Both parties appeal from a judgment for the plaintiff in an action to recover damages for a libel. The defendant’s appeal is based upon the admission of irrelevant and incompetent evidence as to damages; the plaintiff’s is from an order which granted a new trial unless she consented to reduce her damages from $20,000 to $15,000, which she did. The plaintiff lives in Norfolk, Virginia; the defendant is the publisher of a newpaper having a large circulation in New York, the Sunday issue of which, on the date in question — August 11, 1946 — had a circulation in Norfolk and its environs of 32,550 copies. On that day it published an article about a murder which had taken place in Norfolk in May 1938, for which the accused had been convicted of murder in the first degree. It had been a part of the defence that a witness called at the house of the accused and the deceased, his wife, and had talked to the deceased on her porch. The accused had sworn that he had left the house before the time fixed by this witness. Four days after the verdict, the plaintiff executed an affidavit to aid the defence in a motion for a new trial. In order to support the accused’s testimony that he had'left his house before the witness just mentioned had talked to the deceased, the plaintiff swore that from her porch she had seen the accused leave his house and that afterwards she saw the witness talk to the deceased on the porch. The defendant’s article discussed this affidavit, and said that a local detective, who was in charge of the case, had demonstrated that it was impossible from the plaintiff’s porch, where she had said she stood, to see anyone on the accused’s porch. It then proceeded: “he” (the detective) “later learned that the Mattox woman had once been a patient in a mental institution.” These last were the words charged as the libel.

The testimony as to damages, of which the defendant complains, was of several sorts. There was, first, testimony of several witnesses that after the article appeared others had asked the witnesses whether the plaintiff had been in an insane asylum. A witness testified that he had discussed with his family whether the statement in the paper was true; that the family had taken its truth for granted. The plaintiff was allowed to testify that others had asked her about the article; and that when she *900walked in the street, she heard comments of acquaintances, and whispers, inquiring whether she “was the girl” who had “been away.” Other witnesses testified to the plaintiff’s changed appearance and carriage ; that she was no longer cheerful; that she avoided society, appeared nervous, shunned “callers” and stopped attending plays and moving pictures. The plaintiff herself testified to the same things, and added that she went out of town to a “good retreat,” in order to avoid people whom it was hard to convince that she had not “been away”; that she ceased going to church or any “formal gathering”; and that she felt “self-conscious and embarrassed.” Witnesses, including the plaintiff, also testified that she had called in a doctor, who told her not to worry, and to forget the event so far as possible. Finally, the defendant complained of the testimony of the plaintiff and others that she had been absent from her work, there being no special damages alleged.

This being a case of diverse citizenship, we are first to inquire whether the law of New York takes, as the measure, or pattern of liabilities for torts committed elsewhere, the law of the state where the wrongful conduct occurred. We have been unable to find any cases dealing with libel; but the New York courts generally accept the doctrine as to torts that the lex loci delicti is the standard,1 *and there can be no distinction as to libel, provided the law of New York recognizes Virginia as the place where the wrong occurred. On that question there are no New York decisions, so far -as we can find, and elsewhere there has been no authoritative answer, although there has been some discussion, especially in a “Note” in 60 Harvard Law Review 941. The Restatement of Conflicts 2 lays it down that the place of the wrong is “the place of communication,” and that is obviously true, but it does not tell what law should govern when the libel is “communicated” in several jurisdictions. Even though we group all copies of a single issue published in one state as a single tort, as we must, it is possible to view the publication in one state as a wholly separate tort from that in any other, and that has at least the merits of simplicity in theory.3 The difficulty is that in application it would prove unmanageable.

We assume that in any event a plaintiff must recover in one action all his damages for all the publications, wherever made; but, if the publication in each state is a separate wrong, the extent of the liability may vary in the separate jurisdictions : for instance, in the case at bar the law of New York may differ from that of Virginia. It would certainly be an unworkable procedure to tell a jury that they should award damages, so far as they were suffered in State X, according to one measure, and, so far as they were suffered in State Y, according to another. Judge Wyzanski discussed the question in Kelly v. Loew’s, Inc.,4 but did not decide it, because he found the answer in the established law of Massachusetts; and we, too, need not answer it in the case at bar, though for another reason. On the record the plaintiff was not a person of prominence, and it does not appear that she was known outside Virginia; in any event there is no suggestion that she was known in New York. Since -it does not appear that she suffered any damages in that or in any other state, it was not an error for the judge to rule upon the evidence upon the assumption that the only damages were suffered in Virginia; and it is of course irrelevant that in so ruling he was not consciously following our foregoing reasoning.

*901Coming, therefore, to the law of Virginia, although we have found no decisions that it is there libellous per se to say that a person is insane or of instable mind, that is the general law,5 as it probably is also the law of New York,6 and in the absence of evidence to the contrary we are to suppose that the same is true in Virginia. In considering what evidence was admissible in proof of damages we must distinguish between relevancy and competency. The testimony was relevant, so far as it was rationally probative of those consequences of the publication, for which the law of Virginia holds a publisher liable. Liability means that obligation to redress the person wronged, which is imposed upon the wrongdoer because of his commission of the wrongful act; and the extent of the damages is therefore the measure of the liability.7 On the other hand, whether evidence, rationally probative of those consequences, is incompetent — that is, whether it is the sort of evidence which the court of the forum will refuse to accept upon any issue- — is determined by Rule 43(a), Federal Rules of Civil Procedure, 28 U.S.C.A. It will be admitted, if it is admissible under the practice of the state where the case is being tried, or under the practice of federal courts of equity: for the Rules “have attempted to liberalize admissibility of evidence as much as possible by providing always for the widest rule of admissibility.” 8

So far as we know, it is universally agreed that the damages recoverable in libel are the plaintiff’s loss of reputation in t-lie minds of those who know him or know about him, together with this mental suffering as a result of the libel. How this has been interpreted in Virginia in specific cases appears from some decisions of its Supreme Court. In Snyder v. Fatherly,9 the following charge was approved: “You shall take into consideration * * * the probable effect upon those to whose attention they” — the charges— “came, and the probable effect upon the plaintiff’s personal feelings and her standing in the community.” This was only the common rubric, but the court in its opinion added that there was “plenty of evidence * * * showing the injury to the plaintiff’s feelings * * * reflected by the ignominy in which her former associates held her on account of the defamation” 163 S.E. -page 364. In Kroger Grocery & Baking Co. v. Rosenbaum,10 in supporting the verdict as not excessive, the court apparently took into consideration testimony that “it was the general impression in the town of Marion that plaintiff was discharged by defendant for misappropriating money.” In News Leader Co. v. Kocen,11 the court reversed a judgment for the plaintiff because of an erroneous charge; but also indicated that the size of the verdict was a contributing factor. In considering the second question, it recited the plaintiff’s testimony as to her sufferings, in that she had been “mortified by people telephoning her” and by the libel’s being mentioned to her children. Also that she -became hysterical, sleepless, had headaches, bothered her husband, and declined to go to parties. The majority opinion -said as to this testimony: “While the weight to be given this evidence was peculiarly a question for the jury, nevertheless it creates a strong suspicion” that the plaintiff was shamming. In these decisions the courts regarded the liability as extending to such consequences of the libel as — the defendant argues — the law of New York will not include. They did not, however, lay down any general rule, particularly the rule which applies to other torts, and which the Restatement of *902Tarts12 declares should also apply to (lefamation: that is, that the tortfeasor is responsible for all the actual consequences which are reasonably to be anticipated from the publication. However, in James v. Powell,13 the Supreme Court of Virginia did adopt just that doctrine. After adverting to the doctrine in Massachusetts14 that the publisher of a'libel “is not bound to anticipate the general probability of such acts” —republication of the libel — “any more than a particular act by this or that individual,” the court went on to say that, although no paper had been sent to a sub-scriber in the county, the defendant had testified that the papers “were sold to newsboys and to newsstands, that they were' taken to trains and might go into Franklin County * * *. Indeed it was almost certain that sooner or later it would be carried into every county in the vicinity of Danville, and this the publisher must have known. It was a natural and probable result of the publication, and for it Mr. James is liable.” We hold that the law of Virginia does not recognize any limitation of actionable consequences peculiar to cases of libel.; and that all those consequences of the libel to which the plaintiff and other witnesses testified were properly left to the jury as elements of recoverable damages in the case at bar. Therefore that testimony was relevant.

It remains to decide whether any of it was incompetent. The defendant challenges parts of it, principally for two reasons: (1) that in libel the testimony is incompetent of individuals that they thought the publication injurious to the plaintiff’s reputation; it complicates the issues too much, and that the evidence must be confined to general facts; (2) that, in any event, it was hearsay in the case at bar to allow witnesses to testify that third persons had told them of the impressions which the publication left on the declarants’ minds. The first objection appears to have had its origin, in New York at least, in Linehan v. Nelson.15 The defendant’s witnesses in an action for slander had been allowed to testify that the utterance had not affected the plaintiff’s reputation in their minds, and this the court held to be incompetent. Werner, J. (speaking, however, as to this for only four of the seven judges), said that if witnesses, who heard the utterance, were allowed to testify that it did not affect their opinion of the plaintiff, other witnesses must be allowed to testify that it did; and that, especially in actions of libel, this would unduly prolong and confuse the trial. The injury to the plaintiff’s reputation is the aggregate of the detractions which he may suffer in the minds of all those who read the libel; and if it were practical to put before the jury the testimony of all whom it did so affect, that would be the best evidence; and in that event it would obviously be irrelevant how many it did not affect. It is of course quite impossible so to proceed, and there is universal agreement that in its stead evidence is admissible of the standing of the plaintiff, of the extent the newspaper’s circulation and of other relevant general considerations. The question for us is whether this is the only admissible evidence, or whether it may be supplemented by calling individual readers of the libel and asking them what impression it left on their minds. True, if such testimony is not confined at all, the trial may become unmanageable; and if it is unduly confined, the result may distort the truth. Nevertheless, the difficulties are not so great as to demand the total suppression of such direct evidence upon the very kernel of the issue. It is possible to put before the jury a fair sample of the effect upon readers of the libel. That does indeed involve a quantative element, and it may be that the defendant should be entitled to call readers who say that the libel did not impress them unfavorably. Be that as it may, the important thing here is that all such testimony is relevant and that the extent to which it should be allowed is a part of the management of the trial and is *903a question of evidence, stricti juris,16 upon which the law of New York is not conclusive. As an illustration of the views of federal courts, in criminal prosecutions for fraud it is the practice to admit the testimony of individuals that the statement in issue deceived them, although to limit their number.17 Yet the issue in a criminal prosecution does not involve the extent of the injury done by the statement, unlike the issue in libel where the spread of the defamatory matter is the whole basis of one part of the recovery. If kept within proper bounds the actual impact of the libel upon individuals should be admissible, and in the case at bar the judge did not abuse his discretion. We cannot understand the force of the defendant’s argument that it would be difficult to meet this evidence.

The testimony of those witnesses, who told of third persons who had asked them whether the plaintiff had been confined, or had said that they thought she had been so confined, was, indeed, hearsay. The relevant fact was the declarants’ belief, and that was proved only by what they said to the witnesses. Nevertheless, it has long been the generally accepted doctrine in federal courts, and pretty generally elsewhere, that when a person’s feelings or beliefs are relevant, his declaration is competent evidence of their existence. Professor Morgan 18 says that this exception to the hearsay rule had its origin in 1805 in a statement of Lord Ellenborough 19; and after many vicissitudes it has at last shaken off its original and disreputable associate, “res gestae,” and stands upon its own feet.20 An early, if not the first statement of it in the Supreme Court was Insurance Company v. Mosley,21 where Swayne, J. said generally that “wherever the bodily or mental feelings of an individual are material to be proved, the usual expressions of such feelings are original and competent evidence.” True, he discussed the exception in terms of “res gestae,” but whatever he may have meant by that, it is clear that he did not mean that such a declaration was admissible only when it was necessary to make relevant some other conduct of the declarant, or served to explain it. However, he did limit it to some “existing pain or malady,” following Bacon v. Inhabitants of Charlton.22 In Mutual Life Insurance Co. of New York v. Hillman 23 the court extended the doctrine to cover also an intention of the declarant, and at length stripped it of its Shirt of Nessus— “res gestae.” In Fidelity Mutual Life Association v. Mettler,24 it was applied to a declaration of the witness’s family of their belief in the death of the insured, where that belief, as distinct from the d’eath itself, was a relevant fact. Finally, although in Shepard v. United States 25 the declaration was not of the declarant’s feelings, intentions or beliefs, and for that reason was declared incompetent. Cardozo, J. recognized the general doctrine. It must be owned that Throckmorton v. Holt26 opened doubts about the scope of the exception. A bare majority of the court there thought that declarations of a decedent’s testamentary intentions, whether made before or after the date of a paper purporting to be his will, were not competent upon the issue whether the paper was a forgery, or an authentic will. It is not clear just what was the basis of this ruling; apparently 180 U.S. pages 573, 574, 21 S.Ct. pages 482, 483, it did not extend to cases when the declarations were relevant upon the issue of testamentary capacity, or when they were a part of the “res gestae” — for that spectre reappeared. Moreover the court was obviously much influenced by the fact that the validity of a will was at stake. Whatever it meant, it has not been taken *904as overruling Mutual Life Insurance Co. of New York v. Hillmon, supra, which has since then been often cited as still authoritative.27 If Throckmorton v. Holt, supra, still stands for any general proposition, whatever that may be, it is confined to testamentary litigation. Thus, the testimony of those witnesses in the case at bar was competent that others had expressed in one fashion or another the belief that the plaintiff was a person of unsound mind. That this belief arose from the article was an inevitable inference or at any rate, an inference permissible to the jury.

The defendant also objected to the admission of testimony that the plaintiff called in a doctor and that the doctor advised her not to worry and to try to forget the article and its effect. The fact that she called in a doctor was not hearsay, and it was relevant to show the extent of her suffering; quite as relevant as it would have been, had she -testified that she walked her room the night through because of shame. The doctor’s advice, as advice, was not hearsay, except in so far as from it could be inferred some diagnosis by him of her case, which justified his advice. Assuming for argument that, pro tanto, it was ■on that account hearsay, it amounted to no more than that he thought that she was disturbed enough to be told that she must try not to take the matter so 'hard. That is so trivial an incident that it does not ■deserve even the attention which we have .already given it.

The last'obj ection which needs .any notice is the admission of testimony ithat the plaintiff was away from her work for a season. Of this the defendant' complains because the plaintiff made no claim for special damages. The answer is to be found in that part of the charge in which the judge said that she had not asked “for •damages for any medical bills, for any impairment of her earning capacity or any wages lost”; and that “they may not be considered by the jury.” All the offending testimony in question was relevant and competent to show the extent of the plaintiff’s suffering from the libel, and in its other aspect the judge dealt with it in the only way he could, when he told the jury not to take it into consideration in making their award. Nor is there the slightest basis for the assertion in the defendant’s brief that' the jury must have “made allowance” for it because of the size of the verdict. There was ample justification for the original verdict of $20,000. The libelous words were wholly gratuitous, and were obviously added to support the writer’s intimation that the plaintiff’s whole affidavit was a fabrication. If a newspaper chooses to embellish accounts of crime with false slurs upon the good name of blameless persons, there is no conceivable reason why it should not make full redress.

Little need be said about the plaintiff’s appeal. Had she not consented to the reduction of the judgment, the judge would have ordered a new trial because he thought that the verdict was excessive; and from that order she could not have appealed.28 Her position was not bettered by accepting the reduction and maintaining the reduced judgment.29 Judgment affirmed.

On Petition by Appellant for Rehearing.

The petition for rehearing asserts (1) that the petitioner has never been heard on the issues which this court’s opinion held to be decisive of the appeal; (2) that the court erred in disregarding the New York “single publication” rule and selecting the law of Virginia as controlling; and (3) that we misinterpreted the law of Virginia.

*905At the court’s request the respondent filed an answering brief to the petition for rehearing and the petitioner filed a reply brief to the respondent’s answering brief. Consequently the issues have been fully presented on briefs. We think it Unnecessary to grant oral argument.

Our decision held that, since it did not appear that the plaintiff was known outside Virginia or suffered any damages except in that state, it was correct for the judge to make rulings as to the admission of evidence bearing on damages in accordance with the law -of Virginia. The petitioner contends that our decision runs counter to the “single publication” rule which New York has adopted with respect to defamation contained in newspapers, magazines and books. The authoritative exposition of -this rule is to be found in Gregoire v. G. P. Putnam’s Sons, 298 N.Y. 119, 81 N.E.2d 45. At page 123 of 298 N.Y., at page 47 of 81 N.E.2d the opinion recognizes the common law rule that each delivery to a third person of a defamatory article constituted a new publication which gave rise to a new cause of action, notes that the rule had its origin in an era antedating the modern process of mass publication and nationwide circulation, and proceeds:

“That rule also gave scant heed to the public policy which underlies statutes of limitation, long regarded as ‘statutes of repose’ designed to outlaw stale claims. * * * within recent years courts of this State and other jurisdictions have ruled that the publication of a defamatory statement in a single issue of a newspaper * * * is, in legal effect, one publication which gives rise to one cause of action and that the applicable Statute of Limitation runs from the date of that publication.”

The question for decision was whether the plaintiff’s action, which was based on libelous matter in a book published by the defendant, was barred by the New York statute of limitations. A copy of the book was sold in New York within the statutory period but the initial publication of the book long antedated such period. The precise decision was that the statute began to run from the date of the initial publication. Other cases cited in that opinion at page 123 of 198 N.Y. at page 47 of 81 N.E.2d have likewise applied the “single publication” concept in conjunction with the New York statute of limitations to bar recovery for libels initially published here and distributed later either here or in other states or countries. Compare Hartmann v. Time, Inc., 3 Cir., 166 F.2d 127, certiorari denied 334 U.S. 838, 68 S.Ct. 1495, 92 L.Ed. 1763. The rule is well characterized in 48 Col. L. Rev. 932, at 935 where it is said that the single publication rule has been evolved as a practical means of protecting the forum against the multiplicity of suits and indefinite tolling of the statute of limitations which results from the application of the early common law rule of multiple publication to modern methods of mass distribution. No case has come to our attention which indicates that a court of New York, whose law is concededly binding upon the federal court when jurisdiction rests on diverse citizenship, would not apply the law of Virginia with respect to damages recoverable in a libel action under circumstances like those at bar. As noted in our former opinion the single publication concept may be the only workable procedure when a plaintiff suffers damages in several states whose laws provide different measures of damages, but that problem is not here presented since the plaintiff’s reputation was damaged only in Virginia. The court adheres to the view that Virginia law applies in such a case.

The petitioner’s brief has not convinced us that we misinterpreted the Virginia law. Nor do we see reason for adding to our previous discussion of it.

Accordingly we adhere to our opinion and deny the petition for rehearing.