WATERMAN, Circuit Judge.
This is an action brought by the Massachusetts Mutual Life Insurance Com-pan for a declaratory judgment determining the rights of the parties under the provisions of two life insurance policies. The insurer disclaims liability on the ground that the decedent-insured committed suicide. The beneficiaries under the policies, defendants below, contend that the insured’s death resulted from injuries accidentally sustained as a result of a gunshot wound. The jurisdiction of the court below was founded upon a diversity of citizenship, and is not here contested.
From a judgment for the beneficiaries, the insurance company brings this appeal alleging, as grounds for reversal, two errors in the conduct of the trial below. As will appear more fully hereafter, we find no error in the admission or exclusion of evidence that would support a reversal of the judgment below and require us to order a new trial. The judgment, accordingly, is affirmed.
[465]*465Appellant maintains, first, that the district court erred in excluding from evidence certain testimony of decedent’s physician concerning statements allegedly made by decedent to him sometime prior to death. Without objection, the physician, Dr. Frederick T. Volk, testified that for some twelve or thirteen years before the decedent, Elmer T. Brei, died, he had treated Brei for varicose veins of the legs. Two years before his death Brei broke his right ankle and Dr. Volk treated him for this. About eight months after the fracture the tissue around the region of the ankle broke down and developed a large varicose ulcer or open sore. This substantially disabled Mr. Brei, required him to use crutches or a cane to move about, and prevented him from attending to his business, as a result of which, the evidence indicated, he encountered serious financial difficulties.
Dr. Volk further testified that a month before his death, Brei came to Dr. Volk’s office several times for further treatment of his legs. Counsel for the insurance company asked whether at any of these visits Brei discussed his finances or financial condition, and whether any of these statements “were not directly connected with the treatment.” Over objection from the defense, Dr. Volk responded affirmatively to both questions: “He made a statement to me about two weeks — in fact, the last time he was in my office he made a statement to me that he would — if he didn’t get a break, if things didn’t come better for him he would find a way — .” The court thereupon sustained the objection of defense counsel that the statements were a “privileged communication between patient and physician.” 1 We are faced, at the outset, with the question of what law governs the issue of privilege in diversity cases. The parties have assumed that the availability of the patient-physician privilege turns on the law of New York. There is considerable confusion in the decisions as to application of the Erie case2 and the Federal Rules of Civil Procedure to state evidentiary privileges.3 However, the [466]*466weight of authority appears to favor the view that the state rule is to govern 4 and this view has been consistently adopted in this circuit with respect to the patient-physician privilege provided by the New York statute, though without discussion of the issue and largely on the basis of precedents that antedate the Rules Enabling Act and Erie. Engl v. Aetna Life Ins. Co., 139 F.2d 469 (2 Cir. 1943); Stiles v. Clifton Springs Sanitarium Co., 74 F.Supp. 907 (W.D.N.Y. 1947); Munzer v. Swedish American Line, 35 F.Supp. 493 (S.D.N.Y.1940).
We find that reason, as well as authority, supports the view that the state rule as to the patient-physician privilege should govern.5 The privilege reflects a legislatively determined state policy. It is designed to encourage confidential communications between persons in the relationship of patient and physician by protecting these communications from compulsion to reveal them.6 The rule of privilege is unlike the ordinary rules of practice which refer to the processes of litigation, in that it affects private conduct before litigation arises. The usual rules of evidence are concerned with the procedure adducing facts at a trial. Subject to the limitations of the "outcome” test announced in Guaranty Trust Company v. York, 326 U.S. 99, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945) and Ragan v. Merchants Transfer & Warehouse Co., 337 U.S. 530, 69 S.Ct. 1233, 93 L.Ed. 1520 (1949), they are “procedural” and look for their authority to the Federal Rules of Civil Procedure. The patient-physician privilege is more than a rule of procedure since it goes to relationships established and maintained outside the area of litigation, and “affect[s] people’s conduct at the stage of primary private activity and should therefore be classified as substantive or quasi-substantive.” 7
Whether the court below erred in its ruling that the statements made to Dr. Volk were a “privileged communication between patient and physician” therefore turns upon the proper construction of § 352 of the New York Civil Practice Act, to which we now turn:
The Act provides:
Ҥ 352. Physicians, dentists and nurses not to disclose professional information.
“A person duly authorized, to practice physic and surgery, or dentistry, or a registered professional or licensed practical nurse, shall not be allowed to disclose any information which he acquired in attending a patient in a professional capacity, and which was necessary to enable him to act in that capacity, * * 8
[467]*467The purpose of the provision, and of its predecessor statute, 2 R.S. Ch. 7, pt. 3, p. 406, § 73,9 has been stated by the New York court to be
“to protect those who are required to consult physicians from the disclosure of secrets imparted to them, to protect the relationship of patient and physician, and to prevent physicians from disclosing information which might result in humiliation, embarrassment, or disgrace to patients. When the original statute was enacted, it was believed that the benefits which would accrue from its enactment by preventing disclosure by physicians of information gained in consultation and by inspiring confidence between patients and their physicians would outweigh any injustice which might result in particular cases caused by the exclusion of testimony by physicians at trials.”
Steinberg v. New York Life Ins. Co., 263 N.Y. 45, 48-49, 188 N.E. 152, 153, 90 A.L.R. 642 (1933).
In an early construction of the predecessor of present § 352,10 the New York Court of Appeals stated that two conditions must be met before the privilege applies:
“Before information can be excluded under this statute, it must appear that it was such as the physician acquired in some way while professionally attending the patient; and it must also be such as was necessary to enable him to prescribe as a physician, or to do some act as a surgeon. It is not sufficient to authorize the exclusion that the physician acquired the information while attending his patient; but it must be the necessary information mentioned.”
Edington v. Aetna Life Ins. Co., 77 N.Y. 564, 569-570 (1879).
Although appellant concedes that the information here sought to be introduced into evidence was acquired by Dr. Volk “in attending a patient in a professional capacity,” appellant argues that the information fails to meet the second statutory requirement of being “necessary to enable [the doctor] to act in that capacity.” Dr. Volk, appellant points out, was treating the decedent for varicose veins and for complications related to that condition. The statement of Brei, by contrast, is interpreted by the insurer to indicate a suicidal intent.
As we read the New York authorities, the courts of that state, since the time of Edington v. Aetna Life Insurance Co., supra, have interpreted this second or “necessary information” requirement of the statute with substantial liberality. In Feeney v. Long Island Ry. Co., 116 N.Y. 375, 22 N.E. 402, 5 L.R.A. 544 (1889), a physician who had treated the plaintiff shortly after an accident was asked at trial whether he had conversed with the plaintiff about her injuries, and if he made an examination of her. The trial court sustained an objection made [468]*468in behalf of the plaintiff that these questions called for a privileged communication. On appeal, the court, in affirming the action below, quoting from Edington v. Mutual Ins. Co., 67 N.Y. 185, 194 (1876), stated:
“The point made that there was no evidence that the information asked for was essential to enable the physician to prescribe is not well taken, as it must be assumed from the relationship existing that the information would not have been imparted except for the purpose of aiding the physician in prescribing for the patient. Aside, however, from this, the statute in question being remedial, should receive a liberal interpretation, and not be restricted by any technical rule.” 116 N.Y. at 380-381, 22 N.E. at 403.
In Nelson v. Village of Oneida, 156 N.Y. 219, 50 N.E. 802 (1898), the court went a step beyond the presumption created in Feeney. Here, counsel for the defense in a personal injury case sought to prove that the plaintiff had suffered from an umbilical hernia prior to the accident in question. In his offer of proof, counsel stated that a physician who had attended the plaintiff during childbirth some time prior to the accident would testify that he had observed the hernia. It was stated, however, that the physician had not treated the patient for the latter condition. In affirming the trial judge’s exclusion of the evidence, the court stated:
“The witness acquired the information which the defendant desired to elicit from him while attending the patient in a professional capacity, and the discovery of an umbilical hernia was a necessary incident of the investigations made to enable him to act in that capacity.” 156 N.Y. at 222, 50 N.E. at 802.
The exclusion of information incidentally acquired while treating a patient for an unrelated condition has not been limited to testimony concerning physical illnesses. In probate proceedings, the New York courts have repeatedly held that offered testimony concerning the mental state of a testator by the physician who attended the patient in his final illness was inadmissible in evidence. See, e. g., Matter of Cashman’s Will, 159 Misc. 881, 289 N.Y.S. 328 (1936), aff’d 280 N.Y. 681, 21 N.E.2d 193 (1939); Matter of Coddington’s Will, 307 N.Y. 181, 120 N.E.2d 777 (1954).
Whether incidental information comes to a physician as an inevitable incident of his treating the patient for an unrelated condition, or, as in the case before us, is volunteered by the patient, does not appear to be determinative of the availability of the § 352 privilege under New York law. In Meyer v. Supreme Lodge, 82 App.Div. 359, 81 N.Y.S. 813 (1903), aff’d, 178 N.Y. 63, 70 N.E. 111, 64 L.R.A. 839 (1904), aff’d, 198 U.S. 508, 25 S.Ct. 754, 49 L.Ed. 1146 (1905), decedent, shortly prior to his death, allegedly told his physician that he had swallowed rat poison in an attempt to commit suicide. The information was held to be privileged, in part because of its possible use to the physician in prescribing therapy.11 [469]*469And see Eder v. Cashin, 281 App.Div. 456, 120 N.Y.S.2d 165 (1953). At trial in the case of Scheiner v. Metropolitan Life Ins. Co., 236 App.Div. 24, 257 N. Y.S. 783 (1932), however, two physicians were permitted, over the objection of the plaintiff, “to testify to the physical and mental condition of their former patient, the decedent, including suicidal tendencies and threats.” On appeal, the introduction of this testimony was held to be violative of § 352 of the Practice Act, and hence reversible error, despite the fact that there is no indication in the report that the physicians were psychiatrists, or that they had treated the decedent for the mental state evidenced by the suicidal threats.
On the basis of these authorities it is clear, we believe, that the court below properly excluded the testimony of Dr. Volk. Section 352 was designed to encourage patients to repose just such confidence in their physicians as was here involved, and to prevent physicians, in turn, “from disclosing information which might result in humiliation, embarrassment or disgrace to patients.” Steinberg v. New York Life Insurance Co., supra. It is common knowledge that general practitioners of medicine still purport, in this modern day, to treat “the whole man.” The decedent was suffering from a long and chronic illness. His general morale or state of mind was surely related to his physical condition.
If, as appellant appears to believe, decedent’s statements to Volk indicated a suicidal intent, it would have been reasonable for the physician, in his professional capacity, to have undertaken treatment for this unhappy mental condition, or to have referred the patient to an appropriate specialist for treatment. The patient-physician privilege, we must remember, exists for the benefit of the patient during his lifetime and for the prevention of his disgrace after his death, and cannot be taken away because the physician failed to treat an illness disclosed to him by his patient.
The second ground upon which appellant urges reversal of the judgment below concerns the admission into evidence of testimony elicited in response to a hypothetical question posed by counsel for the defense. Proper disposition of this issue requires a summary of the circumstances surrounding the death of the insured.
In the early morning of February 21, 1958, Elmer Brei was found dead at his place of business. Death had resulted from a blast of shotgun pellets which had entered his head near the left temple and had passed through the upper part of the skull. By his side was a 410 gauge Stevens double-barreled “over and under” shotgun; one barrel contained an exploded shell, the other, its hammer cocked, was loaded. The nature of the wound indicated, according to expert testimony, that the muzzle of the gun had been within six to twelve inches of deceased’s head at the time the gun was discharged.
In support of the hypothesis that Brei had died by his own hand, the insurer introduced evidence showing that Brei had been ill and required to walk on crutches for several years, that as a result of this illness and his inability to work he had become “hopelessly insolvent,” and that, in addition, he was required to be both father and mother to five minor children inasmuch as his wife had died two years prior to his own death. Defendants below, arguing the alternative hypothesis of accidental death, introduced evidence that Brei had been in good spirits up to the day of his death, that he left no will or suicide note, that he went about his affairs in a normal way on that day, that he had arranged for a sale to be made that afternoon of the 410 gauge Stevens and an[470]*470other gun, and that he appeared to have been in the process of dismantling and cleaning the two guns at the time of the fatal blast.
In support of their claim that the death gun may have discharged in the course of being cleaned, by falling, butt first, to the floor, defendants called as a witness Robert E. Perrigo, a ballistics expert. Perrigo testified that he had studied the linkage mechanism of the death gun, and had accurately tested the pressure or “pull” required to activate its two triggers. Counsel for the defense then asked the following hypothetical question:
“Q. Mr. Perrigo, assuming a situation where there is a concrete floor with no covering on it or asphalt tile, no linoleum, no covering of any kind over the concrete floor, assuming that this particular weapon you have examined having in mind your test of the front trigger pull and the nature of mechanical linkages connected with it, assuming that the triggers are retracted in a cocked position in the light of your experiments and study of this subject I ask you for an opinion as to whether this particular weapon is capable of being discharged upon falling from the height of approximately three feet and landing butt first on the concrete.”
Over plaintiff’s objection Perrigo testified that although he had not had the opportunity to experiment with the death gun itself by dropping it under the circumstances specified, he had made such experiments with other and similar guns and was of the opinion, on the basis of those experiments and his study of the death gun, that it could be discharged in the manner suggested.
Appellant now claims that admission of this testimony was reversible error, for two reasons. First, it is said, the hypothetical question had no basis-in the evidence, for “there is no evidence that the gun in this case was dropped on its butt from a height of approximately three feet or for that matter was dropped at all.” Again we are faced with a question as to governing law. Under Federal Rule 43(a) if the evidence was admissible under the law of New York it was admissible in the district court. See J. H. Horne & Sons v. Bath Fibre Co., 272 F.2d 8 (1 Cir. 1959); cf. Persons v. Gerlinger Carrier Co., 227 F.2d 337 (9 Cir. 1955). We hold that the evidence was admissible under New York law.12
An answer to a hypothetical question is not false, of course, merely because the question was premised upon an inaccurate assumption. If the jury determines that the condition or premise of the hypothetical question is not established by the evidence then the answer to the question is merely immaterial to-the established facts. The courts of New York, however, and those of other jurisdictions, have held inadmissible answers, to hypothetical questions having no basis, whatever in the evidence, on the ground that their admission would serve merely to confuse, rather than to enlighten the-jury. Thus in White v. Prudential Insurance Co., 120 App.Div. 260, 105 N.Y.S. 87 (1907), decedent-insured had: been found asphyxiated in her apartment. The windows were closed and the gas burners of a kitchen range were-turned on. At trial, an expert witness for the beneficiaries testified in response-to hypothetical questions that if water-had entered the gas pipes, it might have-emerged at the gas jets, extinguishing-a normal flame at that point, and then receded, leaving the unbumed gas to continue to escape into the room. On appeal, the court ruled that the hypothet[471]*471ical questions were wholly speculative and fanciful, there having been no evidence to suggest that water had entered the gas pipes. Admission of the prejudicial testimony into evidence was thus held to constitute reversible error. And see Jones v. Poinciana Realty Corp., 268 App.Div. 877, 50 N.Y.S.2d 689 (1944).
The New York courts have not required, however, that hypothetical questions be founded on factual premises which have been conclusively proved. In Dilleber v. Home Life Insurance Co., 87 N.Y. 79, 81 (1882), the court stated:
“Counsel in framing hypothetical questions to be put to expert witnesses, are not confined to facts admitted or absolutely proved, but facts may be assumed which there is any evidence on either side tending to establish, and which are pertinent to the theories which they are attempting to uphold.”
In Cowley v. People, 83 N.Y. 464, 470 (1881), the rule was stated even more liberally:
“Another question raised is as to the admissibility of the hypothetical questions put to medical experts rsworn as witnesses. The claim is that a hypothetical question may mot be put to an expert unless it ■states the facts as they exist. It is ^manifest, if this is the rule, that in .•a trial where there is a dispute as to •the facts, which can be settled only :by a jury, there would be no room for a hypothetical question. The •very meaning of the word is that ■it supposes, assumes something, for the time being. Each side, in an issue of fact, has its theory of what is the true state of the facts, and assumes that it can prove it to be so to the satisfaction of the jury; and so assuming, shapes hypothetical questions to experts accordingly. And such is the correct practice.”
We believe that the testimony of Perdigo comes within the stated New York rule for appropriate hypothetical questions. Because of the direction of the fatal blast which came from below and to the side of the deceased and proceeded toward the ceiling, both parties appear to agree that the wound was either self-inflicted or resulted from an accidental dropping of the gun on the floor. No other explanation was offered or appears to be credible. Both of the suggested hypotheses depend upon inferences drawn from circumstantial evidence. Each evidentiary fact which tends to discredit the suicide hypothesis supports the theory that the gun was discharged from having been dropped. In defendants’ evidence concerning (1) an absence of suicidal intent, (2) the presence of a dismantled gun and cleaning fluid on the work bench beside which the fatal blast took place, and (3) decedent’s probable awkwardness in handling a gun without the aid of his crutches or cane (neither having been found nearby at the time of the death), we find a sufficient evidentiary basis to justify the court below in admitting the challenged testimony.
Under these circumstances, the reasoning of the Missouri court in a similar case is persuasive:
“The defendant objected to proof that a violent fall might discharge the pistol on the specific ground that there was no evidence of such fall. Neither was there any evidence that the pistol was held in the hand of the insured and discharged by pulling the trigger, which is a necessary element in the proof of suicide. It is, however, a matter of common knowledge that a pistol like the one here described could be so discharged, but is not, perhaps, a matter of common knowledge that a similar weapon could be discharged by a violent fall, and must be proved as a part of the issue in determining in which of these two ways it was discharged. The discharge is admitted, and it is for the jury to determine how it might occur, and evidence tending to assist them in that inquiry was admissible.” Reynolds v. Maryland Casualty Co., 274 Mo. 83, 201 S.W. 1128 (1917).
[472]*472We find no conflict between the approach of the New York courts toward hypothetical questions and that of Dean Wig-more:
“The trial judge should be given discretion to determine how far the counsel can and must properly limit his questions, and how far the jury may be trusted, with the aid of argument, to discover the conditional nature of the opinion.
“* * * There could be no reason in confining the hypothetical question to the undisputed facts, or to the facts ‘proved.’ The former expedient is at least conceivably possible ; though the latter is not. Both are without the slightest ground in logic or policy.” 2 Wig-more, Evidence § 682 (3d ed. 1940).
The second basis of appellant’s attack upon the testimony of the ballistics expert, Perrigo, goes not to the factual premise upon which the hypothetical question was founded, but to the probative value of the expert opinion itself. Appellant contends that inasmuch as Perrigo did not experiment with the death weapon, but relied, for his opinion, upon having dropped only similar weapons, the testimony was inadmissible to prove that the death weapon could have discharged by being dropped. No authorities are offered in support of this contention.
The New York courts have granted broad discretion to trial judges to determine the competence of expert witnesses and the admissibility of their testimony. The leading case is Slocovich v. Orient Mutual Insurance Co., 108 N.Y. 56, 14 N.E. 802 (1888). There, two expert witnesses had been questioned at trial concerning the value of an insured vessel. The first witness, a marine surveyor, had been on board the ship, but not within five or six years of the time of the trial. In affirming the trial judge’s exclusion of the surveyor’s testimony, the court said:
“It was for the trial judge to determine, in the first instance, whether the witness was competent as an expert to testify to the value of this vessel. He had not seen her for five or six years * * *. Under such circumstances, we canpot say that the judge committed any error in excluding the testimony. If the evidence had been received it certainly would not have been entitled to very much weight with the jury. While it would not, we think, have been erroneous to receive and submit the evidence to the jury for what it was worth, we cannot say, as a matter of law, that the judge exceeded the bounds of a reasonable discretion * * *. The rules determining the subjects upon which experts may testify, and prescribing the qualifications of experts, are matters of law; but whether a witness offered as an expert has those qualifications is generally a question of fact, to be decided by the trial judge; and it has been held that his decision in reference thereto is not reviewable in an appellate court.” 108 N.Y. at 61-62, 14 N.E. at 804.
The second expert witness in Slocovieh knew the ship only by reputation apd had never made an inspection of her. He had been permitted by the trial court to testify to her market value. Once again affirming the ruling below, the court said:
“It was not a sufficient objection to the competency of this witness that he had no personal knowledge of the ship. An expert is qualified to give evidence as to things which he has never seen. He may base an opinion upon facts proved by other witnesses, or upon facts assumed and embraced within the case. Questions may be put to him assuming the facts upon which he is asked to base his judgment and express an opinion.” Id. at 64, 14 N.E. at 805. •
Although the rule denying appellate review to all trial court rulings on the competency of expert witnesses has been modified, see, e. g., Zahn v. Gulf Oil [473]*473Corp., 281 App.Div. 834, 118 N.Y.S.2d 471 (1953), the New York courts have repeatedly followed the general rule of Slocovich. In Kenny v. Douglas Manor Ass’n, 252 App.Div. 780, 299 N.Y.S. 384 (1937), plaintiff sought damages for injuries suffered on a water slide at a place of amusement. The trial judge refused to admit into evidence the testimony of an expert witness concerning water slides in use elsewhere. In reversing the judgment below, the court stated:
“The evidence of this expert was struck out, but it does not appear that it was solely on the ground that he was not qualified. Although his experience was somewhat limited, it was error to strike out the evidence, for the weight and sufficiency thereof was for the jury.” 299 N. Y.S. at 385.
In Schlansky v. Augustus V. Riegel, Inc., 9 N.Y.2d 493, 215 N.Y.S.2d 52, 174 N.E.2d 730 (1961), plaintiff sought compensation for damages to his property caused by blasting at a nearby construction site. At trial, an expert witness testified that it was his opinion, on the basis of an inspection of the site, that more explosive powder was used than was necessary. Defendants objected, claiming that the expert’s opinion was “valueless since he did not know the method of blasting or the strength of the charges used or the character of the soil or rock.” Reviewing the judgment below, the Court of Appeals stated, as to defendants’ objections to the expert’s testimony :
“Such a criticism is unjustified. * * * [T]he expert’s qualifications were established and he stated that from the situation which he observed and the facts contained in the hypothetical question he was able to express an opinion with reasonable certainty. That permitted the taking of his opinion and his lack of further information affected the weight but not the admissibility of his evidence.” 9 N.Y.2d at 497, 215 N.Y.S.2d at 54, 174 N.E.2d at 732.
And see Greengard v. Odorono Co., 235 App.Div. 806, 256 N.Y.S. 708 (1932); Meiselman v. Crown Heights Hospital, Inc., 285 N.Y. 389, 34 N.E.2d 367.
The pattern of the New York authorities is thus clear. Had Perrigo made no inspection of the death gun, his opinion based upon experiments with other weapons might well have had insufficient probative value to go to the jury. See Weibert v. Hanan, 202 N.Y. 328, 95 N.E. 688 (1911); Zahn v. Gulf Oil Corp., 281 App.Div. 834, 118 N.Y.S.2d 471 (1953). Perrigo testified, however, that from his study of the death gun he had found its linkage mechanism and manner of operation to be closely similar, if not identical, to those of the weapons with which he had experimented, the experiments upon which his opinion was based. That being Perrigo’s undisputed testimony, we believe that the court below acted within the bounds of proper trial discretion in admitting Perrigo’s testimony into evidence.
The judgment of the court below is affirmed.