The deceased was a brakeman employed in interstate commerce, at the time of his death engaged in piloting two • gondola -ears in the defendant’s yard at Morrisville, Pennsylvania, on a misty afternoon, at five o’clock on November third. The operation was of the familiar kind; cars were being assorted by gravity upon tracks in the yard, of which there were a great many. A locomotive pushed a train up an incline, detached toe desired number in a string, and let them roll down the other side of the “hump,” to be switched to their destination. Each of the-strings was ridden by one or more brakemen, whose duty it was to control their speed and prevent collisions. It was conceded that at some point after the deceased’s ears had been shunted upon .the track where it was to go, he fell from them and was run over by nine-ears which followed his own, and which were ridden by three brakemen. He lay between, the tracks, and whether his own cars had already also passed over him was not clear. The question is whether he was thrown off through the negligent collision of the following nine, or whether in some unknown way he fell off. In the latter event eoneededly the case was not proved.
The plaintiff’s only witness to the event, one Bainbridge, then employed by the road, stood close to the yardmasteris office, near the “hump.” He professed to have paid-little attention to what went on, but he did see the deceased riding at the rear of his cars, whose speed when they passed him he-
Tito slorv is attacked as too uncertain to support » verdict; and it is quíte tme that the witness’s repeated glances are a suspicious circumstance; there seems to have been no reason for his looking first when the nine ears wc-ro entering the switch, and again after a crash which did not of itself seem to him unusual. Moreover his testimony is not too clear in itself; we have stated it in its more favorable interpretation. However, all this was a matter for the jury, and assuming they thought him truthful, they might infer that a collision took place. It does not appear to us impossible, or indeed improbable, that one in his position could tell whether the two strings were together. The intervals between cars in a train are uniform; they may he detected by tho straight sides. Certainly a gap of four or five ear lengths, when the nine cars came to rest, would have been easily observable; and this was tho story of the defendant. What Bainbridge saw, coupled with what ho had heard, if uncontradicted, would be enough to support a finding that the nine ears had collided with the deceased’s and thrown him off. There is no inherent impossibility in the story.
The defendant’s version, hacked by six or seven of its employees, was that tho deceased had fallen off while clambering over his ears from tho rear end. The two gondolas, with nobody aboard, made a light contact with a string still moving ahead of them, whose crew, looking backward, sa-w him lying on the traek. Tho brakeman on the front of the approaching nine ears saw him between the rails, though not in time to cheek them until they had passed over, and clear of, him. There was however no contact between the gondolas and the nine, which, as wo have just said, stopped some four or five car lengths away. Nobody professed to have seen him fall, but if the defendant’s witnesses were believed, ho must have done so; at least, however he got upon the track, it was not through a collision, and the defendant was not at fault. On this showing the judge thought the evidence of any collision too speculative to support a verdict and took the case from the jury. We have to decide only whether this ruling was right.
Left to ourselves with the bare record we should say that the judge’s conclusion was supported by a very strong balance of probability. Not only was the testimony of Bainbridge somewhat suspicious in itself, hut its contradiction was so manifold as to leave little doubt. Nevertheless, the question was really as to whom to believe, and when that is so, we do not understand that the jury may be displaced. There are of course cases where the story of one side is. impossible because of the physical situation, or because it is incredible by common experience. Southern Ry. Co. v. Walters, 284 U. S. 190, 52 S. Ct. 58, 76 L. Ed. 239, was such a ease. Moreover, in such cases the contradicting circumstances may depend upon tho testimony of other witnesses, and be conclusive only if this he assumed to be true. A court implicitly so assumes when it directs a verdict because of incredibility of the testimony, not inherent. Perhaps a direction may he proper, when the witnesses are impeached in character, or discredited upon cross-examination, if contradicted by others whose apparent veraeity stands. And indeed there may even be eases where numbers alone are enough. A bench of bishops should perhaps prevail over a casual interloper, though the other side does not succeed in discrediting him. It is impossible to set any general rule, for the combinations are infinite.
The most that has been said — probably all that can be — is that there comes a point where the evidence no longer justifies any verdict but one. Pleasants v. Fant, 22 Wall. 116, 121, 22 L. Ed. 780; Delk v. St. L. &
To a majority of us it appears that the direction in the case at bar cannot stand without intruding upon the exclusive competence of the jury; without taking from them their power to decide who of those appearing before them are entitled to credence. No doubt if the formula is to be strictly applied that the court should intervene when reasonable men cannot differ, such eases may fall within it. If it be an exception, still it seems to us required by the traditional division of function, that the jury alone may choose between a single witness, whose story is neither inherently improbable, nor contradicted by evidence which cannot reasonably be disregarded, and a large number who contradict him, at least when they are in the defendant’s employ, and in part concerned in the occurrence.
Judgment reversed.