This is a Jones Act, 46 U.S.C.A. § 688, case in which the plaintiff, Haywood, a seaman, failed to convince a jury that the defendant corporation, his employer and the operator of a vessel on which he was working, was responsible for his sustaining an injurious fall. The principal questions which Haywood raises on this appeal are whether the District Court committed reversible error in excluding certain evidence and whether there was any substantial evidence to justify the court's action in submitting the issue of contributory negligence to the jury.

The accident happened on a coal barge of conventional construction. Topside, the outer shell of the barge was a narrow, horizontal surface or decking called the gunwale. This gunwale was 16 inches wide and served as a place to stand and a walkway around the large open central cargo space or 'hopper'. All around the inner edge of the gunwale there projected upward a very low border and barrier of heavy iron construction. This coaming was described as a 'T-rail' similar to ordinary railroad track but not quite as large. It stood about 3 inches high, was spot welded to the gunwale and provided some protection against a misstep or slipping into the open central cargo space. Though in effect the rail was continuous, there was evidence that it was composed of sections of varying length.

Haywood testified that, while standing on the gunwale, he undertook to pull in a line thrown from another vessel. In so doing and without looking down he placed one foot back against the low rail behind him to brace himself. At that juncture he says a section of the rail gave way causing him to fall back into the hopper injuring himself.

Immediately thereafter Haywood left or was assisted from the scene to obtain medical attention and no one else seems to have inspected the point from which he fell. As a result, though the physical condition of the railing immediately after the accident must have been apparent at the time, there is no direct testimony in this record either to corroborate or contradict the plaintiff's statement that a section of the rail gave way under the pressure he exerted against it with his foot. However, in view of such testimony as there was of the nature, construction, and attachment of the railing, a reasonable mind might well have viewed the plaintiff's account standing alone as incredible. Indeed, during the plaintiff's case, and immediately after Haywood had stated his version of the accident, a juror asked about the rail, 'How could it break off?'

In this context, plaintiff's counsel tried to elicit from plaintiff himself and from another seaman, a witness for the defense who had certain barge inspection duties, whether and how barge rails of this type had come loose on other occasions. To each such effort there was vigorous objection, sustained by the court with observation from the bench that what may have happened in ninety or ninety-nine cases out of a hundred was irrelevant to the actual happening in the case on trial. And when the issue was raised again on motion for new trial, the court disposed of the matter in memorandum opinion as follows:

'Plaintiff's contention that the Court erred in refusing to admit evidence on the subject of how barge rails became loose or broken appears highly spurious and unsound. The tendency of rails to break, it would seem to this court, is a subject far afield and irrelevant to a determination of whether the rail in question actually did break. Assuming, however, that the Court was in error in refusing to admit such collateral testimony, the plaintiff, himself, testified that the rail in question did break and gave way, so that the exclusion of the sought-after generalized testimony of the tendency of rails to break, in no way could have been prejudicial to plaintiff's case.' (108 F. Supp. 664.)

In the circumstances of this case we think the court should have permitted the attempted line of questioning. The situation presented during this trial was unusual. Plaintiff had testified to an event which in physical aspect seemed sufficiently out of the ordinary to make the trier of fact skeptical of its occurrence. One juror expressed such skepticism. In these circumstances, we think it was relevant for plaintiff to show past experience with similar rails on similar barges at least for the limited purpose of showing, if he could, that his account was not inherently incredible. We think such evidence would be objectionable only if it tended to confuse and mislead the jury or to waste time unduly on collateral issues. At times evidence of this type is thus objectionable. But here the physical situation seems to have been simple enough and the structures in question sufficiently standardized so that there was no great likelihood of long and time-consuming excursions to show the details of other mishaps and the similarity or dissimilarity of conditions on various barges. And a court can always call a halt to such inquiries if they become unreasonably prolix. Moreover as concerns possible confusion, we think it could have made clear to the jury that the evidence was admitted solely for the purpose of avoiding any impression, here actually emphasized by a juror's question, that plaintiff's explanation of the accident was so improbable as to be unworthy of belief on its face.

We recognize that this entire area of proof in which allegedly analogous experience with similar physical occurrences is tendered as probative of something about a particular and wholly distinct occurrence has troubled the courts. Conflicting decisions and unclear opinions are to be found. But in the main we think the cases go as far as we go here in approving the use of such evidence where there is occasion to show and for the purpose of demonstrating that an apparently out of the ordinary physical occurrence is not as strange as it may seem but rather is well within the realm of possibility and credibility.1 Indeed, numbers of cases go further in admitting such proof of similar occurrences as a basis for a direct affirmative inference that the event in question occurred in the same way.2 We need not and do not go that far. It is enough for present purposes to say that evidence concerning breaking of similar rails on other similar barges was admissible to show that as a physical matter the instant claim of breakage was not so improbable as to seem incredible. As the case developed this issue was crucial. We think the plaintiff is entitled to a new trial in order that he may establish his side of it, if he can.

In addition, in the event of a new trial, we think the mere fact that plaintiff put his foot and his weight against the rail without examining or testing it, or even looking at it, is no evidence of contributory negligence. He was entitled to assume that the rail was strong and secure and would afford normal support. In the present record we find no other basis for submitting the question of contributory negligence to the jury. If no other basis appears on a new trial the issue should not go to the jury.

The judgment will be vacated and the cause remanded for further proceedings consistent with this opinion.