BRAY, J.

In a malpractice action the jury awarded Martin Salgo1 the sum of $250,000 against defendants Leland Stanford Jr. University Board of Trustees, Stanford Uni*564versity Hospitals, 2 and Dr. Frank Gerbode. The trial court reduced the award to $213,355. All defendants appeal from the judgment entered thereon.3//j

Questions Presented

1. Was res ipsa loquitur applicable, and if so, were the instructions thereon proper?4

2. Liability of Dr. Gerbode for negligence of hospital team.

3. Instructions on alleged other negligence of defendant Gerbode.

4. Experimentation and the manufacturer’s brochure.

5. Instructions on (a) duty to call specialist; (b) physician’s duty to disclose; (c) failure to produce evidence.

6. Medical texts as evidence.

7. Reference to malpractice judgments.

Evidence

Dr. Gerbode has been licensed to practice medicine in California since 1937. He specializes in surgery, surgery of the heart, major vessels, and in thoracic surgery, with a special interest in cardiovascular surgery. He is recognized as an outstanding authority and is a professor of surgery at Stanford Medical School. Plaintiff was 55 years of age, with a history of eye condition indicating premature aging. About two or three years prior to the occasion upon which this suit is based, he had developed cramping in his legs upon walking and for approximately a year had been treated with drugs by a physician. This doctor referred him to Dr. Gerbode as a specialist in the surgical treatment of arterial diseases./# December 31,1953, at Stanford Hospitals, Dr. Gerbode examined plaintiff. His chief complaint was cramping pains in his legs, mostly in the calves, causing intermittent limping. This condition had started gradually, becoming increasingly more *565severe. He complained of pain in Ms Mps and lower back on exercise. He also had right side abdominal pain. Dr. Gerbode’s examination found a man who looked much older than his stated age. Both legs were atrophic in the thighs and calves. The right leg was blue. No pulses below the femoral pulse on each side were palpable. There was a weak femoral pulse on the left and none on the right. Upon raising the legs they blanched. This is a characteristic of advanced arterio insufficiency. Dr. Gerbode then diagnosed a probable occlusion of the abdominal aorta which had impaired the blood supply to the legs and other areas and an advanced arteriosclerosis. Dr. Gerbode was uncertain whether the decreased circulation was limited to the legs alone, or to something blocking the blood higher up. Plaintiff’s blood pressure was 180/90, which Dr. Gerbode felt was due to the generalized arteriosclerosis. The latter is a serious disease and one which might cause a stroke in the brain or a coronary ocplusion to the vessels of the heart.

. Dr. Gerbode advised plaintiff that he had evidence of serious circulatory disturbance, that the examination indicated that plaintiff might have a block in his abdominal aorta,' and that there was something else wrong as shown by the pain in his right side and back. Dr. Gerbode told plaintiff of the seriousness of his condition and that plaintiff should enter the hospital for a thorough evaluation of his condition • that one of the things the doctor wished to have done was a study of plaintiff’s aorta, which would entail an anesthetic and an injection of some material in the aorta to localize the block; - also X-rays of his gastrointestinal tract would be taken. Dr. Gerbode stated that if his clinical findings were borne out by the further examination contemplated his condition would be helped by an operation removing and replacing a segment of the aorta. Such an operation would improve the circulation to the legs and back and prolong his life. Dr. Gerbode did not explain all of the various possibilities to plaintiff of the proposed procedures but did say that his circulatory situation was quite serious. Dr. Gerhode reported to the referring physician and recommended the performance of an aortography in order to locate the block and its extent so that proper surgery could be done.f 'A study of the gastrointestinal -tract was. also necessary. An aortography consists of injecting in the aorta an X-ray contrast medium and then taking X-ray pictures of the abdominal aorta and its branches to discover the block, if any. /

*566• At Dr. Gerbode ’s suggestion plaintiff entered the hospital on January 6, 1954..[That afternoon Dr. Gerbode ordered, among other things, X-rays of the chest and abdomen after a barium swallow. The X-rays were taken and showed marked calcification in the abdominal aorta, iliac and femoral vessels. This presence of calcium indicated that the illness was of long duration! ,fDr. Gerbode requested in writing that the aortography be performed by the hospital’s X-ray department.

The normal procedure is for the attending surgeon to tell members of the house staff team who are to perform the procedure basically what the problem is. Dr. Gerbode did this with Dr. Ellis and Dr. Andrews of the staff. Dr. Ellis was to perform the aortographyEllis had five years practice in surgery and was in charge of all special diagnostic procedures at the hospital, such as aortographies, that had to do with the injection of radio-opaque or contrast material in various arteries and blood vessels of the body.

On January 7th Dr. Ellis called on plaintiff in his hospital room and informed him that he was to do the aortography and would do it the next afternoon. He explained that he would inject some material into the aorta and take films at that time to see if they could ascertain the precise condition of plaintiff’s circulatory system. The next afternoon Dr. Ellis saw plaintiff and informed him that the procedure had been postponed until the following day because plaintiff still had some barium in his intestines from the first X-ray study.

On January 6th, Dr. Howard, an anesthetist, saw plaintiff and examined him to determine if he was fit to receive the anesthetic. When the procedure was postponed, Dr. Clark saw the patient on January 7th and informed him the procedure would take place on the next day.

On the afternoon of January 8th Dr. Ellis went to the X-ray room where plaintiff was lying on a table.,/ Present were the anesthesiologist, Dr. Bengle, Dr. Andrews, a radiologist, and several technicians. Dr. Gerbode was there at the beginning of the procedure but gave no instructions and did not participate in the procedure. Plaintiff was already anesthetized and asleep. Dr. Ellis was inserting the needle in plaintiff’s aorta when Dr. Gerbode came in the room. As the patient was apparently in good condition Dr. Gerbode left and did not see the patient again until the next morning. /í-

An aortography is a procedure requiring an anesthesiologist, a radiologist and a surgeon. The function of the surgeon (Dr. Ellis) is to insert the needle necessary for the injection *567of material into the aorta and to discuss with the radiology-department the materials used and the timing. A hollow Number 16 or Number 18 needle is used. The hollow is closed with a metal rod or stylette. It is approximately l/32nd inch in diameter and 6 inches long. The patient is placed on his abdomen, face down on the table, and given a general anesthetic. A sensitivity test is then done to determine if the patient is sensitive to the radio-opaque material to be used. The needle is inserted to the left of the spinal column approximately 3 to 4 inches to the left of the midline of the back underneath the 12th rib. The needle is inserted in an upward direction toward the front of the body so as to enter the aorta which lies in front of the spinal column. The material used here was 70 per cent sodium urokon. This under an X-ray will appear in contrast to the body tissues. One c.c. of it was injected into a vein in plaintiff’s arm. He appeared not to be sensitive to it. After the surgeon feels the needle penetrate the wall of the aorta a metal rod is removed from the needle and blood flows from the aorta through the hollow needle. A syringe is then attached to the needle. In this case 30 c.c.’s of sodium urokon were then injected at a fairly rapid rate. Defendants’ witnesses testified that there was no difficulty in inserting the needle on the first attempt and that it was only inserted once. (Plaintiff contends otherwise, as will hereafter appear.) The injection took only a few seconds and then a series of X-rays were immediately taken by a machine already in position. While the films were still wet Dr. Ellis and the radiologist, Dr. Stone, examined them. They showed that the descending aorta in the abdomen just below the vessels leading to the kidneys was blocked. In a consultation between Drs. Ellis, Stone and Andrews (another radiologist) it was deemed desirable to take additional X-rays in order to determine the extent or length of the block. During this time plaintiff was kept under anesthesia and the needle remained in place, it being the custom so to do while the doctors are determining whether additional X-rays are necessary. This obviates the necessity of again inserting the needle. The doctors hoped that by changing the timing of the pictures in relation to the time of making the second injection they might be better able to delineate the vascular tree. Twenty c.c.’s of sodium urokon were then injected, without changing the position of the needle, and additional X-rays taken, particularly of the body further down than in the first pictures^/The entire procedure seemed to pro*568ceed in a normal manner, and the patient seemed that evening to have recovered from the anesthesia. At 5 o’clock Dr. Gerbode was informed by Dr. Ellis that the procedure had been routine and gone well. The next morning when plaintiff awoke he noticed that his lower extremities were paralyzed. This condition is permanent.

1. Bes Ipsa Loquitur.

The court instructed that the doctrine applied.' If it did not, or if the instructions thereon were improper, the judgment will have to be reversed, even though there should be evidence of negligence of any or all defendants. (Dees v. Pace, 118 Cal.App.2d 284 [257 P.2d 756].)

The application of the doctrine of res ipsa loquitur in malpractice cases is a development of comparatively recent years. Before that time, the facts that medicine is not an exact science, that the human body is not susceptible to precise understanding, that the care required of a medical man is the degree of learning and skill common in his profession or locality, and that even with the greatest of care untoward results do occur in surgical and medical procedures, were considered paramount in determining whether the medical man in a given circumstance had been negligent. But gradually the courts awoke to the so-called ‘ ‘ conspiracy of silence. ’ ’ No matter how lacking in skill or how negligent the medical man might be, it was almost impossible to get other medical men to testify adversely to him in litigation based on his alleged negligence. Not only would the guilty person thereby escape from civil liability for the wrong he had done, but his professional colleagues would take no steps to insure that the same results would not again occur at his hands. This fact, plus the fact that usually the patient is by reason of anesthesia or lack of medical knowledge in no position to know what occurred that resulted in harm to him, forced the courts to attempt to equalize the situation by in some cases placing the burden on the doctor of explaining what occurred in order to overcome an inference of negligence. One other fact contributed to the application of the doctrine, namely, that certain medical and surgical procedures became so common that in many of them the laymen knew that if properly conducted untoward results did not occur,5 and in others medical men *569(when it was possible to get them to admit it) from their specialized knowledge lmew that without negligence the result would have been a good one.

The great difficulty in the application of the doctrine is to determine where to draw the line. To apply it in all cases where an unexpected result occurs would hamstring the development of medical science. No medical man would dare to use new procedures, especially in surgery, because if injury resulted he would be prima facie guilty of negligence. Medical science has developed in leaps and strides in the past few years. Procedures that 40 years or even 10 years ago, would have been considered impracticable and fatal are now being successfully used; for example, surgery upon the heart. Even the procedure used in this case, translumbar aortography where the aorta is punctured and a foreign substance injected in order to determine the location of a suspected block, is one which but a few years ago would not have been attempted but one which is of great value in determining whether or not corrective surgery is needed and advisable. Thus a great responsibility rests upon the courts—to determine the point at which the doctrine will apply in order to be fair to a patient who has received a result which either common knowledge of laymen or of medical men teaches ordinarily would not occur without negligence, and to be fair to the medical men if there is a result which could occur without negligence and which should not impose upon them the presumption of negligence.6

Cases in which the doctrine has been applied follow: Ragin v. Zimmerman, 206 Cal. 723 [276 P. 107]7, Costa v. University of California, 116 Cal.App.2d 445 [254 P.2d 85], and other California cases applying the doctrine to X-ray burns both in diagnosis and treatment; cases such as Timbrell v. Suburban Hospital, Inc., 4 Cal.2d 68 [47 P.2d 737], dealing with burns by hot water bottles, hot compresses or heating pads; Bauer v. Otis, supra, 133 Cal.App.2d 439, injection in the arm *570muscles; Cavero v. Franklin Gen. Benevolent Soc., 36 Cal.2d 301 [223 P.2d 471], death of child during tonsil removal due to erratic and excessive administration of anesthetic.

In these cases application of the doctrine was denied: Farber v. Olkon, 40 Cal.2d 503 [254 P.2d 520], shock therapy ; Engelking v. Carlson, 13 Cal.2d 216, 221-222 [88 P.2d 695], removal and transportation of ligaments in knee (this ease was disapproved in Seneris v. Haas, 45 Cal.2d 811, 827 [291 P.2d 915, 53 A.L.R.2d 124]; Dees v. Pace, supra, 118 Cal.App.2d 284, hysterectomy; Bennett v. Los Angeles Tumor Institute, 102 Cal.App.2d 293 [227 P.2d 473], X-ray burns; Pink v. Slater, 131 Cal.App.2d 816 [281 P.2d 272], removal of scars.

Jf A study of the cases both pro and con on the application of the doctrine in malpractice actions demonstrates that the doctrine is- applicable only where it is a matter of common ' knowledge among laymen or medical men or both that the injury would not have occurred without negligence.Tí (See Seneris v. Haas, supra, 45 Cal.2d 811, 824-825.) Plaintiff contends that there is an additional situation in which the doctrine will apply, namely, where the patient is under anesthesia and injury occurs, particularly to a different part of the patient’s body than the one on which the work was to be performed, and that such application of the doctrine should be made here, because the aorta was the vessel involved and there was evidence that the spinal column was injured. Plaintiff cites Dierman v. Providence Hospital, 31 Cal.2d 290, 292 [188 P.2d 12], Ybarra v. Spangard, 25 Cal.2d 486, 490 [154 P.2d 687, 162 A.L.R. 1258], Meyer v. McNutt Hospital, 173 Cal. 156, 159 [159 P. 436], and Bauer v. Otis, supra, 133 Cal. App.2d 439. However, an examination of those cases shows that while at first blush it appears that it is the mere fact that the patient is under anesthesia that causes the doctrine to apply, actually it is not so, and the doctrine applies because the results were ones which either laymen or medical men know ordinarily do not occur without negligence.^-To apply the doctrine in every case merely because the patient is under anesthesia would put a hopeless burden on the medical profession. It must be remembered that the doctrine goes further than to require the doctor to explain what happened, as, of course, he will have to do to overcome the plaintiff’s charge of negligence,—it infers that he was negligent.

There can be little question but that aortography and its results, because it is a relatively new diagnostic procedure, is not a matter of common knowledge among laymen.,/' (Plain*571tiff contends and one of Ms witnesses testified that trans-lumbar aortography was not used enough in the bay area to constitute routine procedure.) Very few laymen have ever heard of it. So far as laymen are concerned, it cannot be said that it is a matter of common knowledge that injury results only where there has been negligence in its use. Particularly is this so, when it is performed upon a person with the advanced degree of arteriosclerosis possessed by plaintiff. It is a matter outside the realm of the laymen’s experience, and hence common knowledge as a basis for the application of the doctrine does not exist. See Dees v. Pace, supra, 118 Cal.App.2d 284.

- / This brings us to the question of whether there was any professional evidence calling for the application of the doctrinéis, Plaintiff’s medical witness did not testify upon this subject, but all the witnesses agree that paralysis is a rare complication of aortography. This fact does not prove that it normally does not occur in the absence of negligence. “The mere fact in itself that an unfavorable result is somewhat rare does not give rise to” the inference of negligence. (Dees v. Pace, supra, 118 Cal.App.2d 284, 289.) None of the defendants’ witnesses testified directly that the paraplegia would not occur without negligence. Dr. Wylie testified that there are risks attendant upon this procedure, that the risk of vessels constricting or occluding as the result of the drug used in the procedure is one of the risks which must be assumed and that there is little than can be done to guard against it. Dr. Naffziger testified that the risks had to be balanced against the importance to the patient of determining the exact diagnosis and the future treatment necessary.

With the exception of plaintiff’s witness Dr. Edmeads, none of the experts could determine the exact cause of the paraplegia. In effect, they stated it might have been one of three: (1) constriction of the blood vessels in the spinal cord, due to the urokon; (2) direct damage to the spinal cord from urokon in the spinal cord circulation; (3) the plaintiff’s condition, a partially blocked aorta, arteriosclerosis and high blood pressure of several years standing, obliteration of blood vessels and blood supply to legs, was such that sudden and total paralysis could occur at any moment. Their testimony was to the effect that the first two conditions could result from an aortography. Dr. Edmeads from an examination of the X-rays showing the needle in place at the times of *572both injections, opined that the needle at the time of the second injection was near or in an artery supplying blood to the spinal column and that the urokon was injected thereby into the column, and that caused the paraplegia. Defendants disagreed with this diagnosis. L/There was no medical testimony upon which res ipsa loquitur could be based unless it be Dr. Bdmeads’ testimony that the needle may have been inserted in the wrong place. There was no testimony that in aortography, without negligence, a needle could be inserted in a spinal artery. In fact, the testimony was just to the contrary, that there should be no great difficulty in inserting the needle in the aorta. Dr. Bdmeads’ testimony, if believed, would bring the case within the rule of Hierman v. Providence Hospital, 31 Cal.2d 290, 292, Ybarra v. Spangard, 25 Cal.2d 486, 490, and Meyer v. McNutt Hospital, 173 Cal. 156, 159, all supra, that where a patient is under anesthesia and a different part of his body is injured than that which should have been involved in the procedure, and there is no evidence that such a result ordinarily might occur without negligence, the doctrine applies. See Seneris v. Haas, supra, 45 Cal.2d 811, 823: “. . . the jury, under appropriate instructions, should have been permitted to determine whether each of the conditions necessary to bring into play the rule of res ipsa loquitur were present. [vVHere, there was a conflict in the testimony, defendants’ experts testifying in effect that the urokon could have affected the spinal cord even if properly injected in the aorta and that such a situation might have occurred here; plaintiff’s expert testifying in effect that the X-ray showed the needle to have been inserted in the wrong place. The jury were not told that the doctrine could apply only in the event they found that the needle had been inserted in the wrong place. On the contrary, the court instructed the jury that as a matter of law, from the “happening of all the events involved in this case, however, as established by the evidence,” (emphasis added) the inference of negligence arose. The jury were given no opportunity to determine the facts upon which the doctrine would or would not arise. This was prejudicial error.8 |x/' ■

Although there was evidence on other theories of the case that would have supported the action of the jury, nevertheless *573the judgment will have to be reversed because of these prejudicial instructions, as there is no way of telling whether the jury decided as it did because of such improper instructions, or because negligence may have been proved otherwise.

2. Dr. Gerbode’s Liability.

Assuming that there was negligence in the performance of the aortography, we find no evidence which would make Dr. Gerbode liable therefor in the absence of an agreement, express or implied, that Dr. Gerbode himself would perform it, or of evidence supporting the other theories of negligence raised by plaintiff. Beside Dr. Gerbode, four doctors testified that it was not customary for the attending physician to perform or to be present at the performance of an aortography/.'and that it is customary to have such a procedure performed by the hospital personnel who are accustomed to working together in the performance of this and other complicated diagnostic procedures and perform them ^ regularly. There was no contradiction of this testimony. / While Dr. Gerbode ordered the aortogram (and would be responsible for any negligence in prescribing such procedure) he cannot be held liable for the negligence of the team in the actual performance of it, as he neither participated in, nor had the right to direct it.AjWhen a patient is placed in a hospital his attending physician orders many procedures to be undertaken by the hospital staff or employees. Common examples are urinalysis, blood counts and X-rays. Suppose that in extracting blood for a count the hospital personnel negligently infected the patient. It could not be contended that the attending physician was liable for that negligence. The same is true here. The attending physician cannot be held liable for acts over which he had and could have no control.

This discussion is limited solely to the effect of custom, in the absence of an express or implied agreement that the attending physician will direct the procedure.

Plaintiff contends that because the majority of the aortographies performed in the bay area were performed in two hospitals, 89 at University of California Hospital, 168 at Franklin Hospital, 68 at all other hospitals (not including Fort Miley Hospital, the figures for which were not available), a total of 325, that it cannot be said there is a general custom but merely a custom of those two hospitals. This is a non sequitur. The record is not clear as to how many of this 325 were performed prior to plaintiff’s operation. Conserva*574tively at least one-half were. Assuming only 162 as the proper figure, the performance of that many surgical procedures in a given area in a particular way should be sufficient to establish a custom or practice.

Of course, the furnishing by the hospital of a surgical team would not relieve the attending surgeon of his obligation to determine the competency of such team. (One of plaintiff’s contentions is that Dr. Gerhode failed in this duty.)/ But to hold that the attending surgeon who does not participate nor have the right to participate in the procedure is liable for the acts of a competent team supplied by the hospital would be against the best interests of patients generally. The patient by the use of such a team gets the benefit of medical people who have become experts in the particular procedure. That such is not the law is established by Seneris v. Haas, supra, 45 Cal.2d 811, 828-829, where the obstetrician who was engaged in the delivery of a baby was held not liable for the negligence of the accompanying anesthetist supplied by the hospital for negligence in giving the mother a spinal anesthetic. See also Huber v. Protestant Deaconess Hospital Assn., (1956) -Ind.App. - [133 N.E.2d 864]; Wiley v. Wharton, (1941) 68 Ohio App. 345 [41 N.E.2d 255]. In Sherman v. Hartman, 137 Cal.App.2d 589 [290 P.2d 894], we held that the attending physician who left the patient in charge of a hospital nurse while a blood transfusion was still running was not liable for the nurse’s negligence. See also Hallinan v. Prindle, 17 Cal.App.2d 656 [62 P.2d 1075] (hospital surgical nurse negligently placed formalin instead of novocaine on surgical tray and surgeon injected it in patient); Hohenthal v. Smith, 72 App.D.C. 343 [114 F.2d 494] (hospital intern negligent in administering a hypodermoelysis). Gases like Ales v. Ryan, 8 Cal.2d 82 [64 P.2d 409], Key v. Caldwell, 39 Cal.App.2d 698 [104 P.2d 87], Armstrong v. Wallace, 8 Cal.App.2d 429 [47 P.2d 740], and others cited by plaintiff are not in point. They all deal with hospital employees who were under the direct control and supervision of the surgeon performing the operation.

As the case will have to be tried again, we deem it unnecessary to discuss plaintiff’s other claims of liability of defendant Gerhode such as his claim that Dr. Ellis was inexperienced, that Dr. Gerhode knew that fact and was therefore negligent in permitting him to perform the aortography, that defendant in view of plaintiff’s condition was negligent in prescribing an aortography, and further that the evidence showed an ex*575press or implied agreement that defendant would perform it himself. The court’s instructions on this subject left the determination of these matters to the jury depending upon its findings on the controverted facts.

3. Instructions on Other Theories of Dr. Gerbode’s Negligence.

A number of instructions were given upon the question of liability of Dr. Gerbode for negligence of the surgical team (admittedly employees of defendant hospital). Beading them, it is difficult to tell whether some of them were not instructions that as a matter of law Dr. Gerbode was liable for their negligence, if any, or whether the jury were to determine the facts upon which such liability was based. The jury were not told clearly that in considering whether, in the absence of an express or implied understanding that he would participate in the procedure, the attending physician’s liability for negligence of those participating in spite of a custom or practice to the contrary would depend upon whether the attending physician in view of that custom had any right to control the procedure. Two instructions, Number 25 on the “borrowed servant doctrine,” and Number 26 which seems to assume that the team was assisting Dr. Gerbode and under his direction, leave it doubtful even when read with all the other instructions in the case, whether the jury would understand that the instructions only applied in the event the jury should find the facts upon which the rules of law there stated must be based. Plaintiff advanced several theories of Dr. Gerbode’s liability. One was that as attending physician defendant Dr. Gerbode was liable for the acts of the team regardless of the general custom. Most of the other theories were based upon a conflict in the evidence, such as whether there was an express or implied contract that Dr. Gerbode would personally perform or direct the procedure, and whether Dr. Gerbode informed plaintiff of the type of procedure to be undertaken. Those were jury questions.

4. Experimentation and the Manufacturer’s Brochure.

An instruction was given to the effect that if a surgeon seeks fields of experimentation in treating his patients he is accountable for any damages proximately caused by any unskillful treatment of the patient. A further instruction was given to the effect that if urokon was injected in greater amount than that recommended by the manufacturer’s bro*576chure and if the jury found that such injection constituted experimentation, then all participating would be guilty of negligence unless the patient was first warned of the experimentation and consented to it. The first question to be considered hereunder is the admissibility of the brochure. It was published by the manufacturer of the sodium urokon. It stated that for translumbar aortography in an adult, “10 to 15 ec of 70% Urokon is adequate” and that aortography should not be repeated within 24 hours. The parties differ as to the meaning of the language. Plaintiff contends it negatives a second injection. Defendants contend it only negatives a second insertion of the needle, and that the second injection, the needle remaining in place after the first injection, in nowise contravenes the instruction. The uncontradieted evidence was that the second injection if deemed necessary is customary. In the “first run” in plaintiff’s case, 30 e.c.’s, and in the “second run” 20 c.c.’s, a.total of 50 c.c.’s, were injected. Dr. Wylie testified that it was customary to use 50 c.e.’s on the first run; Dr. Williams that the customary dosage was from 30 to 70 c.e.’s; Dr. Stone that 30 e.c.’s were the customary first run dosage; Dr. Abrams that 50 c.c.’s were the custom frequently followed by an additional 50 c.e. ’s on the second run. There was no expert testimony that the amount used was improper. It is rather significant that the brochure recommends the use of 50 c.e. ’s of urokon in another procedure involving injection of the contrast medium for visualization of the heart itself.

Defendants and amicus curiae urge that a manufacturer’s brochure is not admissible in evidence and does not establish a standard of care. They contend drug manufacturers’ recommendations are always conservative and are quickly outdated, that they expect and the custom is that after a material has been available for a period of time, physicians using it rely primarily on their own experience and the published literature of colleagues concerning its use in actual practice. They contend that the miraculous developments which have taken place in the effective use of antibiotics and other drugs might never have been accomplished if physicians were required to follow blindly the suggestions of the manufacturers who prepare but do not use them.

No objection was made to the introduction of the brochure at this trial, but the matter must be determined for the benefit of the court at a retrial. Julien v. Barker, (1954) 75 Idaho 413 [272 P.2d 718, 724], held admissible an instruc*577tion sheet issued by the manufacturer of and accompanying the anesthetic charged to have been negligently administered. The court stated (p. 724): “. . . it is not conclusive evidence of standard or accepted practice in the use of the drug by physicians and surgeons, nor that a departure from such directions is negligent. But it is prima facie proof of a proper method of use, given by the maker, which must be presumed qualified to give directions for its use and warnings of any danger inherent therein. [Citations.] ” Thus, while admissible, it cannot establish as a matter of law the standard of care required of a physician in the use of the drug. It may be considered by the jury along with the other evidence in the case to determine whether the particular physician met the standard of care required of him. The court’s instruction on the subject should have been limited to this effect.

The mere fact of a departure from the manufacturer’s recommendation where such departure is customarily followed by physicians of standing in the locality does not make that departure an “experiment.” There was in this case no evidence of experiment and the instructions concerning “experiment” should not have been given. Instructions without support in the evidence should not be given. (Rodenberger v. Frederickson, 111 Cal.App.2d 139, 142 [244 P.2d 107].)

5. Instructions.

(a) Duty to Call a Specialist.

It is difficult to understand the basis for these instructions. They informed the jury that it is the duty of a general practitioner to call in a specialist if the prudent practitioner would have done so. (These did not deal with the performance of the aortography by Dr, Ellis.) The evidence showed that Dr. Carson, plaintiff’s physician, referred him to Dr. Gerbode because he was a specialist in the field of vascular surgery in which aortography is a diagnostic adjunct. It also showed that he is eminent in that field. He had had much experience with aortography before sodium urokon was used. His supposed ignorance is blamed upon his not knowing what was contained in the brochure, not being familiar with the Standard Nomenclature of Diseases and Operations published by the American Medical Association, and not knowing the effect of urokon injected elsewhere than in the aorta. The uncontradicted expert testimony was that the aortography was sound *578medical practice under the circumstances. There was nothing in the evidence justifying the giving of these instructions.

(b) Duty to Disclose.

Plaintiff, his wife and son testified that plaintiff was not informed that anything in the nature of an aortography was to be performed. Dr^-Gerbode-and-Dr-.—Bllis^contradicted this, although admitting that the details of the procedure and the possible dangers therefrom were not explained. The court gave a rather broad instruction upon the duty of a physician to disclose to the patient “all the facts which mutually affect his rights and interests and of the surgical risk, hazard and danger, if any. ...” A physician violates his duty-(to his patient and subjects himself to liability if he withholds any facts 'which are [necessary to form the basis of an intelligent consent by the patient to the proposed treatment. Likewise the physician may not minimize the known dangers of a procedure or operation in order j to induce his patient’s consent. At the same time, the physician must place the welfare of his patient above all else and this very fact places him in a position in which he sometimes must choose between two alternative courses of action. One is to explain to the patient every risk attendant upon any surgical procedure or operation, no matter how remote; this may well result in alarming a patient who is already unduly apprehensive and who may as a result refuse to undertake surgery in which there is in fact minimal risk; it may also result in actually increasing the risks by reason of the physiological results of the apprehension itself. The other is to recognize that each patient presents a separate problem, that the patient’s mental and emotional condition is important and in certain cases may be crucial, and that in discussing the element of risk a certain amount of discretion must be employed consistent with the full disclosure of facts necessary to an informed consent. (Hunt v. Bradshaw (1955), 242 N.C. 517 [88 S.E.2d 762]; cf. Simone v. Sabo (1951), 37 Cal.2d 253 [231 P.2d 19] ; Schloendorff v. Society of New York Hospital (1914), 211 N.Y. 125 [105 N.E. 92, 52 L.R.A.N.S. 505].)

The instruction given should be modified to inform the jury that the physician has such' discretion consistent, of course, with the full disclosure of facts necessary to an informed consent.

*579(c) Failure to Produce Evidence.

Three instructions were given on this subject, one general and two applying only to the defendants. They should not have been given. There was no evidence in the case to justify the instructions. Defendants produced the witnesses engaged in the procedure and endeavored to explain the cause of the injury. Whether they successfully explained it or negatived the facts upon which the charge of negligence was based was a matter for the jury, and the jury were so told. (There were many instructions on the duty and liability of defendants, unnecessarily repetitive.) There was nothing in the case- to show, whether res ipsa loquitur applied or did not, that there was any suppression of evidence or failure to produce any evidence by defendants.

6. Medical Texts As Evidence.

One of the questions which will undoubtedly be raised at a retrial is the admissibility of medical text books, pamphlets and periodicals. In Gluckstein v. Lipsett, 93 Cal.App.2d 391 [209 P.2d 98]9, we reviewed the eases in California dealing with the subject and held that on direct examination medical text books were not admissible. On cross-examination they were admissible only if the expert witness “has based his opinion either generally or specifically on a medical work or medical works. . . .” (P. 403.) This rule has been in nowise changed. This rule does not permit reading to a witness who has not based his opinion on a medical work, text or brochure, extracts therefrom as a part of a question.

7. Reference to Malpractice Judgments.

It is improper in argument to refer to judgments given in malpractice actions whether brought against doctors or lawyers or to refer to any protection either has against such actions.

The judgment is reversed.

Peters, P. J., and Wood (Fred B.), J., concurred.

A petition for a rehearing was denied November 21, 1957, and respondent’s petition for a hearing by the Supreme Court was denied December 18, 1957. Kaufman, J. pro tem.,* participated therein in place of Spence, J. Gibson, C. J., and Carter, J., were of the opinion that the petition should be granted.

SPOTTSWOOD W. ROBINSON, III, Circuit Judge:

This appeal is from a judgment entered in the District Court on verdicts directed for the two appellees at the conclusion of plaintiff-appellant Canterbury’s case in chief. His action sought damages for personal injuries allegedly sustained as a result of an operation negligently performed by appellee Spence, a negligent failure by Dr. Spence to disclose a risk of serious disability inherent in the operation, and negligent post-operative care by appellee Washington Hospital Center. On close examination of the record, we find evidence which required submission of these issues to the jury. We accordingly reverse the judgment as to each appellee and remand the case to the District Court for a new trial.

I

The record we review tells a depressing tale. A youth troubled only by back pain submitted to an operation without being informed of a risk of paralysis incidental thereto. A day after the operation he fell from his hospital bed after having been left without assistance while voiding. A few hours after the fall, the lower half of his body was paralyzed, and he had to be operated on again. Despite extensive medical care, he has never been what he was before. Instead of the back pain, even years later, he hobbled about on crutches, a vietim of paralysis of the bowels and urinary incontinence. In a very real sense this lawsuit is an understandable search for reasons.

At the time of the events which gave rise to this litigation, appellant was nineteen years of age, a clerk-typist employed by the Federal Bureau of Investigation. In December, 1958, he began to experience severe pain between his shoulder blades.1 He consulted two general practitioners, but the medications they prescribed failed to eliminate the pain. Thereafter, appellant secured an appointment with Dr. Spence, who is a neurosurgeon.

Dr. Spence examined appellant in his office at some length but found nothing amiss. • On Dr. Spence’s advice appellant was x-rayed, but the films did not identify any abormality. Dr. Spence then recommended that appellant undergo a myelogram — a procedure in which dye is injected into the spinal column and traced to find evidence of disease or other disorder — at the Washington Hospital Center.

Appellant entered the hospital on February 4, 1959.2 Thg myelogram revealed a “filling defect” in the region of the fourth thoracic vertebra. Since a myelogram often does no more than pinpoint *777the location of an aberration, surgery may be necessary to discover the cause. Dr. Spence told appellant that he would have to undergo a laminectomy — -the excision of the posterior arch of the vertebra — to correct what he suspected was a ruptured disc. Appellant did not raise"! any objection to the proposed operation nor did he probe into its exact nature.

Appellant explained to Dr. Spence that his mother was a widow of slender financial means living in Cyclone, West Virginia, and that she could be reached through a neighbor’s telephone. Appellant called-his mother the day after the myelogram was performed and, failing to contact her, left Dr. Spence’s telephone number with the neighbor. When Mrs. Canterbury returned the call, Dr. Spence told her that the surgery was occasioned by a suspected ruptured disc. Mrs. Canterbury then asked if the recommended-operation was serious and Dr. Spence replied “not anymore than any other operation.” He added that he knew Mrs. Canterbury was not well off and that her presence in Washington would not be necessary. The testimony is contradictory as to whether during the course of the conversation Mrs. Canterbury expressed her consent to the operation. Ap pellant himself apparently did not converse again with Dr. Spence prior to the operation.

Dr. Spence performed the laminectomy on February 113 at the Washington Hospital Center. Mrs. Canterbury traveled to Washington, arriving on that date but after the operation was over, and signed a consent-form at .the hospital. The laminectomy revealed several anomalies : a spinal cord that was swollen and unable to pulsate, an accumulation of large tortuous and dilated veins, and a complete absence of epidural fat which normally surrounds the spine. A thin hypodermic needle was inserted into the spinal cord to aspirate any cysts which might have .been present, but no fluid emerged. In suturing the wound, Dr. Spence attempted to relieve the pressure on the spinal cord by enlarging the dura —the outer protective wall of the spinal eord^-at the area of swelling.

For approximately the first day after the operation appellant recuperated normally, but then suffered a fall and an almost immediate setback. Since there is some conflict as to precisely when or why appellant fell,4 we reconstruct the events from the evidence most favorable to him.5 Dr. Spence left orders that appellant was to remain in bed during the process of voiding. These orders were changed to direct that voiding be done out of bed, and the jury could find that the change was made by hospital personnel. Just prior to the fall, appellant summoned a nurse and was given a receptacle for use in voiding, but was then left unattended. Appellant testified that during the course of the endeavor he slipped off the side of the bed, and that there was no one to assist him, or side rail to prevent the fall.

Several hours later, appellant began to complain that he could not move his legs and that he was having trouble breathing; paralysis seems to have been virtually total from the waist down. Dr. Spence was notified on the night of February 12, and he rushed to the hospital. Mrs. Canterbury signed another consent form and appellant was again taken into the operating room. The surgical wound was reopened and Dr. Spense created a gusset to allow the spinal cord greater room in which to pulsate.

Appellant’s control over his muscles improved somewhat after the second operation but he was unable to void properly. As a result of this condition, he came under the care of a urologist while *778still in the hospital. In April, following a cystoscopic examination, appellant was operated on for removal of bladder stones, and in May was released from the hospital. He reentered the hospital the following August for a 10-day period, apparently because of his urologic problems. For several years after his discharge he was under the care of several specialists, and at all times was under the care of a urologist. At the time of the trial in April, 1968, appellant required crutches to walk, still suffered from urinal incontinence and paralysis of the bowels, and wore a penile clamp.

In November, 1959 on Dr. Spence’s recommendation, appellant was transferred by the F.B.I. to Miami where he could get more swimming and exercise. Appellant worked three years for the F.B.I. in Miami, Los Angeles and Houston, resigning finally in June, 1962. From then until the time of the trial, he held a number of jobs, but had constant trouble finding work because he needed to remain seated and close to a bathroom. The damages appellant claims include extensive pain and suffering, medical expenses, and loss of earnings.

II

Appellant filed suit in the District Court on March 7, 1963, four years after the laminectomy and approximately two years after he attained his majority. The complaint stated several causes of action against each defendant. Against Dr. Spence it alleged, among other things, negligence in the performance of the laminectomy and failure to inform him beforehand of the ..risk involved. Against the hospital the complaint charged negligent post-operative care in permitting appellant to remain unattended after the laminectomy, in failing to provide a nurse or orderly to assist him at the time of his fall, and in failing to maintain a side rail on his bed. The answers denied the allegations of negligence and defended on the ground that the suit was barred by the statute of lim-** itations.

Pretrial discovery — including depositions by appellant, his mother and Dr. Spence — continuances and other delays consumed five years. At trial, disposition of the threshold question whether the statute of limitations had run was held in abeyance until the relevant facts developed. Appellant introduced no evidence to show medical and hospital practices, if any, customarily pursued in regard to the critical aspects of the case, and only Dr. Spence, called as an adverse witness, testified on the issue of causality. Dr. Spence described the surgical procedures he utilized in the two operations and expressed his opinion that appellant’s disabilities stemmed from his pre-operative condition as symptomized by the swollen, non-pulsating spinal cord. He stated, however, that neither he nor any of the other physicians with whom he consulted was certain as to what that condition was, and he admitted that trauma can be a cause of paralysis. Dr. Spence further testified that even without trauma paralysis can be anticipated “somewhere in the nature of one percent” of the laminectomies performed, a risk he termed “a very slight possibility.” He .feltthat .communication of that..risk-to the patient is not good medical practice because it might deter 'patients. from undergoing needed surgery and might produce' adverse psychological reactions which eoúld preclude the success of the operation.

At the close of appellant’s case in chief, each defendant moved for a directed verdict and the trial judge granted both motions. The basis of the ruling, he explained, was that appellant had failed to produce any medical evidence indicating negligence on Dr. Spence’s part in diagnosing appellant’s malady or in performing the laminectomy; that there was no proof that Dr. Spence’s treatment was responsible for appellant’s disabilities; and that notwithstanding some evidence to show negligent post-operative care, an absence of medical testimony to show causality precluded submission of the case against the hospital to the jury. *779The judge did not allude specifically to the alleged breach of duty by Dr. Spence to divulge the possible consequences of the laminectomy.

We reverse. The testimony of appellant and his mother that Dr. Spence did not reveal the risk of paralysis from the laminectomy made out a prima facie case of violation of the physician’s duty to disclose which Dr. Spence’s explanation-did not negate as a matter of law. There was also testimony from which the jury could have found that the laminectomy was negligently performed by Dr. Spence, and that appellant’s fall was the consequence of negligence on the part of the hospital. The record, moreover, contains evidence of sufficient quantity and quality to tender jury issues as to whether and to what extent any such negligence was causally related to appellant’s post-laminectomy condition. These considerations entitled appellant to a new trial.

Elucidation of our reasoning necessitates elaboration on a number of points. In Parts III and IV we explore the origins and rationale of the physician’s duty to reasonably inform an ailing patient as to the treatment alternatives available and the risks incidental to them. In Part V we investigate the scope of the disclosure requirement and in Part VI the physician’s privileges not to disclose. In Part VII we examine the role of causality, and in Part VIII the need for expert testimony in non-disclosure litigation. In Part IX we deal with appellees’ statute of limitations defense and in Part X we apply the principles discussed to the case at bar.

Ill

Suits charging failure by a physician 6 adequately to disclose the risks and alternatives of proposed treatment are not innovations In American law. They date back a good half-century,7 and in the last decade they have multiplied rapidly.8 There is, nonetheless, disagreement among the courts and the commentators 9 on many major questions, and there is no precedent of our own directly in point.10 For the tools enabling resolu*780tion of the issues on this appeal, we are forced to begin at first principles.11

The root premise is the concept, fundamental in American jurisprudence, that “[e]very human being of adult years and sound mind has a right to determine what shall be done with his own body. . . . ” 12 True consent to what happens to one’s self is the informed exercise of a choice, and that entails an opportunity to evaluate knowledgeably the options available and the risks attendant upon each.13 The average patient has little or no understanding of the medical arts, and ordinarily has only his physician to whom he can look for enlightenment with which to reach an intelligent decision.14 From these almost axiomatic considerations springs the need, and in turn the requirement, of a reasonable divulgenee by physician to patient to make such a decision possible.15

*781A physician is under a duty to treat his patient skillfully16 but proficiency in diagnosis and therapy is not the full measure of his responsibility. The cases demonstrate that the physician is under an obligation to communicate specific information to the patient when the exigencies of reasonable care call for it.17 Due care may require a physician perceiving symptoms of bodily abnormality to alert the patient to the condition.18 It may call upon the physician confronting an ailment which does not respond to his ministrations to inform the patient thereof.19 It may command the physician to instruct the patient as to any limitations to be presently observed for his own welfare,20 and as to any precautionary therapy he should seek in the future.21 It may oblige the physician to advise the patient of the need for or desirability of any alternative treatment promising greater benefit than that being pursued.22 Just as plainly, due care normally demands that the physician warn the patient of any risks to his well-being which contemplated therapy may involve.23

The context in which the duty of risk-disclosure arises is invariably the occasion for decision as to whether a particular treatment procedure is to be undertaken. To the physician, whose training enables a self-satisfying evaluation, the answer may seem clear, but it is the prerogative of the patient, not the physician, to determine for himself the direction in which his interests seem to lie.24 To enable the patient to chart his course understandably, some familiarity with the therapeutic alternatives and their hazards becomes essential.25

*782A reasonable revelation in these respects is not only a necessity but, as we see it, is as much a matter of the physician’s duty. Ijt is a duty to warn of the dangers lurking in the proposed treatment, and that is surely a facet of due care.26 It is, too, a duty to impart information which the patient has every right to expect.27 The patient’s reliance upon the physician is a trust of the kind which traditionally has exacted obligations beyond those associated with arms-length transactions.28 His dependence upon the physician for information affecting his well-being,' in terms of contemplated treatment, is well-nigh abject. As earlier noted, long before the instant litigation arose, courts had recognized that the physician had the responsibility of satisfying the vital informational needs of the patient.29 | More recently, we ourselves have found “in the fiducial qualities of [the physician-patient] relationship the physician’s duty to reveal to the patient that which in his best interests it is important that he should know.” 30 We now find, as a part of the physician’s overall obligation to the patient, a similar duty of reasonable disclosure of the choices with respect to proposed therapy and the dangers inherently and potentially involved.31

This disclosure requirement, on analysis, reflects much more of a change in doctrinal emphasis than a substantive addition to malpractice law. It is well established that the physician must seek and secure his patient’s consent before commencing an operation or other course of treatment.32 It is also *783clear that the consent, to be efficacious, must be free from imposition upon the patient.33 It is the settled rule that therapy not authorized by the patient may amount to a tort — a common law battery —by the physician.34 And it is evident that it is normally impossible to obtain a consent worthy of the name unless the physician first elucidates the options and the perils for the patient’s edification.35 Thus the physician has long borne a duty, on pain of liability for unauthorized treatment, to make adequate disclosure to the patient.36 The evolution of the obligation to communicate for the patient’s benefit as well as the physician’s protection has hardly involved an extraordinary restructuring of the law.

IV

Duty to disclose has gained recognition in a large number of American jurisdietions,37 but more largely on a different rationale. The majority of courts dealing with the problem have made the duty depend on whether it was the custom of physicians practicing in the community to make the particular disclosure to the patient.38 If so, the physician may be held liable for an unreasonable and injurious failure to divulge, but there can be no recovery unless the omission forsakes a practice prevalent in the profession.39 We agree that the physician’s noncompliance with a professional custom to reveal, like any other departure from prevailing medical practice,40 may give rise to liability to the patient. We do not agree that the patient’s cause of action is dependent upon the existence and nonperformance of a relevant professional tradition.

There are, in our view, formidable obstacles to acceptance of the notion that the physician’s obligation to disclose is either germinated or limited by medical practice. To begin with, the reality of any discernible custom reflecting a professional concensus on communication of option and risk information to patients is open to serious doubt.41 We sense the danger that what in fact is no *784custom at all may be taken as an affirmative custom to maintain silence, and that physician-witnesses to the so-called custom may state merely their personal opinions as to what they or others would do under given conditions.42 We cannot gloss over the inconsistency between reliance on a general practice respecting divulgence and, on the other hand, realization that the myriad of variables among patients 43 makes each case so different that its omission can rationally be justified only by the effect of its individual circumstances.44 Nor can we ignore the fact that to bind the disclosure obligation to medical usage is to arrogate the decision on revelation to the physician alone.45 Respect for the patient’s right of self-determination on particular therapy46 demands a standard set by law for physicians rather than one which physicians may or may not impose .upon themselves.47

More fundamentally, the majority rule overlooks the graduation of reasonable-care demands in Anglo-American jurisprudence and the position of professional custom in the hierarchy. The caliber of the performance exacted by the reasonable-care standard varies between the professional and non-professional worlds, and so also the role of professional custom. “With but few exceptions,” we recently declared, “society demands that everyone under a duty to use care observe minimally a general standard.” 48 “Familiarly expressed judicially,” we added, “the yardstick is that degree of care which a reasonably prudent person would have exercised under the same or similar circumstances." 49 “Beyond this,” however, we emphasized, “the law requires those engaging in activities requiring unique knowledge and ability to give a performance commensurate with the undertaking.”50 Thus physicians treating the sick must perform at higher levels than non-physicians in order to meet the reasonable care standard in its special application to physicians51— “that degree of care and skill ordinarily exercised by the profession in [the physician’s] own or similar localities.” 52 And practices adopted by the profession have indispensable value as evidence tending to establish just what that degree of care and skill is.53

We have admonished, however, that “[t]he special medical standards54 are but. adaptions of the general standard to a group who are required to act as *785reasonable men possessing their medical talents presumably would.”55 There is, by the same token, no basis for operation of the special medical standard where the physician’s activity does not bring his .medical knowledge and skills peculiarly .into play.56 And where the challenge to the physician’s conduct is not to be gauged by the special standard, it follows that medical custom cannot furnish the test of its propriety, whatever its relevance under the proper test may be.57 The decision to unveil the patient’s condition and the chances as to remediation, as we shall see.Js ofttimes a non-medical judgment58 and, if so, is a decision out.side the ambit of the special standard. Where that is the situation, professional custom hardly furnishes the legal criterion for measuring the physician’s responsibility to reasonably inform his patient of the options and the hazards as to treatment.

The majority rule, moreover, is at war with our prior holdings that a showing of medical practice, however probative, does not fix the standard governing recovery for medical malpractice.59 .Prevailing medical practice, we have maintained, has_evidentiary value in determinations as to what the specific criteria measuring challenged professional conduct are and whether they have been met,60 but does not itself define the standard.61 That has been our position in treatment cases, where the physician’s performance is ordinarily to be adjudicated by the special medical standard of due care.62 We see no logic in a different rule for nondisclosure cases, where the governing standard is much more largely divorced from professional considerations.63 And surely in nondisclosure cases the factfinder is not invariably functioning in an area of such technical complexity that it must be bound to medical custom as an inexorable application of the community standard of reasonable care.64

Thus we distinguished, for purposes of duty to disclose, the speeial- and general-standard aspects of the physician-patient relationship. When medical judgment enters the picture and for 'that reason the special standard controls, prevailing medical practice must be given its just due. In all other instances, however, the general standard exacting ordinary care applies, and that standard is set by law. In sum, the physician’s duty to disclose is governed by the same legal principles applicable to others in comparable situations, with modifications only to the extent that medical judgment enters the picture.65 We hold that the standard measuring, performance of that duty by physicians, as by others, is conduct which is reasonable under the circumstances.66

*786V

Once the circumstances give rise to a duty on the physician’s part to inform his patient, the next inquiry is the_scope of the disclosure the physician is legally obliged to make. The courts have frequently confronted this problem but no uniform standard defining the adequacy of the divulgenee emerges from the decisions. Some have said “full” disclosure,67 a norm we are unwilling to adopt literally. It seems obviously prohibitive and unrealistic to expect physicians to discuss with their patients every risk of proposed treatment — no matter how small or remote68 — and generally unnecessary from the patient’s viewpoint as well. Indeed, the cases speaking in terms of “full” disclosure appear to envision something less than total disclosure,69 leaving unanswered the question of just how much.

The larger number of courts, as might be expected, have applied tests framed with reference to prevailing fashion within the medical profession.70 Some have measured the disclosure by “good medical practice,”71 others by what a reasonable practitioner would have bared under the circumstances,72 and still others by what medical custom in the community would demand.73 JVe have explored this rather considerable body of law but are unprepared to follow it. The duty to disclose, we have reasoned, arises from phenomena apart from medical custom and practice.74 The latter, we think, should no more establish the scope of the duty than its existence. Any definition of scope in terms purely of a professional standard is at odds with the patient’s prerogative to decide on projected therapy himself.75 That prerogative, we have said, is at the very foundation of the duty to disclose,76 and both the patient’s right to know and the physician’s correlative obligation to tell him are diluted to the extent that its compass is dictated by the medical profession.77

In our view, the patient’s right of self-decision shapes the boundaries of the duty to reveal. That right can be effectively exercised only if the patient possesses enough information to enable an intelligent choice.’ The scope of the physician’s communications to the patient, then, must bef measured by the patient’s need,78 and that need is the information material to the decision. Thus tlie test for determining whether a par*787ticular peril must be divulged is its materiality to the patient’s decision: all risks, potentially affecting the decision must be unmasked.79 And to safeguard the patient’s interest in achieving his own determination on treatment, the law must itself set the standard for adequate disclosure.80

Optimally for the patient, exposure of a risk would be mandatory whenever the patient would deem it significant to his decision, either singly or in combination with other risks. Such a requirement, however, would summon the physician to second-guess the patient, whose ideas on materiality could hardly be known to the physician. That would make an undue demand upon medical practitioners, whose conduct, like that of others, is to be- measured in terms of reasonableness. Consonantly with orthodox negligence doctrine, the physician’s liability for nondisclosure is to be determined on the basis of foresight, not hindsight; no less than any other aspect of negligence, the issue on nondisclosure must be approached from the viewpoint of the reasonableness of the physician’s divulgence jn terms of what he knows or should know to be the patient’s informational needs. If, but only if, the fact-finder can say that the physician’s communication was unreasonably inadequate is an imposition of liability legally or morally justified.81

Of necessity, the content of the disclosure rests in the first instance with the physician. Ordinarily it is only he who is in position to identify particular dangers; always he must make a judgment, in terms of materiality, as to whether and to what extent revelation to the patient is called for. He cannot know with complete exactitude what the patient would consider important to his decision, but on the basis of his medical training and experience he can sense how the average, reasonable patient expectably would react.82 Indeed, with knowledge of, or ability to learn, his patient’s background and current condition, he is in a position superior to that of most others — attorneys, for example — who are called upon to make judgments on pain of liability in damages for unreasonable miscalculation.83

From these considerations we derive the breadth of the disclosure of risks legally to be required. The scope of the standard is not subjective as to either the physician or the patient; it remains objective with due regard for the patient’s informational needs and with suitable leeway for the physician’s situation. In broad outline, we agree that,“[a] risk is thus material when a reasonable person, in what the physician knows or should know to be the patient’s position, would be likely to attach significance to the risk or.cluster of risks in deciding whether or not to forego the proposed therapy.” 84

The topics importantly demanding a communication of information are the inherent and potential hazards of the proposed treatment, the alternatives to *788that treatment, if any, and the results likely if the patient remains untreated. The factors contributing significance to the dangerousness of a medical technique are, of course, the incidence of injury and the degree of the harm threatened.85 A very small chance of death or serious disablement may well be significant; a potential disability which dramatically outweighs the potential benefit of the therapy or the detriments of the existing malady may summons discussion with the patient86

There is no bright line separating the significant from the insignificant ; the answer in any case must abide a rule of reason. Some dangers — infection, for example — are inherent in any operation; there is no obligation to communicate those of which persons of average sophistication are aware.87 Even more clearly, the physician bears no responsibility for discussion of hazards the patient has already discovered,88 or those having no apparent materiality to patients’ decision on therapy.89 The disclosure doctrine, like others marking lines between permissible and impermissible behavior in medical practice, is in essence a requirement of conduct prudent under the circumstances. Whenever nondisclosure of particular risk information is open to debate by reasonable-minded men, the issue is for the finder of the facts.90

VI

Two exceptions to the general rule of disclosure have been noted by the courts. Each is in the nature of a physician’s privilege not to disclose, and the reasoning underlying them is appealing. Each, indeed, is but a recognition that, as important as is the patient’s right to know, it is greatly outweighed by the magnitudinous circumstances giving rise to the privilege. The first comes into play when the patient is unconscious or otherwise incapable of consenting, and harm from a failure to treat is imminent and outweighs any harm threatened by the proposed treatment. When a genuine emergency of that sort arises, it is settled that the impracticality of confer*789ring with the patient dispenses with need for it.91 Even in situations of that character the physician should, as current law requires, attempt to secure a relative’s consent if possible 92 But if time is too short to accommodate discussion, obviously the physician should proceed with the treatment.93

The second exception obtains when risk-disclosure poses such a threat of detriment to the patient as to become unfeasible or contraindicated from a medical point of view. It is recognized that patients occasionally become so ill or emotionally distraught on disclosure as to foreclose a rational decision, or complicate or hinder the treatment, or perhaps even pose psychological damage to the patient.94 Where that is so, the cases have generally held that the physician is armed with a privilege to keep the information from the patient,95 and we think it clear that portents of that type may justify the physician in action he deems medically warranted. The critical inquiry is whether the physician responded to a sound medical judgment that communication of the risk information would present a threat to the patient’s well-being.

The physician’s privilege to withhold information for therapeutic reasons must be carefully circumscribed, however, for otherwise it might devour the disclosure rule itself. The privilege does not accept the paternalistic notion that the physician may remain silent simply because divulgence might prompt the patient to forego therapy the physician feels the patient really needs.96 That attitude presumes instability or perversity for even the normal patient, and runs counter to the foundation principle that the patient should and ordinarily can make the choice for himself.97 Nor does the privilege contemplate operation save where the patient’s reaction to risk information, as reasonable foreseen by the physician, is menacing.98 And even in a situation of that kind, disclosure to a close relative with a view to securing consent to the proposed treatment may be the only alternative open to the physician.99

*790VII

No more than breach of any other legal duty does nonfulfillment of the physician’s obligation to disclose alone establish liability to the patient. An unrevealed risk that should have been made known must materialize, for otherwise the omission, however unpardonable, is legally without consequence. Occurrence of the risk must be harmful to the patient, for negligence unrelated to injury is nonaetionable.100 And, as in malpractice actions generally,101 there must be a causal relationship between the physician’s failure to adequately divulge and damage to the patient.102

A causal connection exists when, but only when, disclosure of significant risks incidental to treatment would have resulted in a decision against it.103 The patient obviously has no complaint if he would have submitted to the therapy notwithstanding awareness that the risk was one of its perils. On the other hand, the very purpose of the disclosure rule is to protect the patient against consequences which, if known, he would have avoided by foregoing the treatment.104 The more difficult question is whether the factual issue on causality calls for an objective or a subjective determination.

It has been assumed that the issue is to be resolved according to whether the factfinder believes the patient’s testimony that he would not have agreed to the treatment if he had known of the danger which later ripened into injury.105 We think a technique which ties the factual conclusion on causation simply to the assessment of the patient’s credibility is unsatisfactory. To be sure, the objective of risk-disclosure is preservation of the patient’s interest in intelligent self-choice on proposed treatment, a matter the patient is free to decide for any reason that appeals to him.106 When, prior to commencement of therapy, the patient is sufficiently informed on risks and he exercises his choice, it may truly be said that he did exactly what he wanted to do. But when causality is explored at a post-injury trial with a professedly uninformed patient, the question whether he actually would have turned the treatment down if he had known the risks is purely hypothetical: “Viewed from the point at which he had to decide, would the patient have decided differently had he known something he did not know?”107 And the answer which the patient supplies hardly represents more than a guess, perhaps tinged by the circumstance that the uncommunicated hazard has in fact materialized.108

In our view, this method of dealing with the issue on causation comes in second-best. It places the physician in jeop*791ardy of the patient’s hindsight and bitterness. It places the factfinder in the position of deciding whether a speculative answer to a hypothetical question is to be credited. It calls for a subjective determination solely on testimony of a patient-witness shadowed by the occurrence of the undisclosed risk.109

Better it is, we believe, to resolve the causality issue on an objective basis: in terms of what a prudent person in the patient’s position would have decided if suitably informed of all perils bearing significance.110 If adequate disclosw. could reasonably be expected to llave caused that person to decline the treatment because of the revelation of the kind of risk or danger that resulted in harm, causation is shown, but otherwise not.111 The patient’s testimony is relevant on that score of course but it would not threaten to dominate the findings. And since that testimony would probably be appraised congruently with the factfinder’s belief in its reasonableness, the case for a wholly objective standard for passing on causation is strengthened. Such a standard would in any event ease the fact-finding process and better assure the truth as its product.

VIII

In the context of trial of a suit claiming inadequate disclosure of risk information by a physician, the patient has the burden of going forward with evidence tending to establish prima facie the essential elements of the cause of action, and ultimately the burden of proof — the risk of nonpersuasion112— on those elements.113 These are normal impositions upon moving litigants, and no reason why they should not attach in nondisclosure cases is apparent. The burden of going forward with evidence pertaining to a privilege not to disclose,114 however, rests properly upon the physician. This is not only because the patient has made out a prima facie case before an issue on privilege is reached, but also because any evidence bearing on the privilege is usually in the hands of the physician alone. Requiring him to open the proof on privilege is consistent with judicial policy laying such a burden on the party who seeks shelter from an exception to a general rule and who is more likely to have possession of the facts.115

As in much malpractice litigation,116 recovery in nondisclosure lawsuits has hinged upon the patient’s ability to prove through expert testimony that the physician’s performance departed from medical custom. This is not surprising since, as we have pointed out, the majority of American jurisdictions have limited the patient’s right to know to whatever boon can be found in medical practice.117 We have already discussed our disagreement with the majority rationale.118 We now delineate our view on the need for expert testimony in nondisclosure cases.

There are obviously important roles for medical testimony in such cases, and some roles which only medical evidence can fill: Experts are ordinarily indispensible to identify and elucidate for the factfinder the risks of therapy and *792the consequences of leaving existing maladies untreated. They are normally needed on issues as to the cause of any injury or disability suffered by the patient and, where privileges are asserted, as to the existence of any emergency claimed and the nature and seriousness of any impact upon the patient from risk-disclosure. Save for relative infrequent instances where questions of this type are resolvable wholly within the realm of ordinary human knowledge and experience, the need for the expert is clear.119

The guiding consideration our decisions distill, however, is that medical facts are for medical experts120 and other facts are for any witnesses- — expert or not — having sufficient knowledge and capacity to testify to them.121 It is evident that many of the issues typically involved in nondisclosure cases do not reside peculiarly within the medical domain. Lay witness testimony can competently establish a physician’s failure to disclose particular risk information, the patient’s lack of knowledge of the risk, and the adverse consequences following the treatment.122 Experts are unnecessary to a showing of the materiality of a risk to a patient’s decision on treatment, or to the reasonably, expectable effect of risk disclosure on the decision.123 These conspicuous examples of permissible uses of nonexpert testimony illustiv. the relative freedom of broad areas of the legij problem of risk nondisclosure from the demands for expert testimony that shackle plaintiffs’ other types of medical malpractice litigation.124

*793IX

We now confront the question whether appellant’s suit was barred, wholly or partly, by the statute of limitations. The statutory periods relevant to this inquiry are one year for battery actions125 and three years for those charging negligence.126 For one a minor when his cause of action accrues, they do not begin to run until he has attained his majority.127 Appellant was nineteen years old when the laminectomy and related events occurred, and he filed his complaint roughly two years after he reached twenty-one. Consequently, any claim in suit subject to the one-year limitation came too late.

Appellant’s causes of action for the allegedly faulty laminectomy by Dr. Spence and allegedly careless post-operative care by the hospital present no problem. Quite obviously, each was grounded in negligence and so was governed by the three-year provision.128 The duty-to-disclose claim appellant asserted against Dr. Spence, however, draws another consideration into the picture. We have previously observed that an unauthorized operation constitutes a battery, and that an uninformed consent to an operation does not confer the necessary authority.129 If, therefore, appellant had at stake no more than a recovery of damages on account of a laminectomy intentionally done without intelligent permission, the statute would have interposed a bar.

It is evident, however, that appellant had much more at stake.130 His interest in bodily integrity commanded protection, not only against an intentional invasion by an unauthorized operation 131 but also against a negligent invasion by his physician’s dereliction of duty to adequately disclose.132 Appellant has asserted and litigated a violation of that duty throughout the case.133 That claim, like the others, was governed by the three-year period of limitation applicable to negligence actions 134 and was *794unaffected by the fact that its alternative was barred by the one-year period pertaining to batteries.135

X

This brings us to the remaining question, common to all three causes of action: whether appellant’s evidence was of such caliber as to require a submission to the jury. On the first, the evidence was clearly sufficient to raise an issue as to whether Dr. Spence’s obligation to disclose information on risks was reassonably met or was excused by the surrounding circumstances. Appellant testified that Dr. Spence revealed to him nothing suggesting a hazard associated with the laminectomy. His mother testified that, in response to her specific inquiry, Dr. Spence informed her that the laminectomy was no more serious than any other operation. When, at trial, it developed from Dr. Spence’s testimony that paralysis can be expected in one percent of laminectomies, it became the jury’s responsibility to decide whether that peril was of sufficient magnitude to bring the disclosure duty into play.136 There was no emergency to frustrate an opportunity to disclose,137 and Dr. Spence’s expressed opinion that disclosure would have been unwise did not foreclose a contrary conclusion by the jury. There was no evidence that appellant’s emotional makeup was such that concealment of the risk of paralysis was medically sound.138 Even if disclosure to appellant himself might have bred ill consequences, no reason appears for the omission to communicate the information to his mother, particularly in view of his minority.139 The jury, not Dr. Spence, was the final arbiter of whether nondisclosure was reasonable under the circumstances.140

Proceeding to the next cause of action, we find evidence generating issues as to whether Dr. Spence performed the laminectomy negligently and, if so, whether that negligence contributed causally to appellant’s subsequent disabilities. A report Dr. Spence prepared after the second operation indicated that at the time he felt that too-tight sutures at the laminectomy site might have caused the paralysis. While at trial Dr. Spence voiced the opinion that the sutures were not responsible, there were circumstances lending support to his original view. Prior to the laminectory, appellant had *795none of the disabilities of which he now complains. The disabilities appeared almost immediately after the laminectomy. The gusset Dr. Spence made on the second operation left greater room for the spinal cord to pulsate, and this alleviated appellant’s condition somewhat. That Dr. Spence’s in-trial opinion was hardly the last word is manifest from the fact that the team of specialists consulting on appellant was unable to settle on the origin of the paralysis.

We are advertent to Dr. Spence’s attribution of appellant’s disabilities to his condition preexisting the laminectomy, but that was a matter for the jury. And even if the jury had found that theory acceptable, there would have remained the question whether Dr. Spence aggravated the preexisting condition. A tortfeasor takes his victim as he finds him, and negligence intensifying an old condition creates liability just as surely as negligence precipitating a new one.141 It was for the jury to say, on the whole evidence, just what contributions appellant’s preexisting condition and Dr. Spence’s medical treatment respectively made to the disabilities.

In sum, judged by legal standards, the proof militated against a directed verdict in Dr. Spence’s favor. True it is that the evidence did not furnish ready answers on the dispositive factual issues, but the important consideration is that appellant showed enough to call for resolution of those issues by the jury. As in Sentilles v. Inter-Carribbean Shipping Corporation,142 a case resembling this one, the Supreme Court stated,

The jury’s power to draw the inference that the aggravation of petitioner’s tubercular condition, evident so shortly after the accident, was in fact caused by that accident, was not impaired by the failure of any medical witness to testify that it was in fact the cause. Neither can it be impaired by the lack of medical unanimity as to the respective likelihood of the potential causes of the aggravation, or by the fact that other potential causes of aggravation existed and were not conclusively negated by the proofs. The matter does not turn on the use of a particular form of words by the physicians in giving their testimony. The members of the jury, not the medical witnesses, were sworn to make a legal determination of the question of causation. They were entitled to take all the circumstances, including the medical testimony into consideration.143

We conclude, lastly, that the case against the hospital should also have gone to the jury. The circumstances surrounding appellant’s fall — the change in Dr. Spence’s order that appellant be kept in bed,144 the failure to maintain a side rail on appellant’s bed, and the absence of any attendant while appellant was attempting to relieve himself — could certainly suggest to jurors a dereliction of the hospital’s duty to exercise reasonable care for the safety and well-being of the patient.145 On the issue of causality, the *796evidence was uneontradicted that appellant progressed after the operation until the fall but, a few hours thereafter, his condition had deteriorated, and there were complaints of paralysis and respiratory difficulty. That falls tend to cause or aggravate injuries is, of course, common knowledge, which in our view the jury was at liberty to utilize.146 To this may be added Dr. Spence’s testimony that paralysis can be brought on by trauma or shock. All told, the jury had available a store of information enabling an intelligent resolution of the issues respecting the hospital.147

We realize that, when appellant rested his case in chief, the evidence scarcely served to put the blame for appellant’s disabilities squarely on one appellee or the other. But this does not mean that either could escape liability at the hand of the jury simply because appellant was unable to do more. As ever so recently we ruled, “a showing of negligence by each of two (or more) defendants with uncertainty as to which caused the harm does not defeat recovery but passes the burden to the tortfeasors for each to prove, if he can, that he did not cause the harm.” 148 In the case before us, appellant’s evidentiary presentation on negligence survived the claims of legal insufficiency, and appellees should have been put to their proof.149

Reversed and remanded for a new trial.