Garland Moore complains of injuries he alleged were caused by a nurse negligently injecting him with medicine intravenously instead of intramuscularly in the prescribed manner. After Moore rested his case, the district court entered judgment upon motions for directed verdicts in favor of Dr. William Guthrie, Moore’s attending physician, and Guthrie Hospital, where Moore was a patient. We hold that the evidence conclusively shows the doctor was free of fault and affirm the district court’s judgment in favor of him. Moore’s case against the hospital stands upon a different footing. The Seventh Amendment requires that this claim must be submitted to the jury, for the evidence against the hospital — viewed in the light most favorable to Moore — affords a rational choice for competing inferences. Atlantic Coast Line R. R. v. Truett, 249 F.2d 215, 217 (4th Cir. 1957) ; Burcham v. J. P. Stevens & Co., 209 F.2d 35, 38 (4th Cir. 1954).
I.
In considering the case against Dr. Guthrie, we start with the proposition: “A prima facie case of medical *368malpractice must normally consist of evidence which establishes the applicable standard of care, demonstrates that this standard has been violated, and develops a causal relationship between the violation and the harm complained of.” Kosberg v. Washington Hospital Center, Inc., 394 F.2d 947, 949 (D.C.Cir.1968). Cf. Schroeder v. Adkins, 149 W.Va. 400, 141 S.E.2d 352, 358 (1965). The evidence is uncontradicted that intramuscular injections of chymar and penicillin were proper medication for Moore and that it was an acceptable practice to allow a nurse to administer the injections in the absence of the doctor. The nurse was not the doctor’s agent, and her negligence, if proved, cannot be imputed to him. Hohenthal v. Smith, 72 App.D.C. 343, 114 F.2d 494 (1940). We do not agree with Moore that the doctor can be held liable under the theory of res ipsa loquitur 1 In West Virginia, res ipsa loquitur is not generally applicable to malpractice cases against physicians. The Supreme Court of Appeals of West Virginia has observed that lack of skill and competency of a physician cannot be inferred from the failure to effect a cure, because the doctor may do everything reasonably expected of a competent physician without achieving a successful result. See Vaughan v. Memorial Hospital, 100 W.Va. 290, 294, 130 S.E. 481, 482 (1925) and 103 W.Va. 156, 163, 136 S.E. 837, 840 (1927). We are confident the West Virginia Court would not apply the doctrine when, as in the case before us, the proof conclusively exonerates the doctor. 2
II.
Under West Virginia law a private hospital may be held liable for injury to a patient resulting from the negligence of nurses employed by it. Duling v. Bluefield Sanitarium, Inc., 149 W.Va. 567, 142 S.E.2d 754 (1965). Moore’s case against the hospital for the negligence of its nurse must show that an intravenous injection constituted a lack of reasonable and ordinary care, that the nurse injected the medicine intravenously, and that th’s was the proximate cause of his injury.
Without contradiction, expert witnesses testified that an injection of chymar into the bloodstream is improper. On the record before us, proof of an intravenous injection would establish negligence, especially in the light of Dr. Guthrie’s specific instructions for intramuscular injection. There is no dispute that the nurse employed by the hospital gave the injection. The principal issues, therefore, are simple questions of fact: did an injection introduce the medicine *369in the bloodstream, and did this cause the plaintiff’s injuries?
In support of its motion for a directed verdict, the hospital urges that evidence disclosed that Moore’s injuries could have resulted either from an intravenous injection or from other causes over which the nurse had no control. It relies upon the rule that where the plaintiff's evidence gives equal support to each of two inconsistent inferences for one of which the defendant is not responsible, judgment must be entered for the defendant. See Pennsylvania R.R. v. Chamberlain, 288 U.S. 333, 339, 53 S.Ct. 391, 77 L.Ed. 819 (1933). It is questionable whether Chamberlain currently assays the right of a litigant to have his case placed before a jury.3 In any event the facts proved by the plaintiff and the inferences that may be drawn from the facts do not provide an appropriate basis for the application of the Chamberlain doctrine, even if its vitality be assumed.
Moore was admitted to the hospital on November 29, 1962 for a hernia correction, which was satisfactorily performed on November 30. The hospital records show that he was given penicillin on November 30 through December 4, and chymar was prescribed on December 3 and December 4. Both drugs were to be administered intramuscularly. The records also show that penicillin and chymar were injected at 9 A.M., December 4, and that at 9 A.M. Moore had a grand mal seizure.
Moore introduced evidence that an intravenous injection of chymar could bring on an immediate reaction or seizure but that an intramuscular injection would not cause a reaction for five or ten minutes. Another of his expert witnesses testified the reaction from an intramuscular injection would take from one to three hours.
A doctor called by Moore testified that the instantaneous reaction reported by the hospital indicated that the injection was given improperly and that the only cause of Moore’s injury was injection of the medicine into the bloodstream. This opinion is evidence to support the allegations that the medicine was negligently injected and that this negligence was the proximate cause of Moore’s injury. The opinion was not conclusively discredited because the doctor in answering a question from the court conceded there would be no connection between Moore’s injuries and an intramuscular injection.4 Although the doctor accepted the court’s assumption of an intramuscular injection for the pur*370pose of answering the question, he was unwavering in his belief that an intramuscular injection could not have produced the instantaneous result reported in the hospital records.
The hospital also urges that Moore’s grand mal seizure could have been caused by other conditions. It points to cross-examination of one of Moore’s experts:
“Q. [W]ill you just look at the jury when you give an answer, what are the — some of the precipitating causes of a grand malseizure ?
“A. Of course, the most common thing is epilepsy, which is by far the commonest thing of grand mal-seizures. Probably the next most common thing is brain tumor, and probably the next is stroke. And then we come down the line to things such as anoxia, instant reaction, low blood sugar will do it. Those are the most common things.”
It is apparent that this testimony is insufficient to defeat Moore's claim. The doctor was not asked whether Moore’s grand mal seizure was caused by the maladies that he mentioned. He was replying to a general question in a general way without specifically referring to Moore.
The defendants also urge that Moore’s seizure could have resulted from anaphylatic reaction to penicillin injected intramuscularly. There are two reasons why this theory should not foreclose the jury from consideration of Moore's case. First, Moore’s doctor testified that an anaphylatic reaction could not result from one injection of penicillin. It could result from a second injection, but in his opinion an interval of about ten days between the first injection and a following injection was required to allow the production of antibodies leading to a reaction. He added that a reaction had been reported after an interval of five or six days. Moore’s seizure occurred only four days after his initial injection. Secondly, the anaphylatic reaction from successive intramuscular injections would not be instantaneous. If the penicillin were given intramuscularly, at least five or ten minutes, and possibly as long as one to three hours, would be required for absorption and reaction. On the basis of this testimony, a reasonable inference can be drawn that Moore’s seizure was not caused by successive intramuscular injections of penicillin.
The hospital’s reliance on Swanson v. Wiesenfeld, 24 N.J.Super. 576, 95 A.2d 161 (1953), is misplaced. In that case a cast was applied to the plaintiff’s fractured leg. Four or five days later when the cast was removed it was observed that her toes were not getting a sufficient blood supply; gangrene set in and she lost her leg. The plaintiff’s expert testified that in his opinion the gangrene was caused by swelling of her leg under the cast which had been put on too tightly. An expert for the defendant testified that the gangrene was caused by trauma arising out of the fracture accident. A verdict for the plaintiff was set aside and the case remanded for a new trial because of error in admitting testimony of the plaintiff’s expert that he assumed swelling occurred, when in fact there was no evidence of such swelling, and assumed that the cast had been applied too tightly, when there was no evidence to this effect. Swanson differs from the case before us because here the instantaneous reaction to the injection provided evidence that the medicine was improperly injected into the bloodstream and negated the contention that it was injected intramuscularly.
Finally, the hospital urges that Moore’s expert witness should not have been allowed to express an opinion that an intravenous injection caused Moore’s injury, because the opinion in part was based upon medical reports concerning Moore which were not in evidence. It is not clear from the record, however, just what information the doctor obtained from the fugitive medical reports. The district judge took the hospital’s objection under advisement and made no rul*371ing upon it. The absence of the medical reports from the record prevents us from determining whether the facts they disclosed were introduced into evidence by some other means. In any event, the question is not likely to arise at retrial. The medical reports may be made a part of the record if they are admissible, or the authors may be deposed, or the information contained in these reports may also be found in other reports and testimony.
The judgment in favor of Dr. William Guthrie is affirmed. The judgment in favor of Guthrie Hospital, Inc., is vacated, and this action is remanded for retrial.