POSNER, Chief Judge.
The plaintiffs (collectively “Khan”) operated a gas station in DuPage County, Illinois, [1360] under a contract with State Oil Company, a distributor of gasoline and related products. The contract provided for the lease of the station (which State Oil owned), and the supply of gasoline and ancillary products for resale, to Khan. Mr. Khan was the actual signatory of the contract, rather than his corporation, which operated the station, so it does not appear that he is complaining about a merely derivative injury to himself, in which event he would not be a proper party. Hammes v. AAMCO Transmissions, Inc., 33 F.3d 774, 777 (7th Cir.1994).
State Oil terminated the contract because Khan failed to pay the agreed-upon rent for the station. The termination precipitated this suit, which, so far as relevant to the appeal, charges price fixing in violation of section 1 of the Sherman Act, 15 U.S.C. § 1, and breach of contract under the common law of Illinois. The district judge granted summary judgment for the defendant on both claims. He ruled that the legality under the Sherman Act of the alleged price fixing was to be tested by the rule of reason rather than by the per se rule, that the plaintiff had presented no evidence on essential elements of a rule of reason case (such as market power), that the study conducted by the plaintiffs’ economic expert was inadmissible, and that without the study the plaintiffs could not even prove injury.
The contract between State Oil and Khan provided that State Oil would establish a suggested retail price for the gasoline (which was sold under the brand name “Union 76”) and would sell the gasoline to Khan for 3.25 cents less than that price. If Khan believed the price was too high he could ask State Oil to lower it and if State Oil complied Khan would be entitled to purchase the gasoline from State Oil at the same margin, that is, at the new price minus 3.25 cents. If State Oil refused to reduce the suggested retail price Khan could still charge a lower price, but his margin would be smaller because he would not be getting a lower price from State Oil. If Khan believed the suggested retail price was too low he could ask State Oil to raise it, thus preserving his margin; but if State Oil refused and Khan went ahead and raised his price anyway, the contract required Khan to rebate the difference between his new price and the suggested price times the number of gallons sold at the new price. The contract thus required Khan to rebate the entire profit from raising his price without his supplier’s permission above the retail price suggested by the supplier.
The provision concerning the charging by Khan of a price below the suggested retail price neither is price fixing nor is germane to the price-fixing charge. A supplier is under no obligation to lower his price to his customer just because the customer wants to resell the supplier’s product for less than the supplier has suggested without sacrificing any of his profit margin. The contract in this case merely disclaims any such unusual obligation, and since the obligation has no basis in antitrust law the disclaimer has no antitrust significance either.
State Oil also denies that the provision in the contract pertaining to Khan’s charging a price above the suggested retail price is a form of price fixing. It points out that Khan was free to charge as high a price as he wishes. This is true in the sense that it would not have been a breach of contract for Khan to raise his price. But the contract made it worthless for him to do so; and, realistically, this was just an alternative sanction to termination, and probably an equally effective one. Generally when a seller raises his price, his volume falls; and if his profit on each unit sold is frozen, the effect of his raising his price will be that he loses revenue: he will sell fewer units, at the same profit per unit. The contract, incidentally, required Khan to buy all his gasoline from State Oil; so he could not merely switch to another brand if he wanted to charge a higher price.
Practices that have the same effect are not always treated the same in law. More precisely, two practices that have one effect in common may differ in their other effects. A merger between competitors and a price-fixing agreement between competitors has the same effect in extinguishing price competition between the parties, but the merger is more likely to produce offsetting cost savings and it is therefore treated more leniently by the antitrust laws. So the fact that State [1361] Oil’s rebate scheme was as effective in deterring Khan from raising his price as a threat to terminate his lease would have been does not dictate that the two practices be treated identically under the antitrust laws. But State Oil has not identified any other relevant difference between the two methods of preventing a dealer from charging more than the suggested retail price. The purely formal character of the distinction that it urges can be seen by imagining that the contract had forbidden Khan to exceed the suggested retail price and had provided that if he violated the prohibition the sanction would be for him to remit any resulting profit to State Oil. There is no practical difference between that form of words and permitting Khan to sell at a higher price but providing that if he does so the profit belongs to State Oil.
So State Oil engaged in maximum price fixing; the next question is whether this practice is illegal per se, meaning that all the plaintiff need prove to prevail is that the defendant engaged in the practice; investigation of its actual economic effects is pretermitted. E.g., Business Electronics Corp. v. Sharp Electronics Corp., 485 U.S. 717, 723, 108 S.Ct. 1515, 1519, 99 L.Ed.2d 808 (1988); Arizona v. Maricopa County Medical Society, 457 U.S. 332, 342-44, 102 S.Ct. 2466, 2472-73, 73 L.Ed.2d 48 (1982). Challenged practices that do not fall within any of the per se categories are subject to the broader-ranging inquiry into effect and motives that goes by the name of the “rule of reason” and that requires the plaintiff to prove that the defendant’s conduct actually (or with a high likelihood) reduced competition. Id. at 343, 102 S.Ct. at 2472; Business Electronics Corp. v. Sharp Electronics Corp., supra, 485 U.S. at 723, 108 S.Ct. at 1519. Price fixing has long been illegal per se. In its usual and most pernicious form, the term refers to an agreement or conspiracy between competing firms to fix a minimum price for their product. By a modest extension it refers also to an agreement between competitors to fix either a minimum or a maximum price for the resale of their product by their dealers. See Arizona v. Maricopa County Medical Society, supra, 457 U.S. at 348, 102 S.Ct. at 2475; Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219 (1951). Why might competitors fix a minimum resale price? In order to make it more difficult for any of them to engage in undetected violations of their agreement to fix their own (that is, the wholesale) prices; a supplier who observed that he was losing sales because his competitor’s dealers were selling the competitor’s product at a low price would know that the competitor was failing to enforce the price-fixing agreement. Pauline M. Ippolito & Thomas R. Overstreet, Jr., “Resale Price Maintenance: An Economic Assessment of the Federal Trade Commission’s Case Against the Corning Glass Works,” 39 J. Law & Econ. 285, 293-94 (1996). Why might competitors fix a maximum resale price? The difference between what a supplier charges his dealer and what the dealer charges the ultimate customer is, functionally, compensation to the dealer for performing the resale service; so by agreeing on the resale prices of their goods competing sellers can reduce their dealers’ margin below the competitive price for the dealers’ service. This is a form of monopsony pricing, which is analytically the same as monopoly or cartel pricing and so treated by the law. E.g., Mandeville Island Farms, Inc. v. American Crystal Sugar Co., 334 U.S. 219, 68 S.Ct. 996, 92 L.Ed. 1328 (1948).
The questionable next step (logically, not chronologically, next) in the evolution of antitrust law was to affix the per se label to contracts in which a single supplier, not acting in concert with any of its competitors, fixed its dealers’ retail prices. Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S.Ct. 376, 55 L.Ed. 502 (1911). Here the economic difference between fixing a minimum resale price and fixing a maximum resale price becomes more pronounced, although most economists believe that neither form of price fixing is pernicious when the supplier is neither the cat’s paw of colluding distributors .nor acting in concert with his competitors. A supplier acting unilaterally might fix a minimum resale price in order to induce his dealers to furnish valuable point-of-sale services (trained salesmen, clean restrooms—whatever) to customers, which they [1362] could not afford to do without a guaranteed margin to cover the costs of the services, because the customers would use the services provided by the full-service dealers but then purchase the product from a competing dealer who could sell the product at a discount because he had not borne the expense of providing the services. Lester G. Telser, “Why Should Manufacturers Want Fair Trade?” 3 J. Law & Econ. 86 (1962); Ippoli-to & Overstreet, supra, 39 J. Law & Econ. at 294 (summarizing this and other theories of benign resale price maintenance).
As for maximum resale price fixing, unless the supplier is a monopsonist he cannot squeeze his dealers’ margins below a competitive level; the attempt to do so would just drive the dealers into the arms of a competing supplier. A supplier might, however, fix a maximum resale price in order to prevent his dealers from exploiting a monopoly position. We do not know anything about the competitive environment in which Khan and State Oil operate—which is why the district judge was right to conclude that if the rule of reason is applicable, Khan loses. But suppose that State Oil, perhaps to encourage the dealer services that we mentioned, has spaced its dealers sufficiently far apart to limit competition among them (or even given each of them an exclusive territory); and suppose further that Union 76 is a sufficiently distinctive and popular brand to give the dealers in it at least a modicum of monopoly power. Then State Oil might want to place a ceiling on the dealers’ resale prices in order to prevent them from exploiting that monopoly power fully. It would do this not out of disinterested malice, but in its commercial self-interest. The higher the price at which gasoline is resold, the smaller the volume sold, and so the lower the profit to the supplier if the higher profit per gallon at the higher price is being snared by the dealer.
Despite these points, the Supreme Court has thus far refused to reexamine the cases in which it has held that resale price fixing is illegal per se regardless of the competitive position of the price fixer or whether the price fixed is a floor or a ceiling. The key precedent so far as the present case is concerned is Albrecht v. Herald Company, 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998 (1968), a damages suit like this where the Court held over a vigorous dissent that the action of a newspaper publisher in fixing a ceiling at which its distributors could resell the newspaper to the public was illegal per se. State Oil seeks to distinguish Albrecht by pointing out that the initiative for the newspaper to take action against the plaintiff distributor had come from another distributor, giving the scheme a “horizontal” flavor. True, but this was not a factor on which the Court relied. It stated its holding broadly: maximum price fixing is illegal per se even if entirely “vertical,” that is, even if the only parties in the picture are a single supplier and a single dealer, as in this case. The Court explicitly rejected the view that “contracts between a single supplier and his many dealers to fix maximum resale prices would not violate the Sherman Act.” Id. at 152 n. 8, 88 S.Ct. at 873 n. 8. The only use the Court made of the involvement of the other distributor was to show that it was not a case in which the supplier, as permitted by the Colgate doctrine (see below), had merely cut off a dealer for failing to adhere to a suggested price. See id. at 149-50, 88 S.Ct. at 871-72. It is not cricket to distinguish a precedent by pointing to a fact mentioned by the court in the previous opinion but clearly given no weight by it. Otherwise no precedents would have any force, for no two cases are exactly alike.
State Oil points out that a supplier has the right to suggest a retail price and terminate a dealer who does not adhere to it. True, United States v. Colgate & Co., 250 U.S. 300, 39 S.Ct. 465, 63 L.Ed. 992 (1919), but irrelevant. In such a case there is no agreement between the parties and so no basis for invoking section 1 of the Sherman Act, which is limited to contracts, combinations, and conspiracies, all of which involve an element of agreement. There was an explicit agreement that Khan could not make money if he sold above the suggested retail price. Had he raised its price above that level, State Oil would have had a contractual right to a rebate.
State Oil’s main argument is that Albrecht is no longer the view of the Su[1363] preme Court. (That argument was squarely-rejected in Caribe BMW, Inc. v. Bayerische Motoren Werke Aktiengesellschaft, 19 F.3d 745, 752-54 (1st Cir.1994), so if we accepted State Oil’s argument we would be creating a circuit split.) State Oil relies on a line' of cases decided after Albrecht in which the Court established the concept of “antitrust injury.” There is no right to maintain a suit under the antitrust laws unless the defendant’s conduct has impaired the kind of interest that the antitrust laws were intended to protect. Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977). And there is no such interest in the maintenance of a monopoly price. If typically and here a resale price ceiling imposed by a seller merely prevents his dealers from reaping monopoly profits, the injury to the dealers from the ceiling— the loss, that is, of monopoly profits — will not support an antitrust suit. The requirement of proving antitrust injury is not waived in per se eases. In Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990), the Supreme Court held that a competitor of dealers prevented by their suppliers from raising their prices could not complain that the restriction was preventing him from raising his own prices.
We have considerable sympathy with the argument that Albrecht is inconsistent with the cases that establish the requirement of proving antitrust injury. In fact, we think the argument is right and that it may well portend the doom of Albrecht. In Jack Walters & Sons Corp. v. Morton Building, Inc., 737 F.2d 698, 706-07 (7th Cir.1984), we said we regarded the continued validity of Al-brecht as an open question, albeit for a different reason: that after Albrecht the Supreme Court had (reversing its previous position) recognized that exclusive dealer territories may be procompetitive. Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977). As we pointed out earlier in this opinion, and as one of the dissenting opinions in Albrecht had pointed out as well, 390 U.S. at 169, 88 S.Ct. at 881-82, a price ceiling is a natural and procompetitive incident to a scheme of territorial exclusivity. The majority opinion in Albrecht had rejected this argument on the ground that price fixing cannot be “justified because it blunts the pernicious consequences of another distribution practice,” 390 U.S. at 154, 88 S.Ct. at 874, namely exclusive territories. We now know that the consequences of that other practice are not (not generally, at any rate) pernicious. So another prop beneath Albrecht has been knocked away, although State Oil does not make the argument from Sylvania, maybe because it does not grant its dealers exclusive territories (the record is silent on the matter).
Yet despite all its infirmities, its increasingly wobbly, moth-eaten foundations (see 8 Phillip E. Areeda, Antitrust Law: An Analysis of Antitrust Principles and Their Application ¶¶ 1635-38 (1989)), Albrecht has not been expressly overruled (or, what would amount to the same thing, a later case, conceded to be indistinguishable, has not been expressly overruled). And the Supreme Court has told the lower federal courts, in increasingly emphatic, even strident, terms, not to anticipate an overruling of a decision by the Court; we are to leave the overruling to the Court itself. Thurston Motor Lines, Inc. v. Jordan K. Rand, Ltd., 460 U.S. 533, 535, 103 S.Ct. 1343, 1344, 75 L.Ed.2d 260 (1983) (per curiam); Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U.S. 477, 484, 109 S.Ct. 1917, 1921-22, 104 L.Ed.2d 526 (1989); id. at 486, 109 S.Ct. at 1923 (dissenting opinion) (“indefensible brand of judicial activism”); see also Olson v. Paine, Webber, Jackson & Curtis, Inc., 806 F.2d 731, 734, 741-42 (7th Cir.1986). Al-brecht was unsound when decided, and is inconsistent with later decisions by the Supreme Court. It should be overruled. Someday, we expect, it will be. The Supreme Court in the Atlantic Richfield case was willing to assume only “arguendo” that Albrecht had been correctly decided, 495 U.S. at 335 n. 5,110 S.Ct. at 1889 n. 5, and it cited at length the critical academic commentary on the case. Id. at 343 n. 13, 110 S.Ct. at 1894 n. 13. And in Arizona v. Maricopa County Medical Society, supra, the Court, while reaffirming the per se rule against horizontal maximum price fixing, noted that Kiefer-Stewart, unlike Albrecht, had been a [1364] case of horizontal maximum price fixing and that “horizontal restraints are generally less defensible than vertical restraints,” 457 U.S., at 348 n. 18, 102 S.Ct., at 2475 n. 18, though elsewhere the opinion quotes at length from Albrecht, and with apparent approval, a denunciation of maximum price fixing not limited to the horizontal variety. Id. at 347, 102 S.Ct. at 2474.
But all this is an aside. We have been told by our judicial superiors not to read the sibylline leaves of the U.S. Reports for prophetic clues to overruling. It is not our place to overrule Albrecht; and Albrecht cannot fairly be distinguished from this case.
It might be distinguishable if there were a class of cases in which maximum price fixing, though wholly vertical, wholly unilateral, did cause antitrust injury, even if Albrecht and the present case were not members of that class. Then, while Albrecht itself might be decided differently today, its principle that such price fixing is illegal per se would have some domain of application. In that event affirmance here would construe Albrecht narrowly but not abrogate it. But State Oil is not able to identify any cases, real or hypothetical, in which the practice condemned in Albrecht could cause an injury to the interests protected by antitrust law. If proof of antitrust injury is required in cases involving the sort of price fixing involved in Albrecht, no such case could be brought, whether by a private plaintiff or by the Department of Justice or the Federal Trade Commission.
More to the point, the Supreme Court’s conception of antitrust injury may be broader than State Oil’s. The Court has never retreated from the proposition that vertical minimum price fixing (resale price maintenance) is illegal per se. See, e.g., Business Electronics Corp. v. Sharp Electronics Corp., supra, 485 U.S. at 725-26, 108 S.Ct. at 1520. Yet, under the dealer-service theory that we sketched, and other theories that have the support of antitrust economists, resale price maintenance does not impair any interest that the antitrust laws interpreted in light of modern economics could be thought intended to protect. It increases rather than reduces competition — as the Court recognized in the Sylvania decision, dealing with the closely related area of territorial and other nonprice restrictions placed by suppliers on competition among their dealers. Yet Sylvania itself reaffirms the per se rule against vertical price restrictions. 433 U.S. at 51 n. 18, 97 S.Ct. at 2558 n. 18. The Court must think that preventing intrabrand price competition harms an interest protected by the antitrust laws even if the restriction increases competition viewed as a process for maximizing consumer welfare and even if a restriction that had similar effects but was not an explicit regulation of price would be lawful. If this is what the Court believes — and it does appear to be the Court’s current position, though not one that is easy to defend in terms of economic theory or antitrust policy — the Court may also think that interfering with the freedom of a dealer to raise prices may cause antitrust injury. We suspect not, but we cannot have sufficient confidence in our view to declare a decision of the Supreme Court that has not been expressly overruled nevertheless defunct, as we would have to do in order to agree with the district court’s ruling that the maximum price provision in the contract between State Oil and Khan was not illegal per se. In Atlantic Richfield, despite the Court’s evident skepticism about the continued soundness of Albrecht, the Court distinguished it on the ground that the dealers subject to a price ceiling imposed by their supplier, as distinct from their competitors, were the intended beneficiaries of Al-brecht. See 495 U.S. at 336-37, 110 S.Ct. at 1890. The implication is that the injury to a dealer like Khan from not being able to raise his price because of a restriction imposed by his supplier is antitrust injury.
But even if Khan, as we believe, is not ruled out of court by the concept of antitrust injury, he had to prove injury in fact to be able to maintain the suit. The only evidence of injury was in the report of his economic expert, so if the judge did not abuse his discretion in excluding the report we must affirm the dismissal of the antitrust count even though we think he was wrong to think that Albrecht made it impossible for Khan to prove antitrust injury. But we think it was an abuse of discretion.
[1365] The judge’s ground for the exclusion was that “the report turns entirely on the experience of one receiver who operated one gas station over a five-month period.” Before Khan was terminated, State Oil issued a notice of termination and, on the ground that Khan was selling inventory in which State Oil retained a security interest without reimbursing State Oil, obtained from a state court an order appointing a receiver to operate the station. He did so for five months. (The record does not reveal what happened at the end of that period. We assume that State Oil either took over the operation of the station itself or obtained another lessee.) The plaintiffs’ expert obtained the receiver’s financial records and determined from the cost and revenue figures in them that the receiver must have disregarded the price ceiling in the agreement with State Oil. The figures showed that the receiver had realized a margin on his sales of gasoline in excess of 3.25 cents. The only way he could have done this was by charging a price in excess of the suggested retail price and not rebating the higher mar-gin generated by the higher price. The expert inferred that had Khan been allowed to raise his price above the suggested retail price, he would have had a higher income, enabling him to pay his rent and therefore avert termination.
Of course competitive conditions may have changed during the time the receiver was operating the station. A dealer’s ability to raise his price depends on the demand for his product; the demand for gasoline in general, or for Union 76 relative to other brands, may have increased after the receiver took over. If so, his ability to maintain a price higher than the suggested retail price would not prove that Khan had had a similar ability. And if Khan could not have maintained a price above the suggested retail price, simply because competition would not have permitted him to do so, then he wasn’t hurt by the contractual provision of which he complains. There would be a violation of the antitrust laws, but no injury.
But this is just to say that the evidence presented by the expert was not conclusive on the subject of injury—was, indeed, very far from being conclusive. That did not make it either inadmissible or devoid of probative value. The only ground on which it could be argued to be inadmissible would be that the expert, although a Ph.D. in economics from a reputable university and an experienced consultant in antitrust economics, and hence qualified to offer expert economic evidence in this case, had failed to conduct a-study that satisfied professional norms. As we have emphasized in cases involving scientific testimony—and the principle applies to the social sciences with the same force that it does to the natural sciences—a scientist, however reputable, is not permitted to offer evidence that he has not generated by the methods he would use in his normal academic or professional work, which is to say in work undertaken without reference to or expectation of possible use in litigation. Braun v. Lorillard Inc., 84 F.3d 230, 234 (7th Cir. 1996); Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 318 (7th Cir.1996); Bammerlin v. Navistar Int’l Transportation Corp., 30 F.3d 898, 901 (7th Cir.1994). The district judge identified no basis, and State Oil can point to none, for supposing that the expert’s report flunked this test. The inference regarding the receiver’s profit margin, drawn from the station’s cost and revenue data, was straightforward, and, so far as appears, was made in just the way that an economist interested in a firm’s profit margins for reasons unrelated to litigation would make it; and likewise the inference that if Khan had enjoyed the freedom that the receiver evidently thought he had he would have charged a higher price, and made more money, than he did. If anything, the economist was overqualified to give evidence that could as easily have been given by an accountant; but overqualification is not yet a recognized basis for disqualification.
Weight is different from admissibility. An expert’s report might be admissible but so lacking in weight as not to block the granting of summary judgment for the other side. Wittmer v. Peters, 87 F.3d 916, 920 (7th Cir.1996); Mid-State Fertilizer Co. v. Exchange Nat’l Bank, 877 F.2d 1333, 1338-39 (7th Cir.1989); Hayes v. Douglas Dynamics, Inc., 8 F.3d 88, 92 (1st Cir.1993). But that is not the situation here. State Oil put [1366] in no evidence. In the absence of contrary evidence, Dr. Bamberger’s inference from his study of the receiver’s experience that Khan had been hurt by the maximum price provision of the gasoline supply contract was sufficiently plausible to defeat summary judgment on the question of injury.
The antitrust claim should not have been dismissed. We turn to the breach of contract claim. Although Khan’s primary complaint is that he was prevented from raising his price, he also complains that there were times when he wanted to lower his price on nonpremium gasoline. As we said at the outset, this claim has no standing as an antitrust claim. But State Oil concedes that its contract implicitly obligated it to suggest retail prices (and adjust its wholesale prices accordingly, to maintain the 3.25 cents margin between wholesale and suggested retail price that the contract guarantees) that were realistic in light of competitive conditions facing Khan. The only evidence Khan presented concerning those competitive conditions was evidence that on sixteen occasions during his operation of the station he had called other dealers and been told that their retail prices were lower than the suggested retail prices fixed by State Oil. ¡Jhis evidence falls far short of establishing a genuine issue of material fact concerning a breach of the contract. There is no evidence that the dealers were actual competitors of Khan or that the prices charged by those dealers were prices for the same grades of gasoline sold by Khan, which, remember, sold a gasoline that has a well-recognized brand name. The evidence is perfectly consistent with an assumption that State Oil’s suggested retail prices were competitively realistic.
So the contract claim was rightly dismissed as a matter of contract law, but we note the absence from the record of any indication of why the district judge, having dismissed the federal claims before trial, retained rather than, as is the norm in such situations, relinquished jurisdiction over the supplemental state law claim. 28 U.S.C. § 1367(c)(3); Brazinski v. Amoco Petroleum Additives Co., 6 F.3d 1176, 1182 (7th Cir.1993). (It does not appear that there is diversity of citizenship between the parties.) The presumption in favor of relinquishment when all federal claims fall out before trial is rebuttable, id., but it should not be lightly abandoned, as it is based on a legitimate and substantial concern with minimizing federal intrusion into areas of purely state law. Since the absence of merit of the supplemental claim in this case is clear as a matter of elementary contractual interpretation, so that the retention of jurisdiction for purposes of dismissing the claim did not require the district judge to speculate about the meaning of state law, we think the judge was right to retain jurisdiction rather than visit upon the parties and the state courts the burden of further litigation. As we said in Bmzinski, citing a slew of earlier cases, if the correct disposition of the supplemental claim is so clear as a matter of state law that it can be determined without a trial and without entanglement in difficult issues of state law, considerations of judicial economy counsel retention and prompt decision, rather than remission to the state comet. Id. But we remind the district courts of the presumption against retaining jurisdiction of supplemental state-law claims when the federal claims are dismissed before trial, and of the concomitant importance of stating the ground on which the court believes in a particular case that the presumption has been rebutted.
The judgment of the district court is affirmed in part and reversed in part, as explained in the opinion, and the ease is remanded for further proceedings consistent with the opinion.
AFFIRMED In PART, REVERSED In PART AND Remanded.