At a little past 7 o’clock in the evening of January 25, 1888, the schooner Nellie V. Rokes, loaded with a cargo of guano, while-sailing up the Swash channel, came in collision with the steam-ship Allianca, outward bound, at a point probably not much over half a mile-below the Romer beacon. The wind was moderate from the south-east. The schooner, until shortly before the collision, was sailing N. W. by N., wing and wing, her main-boom being to port, and the fore-boom and spanker to starboard. When the steamer entered the Swash channel from H to 2 miles above, a snow squall prevented seeing any distance in that vicinity, though it was clear below. The bell buoy at the entrance could not be seen. After going under a slow bell, as the weather became a little clearer, the steamer proceeded at half speed, equal to-about seven knots, steering S. E. § S. The witnesses for the schooner testify that they saw the steamer’s red light, either ahead, or a little on their port bow, from two to three miles distant, next both colored lights,, and then her green light only, all on their port bow; that when the; steamerwas somewhat near, variously estimated at from 200 to 400 yards,, a lantern was shown and swung over the port side of the schooner; that *477this was shortly after repeated; and that then the schooner’s wheel was put hard a-port, under which the schooner swung from 5 to 7 points, so as to head from N. N. E. to N. E., when she was struck hy the steamer’s stern on her port quarter, about 10 feet forward of the taffrail. The steamer’s three witnesses on the bridge testify that a white light was first seen for a few seconds, supposed to be that of a tug coming up, and about a point and a half on the steamer’s starboard bow, estimated at about half a mile distant, upon which the steamer’s helm was put to starboard to give a wider berth; that the steamer swung about a point and a half under her starboard helm; that in the moan time, the white light, which was seen but a short time, disappeared, and the schooner’s green light was then first seen, and very soon after both her colored lights, all on the steamer’s starboard bow; whereupon the steamer’s engines were stopped and backed full speed, and so continued until collision; that the steamer on reversing swung back again to her original heading of S. E. J S.; that the steamer was making stern-way at the collision; and that the schooner at no time after the steamer got headed down the Swash channel was on the steamer’s port bow until she crossed the steamer’s course under a port helm shortly before collision.

There is some conflict in the evidence as to the lights seen, and in what order, and over which bow of the respective vessels. The discrepancies* are not of special importance, except as bearing upon the credibiii ty of the witnesses, and the reliance to be placed upon their recollection and their testimony as it stands. The circumstances are not sufficient to make a case of unavoidable accident, and the steamer, therefore, is to be held answerable for not keeping out of the way of the schooner, provided the latter observed the rules of navigation. The defense of the steamer, in effect, is that the collision was caused solely by the schooner’s improperly and unlawfully changing her course by her port helm. The libelant admits the change, but contends that it was made in extremis, when collision was apparently unavoidable through the steamer’s fault. The case has been carefully prepared and argued, with illustrative charts upon a large scale, which have made comparatively easy a careful examination of the alleged navigation of each vessel, as well, also, as a test of the probable accuracy of the different witnesses.

1. I am satisfied that the immediate cause of the collision was the schooner’s porting. She must have changed at least six points to starboard. In the sworn amendments, put in upon exceptions to the libel, requiring further specifications in this respect, a change from about N. W. by N. to about N. E. is stated, which is a change of about seven points. The mate testifies to a change of from six to seven points; the other testimony makes at least six, except that of the master of the schooner, who, on the trial, for the first time reduced the estimate to five. The deliberate statements made under exceptions to the answer cannot be departed from except on very satisfactory evidence to the contrary, which does not exist in this case. The master further testifies that he thinks a change of five points could be made in sailing a length and a half, or about 200 feet under a port wheel, as the schooner was *478then sailing; while three times that distance would be required for an equal change under a starboard wheel. The mate estimated that the schooner would run about 800 feet in changing five points. While there would be some difference, doubtless, in the schooner’s changes, under a port wheel and a starboard wheel when sailing wing and wing, with the wind one point on the side of the spanker, as in this case, I cannot credit the master’s estimate that she would change five points in going 200 feet, only a quarter of the distance estimated by the mate. Taking into account the mate’s testimony, and all the other circumstances, and the absence of any experiments or facts to verify the master’s estimate, I do not think it possible that this schooner, 130 feet long, loaded with guano, could change six points to starboard in going less than 500 or 600 feet; and, if made in even 500 feet, that would give her a direct offing to starboard of at least 250 feet. This change by the schooner, moreover, caused the steamer to back, and to stop her own way more to the eastward, which, in passing over the considerable interval that then separated the vessels, would have resulted in a difference of relative position abreast at the time of passing, had the schooner not changed her course, of at least 350 feet. In other words, they would have passed starboard to starboard with more than 200 feet of clear water between them had the schooner kept her course. The steamer’s testimony is even less favorable to the schooner. Her testimony and the log show that she was backing altogether about five minutes, of which about two minutes, at least, was before collision. She did not begin to back until the schooner showed both colored lights, after first showing the green light, which indicates that the schooner was then under a port helm; and the schooner, moving at the rate of four knots during these two minutes must have traveled at least 800 feet; and this agrees with the testimony of the mate of the schooner. A change of six points in traveling this distance would have given the schooner an offing to starboard of 425 feet; and upon this estimate, which I think really accords with the weight of evidence, there would have been more than 350 feet of clear water between them in passing had the schooner kept her course. It is not material to the above conclusion just how far to the westward of the Romer light the respective courses of the vessels may have been. All that is material is the fact that their courses crossed only about half a point; and there is nothing to discredit the testimony of either vessel as to her own course, showing that such was the fact. The course of the Allianca, in passing through a channel like the Swash, more than two miles long and only about a quarter of á mile in breadth, for vessels of her draft, was necessarily restricted within very narrow limits of variation.

2. If the above conclusion is correct, that the steamer would have passed from 200 to 350 feet to starboard of the schooner, or even considerably less, had the latter kept her course, it is impossible to charge the steamer with the schooner’s damages. The Kanawha, 28 Fed. Rep. 329. The latter’s change of course upon such facts cannot be justified under the plea of a change in extremis. In the case of The Elizabeth Jones, 112 *479U. S. 514, 5 Sup. Ct. Rep. 468, a similar excuse was made for the porting of the Jones, but disallowed. In the supreme court, Mr. Justice Blatchi’oiid in reference to this point says:

“To be an excusable mistake in extremis, a pardonable maneuver, though contributing to or inducing a collision, when the maneuver would have been faulty if not excusable, it must be one produced by fault or mismanagement in the other vessel. ”

In the present case, the steamer properly starboarded her helm when the schooner’s white light was seen upon the steamer’s starboard bow. The vessels must have been then at least one-third of a mile apart, if not more. That maneuver of the steamer was sufficient to clear the schooner by an ample margin for safety. The distance of the steamer was such that the schooner had no right to assume that the steamer would not clear her by proper maneuvers, and there is no reason for holding that the schooner’s lights were not seen as soon as they were visible, and a sufficient maneuver promptly made. There was no fault or mismanagement, therefore, in the steamer'up to that moment which could have induced the schooner’s change of course; so that upon the rulo laid down in The Elizabeth Jones, the schooner’s change of course could not be excused.

In another point of view, also, the schooner’s porting cannot be justified, namely, that, though their courses were not crossing by more than half a point, the schooner’s change was manifestly in the direction to produce collision, instead of away from it. The green light of the steamer was then visible, as all the testimony on the part of the schooner admits; and a projection of the courses of the vessels upon the evidence shows that it was impossible that the collision could have happened in the manner the weight of testimony shows it did happen, had not the schooner at the time she ported been on the steamer’s starboard bow, and at least a quarter or a third of a mile distant. 1 have no doubt, that at the lime when the schooner ported the steamer’s green light only was visible, and that the schooner was at least a couple of hundred feet to the westward of the line of the steamer’s course. Mere apprehension of danger is no justification for such a change, where, as here, the facts were not such as to justify the schooner in supposing that the steamer either would not or could not avoid collision by her own maneuvers. The Free State, 91 U. S. 204; The U. S. Grant, 6 Ben. 467; The Britannia, 34 Fed. Rep. 553; The Scotia, 5 Blatchf. 227. The change was too great, and made too early to be excused as a change in extremis. The City of New York, 15 Fed. Rep. 625, 35 Fed. Rep. 610.

3. I cannot find any fault established against the sicamor. The steamer was going at half speed only, or about seven knots. The schooner’s white light must have been seen from onc-third to half a mile distant. Assuming that the weather was such that it could not have been seen sooner, I cannot find that half speed was not “moderate speed,” when the light could be seen that distance. There is no proof, and I cannot assume, that the steamer could not have avoided the schooner by reversing as soon as her light was seen, had backing been the proper maneuver. But *480backing was not, then, the necessary or proper maneuver. The steamer rightly adopted a different maneuver; namely, starboarding, which would have given the schooner a wide berth had the latter not wrongfully ported and crossed the steamer’s course. As it was, the steamer must have been nearly, if not quite, stopped at the time of collision; otherwise the steamer, striking at nearly a right angle, would have cut off the schooner’s stern, whereas, according to the mate, she was only “shook a little; not a great deal.” Judged by the rule laid down in The City of New York, 35 Fed. Rep. 609, namely, that “such speed only is moderate as will permit the steamer seasonably and effectually to avoid the collision by slackening speed or by stopping and reversing, within the distance at which an approaching vessel can be seen,” the evidence does not show that the Allianca’s speed was in excess of that.

Nor can I find the steamer in fault for not sooner seeing the schooner’s lights. The steamer’s lights were probably seen from the schooner earlier than the schooner’s lights were visible, by reason, perhaps, of the greater brilliancy of the steamer’s lights, and of the thicker weather about her, that interfered more with her officers’ vision. It seems to me improbable that the steamer’s white or red light was seen at a distance of two or three miles. The red light, first supposed to be the steamer’s, may possibly have been the new red light of the Romer. The testimony of Oumming, the mate, as to what he did after the first light was seen, shows that the time from then to the collision was short, and that the steamer’s green light was seen not long before he showed the lantern which the steamer saw. The testimony of the schooner’s witnesses as to the lights seen, their bearing, and the intervals, is inconsistent, and cannot be adopted.

The position of the vessels, in my judgment, may be approximately stated as follows: When about 700 yards apart, and their courses crossing by an angle of half a point, the steamer was at or near the point of convergence, so that her white light, when first seen from the schooner, would be seen ahead, as the master of the schooner states. The mate was on the port side, which in some degree accounts for his error. Possibly the overlapping of the steamer’s ranges of colored lights was such as to permit both colored lights to be then seen for a short time. Soon after that the schooner’s white light from the lantern was seen, from half a point to a point on the steamer’s starboard bow. The steamer’s red light was then shut in, and her green light only was from that time visible, a little on the schooner’s starboard bow, but so little that it might be called ahead; or it might have been, possibly, actually ahead, or even a trifle on the schooner’s port bow for a moment during any yawing of the.schooner to starboard that might have been permitted; but neither of the steamer’s colored lights otherwise came on the schooner’s port bow, nor was the former’s red light again seen until she crossed the steamer’s course by her own porting. Libel dismissed, with costs.