Jackson, Circuit Judge.
The collision which gave rise to and forms the subject of inquiry in this suit took place in the natural or middle channel of Lake George, at or about 7:30 o’clock a. m., on August 13, 1890, between the propeller Devereaux and the schooner Mary B. Mitchell, which was the first of two schooners in tow of the steam barge Alex*404ander Folsom, and resulted in damage to both of the colliding vessels The owners of the Devereaux libeled both the Folsom and the Mitchell, and, after stating their version of the facts preceding and attending the collision, alleged as faults against said vessels that they were not in charge of a proper complement of competent officers and men; that those in charge were incompetent, and inattentive to their navigation; that they maintained no proper lookout; that they did not check down; that they improperly carried sail in proceeding through the said channel; that the Folsom met and passed the Devereaux at too great speed, causing her to sheer, under the influence of suction; that the Mitchell came up to windward while the Devereaux was on such sheer; that said schooner failed to keep her own proper side of the channel; and that she failed to observe the sheer of the Devereaux, and steer out to the westward. The respondent the Mitchell Transportation Company, as the sole owner of said steam barge and schooner, in its answer, after admitting the collision, and giving its version of the facts relating thereto and causing the same, denied each and all of said alleged faults. The district court found the Folsom and Mitchell in fault for carrying sail and proceeding at too great speed, which caused the Devereaux, in meeting and passing the Folsom, to suddenly sheer, under the influence of what is called “suction;” that the Mitchell suddenly changed her course by sheering to starboard; that the tendency of the Devereaux to sheer, under the circumstances, “being well known to skillful seamen, the master of the Mitchell should have considered it possible, if not probable, on the part of the Devereaux, and have so far guarded against it as to have his own vessel in perfect control, and his wheel on the starboard, so as to have headed his vessel to port, and have been able to put her in that course promptly when the emergency made it necessary.” A decree was accordingly rendered in libelant’s favor for the damage sustained by the Devereaux, which was fixed at the sum of $15,143.88, with interest. From this judgment respondent has appealed, and has assigned as error therein several matters which need not be especially noticed; the principal grounds relied on for reversal being that the Folsom and Mitchell were not guilty of any fault or negligence which caused, or contributed to cause, the collision, and that the district court erred in condemning them.
The appeal involves mainly questions of fact, to be determined, under the settled rules of evidence, from the testimony, which, as set forth in the record, presents the full average of conflict usually found in collision cases, especially in respect to matters of opinion and theory. To review the evidence in detail, or attempt to reconcile the testimony on many points, would be a useless labor. We have given the respective theories of the parties, and the proof on both sides, full and careful consideration and examination, and deem it sufficient to state briefly the material facts of the case, which, in our opinion, are established by the testimony, and the conclusions properly deducible therefrom, as read in the light of the surrounding circumstances.
*405The middle channel of Lake George is about two and five eighths miles in length, with its general course or direction nearly north and south. It has a current of about one mile an hour. The view from one end to the other of the channel is open and unobstructed. At the southern and northern entrances of the channels there are red can or nun buoys, numbered 26 and 46, respectively, on the government’s official list of buoys and stakes. The line of the navigable channel is marked by red spar buoys on the east bank, and by black spar buoys on the west bank, thereof. The first black spar or stake buoy on the west bank, numbered 27, is about 660 feet above, and in a northwesterly direction from, the southern or lower red can buoy. The other marginal red and black spar buoys are about 1,290 feet apart, and nearly opposite each other. The northern portion of the channel, called the “cut,” extending from the northern entrance, at or near the red can or nun buoy No. 46, down to the second elbow between red spar buoy No. 34 and black buoy No. 35, is artificial, having been formed or constructed by dredging, and has a navigable width of 300 feet. From said elbow, or the end of said cut, southward, the channel is natural, with a navigable width of about 180 feet. The waters of the lake on either side of the channel, both artificial and natural, vary in depth from 5 to 9 feet, presenting a broad sheet or surface of water, with the navigable channel extending through the same defined by the lines of said red and black spar buoys on either side, and which are generally located where the depth of water ranges from 9 to 12 feet. The navigable water of the natural channel accordingly varies in depth from 9 to 12 feet on the outer or buoy line to about 25 feet in the center of the channel at or near the point at which the collision took place, thereby indicating, as stated by one or more of the witnesses, that the channel bank slants off towards the center of the channel where the greatest average depth of water is found. The collision occurred between the third and fourth black spar buoys, counting from the red can buoy at the southern entrance of the channel, or between the second and third black spar buoys, counting from the second elbow south of the northern entrance, and was about three quarters of a mile northward from said red nun buoy at the lower entrance of the channel.
It is alleged in the libel, and the proof clearly establishes the fact, that, owing to its narrow width and moderate depth of water, this natural channel presents a dangerous place for vessels proceeding in opposite directions to pass each other, and required the exercise of great care and vigilance to avoid the danger of collision. This dangerous character of the channel, and the liability of meeting vessels to collide therein, was well known to the officers of both the Devereaux and the Folsom. The Devereaux is 270 feet in length and 37 feet beam. She was proceeding southward through the channel, carrying a cargo of 2,060 tons of iron ore. Her gross tonnage was about 1,618 tons. Her draft was about 15 feet (14 feet, 11 inches) fore and aft. Her speed down the channel was about four miles per hour, and was maintained until a few seconds before the actual collision. The Folsom is 185 feet long and *40635 feet beam. She was proceeding northward, and had no cargo. Her draft was 3 feet forward and 11 feet aft, and her gross tonnage was about 940 tons. The Mary B. Mitchell, the first of the schooners in tow, is 212 feet long and 40 feet beam. The Nelson, the second schooner in tow, is 199 feet long and 33 feet beam. These schooners, like the Folsom, had no cargoes, and were both light. The towline between the Folsom and Mitchell was about 500 feet in length; that between the Mitchell and the Nelson was about 400 feet long,—making the entire length of the tow, including the Folsom, about 1,500 feet. The Mitchell’s tow line was on her port side, and ran through the chock, which was about five feet from her stem. The Nelson’s towline was on the starboard side of her bow. The Folsom carried her foresail and mainsail. The Mitchell and Nelson each carried the same, with the addition of one jib sail.
When the Folsom came into the channel at the southern entrance, and after passing the lower red can or nun buoys, and while approaching the first black spar buoy on the west bank, she blew one blast of her whistle for the Devereaux, indicating that she desired to direct her course to starboard, so as to proceed along the east side of the channel, that being her proper course and side, under rule 18, Rev. St. §4233, as the two steamers were meeting end on, or nearly end on, so as to involve risk of collision if each maintained her course. The Devereaux did not accept this signal, but immediately replied with two blasts of her whistle, indicating that she would starboard her wheel, and take the eastern or port side of the channel. Upon the receipt of this signal the Folsom at once starboarded her wheel, and promptly answered with two blasts of her whistle, thus agreeing of the proposition of the Devereaux that the vessels should pass each other on their starboard sides, by each going to port, and the Folsom, under her starboarded wheel, immediately bore gradually over to the west side of the channel, and proceeded up that side, within about 25 feet of the channel bank, her tow following in direct line astern. When these two blast signals were exchanged, and the respective courses of the two vessels were thereby arranged and agreed upon, the Folsom was about opposite the first black spar buoy on the west bank, and the Devereaux was in the “cut,” about three quarters of a mile from the northern entrance thereof, or about opposite the fourth red spar buoy from the northern entrance of the cut, which is about half a mile above the second elbow between red spar buoy No. 34 and black spar buoy No. 35, at or near which latter point the dredged channel or cut terminates, and thl natural channel of the middle passage begins. After the signals were exchanged, the Devereaux proceeded down the center of the channel until within three or four times her length of the Folsom, when she starboarded her wheel, and gradually hauled off towards the east bank of the natural channel, which she approached within about 25 feet. When the Devereaux thus changed her course from the center towards the east side of the channel, the Folsom had previously drawn herself and tow as near the *407west side of the channel as she well could do, or, as one of libelant’s witnesses (Cleveland, mate of the Devereaux, in a position to see) expresses it, “she (the Folsom) couldn’t starboard much more,” and was occupying something less than one third of the navigable channel, with her tow following in her wake directly astern. The distance between the Devereaux and Folsom as they approached and were in the act of passing each other is variously estimated by the witnesses at from 37 to 70 feet. Considering the breadth of beam of each vessel, the respective distance of each from the east and west banks of the channel, and the navigable width of the channel, our conclusion is that they were about 60 feet apart while in the act of passing each other. This conclusion is-supported by the estimates of several witnesses who were in as good position to form as correct an opinion on the subject as others who estimate the distance at from 37 to 40 feet, and is corroborated by other established facts already stated, viz., that tbe Folsom was distant from the west bank 25 feet, which, with her 33 feet width of beam, made her occupy about 58 feet of the navigable channel; that the Devereaux was distant from the east bank about 25 feet, which, with her 37 feet width of beam, made her occupy about 62 feet of the channel, and that the two vessels thus situated covered about 120 feet of the 180-foot channel, leaving a distance between them of about 60 feet. When in the act of passing the Folsom at said distance, the Devereaux took a sudden sheer of about two points to starboard, carrying her towards the stern of the Folsom in a direction across the channel, and across the line of the tow between the Folsom and Mitchell. The master of the Devereaux, upon discovering 'that she was sheering, ordered her wheel hard astarhoard, and gave the engineer four bells, to work her ahead, or give her what is called, in nautical phraseology, a quick or rapid “kick ahead,” in order to straighten her up in the channel again. The master of the Devereaux states that the start of a vessel on a sheer is most quickly detected or discovered by looking astern, and that he first discovered the Devereaux sheering when he happened to be looking astern. But his wheelman testifies that he saw her starting on the sheer, and had commenced turning his wheel before he received the master’s order to starboard. When the four-bell signal to kick her ahead was given, the Devereaux’s engine was making about 20 turns, sufficient to carry her, aside from the current, about three- miles per hour. In obeying the signal the engineer increased the speed of the engine to 65 turns, under the influence of which, and her starboarded wheel, the vessel increased her speed, and regained about one point of her lost course resulting from the sheer, when, seeing that a collision with the Mitchell was imminent, the master of the Devereaux blew a danger signal, and called out to the master of the Mitchell “to let go her towline,”.and thereupon gave his engineer two bells to back, followed quickly by two mpre bells to back strong. These latter orders to his engineer were obeyed; but before the Devereaux’s headway was lost or materially diminished, and before she had regained her lost course, she struck the *408Mitchell nearly end on, some five or six feet from the latter’s stem on her port how, about the center of the chock, cutting the towline in the chock, and causing the stern of the Mitchell to be driven 50 or 60 feet over the west bank of the channel, a little above the third black spar buoy from the lower entrance, and carrying her bow to starboard nearly across the channel. The Nelson following in the wake of the Mitchell, and her master seeing that a collision was inevitable, ported her wheel, came up to the starboard, and struck the Mitchell with her stem a glancing blow aft of her forerigging, when she also swung over on the left bank of the channel, about 50 feet below said third black stake or spar buoy. The Devereaux, after the collision, drew up along the east bank with her stem against the port bow of the Mitchell.
On the morning of the collision the wind was S. E. or S. S. E., and struck the Folsom and her tow on their quarter. The estimates made by the witnesses as to its velocity vary greatly, ranging from 5 to 13 miles per hour. There is equal conflict in the opinions of the witnesses as to its effect on the speed of the Folsom and the two schooners in tow. The wind was not sufficient to make white caps, Tout produced only some ripples on the water, which tends strongly to establish the fact, as testified to by many of the witnesses, that the velocity of the wind was moderate, being what is designated as only a fresh breeze, of sufficient force to drive vessels with sails properly trimmed up against the current of said channel from three to five miles an hour. The testimony of witnesses who were in the best position to observe and to know the fact establishes to our entire satisfaction that the limited sails carried forward of the mainmast by these schooners were not drawing so as to aid in accelerating either their speed or that of the Folsom after entering the channel, and in no way obstructed or interfered with their proper management and ready handling. It is shown by the clear weight of testimony that the limited sails carried were not only not drawing to any appreciable extent, but that the sheets were hanging slack, with the booms held at the port rail by boom tackle. Considering the force and direction of the wind, the character and location of the sails carried, and the influence of the towlines in keeping the schooners directly astern of the Folsom, whose speed was controlled by her engine, rather than by her sails, we are of the opinion that there was little or no tendency on the part of the Mitchell to sheer to starboard or luff to windward, so as to carry her across the course of the Devereaux, as claimed by libelants and stated by some of the witnesses. The testimony fairly predominates against the theory that said schooner actually luffed or came up to windward as the result of carrying such sail as she had about the time or just after the Devereaux took her sheer, and thereby caused the collision. This is a manifest afterthought with the officers of the Devereaux, who, with full knowledge of said alleged tendency to luff, and of all the facts now relied on to establish such luffing, declined to pass on the port side, and voluntarily selected the windward side of the tow. Under such circumstances, it requires something more than theory or speculative opin*409ion to satisfy the court that a questionable and improbable tendency actually operated to produce luffing or such change of course to starboard as rendered the Mitchell responsible for the collision, especially when it is shown by the testimony of witnesses in position to see and know the fact that- she did not change her course. The master of the Northern Queen, occupying an excellent point of observation, and watching the two vessels about to collide, could not say that the Mitchell changed her course. This is substantially confirmed by the mate of the Northern Queen, and by the positive swearing of witnesses on the Folsom and her tow. When the master of the Mitchell saw the Devereaux sheering towards the stern of the Folsom, and in a direction across the channel and the line of her tow, he starboarded his wheel, near which he was standing, so as to bear his vessel off further to port, and then ran forward to the forecastle to watch the movements of the Devereaux, and, upon reaching there, he understood, as others on the Folsom understood, some one on the Devereaux to direct him to port his wheel, and he signaled to his wheelman to do so; but before the wheel made more than two or three turns back from the starboard, (which would hardly have brought it to midship, as it required eight turns to put it from one side to the other,) he gave another order to put the wheel hard astarhoard, and her wheel was hard astarhoard when she was struck. The temporary and almost immediately countermanded order to port her wheel was not executed, and did not result in any change of her course to starboard. It is not material to determine whether any one on the Devereaux did call to the master of the Mitchell to port his wheel. It is admitted that the master of the Devereaux did call to him to let go her towline, which, under the circumstances, might fairly be understood to mean or indicate that the Devereaux would attempt or be forced to cross said line, and, in the view of her taking that course, the master of the Mitchell should properly have signaled his wheelman to port his wheel, so as to open a wider way for the Devereaux to cross her bow. But seeing the Devereaux straightening up instead of crossing, the order to port was immediately changed to hard astarhoard before the schooner had made any change in her course. If the wheel of the Mitchell had been fully ported, so as actually to have changed her course to starboard, it would not, under the circumstances, have been a fault for which she should be condemned, because it would have been a movement or maneuver made in extremis, for which she should not, under well-settled rules, he held responsible. But the testimony does not affirmatively establish any such change of course on the part of the Mitchell, or sustain the charge of the libel that “she did not steer away to the westward, but came up rather towards the course of the Devereaux, thereby presenting her port bow ahead of the Devereaux.”
It is established, by a clear preponderance of the proof, that the Devereaux, in the course of her sheer, passed or came between the Folsom and the Mitchell. The second mate of the Hackett, which was astern of the tow, states that he could see the Devereaux in between them. *410This fact is testified to by several other witnesses, and is not directly contradicted. This change of position on the part of the Devereaux might present to a distant observer of the occurrence the appearance of some change in the course of the Mitchell. The relative positions of the vessels towards each other were changed by the Devereaux’s sheer, and as she straightened up between the Folsom and the Mitchell, heading towards the port bow of the latter, nearly end on, would naturally present to persons at a distance the appearance of some change in the .Mitchell’s course, without any such change having actually taken place. The Mitchell was a good steering vessel, was well appointed and manned, and was violating no law or rule of navigation in carrying the sail she did, so as to impose upon her the burden of showing that her sail did not contribute to cause the collision. The positive testimony of the witnesses that she did not change her course is confirmed and corroborated by the position of the vessel in striking, the place at which the Mitchell was struck, and the positions of the vessels when they drew up after the. collision. After a careful analysis of the testimony, and the facts and circumstances connected with the occurrence, we cannot concur in the conclusion reached by the district court that the Mitchell was in fault or responsible for the collision. The faults imputed to her that she came up to windward; that she failed to keep her proper side of the channel, and failed to observe the sheer of the Devereaux, and steer over to the westward, are not established by the preponderance of the testimony, and the probabilities of the case. It is said by the learned district judge “that the tendency to sheer from suction in that channel by vessels passing under the conditions of this case was so well known by skillful seamen that the master of the Mitchell should have considered it possible, if not probable, on the part of the Devereaux, and have so far guarded against it as to have had. his own vessel in perfect control, and his wheel on the starboard, so as to have headed his vessel to port, and have been able to put her in that course promptly, when the emergency made it necessary.”
If this proposition is correct, should not the master of the Devereaux have considered a sheer on her part possible, if not probable, and have so far guarded against it as to have had his own vessel in perfect control, and her wheel on the starboard, so as to have headed his vessel to port? Instead of doing this, he was proceeding down the channel with his wheel steadied about midships, and did not order it to starboard till after discovering the sheer. It would hardly be a fair or consistent rule to put upon the Mitchell the duty of anticipating and guarding against the Devereaux sheering, and at the same time exonerate the' Devereaux from the obligation of taking precautions to prevent or counteract the alleged well-known tendency to sheer.
It remains to be considered whether the Folsom met and passed the Devereaux at too great speed, or at such undue speed as to divert her from her course, and cause her to sheer to starboard. This is the chief fault relied on to condemn the Folsom, and involves the question of her *411speed, and of its tendency or effect in producing what is called “suction” of sufficient force to sheer the Devereaux from her course. The testimony as to the speed of the Folsom just preceding and at the time of passing the Devereaux consists of the opinions and estimates of witnesses, most of whom were at a considerable distance in front or astern of her. These estimates are in their very nature speculative, uncertain, and not very reliable. They have not the force and character of positive testimony, and are not entitled to as much weight and consideration as the estimates and opinions of those who were upon the Folsom, and in a better position to form a correct judgment on the subject, especially if the latter are corroborated by other established facts and circumstances. The master, mate, wheelman, and engineer of the Folsom place her speed after checking down and entering the channel at from four to four and one half miles per hour. The master and mate of the Mitchell and Nelson give substantially the same estimate. It is established by the most positive testimony of the Folsom’s officers—master, mate, wheelman, first and second engineer, and fireman —that, immediately after the first signal of one blast, the engine of the Folsom was checked down, and half its steam shut off; the natural tendency and probable'effect of which was to reduce her speed nearly one half, as stated by several witnesses. That persons at varying distances in front and rear of the Folsom did not hear the three whistles to check, or did not notice any checking on her part, is such negative evidence as will not warrant the court in reaching the conclusion that six or eight intelligent and. unimpeached witnesses, in a position to know the fact, have testified falsely in stating that she was checked down. The established rule is that the testimony of officers and witnesses as to what was actually done on board their own vessel is entitled to greater weight than that of witnesses on other boats, who judge or form opinions merely from observation. The Hope, 4 Fed. Rep. 89; The Wiman, 20 Fed. Rep. 248, 249; and The Alberta, 28 Fed. Rep. 807, etc. In the application of that rule to the testimony in this case it admits of no doubt that the Folsom did check her speed,-upon entering the channel, down to about four or four and one half miles an hour. This is corroborated and confirmed by other established facts and circumstances. The distance from the can buoy in Mud lake to the place of collision is about 144 miles. The Folsom and her tow weighed anchor and started from the former point between 4 and 4:30 a. m. on the morning of August 13, 1890. The collision occurred about 7:30 a. m., so that the Folsom and her tow were about three hours in traveling said distance of 14f miles. Her speed before reaching the entrance of the channel was greater than after entering the channel and checking down; but, considering the entire distance, and the time in which it was traveled, her average speed did not exceed five miles an hour from her place of starting that morning to the place of collision; and, having checked after entering the channel, it admits of little or no question that her speed against the current was under five miles an hour. This is further con*412firmed by other established facts. When she assented to the two whistles of the Devereaux, she was about three quarters of a mile from the place of collision. The Devereaux was at that time about opposite the fourth red spar buoy from the upper entrance of the cut, or about half a mile above the second elbow from that entrance. From that position to the place of collision the distance is something over three quarters of a mile, so that after exchanging and agreeing upon signals they were nearly equidistant from the place of collision. If 'there was any difference, the Devereaux had the greater distance to travel. It is thus established that the two vessels were moving at about the same relative speed through the water, or so nearly the same speed that the difference was practically immaterial in its effect. The testimony, read and considered in the light of all the attendant and surrounding circumstances, fails to establish any undue speed on the part of the Folsom. It is further established to our satisfaction, by the clear preponderance of the proof, that, when the master of the Folsom discovered the Devereaux start to sheer, he promptly stopped his engine and backed his boat, and did everything that could be done to avoid the collision.
Under these conditions, is the libelants’ theory that the sheer of the Devereaux and consequent collision was caused by suction from the Folsom’s passing át too great speed established? We are clearly of the opinion that it is not. When passing through the water, vessels, in proportion to their size and speed, produce or give rise to displacement waves, which run out, quartering astern from their course, and affect smaller vessels within their reach. The cases are numerous in which larger vessels have been condemned for injuries caused to smaller vessels from such displacement waves. It is also shown by the testimony in this case that'when vessels are passing each other in the same direction there is a tendency on the part of the smaller vessel to be drawn out of her course, and towards the track of the larger, as the latter passes.
In the case of Nestor v. The City of Cleveland, Mr. Justice Brown, then district judge, said that if vessels are going in the same direction, and passing near each other, it (suction) had a very powerful effect to deflect the weaker vessel from her course, and that the suction of two vessels meeting and passing each other was not very powerful, its operation being too short to make any particular effect upon the action of the two vessels, “unless one is much larger than the other.” The theory of suction in meeting and passing vessels is that the current which rushes in astern to fill the displacement of water caused by the larger or more rapidly moving vessel has a tendency to draw the other out of her course when her bow comes within its influence. When it is considered that such current has its direction in the line of the moving vessel, with its greatest force and strength directly astern, its lateral bearing as a drawing and diverting influence cannot, as suggested by Judge Brown, be very powerful. Whatever may be its force, it is clear, from the testimony and from reason, that the smaller vessel is most liable to be affected by it. A relatively greater speed on the part of the smaller ves*413sel may counteract such influence, and may even deflect to some extent the larger vessel, if her speed is sufficiently in excess. But no such fact is established in this case, and the opinion of witnesses, based upon hypothetical statements, not supported by the weight of proof, amounts to practically nothing. That the Folsom, 185 feet long, without cargo, with an average draft of 7 feet, should have drawn or diverted the Devereaux, 270 feet long, carrying a cargo, and with an average draft, fore and aft, of nearly 15 feet, or more than double that of the Folsom, is in itself highly improbable; so much so that it would require the clearest proof to establish the proposition. The displacement of the Devereaux was nearly four times as great as that of the Folsom. When her bow passed, or was in the act of passing, the stern of the Folsom, she was drawing about mur feet more of water than the Folsom’s stern was displacing. This four 1'eet of water was in no way affected by the Folsom’s displacement, and, while it encompassed the bow of the Devereaux, it is difficult to understand how the latter could have been diverted from her course by the Folsom, even if the latter had been going six or seven miles an hour. After the collision the Folsom went up the channel, and passed the Northern Queen within about 30 feet at a greater- rate of speed than she passed the Devereaux, with little or no effect upon her course, although the Northern Queen had less draft of water than the Devereaux, and was not working her engines. But when it is shown that the speed of the Folsom was not excessive, that it did not exceed that of the Devereaux, the theory of the latter having been caused to sheer by'the former’s displacement- of water is exploded.
It is shown by the testimony that when vessels approach too near the bank or bottom, or “smell the land,” as it is called in sailor parlance, they have a strong tendency to sheer towards deep water. Whether the Devereaux’s sheer was occasioned by this influence, we are not now called upon to determine. In view of the fact, disclosed in the record, that there is now pending at Detroit a suit in admiralty by the owner of the Mitchell against the Devereaux to recover the damage sustained by the former on account of said collision, we do not deem it proper to express any opinion as to whether the Devereaux was in fault in not stopping in the wide cut above the second elbow until the Folsom and tow came through the narrow natural channel, instead of meeting and attempting to pass in the latter; or as to whether her sheering, while approaching or passing the Folsom, was due to negligent management or inattention on the part of the Devereaux; or whether her master should have anticipated her sheering, and guarded against it; or as to whether her sheering was caused by too near an approach to the eastern bank of the channel; or whether it was an unavoidable accident. These are questions involved in the suit pending against the Devereaux. For the same reason, we have not undertaken to determine at what precise point the Devereaux commenced taking her sheer, whether before reaching the Folsom or when their bows or stems were abreast of each other, or when her bow had passed the Folsom’s stern; nor have we passed upon the *414question whether the Devereaux, in departing from the statutory rule of passing the Folsom on the port side, took the risk of her ability to pass safely on the starboard hand, as was held in The Titan, 49 Fed. Rep. 479, 480, 1 C. C. A. 324. What we decide in this case is that the libelants have failed to establish, by any fair and satisfactory preponderance of proof, as the burden was on them to do, that the Devereaux’s sheering, and the collision resulting therefrom, was caused by any fault of either the Folsom or Mitchell or both. We have reached this conclusion without considering the new testimony taken by the appellants since the appeal, as we entertain some doubt whether, after an appeal in admiralty to this court, new testimony can be taken, under- existing provisions of law.
The decree of the district court condemning the Folsom and Mitchell is erroneous, and is accordingly reversed, and the cause is remanded to said court, with direction to dismiss the libel at libelants’ costs.