*675Opinion
MOSK, J.We consider issues relating to the liability of a manufacturer and a pharmacy for the production and sale of an allegedly defective drug, stilbestrol (DES). We will decide whether a pharmacy at which the drug was purchased may be held strictly liable for alleged defects in the product (as distinguished from ordinary negligence), and whether a manufacturer which sold 10 percent of DES nationwide may be found to have had a “substantial” share of the market for the purpose of applying the “market share” doctrine enunciated in Sindell v. Abbott Laboratories (1980) 26 Cal.3d 588 [163 Cal.Rptr. 132, 607 P.2d 924, 2 A.L.R.4th 1061].
Plaintiff filed an action for personal injuries allegedly resulting from DES taken by her mother in 1951 and 1952 during pregnancy for the purpose of reducing the risk of miscarriage. The complaint sought damages on the theory of strict liability, alleging that the drug was defectively designed, with the result that plaintiff developed clear cell adenocarcinoma at the age of 23. As defendants, plaintiff joined Exclusive Prescription Pharmacy Corporation (Exclusive) where plaintiff’s mother purchased the DES, and E.R. Squibb & Sons, Inc. (Squibb). The first cause of action alleged that Squibb was the manufacturer of the DES used by plaintiff’s mother. The second count, added after our decision in Sindell, alleged that plaintiff was unable to identify the manufacturer, but that Squibb supplied a “substantial percentage” of DES for use by pregnant women to prevent miscarriage.
Before jury selection began, the court granted Exclusive’s motion for judgment on the pleadings, holding that a pharmacy may not be held strictly liable for dispensing a prescription drug. The court determined that Exclusive rendered a professional service in supplying the DES, that the consumer of the drug was the doctor who prescribed it rather than plaintiff’s mother, and that as a matter of policy the doctrine of strict liability should not be extended to a pharmacy.
In support of her second cause of action plaintiff offered to prove that Squibb sold 10 percent of the DES in the national market. The court ruled that as a matter of law 10 percent of the national market was not a “substantial percentage” within the meaning of Sindell, and it dismissed the second cause of action. The matter went to trial against Squibb on the first cause of action alleging that Squibb had actually supplied the DES taken by plaintiff’s mother.1 The parties introduced evidence on whether Squibb was the manufacturer of the offending drug, and the trial court instructed the *676jury that plaintiff had the burden of proof on this issue. The jury returned a special verdict, finding that plaintiff’s mother had purchased the DES at Exclusive, and that the DES which she purchased was not manufactured by Squibb. The trial court entered judgment in favor of defendants, and plaintiff appeals.
The Action Against Exclusive
In the seminal case of Greenman v. Yuba Power Products, Inc. (1963) 59 Cal.2d 57 [27 Cal.Rptr. 697, 377 P.2d 897, 13 A.L.R.3d 1049], Justice Tray nor, writing for the court, held a manufacturer strictly liable in tort for injuries caused by a defective product which it knew would be used without inspection for defects. In Vandermark v. Ford Motor Co. (1964) 61 Cal.2d 256 [37 Cal.Rptr. 896, 391 P.2d 168], the strict liability doctrine was extended to retailers of defective products.
Plaintiff asserts that a pharmacy which sells prescription drugs is in the same position as a retailer of any other consumer product, and that the reasons advanced in Greenman and Vandermark for imposing strict liability necessarily apply to a pharmacy. Exclusive counters that a pharmacist who dispenses a prescription drug is primarily furnishing a service rather than selling a product, and that the rationale underlying imposition of strict liability does not justify application of the doctrine to him.
Before reaching the merits of these conflicting claims, we observe that pharmacists perform a broad range of tasks, from selling razor blades and dental floss to treating patients in a health care facility by ordering laboratory tests and administering drugs by injection (Bus. & Prof. Code, § 4046, subd. (c)(4)) and acting as consultants regarding medication prescribed for patients at such facilities (Cal. Admin. Code, tit. 22, § 72375). The discussion which follows relates only to the duties in a community pharmacy of a pharmacist who fills prescriptions for drugs on the order of a physician or other medical care provider, and who has used due care in compounding and labelling the drug.
There are no cases in California deciding whether a retail pharmacy is strictly liable for injuries caused by an inherent defect in a drug. In Florida (McLeod v. W.S. Merrell Co. (Fla. 1965) 174 So.2d 736, 739), North Carolina (Batiste v. American Home Products Corp. (1977) 32 N.C.App. 1 [231 S.E.2d 269, 275]) and New York (Bidder v. Willing (1977) 58 App.Div.2d 331 [397 N.Y.Supp.2d 57, 59-60]) a pharmacy is held not to be strictly liable for defects in a prescription drug. These cases rely on section 402A of the Restatement Second of Torts, which declares that unavoidably unsafe products such as drugs are not defective if they are accom*677panied by an appropriate warning (id., com. k), and that a seller is only required to warn of defects of which he knew or should have known (id., com. j).
It is critical to the issue posed to determine if the dominant role of a pharmacist in supplying a prescription drug should be characterized as the performance of a service or the sale of a product. Both parties accept as a general rule that “those who sell their services for the guidance of others . . . are not liable in the absence of negligence or intentional misconduct.” (Gagne v. Bertran (1954) 43 Cal.2d 481, 487 [275 P.2d 15].) This principle was applied in Carmichael v. Reitz (1971) 17 Cal.App.3d 958 [95 Cal.Rptr. 381], the primary authority on which the trial court relied. A doctor who prescribed a drug purchased by the patient at a pharmacy was held not to be strictly liable for injuries resulting from its use. In response to the argument of the plaintiff that the doctor was “in a sense a retailer” of the drug, the court stated that the doctor prescribed the medication only as an aid to effect a cure and was not in the business of selling the drug. The court observed, “[T]he distinction between a transaction where the primary objective is the acquisition of ownership or use of a product and one where the dominant purpose is to obtain services has not been obliterated. Where the services sought are professional in character, the distinction applies a fortiori. ” (Id., at p. 978.) On this theory, a hospital was held not strictly liable as the “supplier” for injuries arising out of the use of a surgical needle which broke during an operation. (Silverhart v. Mount Zion Hospital (1971) 20 Cal.App.3d 1022, 1027 [98 Cal.Rptr. 187, 54 A.L.R.3d 250].)
Magrine v. Krasnica (1967) 94 N.J.Super. 228 [227 A.2d 539], held that a dentist was not strictly liable for injuries caused by a needle which broke during the course of treatment due to a latent defect. The court characterized the difference between a sale and a service as follows: “[T]he essence of the transaction between the retail seller and the consumer relates to the article sold. The seller is in the business of supplying the product to the consumer. It is that, and that alone, for which he is paid. A dentist or physician offers, and is paid for, his professional services and skill. That is the essence of the relationship between him and his patient.” (Id., at p. 543.) (Accord: Babcock v. Nudelman (1937) 367 Ill. 626 [12 N.E.2d 635] [holding optometrists and oculists who examined eyes and sold eyeglasses, exempt from a tax imposed on retailers of tangible goods on the ground that their main objective was to provide a service and that the sale of eyeglasses was purely incidental to the service]; see also Barbee v. Rogers (Tex. 1968) 425 S.W.2d 342.)
As might be anticipated, the parties differ sharply as to whether the main function of a pharmacist is to provide a service or to sell a product. Plaintiff *678asserts that the duties of a pharmacist in filling a prescription do not differ from those of any other retailer: he reads the prescription, fills the container with the proper type and dosage of the medication required, types up a label, attaches it to the container, and exchanges the medication for payment by the customer. In essence, argues plaintiff, a pharmacist is the functional equivalent of “an experienced clerk at a hardware store.”
Exclusive and amici curiae2 paint a dramatically different picture of the role of the pharmacist, characterizing him as a professional who provides an important health service. They point out that with a few exceptions specified by statute, only a physician or a licensed pharmacist may compound or dispense prescription drugs (Bus. & Prof. Code, § 4050 et seq.).3 In order to obtain a license, a pharmacist must have graduated from a four-year college of pharmacy or the department of pharmacy of a university, have one year of practical experience under the supervision of a registered pharmacist, and pass a written examination given by the California State Board of Pharmacy (Board) (§§ 4085, 4087). He must comply with continuing education requirements as a condition to renewal of his license (§ 4098.5), and is subject to rules of professional responsibility and to disciplinary proceedings for violation of those rules (§ 4350 et seq.).
A pharmacist is required not only to assure that the drug prescribed is properly selected, measured and labelled but, according to the Board, he must be alert to errors in prescriptions written by doctors, and contact the doctor in case of doubts or questions regarding the drug prescribed. In addition, the pharmacist may discuss with the patient the proper use of the drug and the potential side effects, and must be aware of the possibility of harmful interaction between various medications which the pharmacist knows the patient is using. According to the Board, about 22 percent of patients are counseled by the pharmacist, who spends one and one-half to two hours a day in such consultation. (Citing Cal. Pharm. (Jan. 1985) 1984 Lilly Digest Rep., p. 33, table 19.)
It seems clear to us that the pharmacist is engaged in a hybrid enterprise, combining the performance of services and the sale of prescription drugs. It is pure hyperbole to suggest, as does plaintiff, that the role of the pharmacist is similar to that of a clerk in an ordinary retail store. With a few exceptions, only a licensed pharmacist may dispense prescription drugs, and as indicated above there are stringent educational and professional require*679ments for obtaining and retaining a license. A pharmacist must not only use skill and care in accurately filling and labelling a prescribed drug, but he must be aware of problems regarding the medication, and on occasion he provides doctors as well as patients with advice regarding such problems. In counseling patients, he imparts the same kind of information as would a medical doctor about the effects of the drugs prescribed. A key factor is that the pharmacist who fills a prescription is in a different position from the ordinary retailer because he cannot offer a prescription for sale except by order of the doctor. In this respect, he is providing a service to the doctor and acting as an extension of the doctor in the same sense as a technician who takes an X-ray or analyzes a blood sample on a doctor’s order.
Nevertheless, it cannot be disputed that a sale in fact occurs. There is an obvious distinction between the doctor who provides a patient with a prescription for a defective drug (Carmichael, supra, 17 Cal.App.3d 958), a dentist who uses a faulty drill (Magrine, supra, 227 A.2d 539) or a hospital that uses a defective needle during surgery (Silverhart, supra, 20 Cal.App.3d 1022), and a pharmacist who fills a prescription. The pharmacist is in the business of selling prescription drugs, and his role begins and ends with the sale. His services are rendered only in connection with the sale, and a patient who goes to a pharmacy to have a prescription filled generally is seeking to purchase the drug rather than to obtain the advice of the pharmacist.
By contrast, the doctor, dentist and hospital in the cases cited above are not in the business of selling the drug or device; they use the product in the course of treatment as one element in their efforts to effect a cure, and furnishing the services does not depend on sale of a product.
Ordinarily, in deciding whether the sale or service aspect of an enterprise predominates, we would confine our consideration to the type of factors discussed above. In the case of a pharmacy, however, we must broaden our inquiry. The Legislature has provided in section 4046 that the practice of pharmacy is not only a profession (subd. (a)), but also a “dynamic patient-oriented health service that applies a scientific body of knowledge to improve and promote patient health by means of appropriate drug use and drug related therapy” (subd. (b)). (Italics added.)
This declaration that a pharmacist provides a health service is similar to section 1606 of the Health and Safety Code, which declares that the distribution or use of whole blood or plasma is a service for all purposes and shall not be construed to be a sale. The object of section 1606 is to avoid application of the doctrine of strict liability to those who make or sell blood or blood plasma, or to those who use it in blood transfusions, thereby *680promoting the constant availability of an adequate supply of blood. (McDonald v. Sacramento Medical Foundation Blood Bank (1976) 62 Cal.App.3d 866, 872 [133 Cal.Rptr. 444].) In keeping with this purpose, it is held that a hospital using blood in a transfusion (Shepard v. Alexian Brothers Hosp. (1973) 33 Cal.App.3d 606 [109 Cal.Rptr. 132]), a blood bank supplying the blood (McDonald) and a manufacturer of blood plasma selling it for transfusion (Fogo v. Cutter Laboratories, Inc. (1977) 68 Cal.App.3d 744 [137 Cal.Rptr. 417]) are immune from strict liability because the Legislature has declared in section 1606 that they are providing a service rather than making a sale.
It is true that section 1606 of the Health and Safety Code provides not only that the distribution and use of blood and blood plasma is a service, but also that they should be deemed not to constitute a sale, whereas section 4046, subdivision (b), does not expressly declare that a sale is not involved in the practice of pharmacy. Nevertheless, the Legislature could not have been unaware when it enacted this provision that a pharmacist’s main function is to fill prescriptions for medications that are sold to the public. Indeed, from a comparative functional aspect, a pharmacist’s role is more aptly characterized as the performance of a service than that of the manufacturer and seller of blood plasma. The Legislature must have intended, therefore, that even though a pharmacist is paid for the medication he dispenses, his conduct in filling a prescription is to be deemed a service, and, like the manufacturer of blood plasma, a pharmacy is immune from strict liability.
We reject plaintiff’s suggestion that because subdivision (b) of section 4046 was enacted at the same time as subdivision (c) of the section (which authorizes a pharmacist to order laboratory tests and give injections in health care facilities) the declaration that the practice of pharmacy provides a “health service” was intended to apply only to a pharmacist acting in that capacity. (Stats. 1981, ch. 989, § 1, p. 3830.) To the contrary, the broad wording of subdivision (b) indicates that it applies to the practice of pharmacy in general, and we so construe it.
There is no definitive legislative history of subdivision (b), and we are not certain, therefore, of the Legislature’s motivation in shielding pharmacies from strict liability. The Legislature may have determined that it is not in the public interest to subject them to strict liability, because (like the need to assure an adequate supply of blood plasma) the wide availability of a full range of prescription drugs at economical cost outweighs the advantage to the individual consumer of being able to recover for injuries on a strict liability basis rather than to be limited to claims arising from negligence.
If pharmacies were held strictly liable for the drugs they dispense, some of them, to avoid liability, might restrict availability by refusing to dispense *681drugs which pose even a potentially remote risk of harm, although such medications may be essential to the health or even the survival of patients. Furthermore, in order to assure thát a pharmacy receives the maximum protection in the event of suit for defects in a drug, the pharmacist may select the more expensive product made by an established manufacturer when he has a choice of several brands of the same drug. As the Board’s amicus brief warns, “Why choose a new company’s inexpensive product, which has received excellent reviews in the literature for its quality, over the more expensive product of an established multinational corporation which will certainly have assets available for purpose of indemnification 10, 20, or 30 years down the line?”
The Legislature might have decided also that, since the doctor who ordered the drug provided by the pharmacy cannot be held strictly liable for its defects and in some circumstances the manufacturer who created the defect can also escape liability, it would be unfair and burdensome to expose the pharmacy alone to strict liability since it may provide the drug only on a doctor’s prescription, which the pharmacist must strictly follow.4 Plaintiff emphasizes that some pharmacies are owned by large chain store operations which are as capable as large drug manufacturers to respond in damages. But most pharmacies are not in this category. According to the Board, approximately 3,385 pharmacy permits are held in California by independent operators of 3 or fewer pharmacies, many of which are single drugstores owned by the operating pharmacist. Only 1,165 permits are held by operators of 4 or more pharmacies, and 642 permits are held by hospital pharmacies.
Finally, plaintiff contends that even if the pharmacist is personally exempt from strict liability because he provides a service, the “merchandising organization which employs him . . . should not be so exempted.” Plaintiff cites no authority in support of this claim, and we perceive no basis in law or rationale for accepting it. The fact that a pharmacy may be owned by an enterprise which also deals in ordinary merchandise does not justify the conclusion that it should be held strictly liable when it performs a service. Moreover, the policy justifying the grant of immunity from strict liability to the practice of pharmacy would only be effectuated if the pharmacy operation itself is exempted.
For the reasons stated above, we conclude that the trial court was correct in granting judgment on the pleadings to Exclusive.
*682 The Action Against Squibb
Plaintiff alleged in her second cause of action that she could not identify the specific manufacturer of the drug taken by her mother and that Squibb “supplied a substantial percentage” of the drug to prevent miscarriage.
By these allegations, plaintiff sought recovery under the “market share” theory advanced in Sindell v. Abbott Laboratories, supra, 26 Cal.3d 588. In that case, the plaintiff charged in her complaint that defendant manufacturers of DES, with knowledge that it might cause cancer in the daughters of the mothers who took the drug, failed to adequately test it for safety or to warn of its dangers. She could not identify the producer of the drug actually ingested by her mother, and for that reason the trial court sustained defendants’ demurrers to the complaint.
We reversed the judgment, reasoning substantially as follows: The general rule is that the burden of proof is on the plaintiff to establish that the injuries she suffered were caused by the conduct of the defendant. This rule is not without its exceptions, however, including the doctrine of “alternative liability” applied in Summers v. Tice (1948) 33 Cal.2d 80 [199 P.2d 1, 5 A.L.R.2d 91]. In Summers, the plaintiff had suffered an injury to his eye after the two defendants had each shot a gun in his direction. The plaintiff was unable to demonstrate which of the defendants was responsible for his injury. It was held that he was not barred from pursuing the action, because the circumstances justified shifting the burden of proof to the two defendants to absolve themselves if they could. This would not be unfair, it was reasoned, because both had acted negligently toward the plaintiff, and they would both escape liability if the plaintiff were forced to choose between them and was unable to isolate which was responsible. We declined to apply this holding without modification to the drug manufacturer defendants in Sindell.
Nevertheless, we held in Sindell, because of policy considerations spelled out in our opinion, that plaintiff and those in her position should not be bereft of a remedy. We stated that the likelihood that any of the defendants caused the plaintiff’s injuries should be measured not by the number of DES manufacturers joined as defendants—only five of two hundred manufacturers were defendants in the action—but by the percentage which the DES sold by each of them for the purpose of preventing miscarriage bore to the entire production of the drug sold by all producers of the identical formula for that purpose. If plaintiff joined in the action the manufacturers of “a substantial share of the DES which her mother might have taken, the injustice of shifting the burden of proof to defendants to demonstrate that they could not have made the substance which injured plaintiff” would be con*683siderably diminished. (26 Cal.3d at p. 612.) We recognized that there were practical problems involved in defining the market and determining market share, but held that these were matters of proof which could not be determined at the pleading stage. (Id., at p. 613.)
The Sindell decision has generated numerous commentaries. (E.g., DES Ruling Shakes Product Liability Field (1980) 66 A.B.A. J. 827; Note (1982) 18 Cal. Western L.Rev. 143; Note (1981) 11 Golden Gate L.Rev. 917; Note (1981) 94 Harv.L.Rev. 668; Note (1982) U. Ill.L.Rev. 1003; Note (1981) 60 Neb.L.Rev. 432; Note (1981) Okla.L.Rev. 843; Note (1981) 8 Pepperdine L.Rev. 1011; Note (1981) 33 Stan.L.Rev. 937; Note (1981) Utah L.Rev. 655; Comment (1981) 26 Vill.L.Rev. 997.) Some of these comments relate to the appropriate composition of the market for application of the market share doctrine and the meaning of the term “substantial share” as used in the opinion. (E.g., Fischer, Products Liability—An Analysis of Market Share Liability (1981) 34 Vand.L.Rev. 1623, 1642-1645; Note (1981) 69 Cal.L.Rev. 1179, 1189-1193, 1197-1199; Comment, The Market Share Theory (1981) 19 Hous.L.Rev. 107, 128-129.)
At the trial level here, plaintiff confined her offer of proof to a national market for DES. Our only inquiry, therefore, is whether, as the trial court determined, a 10 percent share of this market is insufficient as a matter of law to allow plaintiff to proceed on the basis of the market share doctrine.
Plaintiff appears to contend that the doctrine was designed to accomplish a fair approximation of the damages which each defendant DES manufacturer would be required to pay under the principles of comparative fault (American Motorcycle Assn. v. Superior Court (1978) 20 Cal.3d 578 [146 Cal.Rptr. 182, 578 P.2d 899]), and that its applicability is unrelated to whether plaintiff has joined as defendants the manufacturers of a substantial share of the DES market. This claim lacks merit. The major issue decided in Sindell was whether and the circumstances under which a plaintiff in a DES case could avoid application of the usual rule that she had the burden of proving the defendant manufacturer produced the DES which caused her injuries. The opinion makes it clear that in order to shift the burden of proof on the issue of causation in fact, a plaintiff must join in the action the manufacturers of “a substantial share of the DES which her mother might have taken.” (26 Cal.3d at p. 612.) Although we stated that the defendant manufacturers could cross-complain against other DES manufacturers not joined in the action, which might have supplied the injury-causing product, we were careful to qualify the statement with the observation that such pleadings would be filed by defendants only after “plaintiff has met her burden of joining the required defendants.” (Ibid.)
*684We must determine, therefore, whether Squibb’s 10 percent market share is a substantial percentage of the market for the application of the rule laid down in Sindell. Plaintiff, relying on general definitions of the word “substantial” asserts that the term must be defined in the context of a particular case,5 and since Squibb was alleged to be the second largest seller of DES in the country, its 10 percent market share must be deemed substantial in the framework of DES litigation.
We reject this contention because it is contrary to the theoretical justification underlying the market share doctrine. We pointed out in Sindell that a major reason why shifting the burden of proof from plaintiff to defendants was warranted in Summers was that both parties who were or could have been responsible for the harm to the plaintiff were joined as defendants. Thus, there was a 50 percent chance that one of the defendants was responsible for the injury. (26 Cal.3d at p. 602.) We declined to apply an unmodified Summers rationale to the facts in Sindell, because only five of the two hundred manufacturers of the DES which could have harmed plaintiff were before the court, and therefore there was “no rational basis upon which to infer that any defendant in this action caused plaintiff’s injuries, nor even a reasonable possibility that they were responsible.” (Id., at pp. 602-603.) Instead, we concluded that the likelihood that one of the defendants supplied the DES should be determined not by the number of manufacturers joined in the action but by the percentage which the DES sold by each to prevent miscarriage bore to the entire production of the drug sold for that purpose. We held that if the plaintiff joined in the action the manufacturers of a substantial share of the DES which her mother might have taken, the injustice of shifting the burden of proof to defendants to exonerate themselves would be significantly diminished. We declined to declare a specific percentage of the market which would satisfy application of the doctrine, but stated only that it must be substantial.
Since Squibb had only a 10 percent share of the DES market, there is only a 10 percent chance that it produced the drug causing plaintiff’s injuries, and a 90 percent chance that another manufacturer was the producer. In this circumstance, it must be concluded that she failed to meet the threshold requirement for the application of the market share doctrine. The trial court was justified in ruling, therefore, that she could not proceed to trial on the second cause of action.
*685The trial of the first cause of action, which was confined to the question whether Squibb was in fact the manufacturer of the drug taken by plaintiff’s mother, concluded with a jury verdict in Squibb’s favor. Since plaintiff had the burden of proof on this issue, her contention that the trial court erred in instructing the jury to that effect is without merit.
The conclusion that Exclusive may not be held strictly liable and that plaintiff cannot proceed against Squibb on a market share theory makes it unnecessary to discuss other issues raised by plaintiff.
The judgment is affirmed.
Reynoso, J., concurred.
The trial court granted Squibb’s motion to bifurcate the trial into two phases, with the question of product identification to be tried first.
Amici curiae briefs on behalf of Exclusive were filed by the American Pharmaceutical Association and California Pharmacists Association, as well as by the California State Board of Pharmacy and the Califonia Medical Association.
All further statutory references are to the Business and Professions Code unless otherwise noted.
The only discretion allowed the pharmacist is the substitution of a generic equivalent, unless the doctor specifies otherwise. (§ 4047.6.)
In Robinson v. Northern American Life & Cas. Co. (1963) 215 Cal.App.2d 111, 117 [30 Cal.Rptr. 57], the term “substantial” was defined as “not seeming or imaginary, not illusive, real, true, important, essential, material, having good substance .... It means considerable in amount, time, or the like . . . .” Atchison etc. Ry. Co. v. Kings County Water Dist. (1956) 47 Cal.2d 140, 144 [302 P.2d 1], described the word as “a relative term, its measure to be gauged by all the circumstances surrounding the matter in reference to which the expression has been used.”