MERRILL, Circuit Judge:
This ease presents the question wheth-. er appellee Wyeth Laboratories, Inc., a manufacturer of Sabin polio vaccine, should be held to strict tort liability to one who took the drug and contracted polio as a result.
Appellant Glynn Richard Davis took appellee’s Type III polio vaccine at a mass immunization clinic conducted in West Yellowstone, Montana, in March, 1963. At that time he was thirty-nine years old, in good health, and engaged in the lumber business. Within thirty days he evidenced paralysis and other symptoms of polio, and has remained paralyzed from the waist down ever since. He brought suit in the District' Court for the District of Idaho asserting jurisdiction founded on diversity of citizenship and here appeals from judgment for appellee following jury trial.
I. THE BACKGROUND
A. The Vaccine
The vaccine involved is Sabin oral polio vaccine, as developed by Dr. Albert Sabin after many years of research. There are three separate types of Sabin vaccine: Type I, Type II and Type III. Each is designed to immunize the person taking the vaccine from contracting paralytic poliomyelitis from a corresponding type of polio virus. Differing from the earlier Salk vaccine, administered by injection, the Sabin vaccine utilizes live virus.
The use of a live virus polio'vaccine which could be taken orally had been under study throughout the world for a number of years. In the United States it has been determined that the Sabin vaccine strains should be licensed. Licensing was handled by The Division of Biologic Standards of the National Institutes of Health, a part of the United States Department of Health, Education and Welfare. Ultimately the United States licensed three manufacturers of the vaccine, including appellee Wyeth Laboratories. Appellee was licensed to sell Type III vaccine on May 17, 1962.
Licensing in the United States was preceded by worldwide clinical testing on between 700,000 and 1,000,000 people.
The vaccine is licensed for sale only as a prescription drug. It is usually manufactured in what the producers call “lots.” Each lot of the vaccine is man[123]*123ufactured under extremely complex and technical standards devised by The Division of Biologic Standards. The virus used in the vaccine, without regard to who manufactures it, comes from a common source. Appellee obtained from Dr. Sabin a so-called “seed” virus and this original seed, still maintained, is the parent of each separate lot of vaccine manufactured by appellee. Each individual lot is, in turn, run through a number of tests in the manufacturing laboratory.
Following manufacture and satisfactory testing within the laboratory, the lot involved in this case was sent to The Division of Biologic Standards where it was again subjected to rigorous testing. The Government, being satisfied with the result of the test, granted authorization for the release of this lot on January 31, 1963.
B. The Mass Immunization Clinics
In the fall of 1960 an advisory committee was established by the Surgeon General of the United States to review all phases of polio prevention. In February, 1962, the Communicable Disease Center of the Public Health Service, Department of Health, Education and Welfare, issued recommendations and reports of this committee. It was stated:
“With the licensing of type III monovalent vaccine, which is anticipated in the near future, the complete oral polio virus vaccine will become available. Recommendations as to planning and policies which will assure its maximally effective use are now essential for the guidance of the medical profession and official health authorities.”
The goal of this campaign to disseminate the oral vaccine was stated as “the complete elimination of paralytic poliomyelitis from the United States.” Community mass immunization centers were recommended and guidelines were given for setting them up. The following month the Surgeon General, on behalf of the Public Health Service, issued recommendations for the use of the vaccine in the 1962 season, giving further guidance for the conduct of community programs.
That month, March, 1962, representatives of the Public Health Service held a meeting with Idaho public health officials and medical association officers at which a joint release was issued recommending the holding of community clinics. The holding of such clinics in Eastern Idaho was later officially authorized by the Idaho Falls Medical Society and its Public Health Committee, who then selected appellee’s product as the vaccine to be administered. At a subsequent meeting it was decided to include West Yellowstone, Montana, in the Eastern Idaho program since there were no doctors there and the residents relied on medical facilities in- Ashton, Idaho. In the absence of a doctor the administration of the vaccine for the West Yellowstone clinic was delegated to a pharmacist.
The clinics in Eastern Idaho were originally scheduled for the fall of 1962. On September 14, 1962, a statement was issued by the Association of State and Territorial Health Officers through the subcommittee on epidemic intelligence of its committee on infectious diseases. It stated:
“The Sub-committee * * * has reviewed data showing a temporal association between the incidence of paralytic poliomyelitis and the administration of Type 3 oral poliomyelitis vaccine. The Sub-committee believes that the data indicate a causal relationship and show that a small but definite risk attends the use of presently available Type 3 oral polio vaccines. The data further suggest that the risk is almost exclusively limited to adult populations. While the Subcommittee acknowledges the presence of a small risk, it recognizes the tremendous value of oral poliomyelitis vaccines and the detrimental effect the unqualified withdrawal at this time would have on their future use. In view of the very small magnitude of the risk and the enormous potential [124]*124value of oral vaccines, the Sub-committee therefore recommends that the Surgeon General issue a statement which will:
1. Apprise the public of the nature of the risk.
2. Recommend that the non-epidemic use of Type 3 oral vaccine be restricted to preschool and school age children.
3. Recommend that the vaccine continue to be available for epidemic use.
4. Reaffirm the desirability of restricting mass application to the late Fall, Winter and Spring.”
The following day a statement was issued by the Surgeon General respecting his own special advisory committee’s review of occurrences of polio cases associated with the administration of the vaccine. It stated:
“The level of this risk can only be approximated but clearly is within range of less than 1 case per million doses. Since the cases have been concentrated among adults the risk to this group is greater; whereas, the risk to children is exceedingly slight or practically nonexistent.
The Committee therefore recommends that the use of Type III vaccine in mass campaigns be limited to preschool and school age children. Plans for mass programs using Type I and II vaccines in all age groups should continue. Furthermore, Type III vaccine is still indicated for use among adults in high risk groups, which include tourists to hyperendemic areas and persons residing in epidemic areas.
A special report is being prepared and will be sent to the members of the medical and public health profession within the next few days and will be made public.”
The special report followed on September 21, 1962. It stated:
“Present data indicate that for 1962, the paralytic poliomyelitis rate for those under 20 will be approximately 7.6 per million; for those over 20, about 0.9 per million. These rates will represent a record low for the 52 year period since the reporting began.”
It reiterated the recommendation earlier made in the Surgeon General’s statement:
“With the incidence of poliomyelitis at a low level in this country, the Committee therefore recommended that the Type III vaccine be restricted to preschool and school age children and to those adults in high risk groups, such as those travelling to hyperen-demic areas or in areas where a Type III epidemic is present or impending.
Since the vast majority of poliomyelitis cases occur among young children and since children are the principal disseminators of the virus, continued intensive immunization programs among this group are clearly indicated. If this group can be adequately immunized, the spread of the poliomyelitis viruses will be sharply restricted, if not essentially eradicated.”
In December, 1962, a further report was issued. It stated:
“It is therefore recommended: (1) that community plans for immunization be encouraged, using all three types; and (2) that immunization be emphasized for children in whom the danger of naturally occurring poliomyelitis is greatest and who serve as the natural source of poliomyelitis infection in the community. Because the need for immunization diminishes with advancing age and because potential risks of vaccine are believed by some to exist in adults, especially above the age of 30, vaccination should be used for adults only with the full recognition of its very small risk. Vaccination is especially recommended for those adults who are at higher risk of naturally occurring disease; for example, parents of young children,1 pregnant women, persons in epidemic [125]*125situations and those planning foreign travel.
Of greatest importance is the continuing vaccination of oncoming generations.”
With these recommendations before them the East Idaho officials postponed their clinics scheduled for the fall of 1962, but determined to proceed in the spring of 1963. Adults were included in their immunization program.
C. Appellee’s Participation
When Eastern Idaho chose appellee’s vaccine for its clinics, Mr. John Franklin, one of appellee’s salesmen, was assigned to handle the sales and assist in setting up the clinics. He was sent by appellee to Nevada to receive special training in setting up and conducting such clinics. Thereafter he managed the campaign for the Idaho Falls Medical Society. He furnished books to those in charge of clinics, setting forth schedules and procedures to be followed and details of the physical manner in which the clinics were to be set up and also showing sample promotional letters and advertising matter. He arranged for delivery of the vaccine from headquarters at Idaho Falls to the various clinics, including that at West Yellowstone, by the Idaho Falls Jeep Patrol. He arranged for the printing of forms and immunization cards and posters urging “KO Polio” and took charge of sending them to West Yellowstone. He organized meetings and conferred with those in charge of the separate clinics as to the procedures to be followed. For expenses incurred by him in connection with these activities he was reimbursed by the medical society.
Each person who received the vaccine was charged 25$ (although it was given free of charge if the recipient so requested). Appellant paid this amount for his dosage. Funds collected from the clinics were used to pay the medical society’s bill from Wyeth for the vaccine, with the remainder retained by the society.
D. Warnings
The Surgeon General’s March statement was the subject of a news story in a Pocatello, Idaho, newspaper and was read by Franklin. It was his first information respecting the subject and he promptly informed the medical society of the release. The society confirmed the information by a telephone call to the office of the Surgeon General.
The drug when sold to the medical society had a printed insert with each bottle containing 100 doses. This insert contained directions for use and pertinent excerpts from the Surgeon General’s report.
A fact sheet put out by appellee and contained in the book it supplied to clinics was published prior to the Surgeon General’s report and represented the vaccine as completely safe for all ages. A collection of news clippings from Idaho newspapers introduced in evidence by appellant shows not only a complete lack of warning but assurances that the vaccine was safe for all.
No effort was made by Franklin or the medical society to inform the West Yellowstone pharmacist of the existence of risk. The latter did not read the package insert, nor did appellant. The advertising posters made no disclosure of risk and none was made directly to those who took the drug in West Yellowstone. Appellant testified that he had no knowledge of the risk, relied on the posters and was convinced by the campaign’s advertising that it was his civic duty to participate.
II. THE ISSUES AS PRESENTED
Appellant stated claims founded on (1) negligent manufacture, (2) failure to warn of known dangers, (3) strict liability in tort and (4) breach of an implied warranty of fitness. The District Court dismissed all save that of breach of warranty.
We agree with the District Court that there was nothing in the record to create a jury issue as to neg[126]*126ligence in manufacture. On the contrary, the record shows scrupulous attention in the matter of preparation and testing. The resulting product was precisely what was intended.2 For this reason we also reject appellant’s claim that it was error for the -District Court to refuse to give an instruction that under Montana law appellee was held to an implied warranty that there was no “impurity” in the vaccine.
We find no error in the District Court’s choice to present the case to the jury on warranty rather than on strict liability in tort. The law as emerging is tending toward the latter treatment3 but under either approach the elements remain the same. The difference is largely one of terminology.4
We do find reversible error, however, in the manner in which the breach of warranty claim was given to the jury in the light of the court’s dismissal of the claim for breach of a duty to warn.
Appellant contends that the District Court erred in instructing, on warranty, that the test is whether the drug was “reasonably fit and reasonably safe for use by the public as a whole.” 5 He contends that the warranty was that the drug was fit and safe as to him.
As a general proposition we would question the correctness of appellant’s contention. It would, if generally applied, be equivalent to an imposition of absolute enterprise liability whereby all those who suffer unanticipated harm from the use of drugs are compensated out of profits from the sale of such products as a cost of doing business to the manufacturer. ‘' Although there are those who regard such a result as just,6 it has so far been found inappropriate for the courts to impose such a far-reaching change in the law of products liability.
Under the circumstances of this case, however, we conclude that strict liability does attach to sale of the drug to appellant and that the jury should have been so instructed, either by such an instruction as that requested by appellant or otherwise. Our conclusion in this respect is based upon our determination that a duty to warn existed, as to which none of appellant’s requested instructions was given.
[127]*127While appellant alleged negligent breach of a duty to warn as an independent claim, we regard failure to warn, where the circumstances of sale imposed that duty, as exposing the vendor to trict liability in tort7 (or to liability for breach of warranty if that approach is used).
It was stipulated below that Montana law governed this case. We can find no Montana decision in point on the issue of a drug manufacturer’s duty to warn of dangers inherent in its product. Privity of contract between buyer and seller as a prerequisite to recovery in an implied warranty action has long been abolished in that state in cases involving food, and strict liability has been imposed on those who sold it.8 It would seem that the same approach would be adopted by the Montana Supreme Court in cases involving drugs meant for internal use. Faced with the absence of controlling state precedent, we choose to assume that Montana would follow the majority of other states in finding that liability can attach to the sale of drugs, in either tort or warranty, despite lack of privity, and would adopt the views set forth below on the manufacturer’s duty to warn, of dangers in “nondefective” but potentially harmful products.9
III. THE LAW OF STRICT LIABILITY AS APPLICABLE HERE
The clearest statement of the law as it exists today is in our view that set forth in the Restatement (Second) of Torts (1965). Relevant to our case are § 402A and comments j and k. They are set forth in the margin.10
[128]*128The general proposition as there stated (subject to certain exceptions) is that strict liability shall attach to one who sells a product “in a defective condition, unreasonably dangerous” to the consumer.
At the outset we reject appellee’s contention that the rule applies only where unreasonable danger results because of an ascertainable “defect” or “impurity” in the product, and that since this product was precisely what it was intended to be there was no such defect. The true test in a case of this kind is whether the product was unreasonably dangerous.11
Comment j recognizes that to prevent a product from being unreasonably dangerous, direction or warnings as to its use must be given in appropriate cases.
Comment k deals with the unavoidably unsafe product. Here the fact that a product is dangerous does not result in strict liability if, on balance, public interest demands that it be made available notwithstanding its dangerous characteristics. This situation, as the comment notes, is especially common in the field of drugs, and, in particular, new and experimental drugs. We agree with ap-pellee that the Sabin vaccine qualifies for such treatment. As the comment stresses, however, strict liability is avoided in these situations only where sale is accompanied by proper directions and prop[129]*129er warnings.12 Thus we are returned to the problem of the duty to warn.
The duty seems clear where the drug’s danger is directed to a foreseeable and ascertainable class of persons, such as those prone to certain allergies.13 Such a warning constitutes a caution that certain persons should not take the drug; that as to certain persons it is not “fit.” There are many cases, however, particularly in the area of new drugs, where the risk, although known to exist, cannot be so narrowly limited and where knowledge does not yet explain the reason for the risk or specify those to whom it applies. It thus applies in some degree to all, or at least a significant portion, of those who take the drug. This is our case; there seems to be no certain method of isolating those adults who may be affected adversely by taking Type III Sabin vaccine.14
In such cases, then, the drug is fit and its danger is reasonable only if the balance is struck in favor of its use. Where the risk is otherwise known to the consumer, no problem is presented, since choice is available. Where not known, however, the drug can properly be marketed only in such fashion as to permit the striking of the balance; that is, by full disclosure of the existence and extent of the risk involved.
As comment k recognizes, human experimentation is essential with new drugs if essential knowledge ever is to be gained. No person, however, should be obliged to submit himself to such experimentation. If he is to submit it must be by his voluntary and informed choice or a choice made on his behalf by his physician.
In such cases, then, the drug is fit and its danger is reasonable only if the balance is struck in favor of its use. It can properly be marketed only in such fashion as to permit the striking of that balance; that is, by full disclosure of the existence and extent of the risk involved.
When Type III Sabin vaccine was first licensed by the Government in early 1962 and first manufactured and sold by Wyeth, there was no known or foreseeable risk involved in taking it. Thus Wyeth could not initially be expected to warn of unknown dangers. But its responsibility did not end there. When, after further experience, the danger became apparent a duty to warn attached.15 We do not need to fix the precise point at which it attached. Certainly by March, 1963, when appellant took the. vaccine, six months after the Surgeon General’s first report on the subject, it was the responsibility of Wyeth to see that such warning was given.
There will, of course, be cases where the personal risk, although existent and known, is so trifling in comparison with the advantage to be gained as to be de minimis. Appellee so characterizes this case. It would approach the problem from a purely statistical point of view: less than one out a million is just not unreasonable. This approach we reject. When, in a particular ease, the risk qualitatively (e. g., of death or major disability) as well as quantitatively, on balance with the end sought to be achieved, is such as to call for a true choice judg[130]*130ment, medical or personal, the warning must be given.16
Appellee contends that even under such a test no true choice situation is presented here. It asserts that “common sense and knowledge of the mainstreams of human conduct would unavoidably bring one to the conclusion” that appellant would have chosen to take the risk. It says, “Simply stated that proposition is this: A man has less than one in a million chance of contracting the dreaded disease of polio if he takes the vaccine. If he does not take the vaccine his chances of contracting polio are abundantly increased.”
We do not so read the record. The Surgeon General’s report of September, 1962, as we have quoted it, predicted that for the.1962 season only .9 persons over 20 years of age out of a million would contract polio from natural sources. While appellant was the father of two young children, he resided in an area that not only was not epidemic but whose immediate past history of incidence was extremely low. We have no way of knowing the extent to which either factor would affect the critical statistics. Thus appellant’s risk of contracting the disease without immunization was about as great (or small) as his risk of contracting it from the vaccine. Under these circumstances we cannot agree with ap-pellee that the choice to take the vaccine was clear.
We may note further that where the end sought is prevention of disease (and the likelihood of contracting the disease from natural sources is a relevant factor) the situation is a different one from that in which the disease has already struck and the end sought is relief or cure. Risks are far more readily taken in the latter case.
We conclude that the facts of this case imposed on the manufacturer a duty to warn the consumer (or make adequate provision for his being warned) as to the risks involved, and that failure to meet this duty rendered the drug unfit in the sense that it was thereby rendered unreasonably dangerous. Strict liability, then, attached to its sale in absence of warning.
Appellee contends that its duty to warn was met by Franklin’s disclosures to the medical society. It points out that its only direct sale of vaccine was to the medical society and that it was the society’s judgment and not appellee’s to proceed with the clinics.
Ordinarily in the case of prescription drugs warning to the prescribing physician is sufficient.17 In such cases the choice involved is essentially a medical one involving an assessment of medical risks in the light of the physician’s knowledge of his patient’s needs and susceptibilities. Further it is difficult under such circumstances for the manufacturer, by label or direct communication, to reach the consumer with a warning. A warning to the medical profession is in such cases the only effective means by which a warning could help the patient.18
[131]*131Here, however, although the drug was denominated a prescription drug it was not dispensed as such. It was dispensed to all comers at mass clinics without an individualized balancing by a physician of the risks involved. In such cases (as in the case of over-the-counter sales of nonprescription drugs19) warning by the manufacturer to its immediate purchaser will not suffice. The decision (that on balance and in the public interest the personal risk to the individual was worth taking) may well have been that of the medical society and not that of appellee. But just as the responsibility for choice is not one that the manufacturer can assume for all comers, neither is it one that he can allow his immediate purchaser to assume. In such cases, then, it is the responsibility of the manufacturer to see that warnings reach the consumer, either by giving warning itself or by obligating -the purchaser to give warning. Here appellee knew that warnings were not reaching the consumer. Appellee had taken an active part in setting up the mass immunization clinic program for the society and well knew that the program did not make any such provision, either in advertising prior to the clinics or at the clinics themselves. On the contrary, it attempted to assure all members of the community that they should take the vaccine.
We conclude that appellee did not meet its duty to warn.
This duty does not impose an unreasonable burden on • the manufacturer. When drugs are sold over the counter to all comers warnings normally can be given by proper labeling. Such method of giving warning was not available here, since the vaccine came in bottles never seen by the consumer. But other means of communication such as advertisements, posters, releases to be read and signed by recipients of the vaccine, or oral warnings were clearly available and could easily have been undertaken or prescribed by appellee.
Por these reasons we hold that it was error to fail to instruct the jury, either in warranty or in tort, that appellee was strictly liable if its drug caused appellant to contract polio and if appellant’s taking of the drug was without knowledge of risk.20
Reversed and remanded for new trial.