1. The circuit and district courts agreed in holding the Martello to have been in fault for too great speed. In this conclusion we concur.

By the finding of the circuit court that, at the time the horn of the barkentine was heard upon the steamer, the latter was proceeding at a speed of from five and a half to six knots an hour, we are relieved from the necessity of examining the somewhat conflicting testimony upon the question of the steamer's speed. While it is possible that a speed of six miles an hour, even in a dense fog, may not be excessive upon the open ocean and off the frequented paths of commerce, a different rule applies to a steamer just emerging from the harbor of the largest port on the Atlantic coast, and in a neighborhood where she is likely to meet vessels approaching the harbor from at least a dozen points of the compass. Under such circumstances, and in such a fog that vessels could not be seen more than a quarter of a mile away, it is not unreasonable to require that she reduce her speed to the lowest point consistent with a good steerage way, which the court finds in this case to be three miles an hour. The Southern Belle (Culbertson v. Shaw) 18 How. 584; The Bay State (McCready v. Goldsmith) Id. 89.

Further than this, however, the court found (7) that 'about a minute or two after hearing the horn the officers of the Martello saw the barkentine Freda A. Willey looming in sight through the fog,' and that (13) 'as soon as the Willey loomed in sight of those on the Martello, as indicated in the seventh finding, the first officer of the steamship called out 'Hard a-port,' and the lookout reported a vessel on the starboard bow. The captain immediately ordered the helm hard a-port, and the engines reversed full speed.' These findings, taken together, indicate that the Martello took no action to avoid the collision until after she saw the Willey looming up in the fog, which was a minute or two after hearing the horn. Under the circumstances we think the steamer did not act with sufficient promptness. By article 12, subd. b, of the revised international regulations for preventing collisions at sea, prescribed by act of congress of March 3, 1885 (23 State 438, 440), 'a sailing ship under way shall make with her fog horn, at intervals of not more than two minutes, when on the starboard tack one blast, when on the port tack two blasts in succession, and when the wind the wind is abaft the beam three blasts in succession.' The wind at this time was E. by N., and blowing a five to six knot breeze. One blast of the horn, heard upon the steamer's starboard bow, indicated that the sailing vessel was approaching upon her starboard tack, from a direction which would unavoidably take her across the bow of the steamer, unless the speed of the latter were sufficient to carry her beyond the point at which the courses of the two vessels intersected, before the sailing vessel reached that point. The steamship, however, had no right to speculate upon this contingency. Hearing the horn as she did, and being thus apprised of the bearing and course of the approaching vessel, being, as she must necessarily have been, in doubt as to her distance from the steamship, it was the duty of the latter at once to stop until, by repeated blasts of the horn, she could assure herself of the exact bearing, speed, and course of the approaching vessel. The necessity of instantly stopping, or at least of a reduction of speed to the lowest point consistent with the maintenance of steerage way, in the presence of an unknown danger, is one which the masters of steam vessels are slow to appreciate; but the courts have had occasion to enforce it so often it can, as a matter of law, be no longer considered doubtful. The sound of a fog horn upon either bow, if the blast be such as to indicate that the approaching vessel is upon a course crossing that of the steamer, is obviously such a danger. As we had occasion to observe of a somewhat similar collision in The City of New York, 147 U. S. 72, 84, 13 Sup. Ct. 211: 'Upon hearing the fog horn of the bark only one point on her starboard bow, the officer in charge should at once have checked her speed, and if the sound indicated that the approaching vessel was near, should have stopped or reversed until the sound was definitely located, or the vessels came in sight of each other. * * * There is no such certainty of the exact position of a horn blown in a fog as will justify a steamer in speculating upon the probability of avoiding it by a change of the helm, without taking the additional precaution of stopping until its location is definitely ascertained.' The Hypodame, 6 Wall. 216; The Sea Gull, 23 Wall. 165, 177; The Kirby Hall, 8 Prob. Div. 71; The Ceto, 14 App. Cas. 670.

Indeed, the American and English courts are in perfect accord with regard to the duty of the steamer under such circumstances, and we are simply applying to the Martello the law of her own flag as well as ours in holding her to have been guilty of negligence. Thus, in The Kirby Hall, 8 Prob. Div. 71, it was held to be the duty of a steamship, hearing the steam whistle of another steamship in close proximity, in a dense fog, but unable to ascertain her course and position, to stop and reverse her engines so as to take all way off of her, and bring her to a standstill. So, in The John McIntyre, 9 Prob. Div. 135, it was held that while the master of a steamship was not at once bound the moment he heard a whistle, wherever it might be, to stop and reverse his engines, yet if, in a dense fog, he hears the whistle or fog horn of another vessel more than once on either bow and in the vicinity, from such a direction as to indicate that the other vessel is nearing him, it is his duty to at once stop and reverse, so as to bring his vessel to a standstill. In The Dordogne, 10 Prob. Div. 6, it was said to be the duty of a steamer, on hearing the first whistle, to reduce her speed, and, as the vessels get nearer, to bring the ship to as complete a standstill as is possible without putting her out of command, and when the other vessel has come close to, even though not in sight, to stop and reverse the engines. See, also, The Frankland, L. R. 4 P. C. 529; The Ceto, 14 App. Cas. 670; The Ebor, 11 Prob. Div. 25; The Lancashire [1894] App. Cas. 1.

This case is much like that of The Colorado, 91 U. S. 692,—a collision in Lake Huron, in a dense fog, between a propeller and a bark, in which the propeller, though moving only at the rate of from five to six miles an hour, was held to have been in fault in not bringing down her speed to a slower rate, and in giving conflicting orders to the wheel before the position and course of the bark had been ascertained.

2. The Willey is also charged to have been in fault, not only for excess of speed, as to which we express no opinion, but for a failure to provide herself with a mechanical fog horn, as required by article 12 of the revised international regulations, which reads as follows: 'A steamship shall be provided with a steam whistle or other efficient steam sound signals, so placed that the sound may not be intercepted by any obstructions, and with an efficient fog horn, to be sounded by a bellows or other mechanical means, and also with an efficient bell. * * * A sailing ship shall be provided with a similar fog horn and bell.'

The finding of the circuit court that the Willey was not provided with a mechanical fog horn was made under somewhat peculiar circumstances. The opinion of the court was delivered, and the usual findings of fact filed, on July 31, 1889. On the 6th of September, upon request of counsel for the Martello, the court made certain additional findings; but neither in the findings originally requested by the Martello, nor in the findings actually made by the court, nor in the request for additional findings to the number of nearly 100, was any allusion made to the failure of the Willey to provide herself with a mechanical fog horn; but nine months after the last findings had been made, and in July, 1890, the court made an additional finding that 'the horn of the Willey on board of her, and sounded at the time of the collision, was not a horn sounded, or to be sounded, by mechanical means, but was a tin fog horn,'—refusing, however, to find as a conclusion of law therefrom that the Willey was in fault. In view of this lapse of time, and of the fact that the case had been tried upon the theory that the Willey had carried a sufficient horn, we think it was within the discretion of the court to have refused this finding; but, as it appears upon this record as a fact in the case, we are compelled to accord it is proper weight.

Some question was made with regard to the meaning of the words 'similar fog horn,' required upon sailing vessels; but, if steamships must be provided with a horn 'sounded by a bellows or other mechanical means,' and sailing ships with a similar fog horn,' it follows necessarily that it must be sounded in a similar manner. Indeed, as the horn of a steam vessel is usually sounded by steam, we think it is probable that in the use of the word 'bellows' sailing vessels were contemplated. Such has been the construction given to the statute both in this country and in England. The Love Bird, 6 Prob. Div. 80; The Wyanoke, 40 Fed. 702; The Catalonia, 43 Fed. 396; The Bolivia, 1 U. S. App. 26, 1 C. C. A. 221, 49 Fed. 169.

There can be no doubt that the Willey was guilty of a statutory fault in the failure to provide hereself with the fog horn prescribed by the international regulations, and the presumption is that this fault contributed to the collision. This is a presumption which attends every fault connected with the management of the vessel, and every omission to comply with a statutory requirement or with any regulation deemed essential to good seamanship. In The Pennsylvania, 19 Wall. 125, it was said that 'in such a case the burden rests upon the ship of showing, not merely that her fault might not have been one of the causes, or that it probably was not, but that it could not have been.' In this case a bark was condemned for ringing a bell as a fog signal while under way, although in a case arising out of the same collision the judicial committee of the privy council held that, inasmuch as it appeared that the fog horn would not have been heard a sufficient distance to have enabled the steamer to avoid the danger, the bark should not be condemned for a technical failure to comply with the statute. 3 Marit. Law Cas. (O. S.) 477, 23 Law T. (N. S.) 55. In other words, both courts proceeded upon the same legal principle; but in the English court the evidence was considered sufficient to show that the sounding of a bell instead of a fog horn could not have contributed to the collision. To the same effect are Richelieu & O. Nav. Co. v. Boston Marine Ins. Co., 136 U. S. 408, 422, 10 Sup. Ct. 934; Belden v. Chase, 150 U. S. 674, 699, 14 Sup. Ct. 264; The Fanny M. Carvill, 13 App. Cas. 455, in which the court of appeals observed that, 'if you can show that there is a defect in the lights, that vessel must be held to blame, unless she can show that the defect which exists in her lights could not by any possibility have contributed to the collision.' See, also, The Duke of Buccleuch, 15 Prob. Div. 86, [1891] App. Cas. 310.

While no case appears to have arisen in the highest court, either of this country or of England, where this presumption was applied to the absence of a mechanical fog horn, several cases have come before the lower courts in which the requirement has been held to be obligatory, and a noncompliance to create a presumption of fault. The Love Bird, 6 Prob. Div. 80; The Bolivia, 1 C. C. A. 221, 49 Fed. 169, 170; The Trave, 55 Fed. 117; The Wyanoke, 40 Fed. 702; The Catalonia, 43 Fed. 396; The Energy, 42 Fed. 301.

Can it be said in this case that the absence of a mechanical fog horn could not by any possibility have contributed to the collision? We think not. Upon the contrary, it seems to us not improbable that it did. It is apparent that the reason the regulations prescribed a horn blown by mechanical means is that a louder and more prolonged blast can be blown by that method than by the power of the lungs. The evidence is undisputed that the officers of the Martello heard but one blast of the Willey's fog horn before she hove in sight, and that this (her appearance) was only a minute or two after hearing the horn. The finding is that, 'about forty minutes after discharging the pilot, the horn—one blast of a sailing vessel—was heard on the starboard bow. At that time the captain and third officer were on the bridge, a competent lookout was in the cro'nest about 100 feet abaft the stern [stem], the first officer was on the lookout on the forecastle, and the quartermaster was at the wheel.'

'(6) At that time the steamer was heading E. S. E. The wind was about E. by N., blowing about a five to six knot breeze,' etc.

'(11) She [the Willey] was heading north, closehauled, on the starboard tack, sounding her horn at intervals of one or two minutes, and making about four knots an hour.

'(12) While thus proceeding, she thrice heard the steamer whistle of a steamer, answering promptly each time with a single blast of her horn. At this last signal the Martello appeared in sight, bearing about four points on the port bow, and a quarter of a mile away.'

The gist of this is that, while the horn was sounded at intervals of from one to two minutes, and the steamer was to the leeward of the barkentine, but one blast was heard, while three signals of the Martello's whistle were heard upon the barkentine. Now, if the barkentine had been provided with a more powerful horn, it appears to us not only possible, but probable, that more than one blast would have been heard, and the steamer thus apprised of the course and distance of the bark, and of the fact that she was approaching her upon a course that would carry her across the bows of the steamer.

While we hold it to have been a fault upon the part of the steamer not to have stopped when she heard the horn, there was certainly some excuse for her failure to do so, and the master might not unreasonably have supposed that he would hear a second blast before the vessel hove in sight, and thus be able to gauge more accurately her course and distance. Had the officer of the Martello, however, failed to stop or check her speed after hearing two or three blasts of the horn gradually drawing nearer, his neglect would have been so much grosser than it actually was that we cannot presume he would have been guilty of it.

In this particular The Love Bird, 6 Prob. Div. 80, which was a case of collision between the bark Pansewitz and the screw steamship Love Bird, is directly in point. Two blasts of the fog horn were admitted to have been heard upon the steamer. Counsel for the bark argued that, if the steamer had stopped when she first heard the trumpet, there would have been no collision, and hence that the absence of a mechanical fog horn did not contribute to the disaster. On the other hand, it was argued for the steamer that when she first heard the trumpet of the bark she was so near to her that it was impossible to avoid her. Sir Robert Phillimore held that the testimony showed there were three blasts heard on board the steamer nearly ahead; that she proceeded on her course, neither stopping nor reversing her engines, which it was clearly her duty, under the circumstances, to have done. He found, however, that the bark had not sustained the burden of showing 'that by no possibility could the presence of a fog horn have prevented the collision, for it might possibly have given more warning to the other vessel.' He therefore condemned the bark, although the collision occurred within three days after the act requiring a mechanical fog horn came into operation, and although the bark had left her port of departure (Dieppe) before the act took effect. It is true that this case was decided by a single judge, and does not seem to have been appealed, but the opinion seems to us to be founded upon sound legal principles.

After these vssels came in sight of each other they were so far in extremis that it would probably be unjust to impute fault to either of them. The action of the Martello in reversing her engines at full speed was obviously a proper one. Perhaps if, instead of putting her helm hard a-port, she had put it hard a-starboard, and the barkentine had at the same time luffed and come into the wind, the vessels might have escaped each other, or have come together with a glancing blow which would not have proved disastrous. But we impute no fault to the barkentine for her nonaction in that particuular, as she was entitled to hold her course, at least until she received a signal from the steamer to luff.

But, as we think the bark has failed to sustain the burden of proving that the want of a mechanical fog horn could hot have contributed to the collision, the decree of the court below must be reversed, and the case remanded for further proceedings in conformity with this opinion.