The New York (175 U.S. 187)/Opinion of the Court

This collision took place in October, 1891. The navigation of the two steamers was therefore governed by the Congressional Rules and Regulations of April 29, 1864 (13 Stat. at L. 58, chap. 69), reproduced in Revised Statutes, § 4233, and, so far at least as the manaeuvers of the respective vessels took place in American waters, by the supervising inspectors' rules in force in 1891.

The Revised International Regulations of 1885 (23 Stat. at L. 438, chap. 354) apply only to navigation 'upon the high seas and in all coast waters of the United States;' and in § 2, repealing prior inconsistent laws there is an exception of vessels navigating 'the harbors, lakes, and inland waters of the United States.' It is true that in Moore v. ''American Transp. Co.'' 24 How. 1, 16 L. ed. 674, the limited liability act of 1851, which contained an exception of vessels used 'in rivers or inland navigation,' was held, notwithstanding this exception, to apply to vessels navigating the Great Lakes; but the cases are readily distinguishable. In that the exception was 'any canal boat, barge, or lighter, or to any vessel of any description used in rivers or inland navigation.' It was held that the character of the craft enumerated might 'well serve to indicate to some extent, and with some reason, the class of vessels in the mind of the law-makers, which are designated by the place where employed.' But the case was really decided upon the ground of the magnitude of the Lakes, their commerce, their vessels, and the well-known perils incident to lake navigation. It was thought that such commerce deserved to be placed on the footing of commerce on the ocean, and that 'Congress could not have classed it with the business upon rivers, or inland navigation,' in the sense in which we understand these terms. In the present case the exception is specifically of 'vessels navigating the harbors, lakes, and inland waters of the United States.' If the word 'lakes' was not intended to include the Great Lakes it is difficult to see the object of Congress in making use of that word, since nearly all the other navigable lakes, except Lake Champlain, are located within the limits of a single state, and no act was necessary to exempt them, as the power of Congress does not extend to the purely internal or infraterritorial commerce of the country. The Montello, 11 Wall. 411, ''sub nom. United States v. The Montello'', 20 L. ed. 191; Veazie v. Moor, 14 How. 568, 14 L. ed. 545.

The question, however, is one of little practical importance in this case, inasmuch as rule 19 of Rev. Stat. § 4233, is word for word the same as article 16 of the Revised International Rules and Regulations of 1885. Both are as follows: 'If two vessels under steam are crossing so as to involve risk of collision, the vessel which has the other on her own starboard side shall keep out of the way of the other.

The power of the supervising inspectors to adopt rules for the government of steam vessels in passing each other (Rev. Stat. 4412) is limited by § 4400 to steam vessels 'navigating any waters of the United States which are common highways of commerce, or open to general or competitive navigation.' These rules are pertinent to this case only so far as they make it the duty of vessels to indicate by signals of one or two whistles the course they are about to take, and of the other vessel to answer them, and also, in case of vessels crossing each other, within the meaning of article 16, in requiring the obligated vessel to avoid the other by porting and going under her stern. These rules are as follows:

Rule 2. 'When steamers are approaching each other in an oblique direction (as shown in diagram of the fourth situation) they shall pass to the right of each other as if meeting 'head and head' or nearly so, and the signals by whistle shall be given and answered promptly, as in that case specified.'

Rule 3. 'If, when steamers are approaching each other, the pilot of either vessel fails to understand the course of the other, whether from signals being given or answered erroneously or from other cause, the pilot so in doubt shall immediately signify the same by giving several short and rapid blasts of the steam whistle; and after the vessels have approached within half a mile of each other both shall be immediately slowed to a speed barely sufficient for steerage way until the proper signals are given, answered, and understood, or until the vessels shall have passed each other.'

Rule 6. 'The signals by the blowing of the steam whistle shall be given and answered by pilots in compliance with these rules, not only when meeting 'head and head,' or nearly so, but at all times when passing or meeting at a distance of within half a mile, and whether passing to the starboard or port.'

1. We are of opinion that the Canadian statute of 1886 may properly be considered by us.

The question how far this court may take judicial notice of the laws of a foreign country has been the subject of some discussion, and was first considered by this court in the case of Talbot v. Seeman, 1 Cranch, 1, 38, 2 L. ed. 15, 27. That was a case of salvage upon recapture from the French. It became necessary to inquire whether the laws of France were such as to have rendered the condemnation so probable as to create a case of such real danger that her recapture could be considered a meritorious service. To prove this, counsel offered several decrees of the French government, to the reading of which objection was made upon the ground that they were the laws of a foreign nation, and therefore to be proved as facts. In holding that the decree, having been promulgated in the United States as a law of France, was entitled to be read, Mr. Chief Justice Marshall observed 'that the laws of a foreign nation, designed only for the direction of its own affairs, are not to be noticed by the courts of other countries, unless proved as facts, and that this court, with respect to facts, is limited to the statement made in the court below, cannot be questioned. The real and only question is whether the public laws of a foreign nation, on a subject of common concern to all nations, promulgated by the governing powers of a country, can be noticed as law by a court of admiralty of that country, or must be still further proved as a fact. The negative of this proposition has not been maintained in any of the authorities which have been adduced. On the contrary, several have been quoted (and such seems to have been the general practice) in which the marine ordinances of a foreign nation are read as law without being proved as facts. It has been said that this is done by consent; that it is a matter of general convenience not to put parties to the trouble and expense of proving permanent and well-known laws which it is in their power to prove; and this opinion is countenanced by the case cited from Douglas. If it be correct, yet this decree, having been promulgated in the United States as the law of France, by the joint act of that department which is intrusted with foreign intercourse, and of that which is invested with the powers of war, seems to assume the character of notoriety which renders it admissible in our courts.'

The same question as applied to the original rules and regulations was presented to us in the case of The Scotia, 14 Wall. 170, ''sub nom. Sears v. The Scotia'', 20 L. ed. 822, in which we held that, in view of the fact that these rules and regulations were originally adopted by the British orders in council of January 9, 1863, and by Congress in 1864, and had been accepted as obligatory by more than thirty of the principal commercial states of the world, including almost all which have any shipping on the Atlantic ocean, we would take judicial notice of them and treat them as laws of the sea and of general obligation. The duty to take judicial notice of these rules was also recognized by this court in The Belgenland, 114 U.S. 355, 370, ''sub nom. The Belgenlant v. Jensen'', 29 L. ed. 152, 157, 5 Sup. Ct. Rep. 860, in ''Richelieu & O. Nav. Co. v. Boston Marine Ins. Co.'' 136 U.S. 408, 422, 34 L. ed. 398, 403, 10 Sup. Ct. Rep. 934, and in numerous cases in the lower courts. There is nothing in the case of Liverpool & G. W. Steam Co. v. ''Phenix Ins. Co.'' 129 U.S. 397, 32 L. ed. 788, 9 Sup. Ct. Rep. 469, in conflict with this. That did not involve a question of general maritime law, but of a statutory exemption from the consequences of negligence in navigation given by a British act of Parliament. We know of no reason why the rule adopted in The Scotia should not be applied to the Revised International Rules and Regulations. They have also been adopted by most, if not all, the nations which gave their assent to the original rules and regulations of 1863, and the reasons which induced this court to take judicial notice of these rules are equally persuasive here. The reference to the Canadian statute of 1886, used in the district court and printed as a part of the record here, shows it to be, except as to the waters covered by it and as to certain immaterial local regulations, a literal copy of the congressional act of 1885.

But we think that for another reason the act is properly before us. After the case had been appealed to the circuit court of appeals, the libellant moved that court for an order requiring the testimony of a witness to be taken to prove the Canadian statute, and filed in support of this motion affidavits that in the printed record there was no copy of this statute, but that it was introduced in the district court and used and referred to in the arguments upon the rehearing before the district judge; that at that time the libellant offered to prove the statute by oral testimony, but that it was then agreed in open court between the proctors that the testimony of such witness might be dispensed with, and that the statute then in court might be used without technical proof thereof. No order was made upon this motion, but there was a further suggestion to the court of a diminution of the record in that the Canadian statute, which was introduced and used as evidence in the district court, did not appear in the record, and a writ of certiorari was granted 'because the transcript of the record in this case does not contain a copy of the Canada statutes governing the navigation of vessels in the waters of Canada during the year 1891, which was introduced in evidence, as alleged.' In obedience to this writ, the clerk of the district court was ordered to transmit to the circuit court of appeals a certified copy of the Canadian statute. This was done, but the clerk, instead of certifying that it was a part of the record, certified only that he had 'carefully compared the same with the original act as published' (by which we understand as published in the statutes of Canada), 'and find the same to be a true copy of such original and of the whole thereof.' It thus appears that the Canadian statute had been used in the district court by consent of counsel, had been treated as part of the record, and that the copy sent up was a true copy of the statute as published. It is true that the clerk did not formally certify it to be a part of the record, but the fact that it had been so treated was established by the affidavit; and the writ of certiorari upon its face recited the fact that a copy of the statute had been introduced in evidence, as alleged, and required the court below to 'send the record and proceedings, with all things concerning the same, as fully and entirely as they remain of record in said district court.' In view of these proceedings, we think the circuit court of appeals should have accepted the certified copy of the statute as properly in evidence before it.

The only novel feature of this statute, pertinent to this case, is as follows:

'Art. 19. 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, that is to say: 'One short blast to mean 'I am directing my course to starboard;' two short blasts to mean 'I am directing my course to port;' three short blasts to mean 'I am going at 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.'

In this view, the question whether two American vessels running from one American port to another are bound, whenever they cross the boundary line between the United States and Canada, which at this point is the filum aquoe of Detroit river, to conform to the navigation laws of Canada, does not arise in this case. Were all the commerce of the lakes carried on in American vessels the question would be less difficult of solution. But as much of this commerce is Canadian, and it is impossible to tell whether an approaching vessel be American or Canadian, an attempt to apply the laws of the United States in all cases might result in confusion and in great injustice to Canadian vessels, in case the rules and regulations of the two countries differed in any material respect. We are saved, however, consideration of these questions by the fact that the signals and the steering rules of the United States and Canada are practically identical. This fact being once established, the duty of vessels of both nations in meeting each other, either upon American or Canadian waters, is easily understood.

2. In judging of the responsibility for this collision it should be borne in mind that the Burlington and her tow were temporarily occupying from two thirds to three quarters of the navigable channel of the river. The distance between the rear barge and the Canadian bank of the navigable channel is variously estimated, but according to the court of appeals was about 500 feet. It may have been as much as 800 feet, but probably was not more than that. The night was clear and starlit, the weather fine, and the collision could scarcely have occurred except by the fault of one or both vessels.

The Conemaugh, a steamer of 1,609 tons burden, was coming down the American side of the river at her usual speed of about 10 miles an hour, and, when her attention was first called to the obstruction of the Burlington's tow, was about passing what are known as the Kasota piles, which were in fact the remains of a coffer dam once used in raising the steamer Kasota. They were near midchannel, though somewhat upon the American side, and about three quarters of a mile above Smith's Coal Dock. As she was passing these piles, leaving them on her port hand, she received and answered a signal of two blasts from the Burlington, which had come down the river on the Canadian side, and was at that time rounding to at the coal dock on the American side, her tow of four barges making a crescent or semicircle, the outer arm of which was, as above stated, from 500 to 800 feet from the Canadian bank. The length of the tow was about 2,600 feet, the width of the channel about 3,000 feet. The Burlington at this time was exhibiting to the Conemaugh her white masthead and her starboard green light. The first barge in tow was also exhibiting her green light, but the others had not rounded to sufficiently to exhibit their colored lights. After exchanging this signal with the Burlington, the wheel of the Conemaugh was put hard-a-starboard, her speed checked, and her course taken across the stream at almost a right angle with her former course. Upon this course she was exhibiting her green light to vessels ascending the river. After she had 'picked up' or discovered the rear barge her wheel was steadied, and then ported to follow the tow, which by the force of the current was gradually swinging down stream, and would ultimately round to on the American side, astern of the Burlington. As the Conemaugh steadied her wheel to starboard her watch made out below the tow and about a mile distant the white and red lights of the New York, apparently somewhat on the American side of mid-channel, and promptly signaled her with two blasts of her whistle, indicating that she would pass her to the left. No answer was received from the New York. Under such circumstances it would have been more prudent for the Conemaugh to stop and wait a few minutes, until the tow had drifted down and left the channel clear below her; but inasmuch as there was a clear space of 500 feet of navigable water between the last barge and the Canadian bank of the channel, we should hesitate to condemn her for this fault, were there no others contributing more immediately to the collision.

Receiving no answer to her first blast, the Conemaugh, when the two steamers were about three quarters of a mile apart, repeated her signal of two blasts,-the New York then showing her masthead and both colored lights. Again no reply was made by the New York. The Conemaugh, which had then ported and was heading toward the Canadian shore, and about four points from the direct course down the river, gave a third signal of two blasts, the New York continuing to show all three of her lights, and being apparently close to and between the second and third barges of the tow. The New York made no answer to this third signal. The duty of the Conemaugh at this juncture was plain. She should have stopped her engines after the second signal, and, if necessary to bring her to a complete standstill, have reversed them. Nothing is better settled than that, if a steamer be approaching another vessel which has disregarded her signals, or whose position or movements are uncertain, she is bound to stop until her course be ascertained with certainty. The Louisiana v. Fisher, 21 How. 1, 16 L. ed. 29; Chamberlain v. Ward, 21 How. 548, 16 L. ed. 211; Nelson v. Leland, 22 How. 48, 16 L. ed. 269; The Martello,153 U.S. 64, 71, ''sub nom. The Martello v. Willey'', 38 L. ed. 637, 640, 14 Sup. Ct. Rep. 723; The Teutonia, 23 Wall. 77, ''sub nom. Sieward v. The Teutonia'', 23 L. ed. 44; The James Watt, 2 W. Rob. 271; The Birkenhead, 3 W. Rob. 75; The Hermann, 4 Blatchf. 441; The Huntsville, 8 Blatchf. 228; The Hammonia, 4 Ben. 515; The Mary Sandford, 3 Ben. 100; The Arabian, 2 Stuart, Vice Adm. 72. There was peculiar necessity for such action in this case. These vessels were about to meet upon crossing courses, and to pass each other in the narrowest part of the channel. The Conemaugh had three times signaled her wish to take the Canadian side, and pass starboard to starboard. The New York had three times neglected to give her assent to this arrangement. The Conemaugh had construed her failure to reply as an acquiescence in her own signals. The New York might have construed such failure as a refusal to acquiesce. In such a case it was clearly incumbent upon the Conemaugh to stop until the mystery of her silence was explained, and in failing so to do she was guilty of fault. Instead of that, while running under check and under a port helm, she steadied and almost immediately lost the green light of the New York, whereupon she sounded an alarm whistle, put her helm hard-a-starboard, and endeavored to shoot across the bows of the New York. The two steamers were then upon converging courses and about a quarter of a mile apart. Even then, if the Conemaugh had put her helm hard-a-port and reversed her engines she would probably have avoided a collision, although her final error, being apparently in extremis, perhaps ought not to be attributed to her as a fault. But she kept on her course at full speed, with her helm hard-a-starboard, while the New York came up the river, under a port wheel and at full speed, displaying her masthead and red light to the Conemaugh. Just before the collision the wheel of the New York was starboarded, but too late to avert the blow. She struck the Conemaugh on her starboard side near the gangway, and sank her within ten minutes. The place of the collision seems to have been very near the Canadian bank, and about 1,000 feet from and a little upon the port quarter of the Furguson, the stern barge of the Burlington's tow.

The fault of the Conemaugh appears the more flagrant from the fact that the two steamers were crossing vessels within the meaning of rule 19 (Rev. Stat. § 4233), and that the Conemaugh, having the New York upon her starboard side, was bound to keep out of her way. The supervising inspectors' rules require that this manoeuver shall be performed by porting the wheel and passing under the stern of the preferred vessel. But, irrespective of this rule, prudent seamanship ordinarily requires that the obligated vessel shall take a course which, if the preferred vessel perform her own duty, will certainly avoid a collision, viz., port and go astern. If, upon the other hand, she elects to starboard and cross the bows of the other vessel, she incurs the manifest danger of not passing the point of intersection before the preferred vessel strikes her, and is justly considered as assuming the responsibility for the success of her manoeuver. The E. A. Packer, 140 U.S. 360, 366, ''sub nom. The E. A. Packer v. New Jersey Lighterage Co.'' 35 L. ed. 453, 457, 11 Sup. Ct. Rep. 794; The Nor, 2 Asp. M. L. Cas. 264. Of course, there may be such conduct on the part of the favored vessel as would show that she was alone guilty of fault, but the greater safety of porting is so manifest that the circumstances must be quite exceptional to justify a different course. The failure of the Conemaugh's manoeuver in this case only emphasizes her original fault in failing to come to a standstill when her two first signals to the New York were disregarded.

The conduct of the Conemaugh, as we shall hereafter show in the navigation of the New York, was not even consistent with her own theory, which was that she would cross the course of the New York and pass down between her and the Canadian bank. Instead of doing so, however, as soon as she had 'picked up' the stern barges and ascertained their exact location, she ported her helm sufficiently to display to these barges a glimmer of her red light, and as the New York was about the same time starboarding to clear these barges, the result was that neither gave the other sufficient room to pass. These circumstances were most favorable to the collision which almost immediately ensued.

3. Inasmuch as no witnesses were sworn from the New York we are compelled to judge of the propriety of her manoeuvers from the admissions in her answer and from the other testimony in the case. From these it appears that the propeller, a vessel of 1,700 tons, was bound up the river, and, when nearing the point below where the river Rouge empties into the Detroit just above Smith's Coal Dock, she descried the Burlington and her tow beginning to round to from the Canadian side of the river to the coal dock on the American side, exhibiting to the New York her masthead and red lights as well as the red side lights of the barges in tow. The answer avers that thereupon 'the New York blew a passing signal of one blast, at the same time checking her engine and reducing her speed to about 4 miles an hour, and then porting her helm so as to pass under the stern of the last barge. When the New York had arrived at a point abreast of the last barge in tow, a signal of two whistles was heard, but being unable to see any vessel, and noticing only a white light close on the Canadian bank of the river, the signal of two blasts was not answered, as it seemed to be intended for some other vessel, the New York being then close to the Canadian bank, and there not being room enough for any vessel to safely pass between her and the bank.'

If there were no other evidence in the case than these allegations, and the uncontradicted testimony of the Conemaugh that she blew three signals to the New York, none of which were answered, it is sufficient to show the latter to have been guilty of a grievous fault. The night was clear, and there appears to have been no difficulty in seeing the white and colored lights of the Burlington and her tow, and should have been none in seeing the lights of the Conemaugh. No reason is given why the signals of the Conemaugh were not heard, and as the New York was not more than a mile distant from her when her first signal was blown, and considerable less than that when the second signal was blown, her inability to hear them is inexplicable, except upon the theory that no sufficient lookout was maintained, or that such lookout did not attend properly to his duties. Her officers failed conspicuously to see what they ought to have seen or to hear what they ought to have heard. This, unexplained, is conclusive evidence of a defective lookout. The Sea Gull, 23 Wall. 165, ''sub nom. The Sarah Watson v. The Sea Gull'', 23 L. ed. 90; The James Adger, 3 Blatchf. 515; The Fanita, 14 Blatchf. 545; The Sunnyside, 91 U.S. 208, ''sub nom. Miner v. The Sunnyside'', 23 L. ed. 302; Spencer, Collisions, § 175.

The force of this presumption of a defective lookout is greatly strengthened by the fact that the claimant did not see fit to put upon the stand the officers and crew of the New York, who certainly would have been able to explain, if any explanation were possible, why the lights of the Conemaugh were not seen and distinguished or her signals heard. It was said by this court in the case of Clifton v. United States, 4 How. 242, 246, 11 L. ed. 957, that 'to withhold testimony which it was in the power of the party to produce, in order to rebut a charge against him, where it is not supplied by other equivalent testimony, might be as fatal as positive testimony in support or confirmation of the charge.' If the New York heard the signals, it was her duty to answer them. Beyond this, however, the answer admits that a signal of two whistles was heard, and a white light close on the Canadian bank of the river was noticed, but the signal was not answered, as it seemed to be intended for some other vessel. However, the white light in connection with the whistles could only have been the masthead light of a steam vessel, and as there is no evidence that there was any other vessel coming up the river, the signal could only have been intended for the New York. If she were unable to see the colored lights of the approaching steamer, it was her duty to stop until she made them out, or otherwise determine the identity and course of the approaching vessel.

Her only excuse for her omission is that she was the preferred vessel within the 19th American and 16th Canadian rule, and that by the 23d American and 22d Canadian rule it was her duty to keep her course. But the fact that a steamer is entitled to hold her course does not excuse her from inattention to signals, from answering where an answer is required, or from adopting such precautions as may be necessary to prevent a collision, in case there be a distinct indication that the obligated steamer is about to fail in her duty. As was said in the case of The Sunnyside, 91 U.S. 208, 222, ''sub nom. Miner v. The Sunnyside'', 23 L. ed. 302, 307: 'Cases arise in navigation where a stubborn adherence to a general rule is a culpable fault, for the reason that every navigator ought to know that rules of navigation are ordained, not to promote collisions, but to save life and property by preventing such disasters.' See also The Delaware, 161 U.S. 459, 40 L. ed. 771, 16 Sup. Ct. Rep. 516; The Maria Martin, 12 Wall. 47, ''sub nom. Martin v. Northern Transp. Co.'' 20 L. ed. 255. Both the Canadian and American Codes provide that in construing and obeying these rules due regard must be had to all dangers of navigation and to any special circumstance which may exist in any particular case, rendering a departure from them necessary in order to avoid immediate danger. There is another rule pertinent in this connection, namely, rule 21 American, and article 18 Canadian, that every vessel when approaching another vessel so as to involve risk of collision, shall slacken her speed, or, if necessary, stop and reverse. That the obligation to observe this rule attached to the New York under the peculiar circumstances of this case is entirely clear. Her attention had been called to the fact that a steamer was coming down the river between the rear barge and the Canadian bank. The channel was narrow, and the descending vessel had signified her intention to starboard her helm and pass the New York to the left. The New York avers in her answer that there was not room enough for any vessel to safely pass between her and the Canadian bank, but, notwithstanding this, she kept her course toward that bank, and was thus constantly narrowing the channel through which the Conemaugh signified her intention of passing. She averred that her speed in passing the tow was about 4 miles an hour, but the District Judge was of opinion that she maintained double that speed until the vessels came together. However this may be, her failure to answer the whistles of the Conemaugh, or to stop and reverse, after her white light was seen, was wholly inexcusable, and, under the particular circumstances, cannot be justified by her general duty as a favored vessel to keep her course or by anything that was said by this court in The Britannia, 153 U.S. 130, ''sub nom. The Britannia v. Cleugh'', 38 L. ed. 660, 14 Sup. Ct. Rep. 795. The master of a preferred steamer cannot, by blindly adhering to his course, atone for the neglect of other precautions.

We do not wish to say that the New York was under any obligation to assent to the proposed arrangement, although in starboarding and passing close to the two rear barges she did in fact take the exact course she would have taken if she had assented. If she had blown one whistle she would have indicated her intention of pursuing her course under her port wheel as the privileged vessel; while if she had blown two whistles she would have starboarded, as she did starboard, and keep as near the rear barges as she safely could. What we do decide is that the duty to answer a signal is as imperative as the duty to give one. Not only does the 2d rule of the supervising inspectors require of crossing steamers that 'signals by whistles shall be given and answered promptly,' but ordinary prudence demands that an obligated steamer proposing by whistle to deviate from the customary course shall receive an immediate reply, that her wheel may be at once put to starboard or port, as the exigencies of the case may require. A delay of even a few seconds may seriously embarrass her as to the intention of the preferred vessel. This is now made obligatory upon vessels navigating the Great Lakes by the act of February 8, 1895 (28 Stat. at L. 645), the 23d rule of which declares that 'every steam vessel receiving a signal from another shall promptly respond with the same signal, or as provided in rule 26.' If the New York had promptly answered the Conemaugh's signals, probably no collision would have occurred.

The comments we have made upon the failure of the Conemaugh to stop and reverse are equally pertinent to the case of the New York. If she did not hear the whistles of the Conemaugh, she ought to have heard them; but irrespective of this, there was enough to apprise her of her danger in pursuing her course with unabated speed. She knew that she was about to meet in a narrow channel a steam vessel coming down upon her with the added speed given by a current of 2 to 2 1/2 miles an hour. She heard her final signal of two blasts as she was passing the last barge, and should have known that if she continued her course a collision would be inevitable, and yet she did not stop or reverse. Her conduct was inexcusable. The lesson that steam vessels must stop their engines in the presence of danger, or even of anticipated danger, is a hard one to learn, but the failure to do so has been the cause of the condemnation of so many vessels that it would seem that these repeated admonitions must ultimately have some effect. We cannot impress upon the masters of steam vessels too insistently the necessity of caution in passing or crossing the course of other vessels in constricted channels.

But, assuming the theory of the New York to be true, and that as the preferred vessel she was bound to keep her course, under rule 19, the fulfilment of her duty in that regard undoubtedly added to the embarrassments of the Conemaugh. It is averred in her answer that after making the white light of the Conemaugh she continued on her course so as to go around close to the last barge, and when abreast of her quarter starboarded, so as to go close under her stern. For this change in her course she relies upon the case of The John L. Hasbrouck, 93 U.S. 405, ''sub nom. Lyman v. The John L. Hasbrouck'', 23 L. ed. 962, in which we held that the obligation of a privileged vessel to keep her course does not forbid such necessary variations in her course as will enable her to avoid immediate danger arising from natural obstructions to navigation. In that case a sailing vessel descending the Hudson river at West Point was held to have been excused in changing her course to round a projection at that place, but in this case the New York had still from 500 to 800 feet before her before reaching the Canadian bank. Her original porting was undoubtedly to avoid the tow, but there seems to have been no immediate necessity for her starboarding to pass so close to the rear barges, though we should not condemn her upon this ground. See discussion of this in The Velocity, L. R. 3 P. C. 44; The Banshee, 6 Asp. M. L. Cas. 221. While the presence of the tow undoubtedly rendered it necessary for the New York to port, and thus to become a crossing vessel, and a preferred vessel under rule 19, there was no obstruction to her continuing under her port wheel until she had approached so near the Canadian bank as to make it necessary to turn.

The theory of the New York is an inconsistent one-as inconsistent as that of the Conemaugh. She argues that she was under no obligation to assent to the signals of the Conemaugh by starboarding her helm. But she did in fact starboard her helm, and now insists that she did this in discharge of her duty as a preferred vessel to resume her course after she had cleared the obstruction. But without deciding that she was in fault for starboarding, her conduct in so doing adds another to the many reasons why she should have indicated to the descending steamer her proposed course. If the Conemaugh recognized the fact that she were the preferred vessel and bound to hold her course it would naturally confuse her to see the New York suddenly starboarding, exhibit both her colored lights, and point directly toward her, as she must have done. The probable explanation of the course of the New York is that the officer of her deck was so intent upon watching the lights of the barges that he omitted to notice the lights of the Conemaugh until the vessels had approached so near that a collision became extremely probable. The fact that her lights were seen and her signals heard by the crews of the Burlington and her barges and by persons standing upon the coal dock, at a greater distance from the Conemaugh than was the New York, only indicates more clearly that her lookout was either insufficient or incompetent. If he actually saw her and reported her to the officers of the deck, the responsibility is only shifted from the lookout to them.

Our conclusions are that the Conemaugh was in fault:--

For not stopping when the New York failed to answer her signals;

For porting and then starboarding in order to cross the bow of the New York;-and the New York:--

For an inefficient lookout;

For failing to answer the repeated signals of the Conemaugh; and--

For failure to stop, after she made the white light of the Conemaugh, until her course and movements had been satisfactorily ascertained.

4. The final question arises upon the insistence of the underwriters of the Conemaugh's cargo, that they are entitled to a recovery to the full amount of their damages against the New York, notwithstanding the Conemaugh may also be in fault for the collision. They are correct in this contention. Indeed, this court has already so decided in the case of The Atlas, 93 U.S. 302, ''sub nom. Phoenix Ins. Co. v. The Atlas'', 23 L. ed. 863. This was a libel against the Atlas by an insurer of the cargo of a canal boat in tow of the steam tug Kate, whereby the canal boat and her cargo were lost. It was insisted by the claimant that, as the libellant had failed to make the Kate a party, and as both vessels were found to be in fault for the collision, there could be a recovery of only a moiety of the damages. The case of The Milan, Lush. 388, was confidently relied upon as an authority. This court, however, was of opinion that a plaintiff, who has suffered a loss by the negligence of two parties, was at liberty, both at common law and in admiralty, to sue both wrongdoers or either one of them at his election, and 'it is equally clear that, if he did not contribute to the disaster, he is entitled to judgment in either case for the full amount of his loss. He may proceed against all the wrongdoers jointly, or he may sue them all or any one of them separately. . . . Co-wrongdoers, not parties to the suit, cannot be decreed to pay any portion of the damage adjudged to the libellant, nor is it a question in this case whether the party served may have process to compel the other wrongdoers to appear and respond to the alleged wrongful act.' A like ruling was made in The Juniata, 93 U.S. 337, ''sub nom. United States v. Juniata'', 23 L. ed. 930, in which a libel was filed by the United States as owner of the cargo of a flatboat in tow of one of two vessels.

The decree of the Court of Appeals is therefore reversed, and the case remanded to the District Court for the Eastern District of Michigan for further proceedings in consonance with this opinion. Costs will be divided equally.