leagues’ affirmance of the judgment entered after Kaiser Gypsum’s directed verdict. However, I dissent from their reversal of the judgments entered in favor of Bridgestone/Firestone and Wagner. I respectfully suggest the trial court was correct in granting nonsuit on the third cause of action for enterprise liability. My dissent is not based on an analysis of whether or not Wheeler v. Raybestos-Manhattan (1992) 8 Cal.App.4th 1152 [11 Cal.Rptr.2d 109] (Wheeler) should be applied retroactively; rather, it is based on my firm belief that Wheeler itself was wrongly decided. My colleagues in the majority correctly observe that the parties herein do not challenge Wheeler, and Wheeler successfully withstood petition for review in our Supreme Court. *342However, the denial of review is not an expression by our Supreme Court of the correctness of Wheeler,1 and I cannot in good conscience debate the retroactivity of an opinion wrongly decided, even if the parties do not take issue. Wheeler is not only contrary to all well-established authority, but it is not soundly reasoned. Liability for injury caused by exposure to asbestos brake products should not be subject to the market share theory announced in Sindell v. Abbott Laboratories (1980) 26 Cal.3d 588 [163 Cal.Rptr. 132, 607 P.2d 924] (Sindell). I respectfully urge that Wheeler is now ripe for review by our Supreme Court.
Wheeler Is Contrary to Precedent
There is simply no rational basis for Wheeler's ignoring this very court’s decision in Mullen v. Armstrong World Industries, Inc. (1988) 200 Cal.App.3d 250, 257 [246 Cal.Rptr. 32] (Mullen) in which we “align[ed] California, the progenitor of the market share theory of liability, with the great majority of jurisdictions which have declined to extend it to the field of asbestos-related injuries.” The “numerous inherent differences between DES and asbestos” (id. at p. 255) which we found in 1988 still existed in 1992, when my colleagues decided Wheeler. Earlier we recognized that diethylstilbesterol (DES) was produced by hundreds of companies pursuant to one formula and that all DES had “identical physical properties and chemical compositions” (id. at pp. 255-256) and created the same risk of harm to all pregnant women. In 1988 we observed that “ ‘Asbestos products, on the other hand, have widely divergent toxicities . . . (Id. at p. 256.) We recognized then, but apparently not four years later, that “ ‘. . . some asbestos products presented] a much greater risk of harm than others.’ ” (Ibid.) We found in 1988 that “This divergence is caused by a combination of factors, including: the specific type of asbestos fiber incorporated into the product; the physical properties of the product itself; and the percentage of asbestos used in the product." (Ibid., italics added.) And we noted other differences in asbestos products, including the geographical origin of the mineral, the form in which the asbestos fibers are packed, and the concentration of asbestos fibers, citing Celotex Corp. v. Copeland (Fla. 1985) 471 So.2d 533, 537-538.
*343In 1988 we characterized Sindell as involving “a group of 11 defendants all ‘engaged in the business of manufacturing, promoting, and marketing’ a single product ‘produced . . . from an identical formula.’ ” (Mullen, supra, 200 Cal.App.3d at p. 256.) We contrasted asbestos as “a generic designation possessing a rainbow-like diversity and a bewildering array of potential uses.” (Ibid., fn. omitted.) We found in Mullen that “. . . plaintiffs áre attacking an entire industry, not seeking recovery for damages caused by a single fungible product ‘carr[y]ing with it a singular risk factor.’ ” (Id. at p. 257.)
Without distinguishing Mullen, and making only passing historical reference to it, Wheeler ignores its teaching and now finds asbestos, at least in brake pads, to be “fungible.” Wheeler concedes that brake pads “are not manufactured from one single chemical formula like DES” (Wheeler, supra, 8 Cal.App.4th at p. 1156) and dismisses as “irrelevant, of course,” the fact pads come in various sizes and shapes. Wheeler then boldly asserts, without any citation to evidence or authority, that “A single type of asbestos fiber, chrysotile, was used in all the pads, and the amount of asbestos by weight in the pads varied within a limited range.” (Ibid.) Hence, brake pads are now “fungible.” Interestingly, the Mullen court specifically found San Francisco General Order No. 21, which rejects application of Sindell to asbestos cases, “[e]qually worthy of mention” (Mullen, supra, 200 Cal.App.3d at p. 258, fn. 8)—the very order which Wheeler recognizes as based on “a variety of sound reasons” (Wheeler, supra, at p. 1155) and then finds “inapplicable” (Id. at p. 1158) to its facts.
Between 1988 and 1992 nothing changed. Asbestos was still as generic as before, general order No. 21 was concededly as “sound” as when adopted, and no new appellate decision or commentary had appeared. Indeed, Wheeler is devoid of both (1) citation to any authority in support of its holding and (2) any good reason for rejecting Mullen as controlling.
Wheeler Was Wrongly Decided
Wheeler's examination of our Supreme Court’s decision in Sindell, supra, 26 Cal.3d 588 leads it to the following conclusion: “While brake pads are not absolutely interchangeable each for one another and. hence are not fungible from the standpoint of an auto mechanic, they are fungible for the purposes of Sindell by virtue of containing roughly comparable quantities of the single asbestos fiber, chrysotile.” (Wheeler, supra, 8 Cal.App.4th at p. 1156.) That conclusion finds no support either in the alleged facts or in law.
*344First, the Wheeler court bases its conclusion on “fungibility” on the erroneous assumption that brake shoes contain “roughly comparable quantities of . . . chrysotile.” (Wheeler, supra, 8 Cal.App.4th at p. 1156.) The court distinguishes the case before it from an Ohio case (Goldman v. Johns-Manville Sales Corp. (1987) 33 Ohio St.3d 40 [514 N.E.2d 691]) in which the duct tape manufactured by the various defendants varied in asbestos content from 15 percent to 100 percent. The Wheeler court then makes the following fatal leap of faith: “In this case the asbestos content of the brake pads is not identical, but it varies in a much more restricted range. Therefore the risk of harm posed by the products of each manufacturer is more nearly equivalent.” (Wheeler, supra, 8 Cal.App.4th at pp. 1156-1157.)
The so-called “restricted” range referenced by the Wheeler court is, in reality, quite unrestricted. Wheeler concedes that asbestos content in brake products ranges from 40 percent to 60 percent by weight. (Wheeler, supra, 8 Cal.App.4th at pp. 1156-1157.) Such a variance means asbestos content in brake shoes varies by as much as 50 percent from one manufacturer to another—well within the 15 percent to 100 percent variance which caused the Goldman court to exclude applicability of the market share theory of liability. How can a variance of 50 percent between two products reasonably be classified as “a much more restricted range” than Goldman's 15 percent to 100 percent (id. at pp. 1156-1157), let alone meet SindelV s requirement of 0 percent variance?
Second, the Wheeler court notes but ultimately ignores two other key points raised by the defendants: (a) chrysotile fibers may come from various geographic sources; and (b) not all pads are (were) manufactured with the same bonding agents. (Wheeler, supra, 8 Cal.App.4th at p. 1156.) In Mullen, we noted that “ '. . .it has been established that the geographical origin of the mineral can affect the substance’s harmful effects. . . (Mullen, supra, 200 Cal.App.3d at p. 256.) If that factor were significant in our analysis of fungibility with regard to asbestos-related injuries in general, it should certainly have influenced the Wheeler court’s decision about fungibility where asbestos-containing brake pads are concerned. Just as importantly, the differences in bonding agents may well affect the amount of asbestos released over time—a factor which is very significant when the (Wheeler) plaintiffs were mechanics who allegedly worked on worn brakes over some undefined period of time.
In sum, I respectfully suggest that brake pads are not fungible within the meaning of Sindell. As such, the Wheeler court was wrong. To apply Wheeler *345retroactively only compounds the error. I would affirm the judgments of the respected trial court in all particulars.
Respondents’ petition for review by the Supreme Court was denied April 28, 1994. Lucas, C. J., was of the opinion that the petition should be granted.
This has been the law since our Courts of Appeal were created. In People v. Davis (1905) 147 Cal. 346 [81 P. 718] “. . . the Supreme Court flatly rejected the notion that its discretionary denial of a hearing, for undisclosed reasons, could be interpreted as a positive approval, and adoption of the point of law decided by the court of appeal.” (9 Witkin, Cal. Procedure (3d ed. 1985) Appeal, § 776, p. 745.) The latest word on the subject, according to Witkin, comes from the advisory committee on revised rule 28: “ ‘Adoption of the new “review” procedure does not affect this legal doctrine, and denial of review will not be an expression of the opinion of the Supreme Court on the correctness of the judgment of the Court of Appeal or on the correctness of any discussion in the Court of Appeal opinion.’ ” (Id. at p. 747.)