FRANK, Circuit Judge.

Appellants, as trustees in reorganization of the New York, New Haven and Hartford Railroad Company, appeal from a judgment, entered upon a jury verdict, awarding $25,077.35 to the plaintiff in his individual capacity and $9,000 to him as administrator of his wife's estate. The action grew out of an accident which occurred at a grade crossing of the appellant railroad in West Stockbridge, Mass. On December 25, 1940, at about 6:15 P.M., the plaintiff was driving a Ford coupe, with his wife as a passenger, at this crossing, when the car was struck by a locomotive engine, causing severe and permanent injuries to the plaintiff and the death of his wife. The complaint alleged that the railroad was negligent in failing to ring a bell or blow a whistle while approaching the crossing, and in failing to have a proper headlight; in view of the verdict, no issue is raised as to appellants' liability if the rulings on evidence and the charge to the jury were proper, and the alleged errors pertain exclusively to these matters. The alleged errors are four in number, and will be taken up seriatim.

1. Appellants urge that the judgment must be reversed because of the court's refusal to admit in evidence a statement signed by the locomotive engineer who was driving the engine when the accident occurred; the statement is in question-and-answer form and represents a stenographic record of an interview, two days after the accident, between the engineer and an assistant superintendent of the railroad. Present at the interview were two other employees of the railroad, and a Mr. Christie, of the Massachusetts Public Utilities Commission. The latter took only a minor part in the interview. Appellants offered merely the engineer's statement, and offered to prove that "this statement was signed in the regular course of business and that it was the regular course of such business to make such statement." The engineer was dead at the time of the trial. The statement was excluded, upon appellee's objection. Since the statement purportedly represents the engineer's version of the accident, it is urged that its exclusion was prejudicial to the defense. Its exclusion was proper, says appellee, because it offends the hearsay rule.

The engineer's report would clearly be excluded under the common law rule. It does not come within the exceptions as to declarations by a deceased witness. Shepard v. United States, 290 U.S. 96, 54 S.Ct. 22, 78 L.Ed. 196; People v. Sarzano, 212 N.Y. 231, 106 N.E. 87. Nor is it the kind of record that falls within the common law exception as to memoranda made in the regular course of business. For the courts — as an inherent and integral part of the "regular course of business" exception to the hearsay rule — have always imposed this requirement, which the engineer's statement here clearly fails to meet: The person making the record, or supplying the information on which it is based, must have had no peculiarly powerful motive to misrepresent; such a motive, if it exists must be relatively minimal and marginal. Wigmore, speaking of records made in the regular course of business, says: "It is often added that there must have been no motive to misrepresent. This does not mean that the offeror must show an absence of all such motives; but merely that if the existence of a fairly positive counter-motive to misrepresent is made to appear in a particular instance, the entry would be excluded."

This motive factor has often been stressed in the decisions. In Conner v. Seattle, R. S. Ry. Co., 56 Wn. 310, 105 P. 634, 635, 25 L.R.A., N.S., 930, 134 Am.St.Rep. 1110, the facts were substantially the same as those in the instant case. There a report of an accident was made in writing by the conductor of a street car, involved in the accident, immediately following the accident, and was soon thereafter given to the defendant Street Railway Company in compliance with its rules. It was urged that the report was admissible "as an original entry made in the due course of the business of the company and made contemporaneously with the transactions recorded." In sustaining the exclusion of this evidence, the court said: "For the sake of argument, we may admit that the report was made in due course and in compliance with a rule and custom universally followed. Yet we are quite unable to see how the statements made in such report can escape the objection of being self-serving, in so far as they were favorable to appellant's contentions (and, of course, it was because they were so favorable that they were offered to support its contentions), being made by appellant's agent and in its interest concerning facts which the agent at the time of making them knew would most likely become matters of dispute and drawn into litigation. Indeed, it is evident that the very making of the report upon the facts surrounding the accident was prompted by the possibility of the respondent claiming damages and suing the appellant therefor. * * * In this case the record of the facts, in the form of the conductor's report, was made for the very purpose of aiding appellant in possible future litigation with the respondent." The court distinguished an earlier case where the question of fact was whether or not a woman was a passenger upon a certain car during a certain trip, she having testified that she had paid her fare by transfer slip; the conductor's trip report was there held admissible as having been made in the regular course of business. Speaking of that case, the court, in the Conner case, said: "We think a careful reading of that decision will show that the court did not regard the report as self-serving, for the reason it was not made under circumstances when there were any inducements whatever to record the facts other than as they actually occurred at the time. It was nothing more or less than a simple matter of bookkeeping in the usual course of business, without any thought of future litigation drawing the facts so recorded in question. It was by reason of the absence of such considerations at the time of making the report that it was there admitted in evidence." In Bloom v. Union Railway Company, 165 App. Div. 257, 150 N.Y.S. 779, and North Hudson Ry. Co. v. May, 48 N.J.L. 401, 5 A. 276, the courts reached the same result on similar facts.

In the Conner case, the conductor's report was (1) made pursuant to a rule imposing a duty to make it and (2) was made in the "regular course of business" — using those words in their colloquial sense. But the court refused to give them such a colloquial meaning, since, if it did so, the foundation of the "regular course of business" exception would disappear. Those words had come to be a short-hand expression or symbol for a doctrine, the essence of which is the reliance on records where the circumstances in which they were made furnish sufficient checks against inducements to misstate to make them trustworthy, give them "some badge of truthfulness."

That basic concept is recurrently expressed in the cases. In Freedman v. Mutual Life Ins. Co., 1941, 342 Pa. 404, 21 A.2d 81, 85, 135 A.L.R. 1249 hospital records were held admissible where there were present "no contemplative motive for falsification." In Re Fennerstein's Champagne, 3 Wall. 145, 147, 18 L.Ed. 121, the court emphasized the ingredient that "there was no motive to falsify." In Poole v. Dicas, 1 Bing., N.C. 649, 131 Eng.Rep. 1267, Tindal, C.J., said: "The clerk had no interest to make a false entry: If he had any interest, it was rather to make a true entry; it is easier to state what is true than what is false; the process of invention implies trouble, in such a case unnecessarily incurred; and a false entry would be likely to bring him into disgrace with his employer. Again, the book in which the entry was made was open to all the clerks in the office, so that an entry if false would be exposed to speedy discovery." In other cases, the absence of a motive to misrepresent is said to be "condition" of admissibility. See Polina v. Gray, L.R. 12 Ch.Div. 411, 429-430; Lassone v. Boston L.R. Co., 66 N.H. 345, 24 A. 902, 903-906, 17 L.R.A. 525, and cases there cited; Malone v. L'Estrange, 2 Ir.Eq.R. 16. In Lord v. Moore, 37 Me. 208, 220, the requirement was said to be that the entrant's situation "was such as to exclude all presumption of his having any interest to misrepresent the fact recorded." And Gray, J., in Kennedy v. Doyle, 1865, 10 Allen, Mass., 161, 167 says that not only must there be "no interest to misrepresent," but also that the entry must be made "before any controversy or question has arisen."

The statement of the engineer may be compared with the "protests" of mariners, which, although made by them as a matter of duty and in the regular course of their business, are inadmissible on behalf of their ships because the courts have always recognized that these reports will be biased and partisan disclaimers of responsibility for a disaster. See, e.g., Merriman v. The May Queen, Fed.Cas.No. 9,481. In Hand v. The Elvira, Fed.Cas. No. 6015, the court said that in such a document "the waves are almost always mountain high, the winds never less than a hurricane, and the peril of life generally impending."

This court has often emphasized the element of trustworthiness as the foundation of the "regular course of business" exception. For instance, in United States v. Cotter, 2 Cir., 1932, 60 F.2d 689, 693 where certain bank records were held admissible, we said that the accuracy of the records "is essential to the very life of [the bank's] business" and spoke of " the probable correctness of ordinary bank books," pointing out that "the danger of mistake is slight." In United States v. Becker, 2 Cir., 1933, 62 F.2d 1007, 1010, we said that "if challenged, the party offering the documents must prove that the system is such as prima facie to be reliable." These two cases, in turn, relied upon Massachusetts Bonding Insurance Co. v. Norwich Pharmacal Co., 2 Cir., 1927, 18 F.2d 934, 937 in which we referred to records that "are in practice accepted as accurate upon the faith of the routine itself, and of the self-consistency of their contents."

The same "principle of a circumstantial guarantee of trustworthiness" — involving the absence of any vigorous motive to misrepresent — is inherent in virtually all the exceptions to the hearsay rule, such as declarations about private boundaries, statements or records concerning family history, spontaneous declarations, and dying declarations. We refer to some of the authorities in a footnote. In them, frequent reference is made to the non-existence of a controversy, likely to lead to litigation in which the declarant has a personal interest that would be likely to negate a fair degree of unprejudiced sincerity.

Wigmore has summarized, approvingly, the following reasons justifying the several exceptions to the hearsay rule: "(a) Where the circumstances are such that a sincere and accurate statement would naturally be uttered, and no plan of falsification be formed; (b) where, even though a desire to falsify might present itself, other considerations, such as the danger of easy detection or the fear of punishment, would probably counteract its force; (c) where the statement was made under such conditions of publicity that an error, if it had occurred, would probably have been detected and corrected." And, with particular reference to those "regular course of business" memoranda which are, he says, justifiably admissible although hearsay, he writes: "In the typical case of entries made systematically and habitually for the recording of a course of business dealings, experience of human nature indicates three distinct though related motives which operate to secure in the long run a sufficient degree of probable trustworthiness and make the statements fairly trustworthy: (1) The habit and system of making such a record with regularity calls for accuracy through the interest and purpose of the entrant; and the influence of habit may be relied on, by very inertia, to prevent casual inaccuracies and to counteract the possible temptation to misstatements * * * (2) Since the entries record a regular course of business transactions, an error or mis-statement is almost certain to be detected and the result by those dealing with the entrant; mis-statements cannot safely be made, if at all, except by a systematic and comprehensive plan of falsification * * * (3) If, in addition to this, the entrant makes the record under a duty to an employer or other superior, there is the additional risk of censure and disgrace from the superior, in case of inaccuracies — a motive on the whole the most powerful and most palpable of the three."

It is clear then, that the words "regular course of business," as used in the decisions, have always included the concept that the circumstances must be such as to safeguard against any crude bias on the part of persons making the records or supply the information and against any great likelihood that the records may have been fabricated by interested persons for the primary purpose of use in litigation which is in prospect at the time. The mere fact that such entries were made with a view to perpetuating evidence is not sufficient to show such bias as to exclude them. But it is beyond question that a requirement in a business that reports should regularly be made which, by their very nature, are highly likely to be biased, did not bring such reports within the meaning of the words of art, "regular course of business." That the defendant railroad here had a regulation requiring its employees, when they were the actors in accidents, regularly to make reports of such accidents for use in probable litigation, did not suffice to include such reports within the "regular course of business," as those words have always been understood by lawyers and judges. For the "regularity" which justifies the exception is the kind which tends to "counteract the possible temptation to mis-statements," as Wigmore has noted. It follows that the phrase "regular course of business" never covered a regular practice of making records with the purpose of supplying evidence in a highly probable law suit, when those records are made by persons with every "possible temptation to mis-statements." We have found no case holding or even suggesting that, absent a statute changing the common law rule, such a statement as the engineer's is admissible, loaded as it is with motives to misrepresent the facts.

The question, then, is whether this evidence, so plainly barred at common law, was made admissible by federal legislation, i.e., U.S.C.A., Title 28, § 695, enacted in 1936. This statute renders admissible "any writing or record, whether in the form of an entry in a book or otherwise, made as a memorandum or record of any act, transaction, occurrence, or event," if the trial judge shall find "that it was made in the regular course of any business, and that it was the regular course of such business to make such memorandum or record."

The words, "regular course of business," twice employed in the legislation, are not colloquial words but are words of art, with a long history, and, as we have observed, often theretofore judicially interpreted. Consequently, they should be given that settled meaning when incorporated in a statute, absent a contrary legislative intention clearly expressed in the statute or in its legislative history.

In Case v. Los Angeles Lumber Co., 308 U.S. 106, 115, 60 S.Ct. 1, 7, 84 L.Ed. 110, the Supreme Court said: "The words 'fair and equitable' as used in § 77B, sub. f [11 U.S.C.A. § 207(f)] are words of art which prior to the advent of § 77B had acquired a fixed meaning through judicial interpretations in the field of equity receivership reorganizations. Hence, as in case of other terms or phrases used in that section, Duparquet Huot Moneuse Co. v. Evans, 297 U.S. 216, 56 S.Ct. 412, 80 L.Ed. 591, we adhere to the familiar rule that where words are employed in an act which had at the time a well known meaning in the law, they are used in that sense unless the context requires the contrary. Keck v. United States, 172 U.S. 434, 446, 19 S. Ct. 254, 258, 43 L.Ed. 505." In Keck v. United States, the court said: "These conclusions arising from a consideration of the text of the statute are rendered yet clearer by taking into view the definite legal meaning of the word 'smuggling.' That term had a well-understood import at common law, and in the absence of a particularized definition of its significance in the statute creating it, resort may be had to the common law for the purpose of arriving at the meaning of the word. Swearingen v. United States, 161 U.S. 446, 451, 16 S.Ct. 562 [40 L.Ed. 765]; United States v. Wong Kim Ark. 169 U.S. 649, 18 S.Ct. 456 [42 L.Ed. 890]."

In a given context, words often come to have a meaning which they do not have in other contexts. What "apostles" mean in an admiralty rule would surprise a theologian. To a mathematician, pi is not a Greek letter, no more than F.O.B. to a railroad company is a chance selection from the English alphabet. "Unearned" in the insurance trade, as used in connection with premiums, does not indicate what it does mean to a man on the street. The phrase "divided into watches" in a Seamen's Act, 46 U.S.C.A. § 673, "is," says the Supreme Court, "to be given the meaning which it had acquired in the language and usages of the trade to which the Act relates * * *" and not "the common or ordinary meaning of the words." To the manufacturers of drinks, beer is not a carbonated beverage although it is to a chemist. "Signed by the party to be charged or by his agent" is a phrase with a history it cannot easily shake off, when employed in a statute. The words "equity receivership" do not go virginally into a corporate reorganization statute. Each trade has its peculiar jargon and courts rely on that jargon when it finds its way into a statute dealing with that trade.

And so with "regular course of business" as applied to records or memoranda in an evidence statute. To a layman, the words might seem to mean any record or paper prepared by an employee in accordance with a rule established in that business by his employer. But according to the jargon of lawyers and judges those words, in discussions of evidence, have always meant writings made in such a way as to afford some safeguards against the existence of any exceptionally strong bias or powerful motive to misrepresent.

Those words came into the statute saturated with history. They connote — to recall Wigmore's comments — (1) a regularity serving "to counteract the possible temptation to misstatements"; (2) a situation which would lead to detection of falsification, so that mis-statements "cannot safely be made"; (3) a relationship, when a writing is made by an employee under a duty to his employer, which includes the "risk of censure and disgrace" for mis-statements. It would, therefore, require unequivocal expressions in the statute or its legislative history to yield an interpretation of those words, defying their history, which would render admissible a memorandum made in circumstances that disclose the strongest likelihood of the existence of a motive to misrepresent and the least probability of censure from the employer who imposed the duty to make it, if the memorandum misdescribes the facts so as to favor him.

The statute was a so-called "Model Act," proposed for uniform adoption by the several states and the federal government. New York enacted it in 1928. In the next year, long before Congress also enacted it, the New York statute was interpreted in New York in a case closely resembling the instant case and similar cases above cited; In Needle v. New York Railways Corp., 1929, 227 App. Div. 276, 237 N.Y.S. 547, 549, a policeman, in the course of his regular duties, made a report of a street railway accident. As he had not witnessed the accident, his report was based on the oral statements of others, including, as the court said, that "of the interested motorman, who, instead of being so placed as to be presumed to be without a motive to falsify in helping to make the record, had every reason to give a biased and false report." It was held that the policeman's report was not admissible under the Model Act.

The court in the Needle case gave as another ground for its rejection that the report was based upon the statements of others than the motorman who were under no duty to make them and that the policeman's hearsay statement was founded upon the hearsay of others and not upon his own knowledge. With that aspect of the decision we are not concerned, for it has no bearing here. When herein we refer to the Needle case, we mean, unless we state otherwise, that part of the decision rejecting the motorman's statement because of the presence of that same motive to misstate which caused the rejection of the conductor's statement in Conner v. Seattle R. S. Ry. Co., supra, and similar cases.

It was in 1936, seven years after the Needle decision, that Congress adopted the Model Act. No change in its verbiage was suggested or was made to indicate an intention to deviate from that reasonable New York interpretation. It is a general rule that where a statute has been previously enacted in another jurisdiction, interpretations by its courts before its enactment in another jurisdiction are to be followed because the statute "generally is presumed to be adopted with the construction which it has received." Holmes, J., in James v. Appel, 192 U.S. 129, 135, 24 S.Ct. 222, 223, 48 L.Ed. 377. And constructions adopted by other jurisdictions are peculiarly persuasive where the statute is designed to be "uniform." Union Trust Company v. McGinty, 212 Mass. 205, 98 N.E. 679, Ann.Cas. 1913C, 525; Forgan v. Smedal, 1931, 203 Wis. 564, 234 N.W. 896; Stewart v. Hansen, 1923, 62 Utah 281, 218 P. 959, 44 A.L.R. 340. Of course this rule, like all rules of statutory construction, is not inflexible. We do not, therefore, necessarily feel bound to apply, to the federal statute, all the interpretations of the New York courts, if we find them highly unreasonable. But the ruling in the Needle case, we think, should be followed as entirely reasonable, absent most persuasive contrary arguments.

The draftsmen of the Model Act had not, before Congress enacted it, indicated that they meant that it should render admissible memoranda such as that excluded in the Needle case, i.e., those made in circumstances where the "regularity" lacked all possible safeguards of reliability. The draftsmen and sponsors of that Act consisted of a Committee (headed by Professor E.M. Morgan) appointed by the Legal Research Committee of the Commonwealth Fund, which reported its findings and recommendations in a volume, The Law of Evidence: Some Proposals for Its Reform, published in 1927. It was there said that the problem sought to be solved by the proposed statute was "the need of inducing the courts to give evidential credit to the books upon which the mercantile and industrial world relies in the conduct of business." And the chief criticism made by the Commonwealth Committee of the existing common law rule was that, because each clerk or bookkeeper involved in any transaction must be called or accounted for, any break in the very elaborate chain of a typical business system was fatal and thus, practically, rendered such evidence inadmissible in many cases. In support of this criticism, that Committee sets out in its report the chain of events in a large business house, showing that scores of persons, many of them unidentifiable, work on an order at various stages from the time it is sold until the customer is billed. One reading the report of the Committee might, therefore, reasonably assume that perhaps its chief purpose was the desire to avoid the necessity of proving each link of such a chain. At any rate, he would surely not think that the statute was designed to make admissible documents, like the strongly motivated engineer's statement involved here, which (quoting again from the statute's sponsors, Morgan et al.) in no remote way resemble the kind of record "upon which the mercantile and industrial world relies in the conduct of business."

Certainly no lawyer reading the statute would suppose that its draftsmen intended that the hearsay statements which it renders admissible should be attended by nothing whatever to guard against misstatements. He must assume that the phrase, "regular course of business," was inserted — and twice — with some purpose. Acquainted with the hearsay rule, he must assume that the purpose was to retain some at least of the assurances which that phrase has always symbolized. But to interpret it so as to admit in evidence the engineer's statement here would be to strip that phrase of every vestige of its established connotation. For there can be no slight shadow of any guaranty against a "temptation to misstate" if the words as to memoranda "made in the regular course of business" refer to such reports of accidents — reports required by employers of employees who are participants in the accidents — since, almost inevitably, the inclination of those employees will be to describe such accounts so as to make it appear that those employees and their employer were not at fault.

No one knew better than the sponsors of the Model Act — men like Wigmore and Morgan — the traditional significance of "regular course of business." There can be no doubt that their intention was to widen the exception to the hearsay rule relating to such writings. But it is equally without doubt that they did not intend to abolish the exception and to substitute another, by giving that phrase a meaning precisely opposite to that which they well knew was its recognized meaning. If that had been their intention, they would surely have said so, either in the language of the Act itself or in their Report, in order to avoid misleading the lawyers in the legislatures asked to enact that statute. There is nothing whatever in the Report of the Commonwealth Committee even faintly intimating any purpose completely to do away with every one of the traditional safeguards against a motive to misstate in statements made in "the regular course of business." Nor is there anything in any subsequent comments of any members of that Committee showing that they had any such intention.

And we cannot impute such an unusual intention to Congress. If that was what Congress meant to do, "it would have been easy to say so." It would have been still easier to have omitted altogether the historic words. We must assume that Congress used them deliberately with recognition of their history. And we must so interpret them. We have no right to be generous with other people's words.

The statute, in describing the instruments which it renders admissible, says that they may consist of "any writing" whether "in the form of an entry in a book or otherwise." This was in line with the purpose to broaden the exception: Those words serve to remove any possible doubts as to the form of the writing, and of course we recognize and will give full effect to that intention. We say at once that the engineer's statement was not inadmissible merely because it was not an entry in a book or merely because it was otherwise informal in character. But, whatever their form, the statute requires that memoranda be in "the regular course of business." And the same comments apply to the statutory provision, "The term 'business' shall include business, profession, occupation, and calling of every kind," and to the provision that, "All other circumstances of the making of such writing or record, including lack of personal knowledge by the entrant or maker, may be shown to affect its weight, but they shall not affect its admissibility." They were, also, in accord with the purpose of broadening the exception. But they disclose no purpose to exterminate the inherent character of the exception, i.e., to remove all safeguards theretofore connoted by "regular course of business." In the instant case, we may add, there was no lack of personal knowledge by the maker of the statement, the engineer, and the railroad business was undeniably within the statutory scope.

There are many persons who believe that the hearsay rule should be wiped out. Perhaps they are right. But, in interpreting a statute embodying a limited amount of reform, we must not allow our personal preferences for a more extensive reform to govern our decision. It is our function to find out what Congress intended. If it did not go or want to go as far as we may think desirable, we are not justified in re-shaping the legislation to suit our personal wishes. Of course, there has always been and there always will be judicial legislation. But it should be cautious — "interstitial," Holmes called it. With respect to jury trials, there is still some room for judicial innovations. However, some legal rules — and the hearsay rule is one — are too well established to permit the judiciary to modify them substantially. What Holmes said of consideration is pertinent: "A common-law judge could not say, 'I think the doctrine of consideration a bit of historical nonsense and shall not enforce it in my court.'" And that remark is peculiarly apposite when Congress has acted to introduce specific changes in the hearsay rule; the judges should not take such specific legislative action as giving them a license to legislate at large in that field. This is not to say that courts should be niggardly in giving effect to a new policy set forth in a statute merely because the legislature has carelessly expressed itself. As observed by Holmes, J., in Johnson v. United States, 1 Cir., 1908, 163 F. 30, 32, 18 L.R.A., N.S., 1194: "The Legislature has the power to decide what the policy of the law shall be, and if it has intimated its will, however indirectly, that will should be recognized and obeyed. * * * It is not an adequate discharge of duty for courts to say: We see what you are driving at, but you have not said it, and therefore we shall go on as before." Mr. Justice Cardozo, citing the Johnson case, said in Van Beeck v. Sabine Towing Co., 300 U.S. 342, 351, 57 S.Ct. 452, 456, 81 L. Ed. 685, "There are times when uncertain words are to be wrought into consistency and unity with a legislative policy which is itself a source of law, a new generative impulse transmitted to the legal system." But that does not mean that the courts are free to read into a statute something at which the legislature was not driving.

The limited objective at which Congress was in fact driving, is made clear by the Report of the Senate Judiciary Committee reporting out, with recommendations for passage, the bill which became 28 U.S.C.A. § 695. The Report (with the exceptions of recommendations for a few unimportant verbal changes in the draft of the Model Act) consists entirely of a letter from the Attorney General to the Chairman of the Committee and a memorandum referred to and enclosed with that letter. As we are bound to infer that the Committee and Congress, in enacting § 695, relied on that letter and that memorandum, their contents are significant. The letter reads as follows:

"Modern developments have rendered obsolete the common-law rule governing the admissibility of certain types of documentary evidence. Yet at times the application of the rule has resulted in a miscarriage of justice and has stood in the way of a successful prosecution of meritorious criminal cases. The old common-law rule requires that every book entry be identified by the person making it. This is exceedingly difficult, if not impossible, in the case of an institution employing a large bookkeeping staff, particularly when the entries are made by machine. In a recent criminal case the Government was prevented from making out a prima-facie case by a ruling that entries in the books of banks, made in the regular course of business, were not admissible in evidence unless the specific bookkeeper who made the entry could identify it. Since the bank employed 18 bookkeepers, and the entries were made by bookkeeping machines, this was impossible. The United States Circuit Courts of Appeals for the Second, Fourth, Seventh and Eighth Circuits, and many district courts, as well as a number of the State courts, have recognized the necessity for modifying the rule and have adopted the doctrine that in order to make it admissible in evidence, it is sufficient to show that the entry is contained in a book of regular entries maintained in the establishment, without producing the particular person who made the entry and having him identify it. Owing to the failure of some Federal courts, however, to adopt the modern rule, legislation appears to be necessary to secure uniformity in this matter, and to keep the rules of evidence in line with modern developments. I enclose a draft of bill to accomplish the above-mentioned purpose, and shall be glad if you will introduce it and lend it your support. I also enclose a memorandum, dated January 28, 1936, discussing the questions involved in greater detail."

That memorandum, printed in full in the Senate Committee's Report, cites and quotes from earlier decisions of this and other courts, relaxing "the regular course of business" exception in one respect only, i.e., that where there is a regular system of making entries, and the system is such as to be "likely to ensure accuracy," there is no necessity of introducing the evidence of the entrants. In each of those cases, the court said that an essential condition of the admission of such hearsay is the "circumstantial guaranty of trustworthiness" consisting of the accuracy ensured by the nature of the regular system of entries.

The memorandum stated: "While this modification of the common-law rule has been adopted by most of the Federal courts, as well as by many of the States, nevertheless there are some courts which do not follow it, and for that reason legislation on the subject appears highly desirable * * * A draft of bill is submitted herewith designed to make uniform in the Federal courts the modern rule now followed generally by the Federal courts and many State courts to which reference has been made."

It is apparent that the intention of the Attorney General — and, therefore, of the Senate Committee and of Congress which adopted his explanation as theirs — was to bring the decisions of all the Federal courts into line with the "modern" prestatutory decisions of this court and the Circuit Courts of Appeals for the Fourth, Seventh and Eighth Circuits. There is no slight indication of any intention of going far beyond the purpose of the legislation as explained by the Attorney General and to remove the essentials of the "guarantee of trustworthiness" which had been recognized by this and other courts when in their decisions they had enunciated the "modern rule." Had such been the intention, surely the Senate Committee Report would have at least so intimated and not preserved utter silence concerning it.

As shown by the cases we previously cited, no one would have questioned the correctness of the Needle decision, absent a statute over-ruling it. We have discovered no case, in any jurisdiction where the Model Act has been enacted, holding or intimating that the doctrine expressed in the Needle case has been expunged by that statute. We go further: We know of no decision construing that statute and of no comments by anyone to the effect that the Model Act renders admissible reports by employees (under regulations of their employers requiring such reports) of accidents in which those employees have been active participants. It is suggested that Morgan and Wigmore have said that it was intended that such reports should be admissible. But we have been unable to find that either of them has ever published any such comments, i.e., that they have ever discussed the problem which is here before us, in a case arising under the statute, either vis a vis the Needle case or otherwise.

It is true that another New York case, Johnson v. Lutz, 253 N.Y. 124, 170 N.E. 517, 518, has been criticized by these commentators. There a policeman's report of an accident, based on the unsworn statements to him of bystanders, was excluded, as not admissible under the Model Act, where there was no showing, the court said, "whether they saw the accident and stated * * * what they knew, or stated what some other persons had told them." The criticism of the case (as erroneously narrowing the scope of the statutory provision that lack of personal knowledge of the "entrant or maker" of the record shall not affect its admissibility) loses some of its force when it is noted that the A.L.I. commentator mistakenly states that the policeman's report was "based upon statements made to him by persons who claimed to have seen the accident," and that Wigmore makes a similar error in stating the facts. The decision, moreover, was by a unanimous court, including Judge (now Chief Judge) Lehman, who had previously indicated that the regular entry exception ought to be liberally construed, and Chief Judge (later Mr. Justice) Cardozo, who had not only written in a similar vein, but was also a member of the Legal Research Committee of the Commonwealth Fund, which sponsored the Model Act.

There is no occasion for us here to consider the merits of the Lutz decision. Indeed, we may, arguendo, assume it to have been wrong. For it has no bearing whatever on the case at bar. And if it be thought wrong, it is clearly distinguishable from the Needle case, supra, where the report was excluded because of the motorman's probable bias even though (a) he was, of course, familiar with the facts, (b) he was probably under a duty to state the facts to the investigating policeman, and (c) the policeman, acting officially, was disinterested. The Needle case was decided before the Court of Appeals decided Johnson v. Lutz, and there can be no doubt that the court which decided the Needle case, excluding a policeman's report, would be even more ready to exclude the company's document here. It would do so not because of the reasons given in the Lutz case, and criticized by Wigmore and Morgan, but because of the existence of that strong motive to misrepresent on the part of the interested motorman, which, prior to the statute, had led to decisions excluding accident reports by conductors or similar employees in similar circumstances.

Morgan so far as we can discover, has never criticized the Needle case. Wigmore has done so briefly in a footnote thus: "Highway injury; a police blotter containing a report of the police officer who did not see the affair but took the statement of various persons on arriving, excluded, following Johnson v. Lutz; again directly contrary to the express words" of the statute. It is clear from his description of the facts of the Needle case that Wigmore dealt with them as if they were the same as those in the Lutz case. He overlooked entirely the crucial fact — differentiating Needle sharply from Lutz — that among "the various persons" in Needle was the highly "interested motorman" who "had every reason to give a biased and false report." That part of the opinion which turns on those facts Wigmore has never mentioned or criticized. It cannot possibly be said, therefore, that he has regarded that aspect of the case as "contrary to the express words" of the Act. It is difficult to believe that, had he noted that distinguishing factor, he would have criticized the decision. For as we have seen, he has underscored the absence of "a motive to misrepresent" as an integral and essential part of the "regular course of business" exception. Clearly every one of those guarantees of trustworthiness, justifying, according to Wigmore, the admission of such hearsay, in proper circumstances, was absent in the Needle case and is absent here.

With all that in mind, we construe the statute as not making admissible the engineer's statement which, by its very nature, is dripping with motivations to misrepresent. Accordingly, we decide this and no more: The statute does not permit the introduction in evidence of a hearsay statement in the form of a written memorandum or report concerning an accident, if the statement was prepared after the accident has occurred, where the person who makes the memorandum or report knows at the time of making it that he is very likely, in a probable law suit relating to that accident, to be charged with wrongdoing as a participant in the accident, so that he is almost certain, when making the memorandum or report, to be sharply affected by a desire to exculpate himself and to relieve himself or his employer of liability.

We do not hold that the engineer's statement is inadmissible under the statute merely because it (1) was prepared to perpetuate evidence or (2) was made after litigation was imminent, or (3) was not a formal document or an entry in a book.

The decisions of this court, both before and since § 695 was enacted, have consistently high-lighted the absence of a powerful motive to misstate as a necessary factor to render admissible memoranda made in the regular course of business. Our decisions of that kind, when no statute was applicable, we have already cited. In Pressel v. New England Transportation Co., 2 Cir., 1937, 91 F.2d 1019, a case arising under the New York statute, Civil Practice Act, § 374-a, before the federal statute was operative, we sustained the lower court in rejecting a statement, citing the Lutz and Needle cases. Under the federal statute, § 695, in Hunter v. Derby Foods, 2 Cir., 1940, 110 F.2d 970, 973, 133 A.L.R. 255, where a death certificate made by a coroner in the course of his official duty was held admissible, we referred to the statute and also pointed out that the certificate would be admissible in the New York courts under a New York statute, Civil Practice Act § 367, citing Scott v. Empire State Degree of Honor, 204 App. Div. 530, 198 N.Y.S. 535. We also cited Wigmore, § 1671, which is part of his chapter on the admissibility of official statements; he there says, § 1632, that "an official duty exists to make an accurate statement, and * * * this special and weighty duty will usually suffice as a motive to incite the officer to its fulfilment * * * It is the influence of the official duty, broadly considered, which is taken as the sufficient guaranty of trustworthiness, justifying the acceptance of the hearsay statement." In Ulm v. Moore-McCormack Lines, 2 Cir., 1941, 115 F.2d 492, 495, a marine hospital record, on a form of the United States Public Health Service, executed in the regular course of duty by an attending doctor, was held by us to be admissible. We there discussed § 695 and Wigmore's criticisms of the Lutz case. Interpreting Wigmore's views of the Model Act, we said that its "objective, as Wigmore so lucidly explains, was to do away with the technical rulings which excluded records ordinarily used in business transactions when not formally identified by the makers." We referred to other New York decisions (not however, mentioning the Needle case), and added: "But whatever should be the judicial attitude toward this statute, we do not think the cited New York cases are in point on the immediate issue here." That decision was reaffirmed on rehearing in Ulm v. Moore-McCormack Lines, 2 Cir., 1941, 117 F.2d 222; there we referred with approval to Borucki v. MacKenzie Bros. Co., 125 Conn. 92, 93, 3 A.2d 224, where the Connecticut court, in turn, quoted with approval from the Lutz case. In Reed v. Order of United Commercial Travelers of America, 2 Cir., 1941, 123 F.2d 252, we held that a hospital record of an attending doctor's diagnosis of a patient's condition was admissible under § 695. Coroners' records or hospital records may be made for the purpose of perpetuating testimony, but they are, ordinarily, not made by persons with impelling motives to misrepresent.

Certainly nothing in our decision in United States v. Mortimer, 2 Cir., 1941, 118 F.2d 266, 269, 270 is inconsistent with our decision here. There we held admissible charts, purporting to show defaults in the payment of taxes, which had been prepared by a prosecution witness, Karcher, who was an experienced public accountant, assisted by several aides of whom only one in addition to Karcher took the stand. We held that those charts were not rendered inadmissible because all of Karcher's aides were not called to testify, especially as there was testimony as to the manner in which the charts had been prepared. And we rejected the argument that they were inadmissible under § 695 merely because they were made in preparing evidence for the trial. In so ruling, we said: "There are numerous cases holding admissible on the testimony of a supervising agent statements compiled from voluminous records according to a method at once practicable and offering reasonable guaranty of accuracy, even though the supervisor had not examined each record himself." After referring to Section 695 and to cases involving the use of bank records, we went on to say: "Likewise, accuracy is the life of an accountant's business, but the multitude of records cannot be checked by any one person alone. And here the system followed was not merely likely to insure accuracy, but apparently did so, since the other side, far from discrediting the records, actually supported them. The trend in the courts is unmistakably to follow the methods of ordinary business in assuming the validity, until discredited, of records daily accepted in commercial routine." There, as in all our other decisions construing and applying § 695, we were careful to ascertain that the "regularity" in the "regular course of * * * business" was such as to afford some "reasonable guaranty of accuracy" or something to show an absence of a vigorous motive to misstate.

We are, then, in no way to be understood as initiating restrictive interpretations of the statute or as retracting or modifying the favorable constructions we have given it in our previous decisions. Our decision here is no less liberal than the decisions of other state or federal courts interpreting the Model Act. For, to repeat, we know of no case in any court holding, or even intimating, that such an obviously motivated record as that here before us is admissible under that Act.

It is urged that if we stress to the extent we have done here the traditional limitation to exceptions to the hearsay rule — i.e., the absence of a strong motive to misrepresent — we will be reverting to a notion which went out of favor a century ago. Page after page of Wigmore's treatise goes to show that that limitation is not thus outmoded either in judicial decisions or according to Wigmore's views as to correct practice. Two years ago, this court, in Meaney v. United States, 2 Cir., 112 F.2d 538, 540, 130 A.L.R. 973, held that narrative statements made by a patient to his physician in describing his "history" are admissible, provided the trial judge decides that "the patient was consulting the physician for treatment and for that alone"; we said (per Judge Learned Hand) that the warrant for the admission, in such circumstances, is that the patient has a "motive to speak the truth" because "his treatment will in part depend upon what he says." And our recent decisions, above discussed, construing § 695, are in accord with that view.

It is also suggested that the rejection, many years ago, of the rule disqualifying interested witnesses from testifying, subject to cross examination, destroyed this rationale of exclusion of hearsay statements, not subject to cross examination, when made with strong motives to misrepresent. The two kinds of evidence are surely of a markedly different character, as has been recognized by the courts which, for about a century, have admitted the first while excluding the second.

It is further suggested that, if we hold that the extent of the motivation here is so great as to preclude adequate trustworthiness, we will be erecting an unworkable standard, as it will involve questions of degree. Even if it did, that would be nothing new, for the cases cited show that, for decades and up to now, the courts have been able to apply such a standard; in Massachusetts Bonding Insurance Company v. Norwich Pharmacal Co., 2 Cir., 18 F.2d 934, 937, we said: "The question, as we view it, like many other questions as to the competence of evidence, is of degree, and is not susceptible of absolute regulation." But our decision here raises no such problem; we are not leaving the extent of the disqualifying motive under § 695 at large or entrusting discretion with respect to it to the trial judge; for, as we have said, we decide merely this: The statute does not render admissible a hearsay statement made by an employee under standing orders from his employer to make reports of accidents in which the employee is a participant, where the primary purpose of the employer, obvious from the circumstances, in ordering those reports is to use them in litigation involving those accidents.

To avoid possible misunderstanding, this should be added: Appellants did not suggest in the court below or in this court that the presence of Mr. Christie, a representative of the State Utilities Commission, when the engineer was interviewed, endowed the engineer's statement with an official status. The sole offer of proof at the trial concerning this matter reads as follows: "The defendants offer in evidence the statement of the engineer, who the proof indicates is now dead, a statement taken in the regular course of business the defendant claims, after the accident happened. The statement was signed by the engineer, and is marked for identification as Exhibit J, under § 695, 28 U.S.C.A. The defendants offer the proof also that this statement was signed in the regular course of any (sic) business, and that it was the regular course of such business to make such statement." In appellants' brief in this court, appellants state their argument as to the alleged error in excluding this statement, as follows: "The defendants offered this statement in evidence under the provisions of 28 U.S.C.A. § 695. Defendants offered proof that the statement was signed in the regular course of business and that it was the regular course of such business to make the statement." The brief then quotes § 695 and continues: "This court has passed upon this section in several cases * * *," citing authorities interpreting that statute. § 695 does not differentiate, in any way, between reports made by public officials and reports made by others. Appellants have not in their briefs referred to the Massachusetts statute requiring its Utilities Commission to make reports as to railroad accidents, nor to the presence of Mr. Christie, the Commission's representative, at the interview with the engineer.

It is thus plain that the engineer's statement has not been asserted by appellants to be, or offered by them as, the report of a public official, or as being admissible as part of such a report. Moreover, we do not know whether or not Mr. Christie was a designated official investigator. If he was, we do not know whether or not he made his official report after attending this interview. If he did, we have no way of knowing what reliance, if any, he placed upon the answers here given by the engineer. Perhaps, if he made a report, it was damaging to appellants' case. The admissibility of an unoffered and perhaps non-existent official report does not serve to make admissible a statement which is not an official report, so that exclusion constitutes reversible error.

Even if we were to assume — contrary to fact — that appellants were in this court urging the admission of a public officer's report or urging that the engineer's statement had something of the status of a public official's report, there would be this added difficulty: It would have been incumbent upon appellants to have brought such matters to the attention of the trial court and to have preserved their offer of proof, made on that basis, for appeal. Such an offer of proof, while provided for by Federal Rules of Civil Procedure, rule 43, 28 U.S.C.A. following section 723c (cf. 3 Moore's Federal Practice, 3076, 3077), is not absolutely essential, if it is otherwise entirely clear what the alleged error is. Meaney v. United States, supra. But where, unlike the Meaney case, the significance of the excluded evidence is not obvious, we could not be expected to reverse on the mere possibility that the exclusion was harmful. Gantz v. United States, 8 Cir., 127 F.2d 498, 503; Herencia v. Guzman, 219 U.S. 44, 46, 31 S.Ct. 135, 55 L. Ed. 81. That question, however, is not before us, since, as we have seen, appellants, on this appeal, have urged merely that the statement was admissible under § 695, and have made no effort whatever to tie it to the tail of an official kite.

It is intimated that, as the defendant offered to prove that the engineer's statement was "signed in the regular course of business and that it was the regular course of business to make such statement," we must assume that appellants offered to prove and could have proved that the statement came within the statute as we have interpreted it. But the engineer's statement is in the record, and we know, from the evidence, that the engineer was a party to the accident. No proof, then, was possible that he did not have the peculiarly strong motive to misrepresent of the kind which, we hold, precludes its admission.

Finally, it is argued that the death of the engineer, making his testimony unavailable, should induce an unusually liberal interpretation of the statute. But, if the statement otherwise came within the statute, it would be admissible regardless of whether the engineer were still alive. As the statute says nothing as to death as a basis for admission, that fact adds nothing, unless at common law it is an independent ground for admission. As we have seen, it is not. There is a suggestion that the American Law Institute is this year recommending the enactment of a new Code of Evidence which — if Congress were to adopt it — would unquestionably render admissible the engineer's statement solely because he is dead. But a proposed statute gives courts no authority. It is for us to construe and apply the existing enactment, not one which would enlarge it and which is, as yet, but a wish. Nor is it our function to write into our decisions the provisions of that proposed Code, several of which are at variance with Supreme Court decisions. Moreover, if the A.L.I. Code were in effect, it would give the trial judge discretion to exclude evidence such as the engineer's statement here, "if he finds that its probative value is outweighed by the risk that its admission will * * * create substantial danger * * * of misleading the jury." By this simple provision, there would be solved a problem bothersome to some of our evidence-law reformers, i.e., how to liberalize the hearsay rule extensively in non-jury cases without automatically extending the liberalization in its entirety to jury cases. But if appellants' construction of § 695 were correct, the trial judge here would have had no such discretion.

2. One of plaintiff's witnesses testified on cross examination that, before this trial, he had given a written statement to plaintiff's attorney. Defendant's attorney, at the trial, asked for that statement. He was admonished by the trial judge that, if he looked at it, it would be admissible, should plaintiff's counsel then choose to offer it. In view of that ruling, defendant's counsel withdrew his request, but under protest. Error is assigned as to that ruling.

The doctrine that such a request makes such a document admissible has a long history. The courts of some jurisdictions still adhere to it; others have rejected it. The New York courts are in the latter class. See, e.g., Smith v. Rentz, 131 N.Y. 169, 30 N.E. 54, 15 L.R.A. 138. But Federal Rules of Civil Procedure, rule 43(a) provides that "the rule which favors the reception of the evidence governs." And the doctrine, supporting the ruling of the trial judge here, was approved, more than fifty years ago, by Judge Lacombe, sitting in the Circuit Court for the Southern District of New York, in Edison Electric Light Co. v. United States Lighting Co., 1891, 45 F. 55, 59. Relying on that decision, the doctrine was applied in McCarthy v. Palmer, D.C. 1939, 29 F. Supp. 585 (affirmed on other grounds by this court in 2 Cir., 113 F.2d 721) despite the intervening adoption of the Federal Rules of Civil Procedure.

That doctrine has never been endorsed by the United States Supreme Court. It grew up as a branch of what was once a "fixed principle" of the common law "that a party was to be kept in the dark as to the tenor of evidence in his opponent's possession." Several explanations of that "principle" have been given. It is said that it stems from the "sporting theory of justice," according to which a trial is regarded not as a means of ascertaining the true facts of the case but as a game of wits between opposing counsel. Another explanation is that the "principle" was founded on the belief that to surrender one's evidence before its disclosure at the trial would enable an unscrupulous adversary to fabricate counter-evidence. These explanations — like most efforts to explain habits and customs — are, probably, alone or together, only partially correct.

A counter "principle" has asserted itself; a movement for liberal pre-trial discovery developed and has gained more and more momentum. The resistance to that movement has been ascribed by Wigmore in large part to fear that "the reduction of litigation to a small compass, in time and expense, would diminish the emoluments of the professional men at law — whether as attorneys or counsellors or as other officers of court depending upon the number and amount of fees." Doubtless such an economic factor has been a partial element in the opposition to the reform. But here, as in many other situations, it will not do to overstress the pecuniary element. In all aspects of life there is resistance to change, sometimes rational and sometimes not. There is a social as well as a physical inertia. Social inertia, when irrational, is usually not explicable solely in terms of pecuniary motives; often such motives are absent or subsidiary. The non-economic irrational components of such inertia are multiple. There is a feeling of pleasure in identification with customary forms; there are emotional disturbances in the face of innovations calling for new adjustments: the new is dangerous, and "not to venture is not to lose." Our Declaration of Independence expressed a profound truth: "All experience hath shewn that mankind are more disposed to suffer, while evils are sufferable, than to right themselves by abolishing the forms to which they are accustomed." There are non-economic vested interests — vested interests in prestige, or in illusions, for example. Physiological and psychological bases for hostility to change have been suggested. Perhaps the anthropologist, dwelling on the way in which some customs irrationally persist, comes closest to supplying an answer: "Culture traits," which originally may or may not have had any utility, sometimes endure regardless of whether they have present social value. "In a certain island in Oceana," writes Benedict, "fish-hooks are currency and to have large fish-hooks came gradually to be the outward sign of great wealth. Fish-hooks are therefore made very nearly as large as a man. They will no longer catch fish, of course. In proportion as they have lost their usefulness, they are supremely coveted." Benedict remarks that all societies, including our own, have their equivalents of such fish-hooks.

With special reference to proposed changes in legal rules, there should not be neglected the devotion found in some members of any professional group to the established professional rituals. One need not go as far as does Seagle in underscoring the effects of professionalism on the "law" to agree with him that many lawyers oppose innovations in legal techniques without regard to their social desirability, and devise rationalizations to support their prejudices. We hear some of them complaining that the new Federal Rules of Civil Procedure, with their hospitality to pre-trial discovery, have engendered fraud and perjury. The answer is that no one knows. Unfortunately, there were perjury and coaching of witnesses in the old days; no data is available to show whether those evils have waxed or waned in these newer days. Some lawyers also grumble, saying that it is "unfair" that a lawyer who has diligently prepared his case should be obliged to let counsel for the adversary scrutinize his data. But the reformers are surely right in replying that "unfairness" to a diligent lawyer is of no importance as against much-needed improvement in judicial ascertainment of the "facts" of cases; the public interest in such ascertainment is paramount. The supporters of liberality in discovery assert that it tends to bring about settlements and to reduce litigation. Whether that is true, again we do not know; the experiment is now in full operation but the returns are not yet in.

At any rate, the old "fixed principle" of keeping the opponent in the dark as to the tenor of the evidence in one's possession is now out of date. The appendant rule here in question is equally so. It is as anachronistic as the buttons on the sleeve of a man's coat; but such a legal rule is more important than coat-sleeve buttons. As it cannot be reconciled with the liberality as to depositions and discovery contained in the new Rules, we reject it. The judge should have dealt with the request as if it had arisen under Federal Rules of Civil Procedure, rule 26(b).

Nevertheless, we do not reverse here for error in the trial judge's ruling, for these reasons: (1) The written statement of the witness could, at most, have been used for purposes of impeachment. As that statement is not in the record before us, it is impossible for us to know whether it contained any remarks contradicting the witness' testimony at the trial so that it would have served for impeaching purposes. If counsel wanted to assign error, he should have asked the trial judge to certify that statement to us, as part of the record on appeal. Since the statement is not before us, the result, if we were to reverse, would be to send the case back on the mere chance that the statement may contain matter which would have led to such an impeachment of the witness as materially to affect the jury's verdict. A verdict should not be so lightly disturbed. (2) Moreover, we cannot say that the trial judge or appellants' counsel was unreasonable in relying on Judge Lacombe's decision in the Edison Electric case. (Certainly appellants' counsel was not surprised, since it happens that he had, on behalf of the same clients he represents here, successfully persuaded the judge to render the decision in McCarthy v. Palmer, supra.) In the circumstances, it would be unwise to overturn a verdict because of the erroneous ruling on this point.

3. The next exclusionary ruling complained of is the court's refusal to permit Adams, a civil engineer in appellant's employ, to describe certain observations made by him at the scene of the accident but long after its occurrence. Appellee had testified that the accident occurred on a dark but clear night, and that although he had come to a full stop about 15 or 20 feet from the track before proceeding, he had not observed the engine. Adams was allowed to testify from a map prepared by him as to the physical location of various objects at or near the crossing. He then testified that he went to the crossing some ten months after the accident, both during the daytime and on "a clear night without a moon, as I recall it," and observed the scene. The judge did not allow him to testify to what he could see when standing at various distances from the crossing, and appellant urges that this constitutes prejudicial error.

It should be noted that Adams was testifying as an observer rather than as an expert, and that even an expert may not state his opinion as to matters of common knowledge. First Trust Co. v. Kansas City Life Ins. Co., 8 Cir., 79 F.2d 48, 54; Farris v. Interstate Circuit, 5 Cir., 116 F.2d 409, 412; United States Smelting Co. v. Parry, 8 Cir., 166 F. 407, 410-415. Nevertheless, his observations probably would have been useful, and should have been admitted, if it had been shown that he could testify to observations made under identical circumstances as those prevailing on the night of the accident. But it does not appear that the visibility was the same, that he was in a similar automobile, that the engine he observed was moving at the same speed or had a head-light of comparable power or design. Unless these conditions were comparable, it can hardly be argued that exclusion of his observations was a fatal error. We are confirmed in this view by the fact that the jury was not thereby deprived of any guide to gauge the visibility. For Adams was allowed to testify fully to what the physical conditions at the crossing were, and both sides introduced photographs which indicated the lay of the land. Under the circumstances, the jury could have made an intelligent estimate themselves, so that it cannot be argued that the exclusion of Adams' observations left them to grope in the dark.

4. The only remaining issue is whether the judge was correct in charging that the burden of proving contributory negligence was on the defendant. In so charging, he was following Federal Rule of Civil Procedure 8(c). It is argued that we should disregard that Rule because burden of proof is a matter of "substance," and hence cannot be altered by court rule. There is no necessity here of considering the argument. For if we were to reject the Rule, we would then turn to the decisions of the New York courts, including those relating to conflict of laws. Erie R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188, 114 A.L.R. 1487; Klaxon Co. v. Stentor Electric Mfg. Co., 313 U.S. 487, 61 S.Ct. 1020, 85 L.Ed. 1477. While, with respect to intra-mural transactions, New York courts hold that the burden of proof is on the plaintiff, in a case such as this, they would apply, as a matter of conflict of laws, the Massachusetts law. Fitzpatrick v. International Ry. Co., 252 N.Y. 127, 169 N.E. 112, 68 A.L.R. 801. And it happens that the Massachusetts rule coincides with Rule 8(c). See General Laws of Massachusetts (Ter.Ed.) § 85, c. 231.

The judgment is affirmed.