PETERS, J.
Plaintiff Eunice D. Clark brought this action against Dr. Selmants, an anesthesiologist, Dr. Gibbons and his partner Dr. Horn, orthopedic surgeons, and Sutter Community Hospital of Sacramento for damages for injuries resulting from an operation allegedly negligently performed in Sutter Community Hospital by Drs. Gibbons and Selmants. The jury returned verdicts of $27,500 against all of the doctors and exonerated the hospital.1 Motions for a new trial were denied, and the doctors have appealed from the judgment.
Defendants’ contentions are that the verdicts are not supported by sufficient evidence of negligence, and that the trial court committed reversible error by giving conditional res ipsa loquitur instructions. Both contentions are unsound.
It must be kept in mind that in determining whether sufficient evidence of negligence was adduced against the doctors,2 the power of an appellate court begins and ends with a finding that the record contains some substantial evidence, contradicted or uncontradicted, which supports the conclusion reached by the jury. The record must be read in the light most advantageous to the plaintiff. All conflicts must be resolved in her favor; and all legitimate and reasonable inferences must be indulged in to uphold the verdict, if that is possible. (Crawford v. Southern Pac. Co., 3 Cal.2d 427, 429 [45 P.2d 183].)
The record discloses that on October 30, 1960, at approximately 2 p.m., plaintiff Eunice Clark, who was 41 years old, obese and in good health, suffered a fractured right ankle when she slipped and fell on a waxed floor in her home. She was taken by ambulance to the Sutter Community Hospital and examined by her physician Dr. Smith. After viewing X-rays of the fracture, Dr. Smith told her that it was a ease for an orthopedic surgeon. Mrs. Clark requested the services of Dr. Gibbons who had previously treated her husband and who [403]*403happened to be in the hospital at that time. Dr. Gibbons determined from viewing the X-rays that plaintiff had a severe trimalleolar fracture of the right ankle and that an open reduction, i.e., reduction by surgery, should be performed as soon as possible. He informed plaintiff that he would prefer to operate that night since the ankle would be stiff by the next morning. Plaintiff agreed.
Dr. Selmants, one of the anesthesiologists on the hospital staff, undertook to administer the anesthesia. Plaintiff gave the doctor a preanesthesia history which revealed that she had eaten between 1 and 2 p.m. Dr. Selmants believed that a general anesthetic should not be given to one who had eaten in the 12 hours preceding surgery, and that a general anesthetic was dangerous for a patient who had eaten within six or seven hours of surgery. Dr. Selmants concluded that plaintiff was not a safe subject for a general anesthetic. He told plaintiff that she should be given a spinal anesthetic, and she agreed to a spinal, although she said that she would prefer a general. Dr. Gibbons concurred in the decision to give a spinal.
Dr. Selmants selected the agent to be used for the spinal. He testified that it was the anesthesiolgist’s duty to know the time required for surgery and that he did not ask Dr. Gibbons how long the operation would take, because he knew from previously working with Dr. Gibbons that the surgeon averaged two hours for usual reductions of this kind. This case was not unusual, he believed, and would accordingly require two hours. The agent that Dr. Selmants chose was 10 milligrams of pontocaine. It was designed to maintain a level of T-10 for two hours plus or minus 15 minutes, and it was predictable in intensity and duration.
Dr. Gibbons testified that plaintiff’s fracture was severe, and that he expected the operation to take from two to three hours. The anesthesiologist said that the anesthesia he selected could not be used unmodified for surgery of from two and a half to three hours, but that by adding epinephine to the pontocaine, he could have produced an agent with a predictable duration of over four and a half hours.
Prior to surgery plaintiff was nervous and anxious. She was not more upset however than any person with an ankle injury and she was calmed by injections of vistaril and nembutal and taken into surgery at approximately 8 p.m.
Dr. Selmants then administered the spinal anesthetic and made pinprick tests to assure that the proper level of anesthesia had been achieved. After the anesthesia was injected [404]*404and the numbness started up her legs, plaintiff felt that she could not breathe and her voice became squeaky. Dr. Selmants stated that a patient should not suffer from shortness of breath at a level of T-10, and that plaintiff’s difficulty in breathing could have been “some undue effect [from] the way the anesthetic was given. ’ ’
When the anesthesia reached the level necessary for surgery, Dr. Gibbons commenced the operation. The level of anesthesia remained adequate at first; but after about one hour the doctors noticed from plaintiff’s unconscious movements that the anesthesia was beginning to wear off. At this point Dr. Gibbons had completed all of the reduction except for reduction of the posterior fragment of the tibia. Dr. Selmants believed that 20 minutes more were needed to complete the operation, but Dr. Gibbons testified that no less than another hour would have been required.
Dr. Selmants could have extended the surgical anesthesia without harm to the patient. He stated that intravenous demerol could have been used for that purpose and that there was no particular reason not to use it, although the extension obtained might still have been insufficient to complete the operation. Also, another spinal could have been given. Dr. Gibbons stated that he did not think that plaintiff “would have been up to” another spinal and that in turning her for the spinal all the propping and draping would have to be undone and this might subject her to a risk of infection.
In any event, the operation was terminated, the incision was closed, and a cast was applied to plaintiff’s ankle in an attempt to reduce the posterior fragment by external pressure. Dr. Gibbons’ operative report mentioned that the operation was not completed because the anesthetic did not last for the required length of time. Dr. Selmants failed to note those facts, although he was supposed to make an accurate report of how the anesthesia proceeded with relation to the needs of the operation.
The decision to terminate surgery was made primarily by Dr. Gibbons. He was in control of the surgery and could have asked for an extension of the anesthesia. He had stated in his deposition that he became upset when the anesthesia began to wear off and that he did not consult with Dr. Selmants about extending the anesthesia but just said “I think we will quit for tonight and do this another time.” At the trial, however, both he and Dr. Selmants testified that they discussed the question whether the anesthesia should be continued or the [405]*405operation terminated and that Dr. Selmants agreed with the decision to terminate.
Dr. Gibbons testified that, when he made the decision to terminate, he expected to complete the operation later, but did not do so because blebs, infected blisters of the skin, developed; that it was very common for blebs to accompany an injury of this nature, that plaintiff’s blebs healed at the normal rate, but that healing of blebs at the normal rate did not permit a second operation within the time when a second operation would have been of any value.
Three days after the operation Dr. Gibbons noticed that the posterior fragment had slipped back to some extent and that another open reduction was required. A second operation could not be performed, however, because pressure and resultant swelling, as pointed out, had caused blebs, which presented a serious risk of infection if the skin were cut.
Dr. Gibbons charged plaintiff less than the normal operating fee because his operation was unsuccessful. She now suffers from osteoarthritis in the ankle joint, which is painful. The arthritis might have resulted from the fracture even if a perfect union had been achieved. However, there is expert testimony that the chances of getting arthritis were increased by the failure to achieve a complete reduction; and the defendants do not claim that the evidence is insufficient td show that the arthritis was due, at least in part, to the failure to complete the operation.
Nothing but a fusion, which would impair the ankle’s up and down movement, could now give plaintiff a pain free ankle. Dr. Gibbons’ partner, Dr. Horn, offered to fuse the ankle for a token fee of $100.
Plaintiff’s position is that the jury received sufficient evidence to find that the imperfect reduction causing the present injury was the result of (1) the negligence of Dr. Selmants in selecting and administering an anesthetic which wore off before the operation was completed and (2) the negligence of Dr. Gibbons in (a) not informing Dr. Selmants that the operation might well take longer than the two hours which Selmants expected the operation to take and (b) terminating the operation prematurely rather than ordering an extension of the anesthesia.
The evidence upon which plaintiff primarily relies to show negligence in the selection or administration of the anesthetic is the following testimony of Dr. Selmants: “Q. . . . if proper care is used, in the usual course of events, anesthetics [406]*406like this don’t run out or wear out, do they—surgical anesthetics ? A. No, sir.
And I said ‘no, sir.’ What I mean is, they can. There is no control. There is—there is—there is a variable, as anything else we do in medicine. There is nothing exact that guarantees this will happen, this will not happen, how long this will last. You cannot ever predicate what you’re doing on the basis that it’s going to be 100 percent; you have a certain area of predictability, and that’s what you go on. Q. Now, my question was—you see, you misunderstood. I’ll say it again. My question was, if proper care is used in a situation like this, anesthetics like this usually do not run out, do they? A. Yes, usually they do not run out.” Dr. Gibbons stated in his deposition that if the anesthesiologist uses proper care and obtains proper information about the case, he can make a spinal anesthesia last long enough for an operation of this kind.
Later in direct examination Dr. Selmants and Dr. Horn testified that it is common knowledge in the field of anesthesia that there is always an inherent risk that, even when due care is used, a spinal anesthesia will not last as long as contemplated. Dr. Selmants explained that immediately upon the injection of any spinal anesthetizing agent a process of detoxification commences within the area of the nerve root blockage which ultimately causes the anesthesia to wear off. The speed of detoxification in any individual depends primarily upon the amount of circulation in the nerve area and the amount of myelin covering the nerve itself.
The evidence, independently of the doctrine of res ipsa loquitur, is sufficient to support the verdict against both doctors. So far as Dr. Selmants is concerned, while the evidence shows that he and Dr. Gibbons were not negligent in making the initial decision to give a spinal rather than a general anesthetic,3 there is evidence that Dr. Selmants was negligent in selecting the agent to be used for the spinal. The orthopedic surgeon expected the operation would take from two to three hours to complete. Anesthetics were available that would have lasted at least that long without undue danger to the patient. Dr. Selmants did not use those anes[407]*407thetics because he did not expect that the operation would take as long as three hours. Dr. Selmants admitted, however, that it was his responsibility to Imow the needs of the surgeon and that he did not inquire how long the surgeon would take in this ease. Because he underestimated the probable length of the operation, he gave an anesthetic which was designed to last a maximum of two hours and 15 minutes and a minimum of one hour and 45 minutes. The jury could have found that this choice of anesthetic was negligent.
In addition, Dr. Selmants testified that anesthetics do not usually wear off prematurely if proper care is used. Premature termination can be caused by improper administration of the agent, and plaintiff’s initial reaction to the injection, her squeaky voice and difficulty in breathing, is evidence that the anesthetic was improperly administered. Dr. Selmants failed to note the premature termination in his records relating to the progress of the anesthesia, as he should have done. The jury could infer from such omission that he possessed some guilty knowledge regarding the termination. Prom the symptoms of improper administration and the inference of guilty knowledge, the jury could properly imply that Dr. Selmants had administered the anesthetic improperly.
Although Dr. Selmants testified that inadequate anesthesia could have resulted from several unpredictable causes and that he used due care in the administration of the anesthetic, the credibility of this testimony was properly left to the jury.
There is also sufficient evidence of negligence to support the verdict against Dr. Gibbons. Dr. Gibbons could have been found responsible for the improper selection of an anesthetic. Although the anesthesiologist ordinarily has the duty to ask the surgeon about the projected length of the operation, Dr. Selmants testified that the surgeon should speak to the anesthesiologist if the operation is going to be unusually long. Dr. Selmants testified that Dr. Gibbons averaged two hours in operating on trimalleolar anide fractures. Dr. Gibbons testified that Mrs. Clark’s fracture was one of the most severe trimalleolars he had ever seen. Under those circumstances the jury could find that Dr. Gibbons expected an unusually long operation and was under a duty to warn Dr. Selmants of the need for an anesthetic that would last more than two hours.
Dr. Gibbons stated in his deposition that he became upset when the anesthesia began to wear off and, without consulting [408]*408Dr. Selmants about the possibility of extending the anesthesia, said that the surgery would be terminated now and completed at a later time. He also admitted that healing of the skin at the normal rate did not permit a second operation to be performed within the necessary time.
Thus, from Dr. Gibbons’ own statements, the jury could have found that in making the decision to terminate he acted rashly knowing the dangers, but without considering the possibility of extending the anesthesia and under the erroneous belief that the completion of the reduction could be accomplished by a second operation. Both the possibility of extending anesthesia and the possibility of further surgery are clearly major factors which should have been considered by a surgeon confronted with inadequate anesthesia. Under these circumstances, the jury could conclude without any other expert testimony that in making the decision to terminate surgery, Dr. Gibbons did not exercise that care and skill ordinarily practiced by other specialists in orthopedic surgery under similar circumstances.4
In addition, the facts of this case warrant the use of the conditional res ipsa loquitur instructions. As a general rule, res ipsa loquitur applies where the occurrence of the injury is of such a nature that it can be said, in the light of past experience, that it probably was the result of negligence by someone and that the defendant is probably the person who is responsible. In determining whether such probabilities exist with regard to a particular occurrence, the courts have relied both on common knowledge and on expert testimony. (Davis v. Memorial Hospital, 58 Cal.2d 815, 817 [26 Cal.Rptr. 633, 376 P.2d 561] ; Siverson v. Weber, 57 Cal.2d 834, 836 [22 Cal.Rptr. 337, 372 P.2d 97].)
[409]*409The doctrine of res ipsa loquitur is a doctrine fundamentally predicated upon inferences deducible from circumstantial evidence and the weight to be given to them. (Quintal v. Laurel Grove Hospital, supra, 62 Cal.2d 154, 163.) As stated in Fowler v. Seaton, 61 Cal.2d 681, 686-687 [39 Cal.Rptr. 881, 394 P.2d 697] :
"One of the frequently quoted statements of the applicable rules is to be found in the opinion of Chief Justice Erle in Scott v. London & St. Katherine Docks Co. (1865) 3 H. & C. 596, quoted in Prosser on Torts (2d ed. 1955) section 42, at page 201, as follows: 'There must be reasonable evidence of negligence; but where the thing is shown to be under the management of the defendant or his servants, and the accident is such as in the ordinary course of things does not happen if those who have the management use proper care, it affords reasonable evidence, in the absence of explanation by the defendants, that the accident arose from want of care. ’
“Of course, negligence and connecting defendant with it, like other facts, can be proved by circumstantial evidence. There does not have to be an eyewitness, nor need there be direct evidence of defendant’s conduct. There is no absolute requirement that the plaintiff explain how the accident happened. Res ipsa loquitur may apply where the cause of the injury is a mystery, if there is a reasonable and logical inference that defendant was negligent, and that such negligence caused the injury. (Prosser on Torts, supra, at p. 204.) ” (See also Quintal v. Laurel Grove Hospital, supra, 62 Cal.2d 154, 164-165.)
More than 20 years ago in Ybarra v. Spangard, 25 Cal.2d 486, 489 et seq. [154 P.2d 687, 162 A.L.R. 1258], this court had occasion to consider the application of the doctrine to cases where injury was received by a medical patient while unconscious under the influence of anesthesia. It was stated:
“There is, however, some uncertainty as to the extent to which res ipsa loquitur may be invoked in cases of injury from medical treatment. This is in part due to the tendency, in some decisions, to lay undue emphasis on the limitations of the doctrine, and to give too little attention to its basic underlying purpose. The result has been that a simple, understandable rule of circumstantial evidence, with a sound background of common sense and human experience, has occasionally been transformed into a rigid legal formula, which arbitrarily precludes its application in many cases [410]*410where it is most important that it should be applied. If the doctrine is to continue to serve a useful purpose, we should not forget that ‘the particular force and justice of the rule, regarded as a presumption throwing upon the party charged the duty of producing evidence, consists in the circumstance that the chief evidence of the true cause, whether culpable or innocent, is practically accessible to him but inaccessible to the injured person. ’ . . .
“. . . [I]t is difficult to see how the doctrine can, with any justification, be so restricted in its statement as to become inapplicable to a patient who submits himself to the care and custody of doctors and nurses, is rendered unconscious, and receives some injury from instrumentalities used in his treatment. Without the aid of the doctrine a patient who received permanent injuries of a serious character, obviously the result of someone’s negligence, would be entirely unable to recover unless the doctors and nurses in attendance voluntarily chose to disclose the identity of the negligent person and the facts establishing liability. (See Maki v. Murray Hospital, 91 Mont. 251 [7 P.2d 228].) If this were the state of the law of negligence, the courts, to avoid gross injustice, would be forced to invoke the principles of absolute liability, irrespective of negligence, in actions by persons suffering injuries during the course of treatment under anesthesia. But we think this juncture has not yet been reached, and that the doctrine of res ipsa loquitur is properly applicable to the ease before us.” (Ybarra v. Spangard, supra, 25 Cal.2d 486, 489-491.)
In Ybarra, it was pointed out that in a modern hospital a patient is quite likely to come under the care of a number of persons in different types of contractual relationships with each other, including physicians and surgeons, anesthetists, and nurses; that every defendant in whose custody the plaintiff was placed for any period was bound to exercise ordinary care to see that no unnecessary harm came to the plaintiff; that, although at the trial some of the defendants might be found liable and others absolved, this would not preclude application of the doctrine; and that, since each of the defendants in acting together to provide the medical treatment at one time or another was in control of the various agencies which might have harmed plaintiff, they should have the burden of initial explanation. (25 Cal.2d at pp. 491-492.)
Ybarra involved an injury which may not have been received during the operation, but Leonard v. Watsonville [411]*411Community Hospital, 47 Cal.2d 509, 514 et seq. [305 P.2d 36], involved an injury during the operation and followed Ybarra in holding that where the conditions of the doctrine are satisfied all persons who had any control over the patient’s body or the instrumentalities causing injury may properly be called upon to meet the inference of negligence by giving an explanation of their conduct. In Leonard it was further held that the inference of negligence arising under the doctrine is dispelled as a matter of law with regard to a particular doctor only where other evidence establishes as a matter of law that he is free from negligence. The evidence establishing the absence of negligence in such a case must be clear, positive, uneontradicted and of such a nature that it cannot be rationally disbelieved.
Ybarra and Leonard establish that, if the conditions giving rise to the doctrine are present when the medical personnel are treated as a group acting in concert and they collectively have access to the chief evidence as to the cause of the injury but the plaintiff does not, a single doctor may not escape the inference as a matter of law merely by showing that as to him alone it is more probable than not that he was free from fault. The basis of the application of the doctrine to all defendants in the cases is that the medical personnel acted as a group and that collectively, without regard to what any one may individually know, or did, they are in a position to explain the cause and produce the chief evidence bearing on the question whereas the plaintiff is not. To avoid the inference as a matter of law an individual doctor must go beyond showing that it was unlikely or not probable he was negligent and must establish that he is free from negligence by evidence which cannot be rationally disbelieved. Falling short of such a showing, it remains for the jury to determine whether the inference arising from the doctrine has been rebutted as to any particular doctor.
The fact that the patient may have received a local anesthetic rather than a general anesthetic does not eliminate the duty of explanation of those who had control over the procedure where the chief evidence as to the cause is accessible to them but not to the plaintiff. The plaintiff’s lack of knowledge may exist not only where he is totally unconscious but also where he is partially unconscious and largely, if not entirely, unaware of what the medical personnel are doing.
The conditions giving rise to the doctrine here existed. This problem was recently discussed in Quintal v. Laurel Grove [412]*412Hospital, supra, 62 Cal.2d 154, a ease involving injuries during an operation. There the plaintiff suffered a cardiac arrest during the administration of a general anesthetic, and it was held that an instruction on conditional res ipsa loquitur was proper even though the medical experts testified that a cardiac arrest, although a rare occurrence, is a known and calculated risk in the giving of a general anesthetic and though there was no expert testimony that when cardiac arrests do occur, they are more likely than not the result of negligence. There was evidence that a method of meeting the unusual risk existed. Experts testified that when due care is used, cardiac arrests do not ordinarily occur, and, in addition, evidence was presented of fever and apprehension of the patient before administration of the anesthetic which tended to show that the cardiac arrest in that case was caused by negligence of the doctors.
Thus, we recognized in Quintal that proof that when due care is exercised an injury rarely occurs, accompanied by other evidence indicating negligence, may be sufficient to warrant an instruction on conditional res ipsa loquitur. (See also Ragusano v. Civic Center Hospital Foundation, 199 Cal.App.2d 586, 593-594 [19 Cal.Rptr. 118].) This is particularly true where, as in Quintal and in the present case, the injury occurred as the result of a normal procedure such as the administration of an anesthetic, rather than from a complex operation.
It is true that evidence that an accident rarely occurs when due care is used does not without more indicate that a particular occurrence is more likely than not the result of someone’s negligence. (Siverson v. Weber, supra, 57 Cal.2d 834, 839.) In Siverson it was stated:
“To permit an inference of negligence under the doctrine of res ipsa loquitur solely because an uncommon complication develops would place too great a burden upon the medical profession and might result in an undesirable limitation on the use of operations or new procedures involving an inherent risk of injury even when due care is used. Where risks are inherent in an operation and an injury of a type which is rare does occur, the doctrine should not be applicable unless it can be said that, in the light of past experience, such an occurrence is more likely the result of negligence than some cause for which the defendant is not responsible.” (57 Cal.2d at p. 839.) But in Siverson there was no evidence of a negligent act of a type that could have caused the accident, and none of the [413]*413witnesses “testified that anything was done during the operation which was contrary to good medical practice. ’ ’ (57 Cal.2d at pp. 838-839.) The court refused to permit an instruction on the doctrine where the only basis for it was evidence that the injury suffered by the patient rarely occurs as a result of the surgical procedure.
The likelihood of a negligent cause is increased if the low incidence of accidents when due care is used is combined with proof of specific acts of negligence of a type which could have caused the occurrence complained of. When those two facts are proved, the likelihood of a negligent cause may be sufficiently great that the jury may properly conclude that the accident was more probably than not the result of someone’s negligence.
That a doctor has done a negligent act of a type that could have caused the accident, which does not ordinarily occur in the exercise of due care, greatly increases the probability that it was his negligence that caused the plaintiff’s injury. Thus, the low incidence of accidents when due care is used plus negligent conduct of a type which could have caused the occurrence may make it probable that the occurrence was the result of someone’s negligence and that the defendant is probably the person who was responsible. Those are the requirements for applying res ipsa loquitur.
The administration of an anesthetic is now a normal and tested procedure. Dr. Selmants stated that one of the reasons for selecting pontocaine in the instant ease was its predictability as to duration. Medical experts testified that spinal anesthetics do not usually run out prematurely if proper care is used, and in explanation of this conclusion, Dr. Selmants stated only that the predictability of such anesthetics was not 100 percent. Dr. Gibbons stated in his deposition without qualification that if an anesthesiologist uses proper care and obtains proper information about the case, he can make a spinal anesthesia last long enough for an operation of this kind.
There is evidence that the injury here was caused by the anesthesia wearing off prematurely and that Dr. Selmants was responsible for selecting and administering an anesthetic which would be adequate for the length of surgery required. Dr. Selmants testified that he did not consult with the surgeon regarding the length of the operation and that he used an anesthetic which, according to the testimony of the surgeon, was inadequate for the estimated length of the oper[414]*414ation. In addition, Dr. Seim ants stated that plaintiff’s unusual reaction to the anesthetic could have been caused by “some undue effect the way the anesthetic had been given. ’ ’
There is also evidence, as we have seen, from which the jury could conclude that Dr. Gibbons in the exercise of due care should have advised Dr. Selmants of the anticipated extraordinary length of the surgery and that the former acted rashly in determining to terminate the surgery. It bears emphasis in this connection that the doctors were aware that the procedure should be performed and completed as soon as possible.
This evidence, taken as a whole, along with the evidence that there was a reasonable method of handling the risk when it occurred is certainly sufficient for the jury to find that the injury was probably the result of negligence of someone and that the defendants were probably the persons responsible.
Accordingly we conclude that it was proper to instruct the jury on the doctrine of res ipsa loquitur. It is not claimed that the form of the conditional instructions given was improper.
The judgment is affirmed.
Mosk, J., Burke, J., and Peek, J.,
Retired Associate Justice of the Supreme Court sitting under assignment by the Chairman of the Judicial Council.