50 had been committed to both. The net increase traceable with any semblance of accuracy to the acquisition is two first-layer middlemen . . . .” 286 F. Supp., at 422.
As to difficulties that a divested Autolite might have in establishing an independent distribution system, the District Court mentioned only one:3 if Ford were to offer its own plugs to its car dealers at a fairly low price, one which independent jobbers could not meet, Autolite *589would have difficulty independently establishing its distribution system. The jobbers would be less interested in handling Autolite’s line since the Ford dealers would not want Autolite at the jobbers’ price and, with this demand cut out, the jobbers would be less interested in pushing Autolite generally.
There is another set of relevant facts found by the District Court. The District Judge found that “there is a rising wind of new forces in the spark plug market which may profoundly change it.” 315 F. Supp. 372, 377. On the basis of the testimony of an executive of one of the producers of plugs for private labels, the court found that the private-brand sector would grow during the next 10 years. This highly speculative observation of the District Court was based on a finding that the mass merchandisers are beginning to enter the plug marketing field in force. Not only do the mass merchandisers market private-brand plugs over the counter, but they are also building service bays. And in these bays many carry only their own proprietary brand of spark plugs. This witness predicted that the mass merchandisers would increase their share of the aftermarket from 4.4% to 10% by 1980. He further predicted that oil companies would enter the replacement market, resulting in a total of 17% of the replacement market being supplied by private-label plugs by 1980. The court concluded that these forces “may well lead to [the market’s] eventual deconeentration by increasing the number of potential customers for a new entrant into the plug manufacturing business and reducing the need for original equipment identification.” 315 F. Supp., at 378.
In its separate opinion on remedies, the District Court correctly stated the relevant law; the purpose, and limit of antitrust remedies, is to
“free these forces [within the market] from the unlawful restraint imposed upon them so that they *590may run their natural course.” 315 F. Supp., at 377.
The violators may not be required to do more than return the market to the status quo ante. See United States v. Paramount Pictures, Inc., 334 U. S. 131, 152-153 (1948); Reynolds Metals Co. v. FTC, 114 U. S. App. D. C. 2, 309 F. 2d 223 (1962) (Burger, J.). Applying this general provision to the instant situation, the District Court correctly stated:
“The court wishes to note here that although it finds that divestiture is the only effective remedy, it does not agree with the Government that the remedy should be affirmatively designed to ‘break the OE tie.’ The remedy is designed to correct the violations of Section 7 found by the court. The OE tie, as such, does not violate Section 7.” 315 F. Supp., at 378.
The District Court then concluded that, in addition to divestiture of the Autolite plant and trade name, certain injunctive provisions were required “to give [Autolite] an opportunity to establish its competitive position.” Ibid. It therefore ordered that Ford be prohibited from manufacturing spark plugs for a period of 10 years. It further ordered that for a period of five years Ford would be required to purchase one-half of its total annual needs of spark plugs from Autolite, bearing the Autolite label. For this five-year period Ford was also ordered not to use or market a spark plug under a trade name owned by or licensed to it. The effect of these orders was twofold. They assured Autolite of a purchaser for a large part of its production for five years. And they prevented Ford from immediately entering the competition for a share of the aftermarket with a plug under its own name; it could not even label a plug under its own name for five years and could not manufacture its own plug for 10 years. *591Given the findings of the court that even with the status of supplier of original equipment (with the company’s own brand name on plugs) to a major auto manufacturer it would take a new entrant into the spark plug market five to eight years to establish a position for its brand in the replacement market, the District Court’s orders assured that Ford could not begin to have brand-name success in the replacement market for at least 10 to 13 years.4
In my view these drastic remedial provisions are not warranted by the court’s findings as to the grounds on which Ford’s acquisition violated the antitrust laws. Further, in light of the District Court’s own factfindings, these remedies will have short run anticompetitive impact and they give no assurance that they will succeed in allowing Autolite to establish its competitive position.
The remedial provisions are unrelated to restoring the status quo ante with respect to the two violations found by the District Court, the ending of Ford’s status as a potential entrant with a moderating influence on the market and the foreclosure of a significant part of the plug market. Indeed, the remedies may well be anti-competitive in both respects. First, the District Court’s order actually undercuts the moderating influence of Ford’s position on the edge of the market. It is the *592possibility that a company on the sidelines will enter a market through internal expansion that has a moderating influence on the market. By prohibiting Ford from entering the market through internal expansion, therefore, the remedy order wipes out, for the duration of the restriction, the pro-competitive influence Ford had on the market prior to its acquisition of Autolite. Second, the Court’s order does not fully undo the foreclosure effect of the acquisition. Divestment alone would return the parties to the status quo ante. Ford would then be free to deal with Autolite or another plug producer or to enter the market through internal expansion. Yet the Court has ordered Ford to buy at least half its requirements from Autolite for five years. Thus, the order itself forecloses part of Ford’s needs from the forces of competition.
The above problems might be minor if the District Court’s remedy were justifiable in terms of returning Autolite to the status quo ante by overcoming some harm to its ability to compete accomplished by Ford’s acquisition. But on this issue the District Court opinion and the majority of this Court are confused. Although the District Court asserted that Autolite needed the aid of its injunctive remedies to establish its competitive position, the court made no findings in its remedy opinion as to the source of Autolite’s competitive weakness. Therefore it never reached the issue whether the source of weakness had anything to do with the violations attributed to Ford. Instead, the court’s opinion proceeded from the recognition of competitive problems immediately to the prescription of a remedy.
In fact, a fair reading of the findings of the District Court shows that the acquisition did not injure Auto-lite’s competitive position. Autolite’s OE status was continued and its share of the aftermarket was increased from 12.5% to 19%.. Thus, its trademark is at least as strong now as when Ford acquired the company. Nor *593did the acquisition and holding of Autolite injure its distribution system. The District Court found that Autolite did not own a distribution system. It merely had short-term contracts with jobbers who distributed its plugs to those who install them in cars or sell them to the public. Almost all of these jobbers had concurrent distribution relations with Ford. In fact, between 1961 and 1966 Ford tripled the number of jobbers handling Autolite plugs. From the opinion below, it appears that Ford has done nothing that will prevent an independent Autolite from seeking to maintain these distribution channels. The only possible finding of injury to be squeezed out of the acquisition relates to the fact that Autolite has been shorn of its status as OE supplier of Chrysler. But this is inconclusive. Autolite had nothing more in its position as OE supplier to Chrysler than it would if Ford voluntarily chose to use Autolite plugs after the divestment: a relationship based on short-term contracts the auto manufacturer could refuse to renew at any time.
The findings of the District Court indicate that Auto-lite’s precarious position did not result from its acquisition by Ford. Prior to the acquisition both Champion and Autolite were in a continually precarious position in that their continued large share of the market was totally dependent on their positions as OE suppliers to auto manufacturers. The very factor that assured that they faced no serious competition in the short run also assured that in the long run their own position was dependent on their relationship with a large auto manufacturer. Thus, the threat to Autolite posed by a simple divestiture is the same threat it had lived with between 1941 and 1961 as an independent entity: it might be left without any OE supply relationship with a major auto manufacturer, and therefore its market position based on this relationship might decline drastically.
*594Today’s opinion errs when it states, ante, at 571, that the District Judge found the OE tie the “key to the solution” of this problem. Although the court indeed found this tie a pervasive factor in the market, it also found that the phenomenon was not created by Ford and that it did not constitute a § 7 violation. Therefore the Court errs in justifying the ancillary remedies as necessary to overcome the OE tie. Even if such a remedy might overcome the OE tie, which I question, there is no justification for burdening Ford with the restrictive order.
Further, the only conclusion to be drawn from the trial findings is that the remedy is unlikely to result in a secure market position for Autolite at the end of the restricted period. Once again it will be dependent for its survival on whether it can maintain an OE supply status. The District Court’s suggestion that Autolite can find a niche supplying private-brand labels is unpersuasive. It cannot be predicted with any certainty that these sales outlets will grow to the extent predicted by one person in that line of the business. Further, even if they do, this is no assurance of Autolite’s survival. There are already several companies in the business of producing plugs for private labels. Autolite will have to compete with them. The results will not be helpful. One possibility is that Autolite would completely monopolize the private-brand market to the extent of about 17% of the replacement market. This is as uncompetitive as it is unlikely. The more reasonable likelihood is that Autolite might be able to gain a position producing, for instance, 5% of the replacement market plugs. But this would be useless because the District Court’s findings make clear that Autolite’s fixed-production plant cannot supply such a small share of the market at a profit.
In the final analysis it appears to me that the District Court, seeing the immediate precariousness of Autolite’s *595position as a divested entity, designed remedies to support Autolite without contemplating whether it was equitable to restrict Ford’s freedom of action for these purposes or whether there was any real chance of Auto-lite’s eventual survival. I fear that this is a situation where the form of preserving competition has taken precedence over an understanding of the realities of the particular market. Therefore I dissent from today’s affirmance of the District Court’s harshly restrictive remedial provisions.5
Of course, the decline would take a number of years, since it would be spread over the life of the cars on the road bearing the producer’s plugs as original equipment — probably five to eight years.
Mr. Justice Stewart, concurring in the result, relies on factual assumptions that seem to me directly contrary to findings made by the District Court. While that court found future developments might arise in the plug market that would enable an independent Autolite without OE status to survive, it also found that an independent entry by Ford in 1960, or even as of the date of the projected divestiture, would have left Autolite doomed because the market would not yet be ready to offer it an independent niche. By slighting these findings, Mr. Justice Stewart is able to avoid the question whether Ford should have to bear the burden of maintaining Autolite’s life until a time when market changes might support it when it is clear that an earlier independent entry by Ford would have left it moribund. He further overlooks the problems discussed below as to the unlikelihood of Autolite’s success, its fixed-production needs versus the small size of the market free of the OE tie.
The District Court made no mention of whether a divested Autolite would have the six regional offices and personnel that it had in 1960. Given the District Court’s solicitude for Autolite’s health, I can only assume that it expected Autolite to be sent out with whatever it had brought in.
The majority opinion errs in its evaluation, ante, at 577, of the effect of the restrictions on Ford’s ability to establish itself in the aftermarket. The District Court opinion makes clear that gaining a position in the replacement market takes five to eight years after the brand of plugs is first installed as original equipment: 18 months to three years before the first cars need plug replacements plus several annual car populations requiring this brand before service centers would be motivated to stock it. Thus, the prohibition against Ford’s using its own name for five years delays the beginning of an independent Ford entry and results in assuring that Ford could not gain a position in the aftermarket for 10 to 13 years after the effective date of the divestiture.
This case illustrates the unsoundness of the direct appeal permitted in cases of this kind under 15 U. S. C. § 29. In a factually complicated case like this, we would be immeasurably aided by the screening process provided by a Court of Appeals review. Limited expediting of such cases, under the discretion of this Court, would satisfy all needs justifying direct review in this Court.