LACOMBE, Circuit Judge.
The facts in this case will best be presented by stating the findings of the district judge:
On December 7, 1890, the Lisbonense, a steamer 270 feet in length, 32 feet beam, and drawing 191-2 feet, was hound into the port of New York on a voyage from the Brazils. La Champagne, a steamer 503 feet long, 51 1-2 feet beam, and drawing 25 1-2 feet of water, was hound out from the same port. The latter, on account of her deep draught, was proceeding out around the Southwest Spit by way of the main ship channel and Gedney’s channel. - The former was coming in by way of the south channel and the swash, the axis of which crosses the main ship c hannel at an angle on the southwest side of 10 3-4 points. Each vessel was in charge of a pilot, and the master of each was on deck. The night was clear, but dark. The tide had been running ebb two hours, setting to the east or southeast with a velocity of between 2 and 3 knots. The wind was fresh from the northwest. La Champagne, in consequence of fog, had been detained over night in the lower bay, and resumed her voyage about 4 A. M. She passed the meridian of the Sandy Hook lights at 5:20 A. M. As she passed she burned her private signal torches, and soon after burned a blue light for the pilot boat outside to receive her pilot. Just prior to the burning of these signals, while she was on the regular main channel course, which is E. by N. 1-4 N., the red and white lights of the Lisbonense were seen about two points on the starboard bow. About the same time the green and white lights of La Champagne and her signal torches were observed by the Lisbonense, without knowing or noticing their precise colors or character, or understanding what line or what vessel they indicated. The Lisbonense was then on the usual course up the south channel about N. W. 3-8 N. The vessels at the time of sighting were about 2 miles apart, and respectively a mile and a mile and a half from the point of collision. The speed of La Champagne was about 12 knots over the ground; the 101-2 knots derived from her engines, running at 35 revolutions, being supplemented by wind and tide. The speed of the Lisbonense, allowing for retardation by wind and tide, was a little less than 71-2 knots. The captain of La Champagne, upon sighting the Lisbonense, observed her carefully with the alidade to see whether her bearing changed or not in reference to the necessity [296]*296of taking precautions against collision.. No change was apparent. When the phot of La Champagne sighted the Lisbonense, he ordered the engine to slow, (an order which was not transmitted, though he believed it to be carried out,) and “ported a little to — as we call it— ‘shave’ along the southerly side of the channel, and about that time they (the Lisbonense) Mowed the whistle.'’ This was immediately answered by one whistle- from La Champagne. The vessels were then about three fourths of a mile apart. The district judge finds that La Champagne “ported her helm so as to change her heading about one quarter of a point to starboard.” That she did make such change is plain upon the evidence, but it seems equally plain that such change and the subsequent steadying were complete before the exchange of whistles. The evidence of the captain of La Champagne would support the inference that this was after the signals, but is not so positive as that of the phot, who places it before, while the cross libel alleges that La Champagne did not then port, (i. e. after signaling,) and the answer of the latter does not aver that she did then port, but alleges that continuously “until the collision her course was that of the line of the (main channel.)” We are satisfied, however, as was the district judge, that the course of La Champagne was in the southerly part of that channel, and that in an ebb tide, with northwest wind, to have headed more to the starboard would have exposed her to serious risk of going ashore. She kept herself on the right-hand side of the channel as much as was prudently possible, but she put herself on that side before the signals were exchanged. The Lisbonense ported “so as to change her heading half a point or a point only, whereupon she steadied her helm upon the proper course to go up the swash channel.” The district judge finds that “thereby she resumed substantially the same course she was on before.”
The appellant assigns tMs finding as error, contending that she swung from half a point to a point to starboard, and was steadied on that new course. An examination of the evidence and of the chart satisfies us that the district judge was correct in finding that upon giving her signal she made no substantial change in her course. Disconnected references to the evidence of those on board might seem to support the appellant’s contention; but, taken as a whole, that evidence indicates that she adhered to her intention — manifest from the moment she entered the south channel — to hold her course by the south and swash channel lights, (the two channels are in the same course,) and the location of the collision confirms this conclusion: Had she steadied on the course to starboard, she would have crossed the axis of the main channel to the eastward of the place of collision, proceeding, not as her cross libel avers, “up towards the swash channe'l,” or, as her captain puts it, “upon the proper course to go up the swash channel,” but toward the Eomer shoals. The captain of La Champagne continued carefully to watch the Lisbonense, and, judging maneuvers necessary, stopped his engines, and a few seconds afterwards reversed at full speed, at the same time giving a signal of three whistles, which announced his reversal to the Lisbonense, and soon after repeated the same signal. The Lisbonense was all the [297]*297timo going at full speed and she continued at full speed until collision. The district judge finds tbe distance between the two vessels when the three whistles of La Ohampagne were blown to be from 1,500 to 2,000 feet. Just before collision the Lisbonense^ helm was put hard astarboard to ease the blow, which nevertheless^ cut into her counter about two feet. Tbe district judge found the Lisbonense in fault because, having given a signal of one whistle, she did not conform her course thereto, and field La Champagne free from fault. When the vessels sighted each other, La Champagne had the Lisbonense on her starboard hand.
We do not understand that either side contends that the locality was not within coast waters, (covered by chapter 354 of the Laws of 1885,) where the revised international rules are to be followed, in which respect the case resembles that of The Aurania and The Republic, 29 Fed. Rep. 98, and does not call for an exact determination of the line where the “coast waters” end and the “harbor” of New York begins. The sixteenth article of those rules provides that if two ships under steam are crossing, so as to involve risk of collision, the ship which has the other on her own starboard side shall keep out of the way of the other . Had she been in the open ocean, La Champagne might have done >his by coming to starboard, so as to allow the Lisbonense to go by,--a maneuvre, as her captain says, “so simple that there is no need of mentioning it,” or she might have slowed or stopped so as not to reach the course of the Lisbonense till the latter had passed on, out of danger. We are satisfied, however, that the existing conditions of wind and tide, the depth of water in the south channel, the length and draught of La Champagne, were special circumstances involving such immediate dangers of navigation as would (under article 23) warrant her master in avoiding the particular maneuvre which compliance with the sixteenth article would naturally suggest. We are not satisfied, however, that down to the time when whistles were exchanged the navigators of the Lisbonense either knew, or were fairly chargeable with knowledge of, the fact that the existence of such special circumstances was going to prevent La Champagne from c onforming her navigation to the requirements of article 16. We are not prepared to assent to the proposition contended for by the appellee, that, because deep-draught steamers find it necessary to take the main and Gedney channel course, it is to be assumed that all steamers whose lights indicate that they are coming down the main channel are of deep draught, and as such, under the special circumstance rule, entitled to right of way over incoming steamers which are on their own starboard hand. Such may be the custom of the port, but it is not proved to be so in this case. Moreover, whatever doubt as to the character of the outgoing vessel might arise from the circumstance that she was in the main channel was in this case set at rest as soon as she signaled her intention to take a course which deep-draught vessels could not follow. The authorities cited by appellee giving right of way to a vessel going with the tide do not apply, as these steamers were not navigating in the same channel.
If La Champagne were to navigate under article 16, the duty [298]*298of the Lisbonense was plain. Article 22 required her to hold her course. After sighting, but before signaling, the pilot of La Champagne, as above stated, ported a little, thus placing her as much on the right-hand side of the channel as was prudent, in view of the set of the tide, the direction of the wind, and the other special circumstances. Both vessels thus proceeded on their respective courses, — the navigators of La Champagne aware of the fact that special circumstances would prevent her maneuvering under the Lisbonenses stern by a swing to starboard, and expecting, apparently, that the Lisbonense would yield La Champagne the right of way, (she being the larger vessel,) possibly by going to port, or by slowing or stopping, (as the Lisbonense had the tide against her, and could keep steerage way at a lower speed,) before she reached the course of La Champagne through the main channel; the navigators of the Lisbonense ignorant of the ■ special circumstances, and therefore confident that under article 15 she had the right of way. Unless in some way the navigators of the Lisbonense were apprised of the fact that La Champagne could not fulfill the obligations of that article, and modified their navigation accordingly, or unless La Champagne made some change in her own navigation, collision seemed inevitable; and up to the moment when there was a change of situation each expected the other to give way. The initiative was taken by the Lisbonense, which sounded a one-blast whistle. Under the situation as she understood it, her duty was plain, — she was to keep her course. The international rules have provided for no signal whereby a vessel which intends to hold her course shall notify another of that intention. Article 19 provides as follows:
“In taking any course authorized or required by these regulations, a steamship under way may indicate that course to any other ship which she has in sight by the following signals on her steam whistle, namely: One short blast to mean, T am directing my course to starboard;’ two short blasts to mean, ‘I am directing my course to port;’ three short blasts to mean, T am going full speed astern.’ The use of these signals is optional, but if they are used the course of the ship must be in accordance with the signal made.”
Under tlie act of February 28, 1871, (section 4412, Eev. St. U. S.,) and its various amendments, the board of supervising inspectors have established regulations to be observed by steam vessels when passing each other in harbors and inland waters, which include rules for giving signals by blasts of the steam whistle. “One short blast” and “two short blasts” are among the signals thus provided for, and the giving of such signals when steam vessels are approaching each other is made compulsory. The meaning of those signals, however, is not the same as under the international rules. One short blast indicates only an intent “to pass to the right or port side of the other;” two short blasts, an intent “to pass to the left or starboard side of the other;” but neither implies, ex necessitate, that such passing shall be accomplished by any change of course. So that . practically, in many cases, the signal simply means, “I have the right of way, and mean to keep it,” or, “I waive my right of way, and will let you pass ahead of me,” or, “I ask you to waive your right of way, and let me keep on so as to cross your bows.” To these signals [299]*299the inspectors’ rules require an answer from the other vessel,- — assent to the proposed navigation being expressed by a like signal to that received, dissent by an unlike signal. The international rules, it will be observed, call for no merely answering signals, and require the vessel giving a signal to alter her course in accordance with the signal made; the inspectors’ rules provide for answering signals, but do not require that a change of course shall accompany a given signal. Thus, with singular fatuity, it has been provided, by authority competent to regula,te the giving of signals by steam vessels bound in and out of seaboard harbors, that the same signal shall have a different meaning when sounded on one side or the other of a line which is certainly invisible, and possibly cannot be definitely located except by further legislation. In 1886 attention was called to this unfortunate condition of affairs by the learned district judge of the southern district of New York, in the case of The Aurania and The Republic, 29 Fed. Rep. 98. That it adds to the perils of navigation in and out of seaboard harbors is manifest. Certainly, there can be required no stronger proof of prudence, care, and skill in their profession on the part of those navigators, who in the face of such indeterminate regulations have for seven years piloted millions if human lives over the debatable waters which intersect the fairways from the open ocean to the piers of New York, without some catastrophe so serious as to make the responsibility for the continuance of such conditions unpleasant to contemplate. :
In this particular case, however, both 3ides concede that the maneuvers attending the collision were had in waters where the international rules control. It does not appear that the navigators of either ship thought otherwise; but, reading their evidence, it seems impossible to escape the conviction that the habit of navigating under, this double set of rules operated to confuse the minds of one, if not of both, pilots, as to their obligations to sound whistles, and as to the meaning of the whistles they did sound. The pilot of the Lisbonense testified: “I blew one whistle to let the steamer know I intended to pass ahead of him, * * * (that I) had the right of way,” and “I understood his answer of one whistle to mean, -All right.’ ” In answer to further questions as to custom among seafaring people in this port, he also testified that when “vessels are crossing — two vessels meeting end on, almost — the man that wishes to pass to the starboard side blows one whistle, and the other vessel responds with one. That means both vessels pass to starboard, — keep to the right, — • each vessel to keep to the right.” He added that;, as he understood the situation, he “had the right of way; * * * that was why he blew one whistle,’’ so that La Champagne should “go clear of him,” either by porting or by “slowing down and giving him a chance, — not crowding him.” Evidently, he sounded his one-blast whistle with the understanding that it meant only what it does where the inspectors’ rules apply, and did nob require him to direct his course to starboard. The international rules, however, under which he was navigating, authorized him to blow such whistle only when he was directing his course to starboard under their authority or requirements; but under those very rules being on the starboard side of La Cham[300]*300pagne, and not advised of peculiar circumstances affecting that vessel, he was not authorized to direct his course to starboard, but was, on the contrary, required to keep his course. To have thus changed his course, though a fault under the rules, would probably not have been by itself one contributing to the collision, as it would have .brought the Lisbonense further to the eastward than she would have gone had she kept her course. Having given the one-blast signal, however, the Lisbonense was bound by article 19 to direct her course to starboard; and her failure substantially to make such change was a fault contributing to the collision, because by not so navigating she failed to reach the course of the other vessel as far ahead of her as by her own signal she engaged to do. Had she substantially so changed her course, the collision would not have happened. We '.therefore concur with the district judge in holding the Lisbonense in fault.
The pilot of La Champagne also was apparently confused as to the meaning of the signals,he received and gave. He testified: .
“[The Lisbonense] blowed one whistle, signifying she wanted to cross my bow and pass me oh the port hand. I acknowledged the signal, which is courtesy, * * * signifying that I understood what he was going to do. * * * It is the courtesy of captains of steamships and steamboats about this harbor always to answer a signal given by another ship, to let them know . you understand what they are doing. . [I] answered for that, and no other, reason, and gave no order because of having answered.”
He added that he thought it was proper for the Lisbonense to come . to port, and that she should have given a two-blast signal. Still, he “accepted the signal when it was given, for courtesy’s sake, which is the practice of the port.” Evidently, he understood the signals as ’ meaning what they do under the inspectors’ rules, namely, a proposition from the Lisbonense to pass him on his port side, and an assent by himself to such maneuver, not appreciating the fact that when La ' Champagne gave her one-blast signal the nineteenth article required him to direct her course to starboard. The captain of La Champagne testified that he understood the signal of the Lisbonense to signify, “I remain on the right side of the channel,” and that La Champagne’s signal imposed on her the duty of “keeping the right side of the channel.” While we are satisfied that La Champagne did keep as far to the starboard as, under the special circumstances of the case, she could, with prudence, we do not find from the evidence that subsequently to the exchange of signals she directed her course to starboard, as article 19 required her to do to keep in accord with her own ' announcement of intention. Even if her momentary sheer to starboard took place, as the district judge found, after the exchange of ■ whistles, it was no more a substantial directing of her course to star- ; board than was the similar momentary sheer of the Lisbonense, and in no sense a compliance with the promise of her signal. Had she directed her course to starboard the collision would not have happened; and while we cannot hold her in fault for not doing so, under the spe- ' cial circumstances, we are of the opinion that, knowing of their existence, she was in fault for announcing to a vessel not possessed of the same knowledge a change of course which she could not carry out. [301]*301The Lisbonense, receiving such an answer to her own whistle, was entitled to assume that the vessel giving it was not so affected by special circumstances that she could not maneuver so as to keep out of the way of the Lisbonense by directing her course to starboard. Even in the sense in which ihe signals were interpreted by both pilots, La Champagne's signal whistle was a promise not to interfere with the lisbonense crossing her bows, whether by changing her own course to starboard or by chocking her speed; and the pilot of, La Champagne undertook to check her speed even before he gave the signal, by ordering the engine to slow, — an order which, for some unexplained reason, was not carried out. Had La Champagne conformed her navigation to the promise of her signal, under either interpretation of it, the collision would not have happened.
The captain of La Champagne testifies that her pilot told Mm he was on ihe point of giving a, i.wo-blast signal when the Lisbonense whistled. Such a signal would have indicated, certainly, that he wa,\ not coming to starboard; thai his purpose was to cross the bows of; the Lisbonense, which, under the special circumstances, he understood he had the right, to do, a nd to widcli purpose the navigation of La Champagne adhered after die exchange of signals. When he re-; ceived the one-blast signal from the Lisbonense he had the option1 either to answer with two blasts, to keep silent, or to answer, as he' did, with one. In the first case he would have distinctly advised the Lisbonense that, despite article 1(5, he considered the situation such' that La Champagne could not keep out of her way; that he did not ini end to do so by going to starboard; on the contrary, that he was going to sheer to port, — and f hereafter the navigator of the Lisbonense would be charged with knowledge that La Champagne claimed to be navigating under the special circumstance rule, and expected Mm to keep out of her way. Had La Champagne kept silent, the Lisbonense would at least have been warned by that circumstance that there was some uncertainty as to what the former intended to do. But by answering with one blast she announced her intentiou to do the very thing she could not do. This was a fault. “Courtesy” might require an answer to a signal, but certainly it did not call for an answer which, under the rules governing her navigation, promised a maneuver which special circumstances forbade her carrying out. Irrespective, therefore, of any faults in their subsequent navigation, the collision is to be attributed primarily to the giving by both vessels of signals, to the promise of which the navigation of neither was substantially conformed.
Although the new proofs taken in the circuit court have neither altered nor made more certain the conclusions reached upon the apostles, the exception to their admission calls for an expression of opinion from the court. As 1.0 the record of proceedings before the French consul, so much of it as contains the statements of the master is admissible under the authority of The Potomac, 8 Wall. 590. In so far as it contains the statements of others, it would be admissible only in contradiction of testimony given by them in the case at bar, and to which tlieir attention liad been called when themselves under examination. The statements before the consM of all others than [302]*302the master should therefore he stricken out, and a similar disposition made of the proof as to what appeared on the engine-room dial. Ho one disputed that La Champagne could run at a less speed than 10 knots, and, even if it were disputed, the dial face was no proof of her possible speed.
The appellant excepted to the exclusion of the following question put to the captain of La Champagne: “What information or instruction did you intend the master of -the Lisbonense to gather from your answer of one whistle to his one whistle?” The exception is unsound. The question was immaterial. The issue was one of negligence in answering the whistle with a single blast. The fact of what he meant by his answer has nothing to do with that issue. The statute, as construed by the court, defines the meaning of the whistle; and, in determining whether he was negligent in sounding it, his conduct must be tested by that definition, not by what he intended his signal should mean.
The decree of the district court is reversed, with costs of this appeal to the appellant, and the cause remanded to that court for further proceedings in conformity with the decision of this court.
WALLACE, Circuit Judge.
When the vessels discovered one another, they were about two miles apart, — La Champagne on a course W. by S., the Lisbonense on a course N. W. 3-8 ST.; the former going at a speed of about 12 knots with the wind and tide, the latter at a speed of about 7 1-2 knots; La Champagne having the Lisbonense about two points on her starboard bow. At this time La Champagne was about 1 1-2 miles from the place of collision, and the Lisbonense was about 1 mile. When the vessels had approached to within about a mile of one another, the Lisbonense gave La Champagne a signal of a single blast of the steam whistle, and altered her course half a point or a point to starboard. La Champagne’s pilot understood this signal to mean that the Lisbonense proposed to pass across the bows of La Champagne, and he answered the signal by one of a single blast from La Champagne; and the pilot of the Lisbonense understood the signal of La Champagne as an assent to the proposition of the Lisbonense. Owing to the conditions of the wind, tide, and the channel way, La Champagne could not alter her course to starboard. Her pilot knew this when he answered the signal of the Lisbonense, and intended to consent to allowing the Lisbonense to pass ahead of La Champagne. The master of La Champagne, however, supposed that his own pilot intended La Champagne to pass ahead of the Lisbonense, and that the Lisbonense would give way, and he did not understand the signals as intended to indicate any different movements. After the exchange of signals La Champagne proceeded, without changing her course or speed, until the vessels were about 700 yards apart, when her master, seeing that the bearings of the vessels did not change, and that they were drawing rapidly together, so as to render the risk of collision serious, and acting on his own judgment, ordered her engines stopped, and in about 20 seconds ordered them backed. These orders were promptly executed, and at the same time La Champagne gave the Lisbonense a signal of three [303]*303blasts of her whistle. The Lisbonense in the mean time had kept on at full speed, but altered her course to port, so as to bring her back upon her course prior to the exchange of signals, and, when overlapping the hows of La Champagne, put her helm hard a-starboard to throw her stern further away from La Champagne, but the latter struck her about 20 feet forward of the stern. Somewhat before the exchange of signals, in view of the approach of the lisbonense, the pilot of La Champagne had ordered her engines slowed. By some inadvertence or misunderstanding this order was not communicated to the engineer, but the pilot was not aware of the omission.
Upon these facts I agree with the conclusion of the majority of the court, that both vessels were guilty of fault contributing to the collision; but I differ with them in attributing as the fault the failure of each to observe the requirements of article 19, by directing her course to starboard after having giveD the other one blast of her steam whistle. Neither pilot supposed that the signals were given conformably with that article, but each supposed that the signal of the other was one given pursuant to the ruts of the supervising inspectors. Neither was misled by the signal of the other, but each understood it according to its intended significance. When the signals were exchanged, there was ample time and distance for the vessels to regulate their respective movements so as to avoid a collision. Under these circumstances, it seems to me quite immaterial that neither directed the course of his vessel conformably with the requirements of the signal of article 19.
When the vessels discovered each other, they were on crossing courses, and under the operation of the rule which required La Champagne to avoid the Lisbonense, and required the Lisbonense to keep her course. At the distance from the intersecting point in theii courses at which each vessel was at the time, if both maintained their respective rates of speed, both would arrive at the intersecting point at substantially the same moment; and this fact must have been obvious to each before signals were exchanged, because during the intervening distance there could not have been any change of the bearings of their respective lights. Thus, for some little time before the signals were exchanged, the pilot of each vessel was bound to know that the situation demanded great vigilance, and that any infraction of the rules of navigation on ids own part would embarrass the other, and might lead to a misunderstanding and precipitate a collision. The situation of La Champagne was such that she could not undertake to fulfill her duty of avoiding the Lisbonense by altering her course to starboard, and passing astern of the Lisbonense; but this did not absolve her from the duty of avoiding the Lisbonense by some other maneuver, if any other was practicable. The choice was open to her of altering her course to port and going across the bows of the Lisbonense, if that could he done safely, or of reducing her speed until the Lisbonense should pass beyond the point where there was danger of collision. The pilot of La Champagne very prudently ordered her engines slowed to enable her to give way to the Lisbonense, if he found that to be the safer way. Apparently, before he had fully determined what course to pursue, the Lisbonense gave [304]*304a signal of a single blast, and this signal he immediately answered by a like signal. Both pilots understood these signals to mean that the Lisbonense intended to pass in front of the bows of La Champagne. The fact that these signals were exchanged is persuasive evidence that in the judgment of both pilots, at the time, that was the preferable way of avoiding a collision, in view of the existing situation. After these signals were exchanged, La Champagne could not prudently attempt to' pass in front of the Lisbonense, and the only thing she could do properly, as she could not alter her course to starboard, was to reduce her speed, and proceed cautiously, until it should appear that the Lisbonense had fully passed beyond the point of danger. La Champagne did not reduce her speed. Her pilot probably miscalculated it, supposing his order to slow had been obeyed. If that order had been obeyed, it would not have been necessary to stop and reverse before the time when the master of La Champagne ordered this to be done. As it was, the pilot did not give these orders in due season, and the master found it necessary to interpose and give them himself, taking the responsibility of navigating the vessel. He delayed too long, relying on the judgment of the pilot.
If the Lisbonense had not changed her course after the exchange of signals, and La Champagne had stopped and backed when she did, the collision would not have taken place. The change of course by the Lisbonense back to port brought the vessels considerably nearer together. It is impossible to determine just when this change of course was made. Her duty was to keep her course until danger of collision had been passed. That duty on her part was as imperative, after the exchange of signals, as was the duty on the part of La Champagne to do all in her power to avoid the Lisbonense. If there were any circumstances in the situation which rendered it impracticable for the Lisbonense to maintain the course she was on when she gave signal to La Champagne, she should not have given that signal It may be believed that if either vessel had done her whole duty, pursuant to the rules of navigation, there would not have been a collision;. but it cannot be demonstrated that the failure of either to follow the rules was innocuous. There was but little time for observation of the other by either vessel after the Lisbonense made her last change of course to port. Each was then in a situation which was full of peril. All that is manifest is that if La Champagne had stopped and backed in season, or if the Lisbonense had not, by altering her course to port, brought herself nearer to La Champagne, the collision would not have occurred. The case is therefore one for a division of the loss.-