EACOMBE, Circuit Judge.
The record is very voluminous. An examination of the briefs of the various appellants indicates that the important questions presented upon the appeals are as follows: First. Should not the petitioner be compelled to pay to the trustee the following sums: (a) The freight and passage money received by the petitioner for the passage from New York to Havre, (b) The freight and passage money received for-the preceding passage from Havre to New York, (c) The amount of money received from the French government as compensation for carrying the mails on the two passages. Second. Should the claims for damages by reason of .loss of life have been excluded? Third. Was the Bourgogne in fault for the collision? Fourth. Was the disaster in question “done, occasioned, or incurred without the privity or knowledge of the owners”?
A preliminary question is raised by motion to dismiss the appeals, which was, by direction of this court, heard with the main [435] argument. Petitioner contends that this court has no jurisdiction of the appeals of the several parties, because the same were not taken within the statutory period after the entry of the decree by Judge Townsend. It is not disputed that appeals from the decree made by Judge Thomas were all taken in time. We are of the opinion that the situation is similar to that which arises so frequently in patent causes, where the court, upon pleadings and proofs, finds that the patent is valid, that the prior art discloses no anticipation and no prior public use, and that the device of defendant infringes the patent, and sends it to a master to^ investigate and report upon profits and damages. From final decree, after the master’s report is in, disposing of the whole cause, appeal is taken, and upon that appeal, although taken long after the entry of the interlocutory decree, the decision of the Circuit Court as. to validity, prior use, etc., is reviewed.
1. As to the freight and passage money from New York to Havre. In the case of The Main, 152 U. S. 122, 14 Sup. Ct. 486, 38 L. Ed. 381, it was held that the words “freight pending,” in section 4283, Rev. St. U. S. [U. S. Comp. St. 1901, p. 2943], and “freight for the voyage,” in section 4284, were “evidently intended to represent the earnings of the voyage, whether from the carriage of passengers or merchandise. If these words were used instead of the words ‘freight for the voyage,’ it would probably more accurately express the intent of the Legislature.” Very many authorities are cited in the briefs bearing upon the question when freight or passage money is earned, and what is the effect of a disaster resulting in total loss of merchandise or passengers. They are not especially persuasive, because in none of them except The Corona (Pacific Coast Co. v. Reynolds, 114 Fed. 877, 52 C. C. A. 497, certiorari denied 187 U. S. 640, 23 Sup. Ct. 841, 47 L. Ed. 345) was the contract of carriage the same as the one now before us. In that case the Court of Appeals for the Ninth Circuit, after referring to the general rules that freight is not earned till the goods are carried to and delivered at the place of destination, and that freight paid in advance may, in the absence of a special agreement to the contrary, be recovered back if the voyage be broken up, adjudged that an item of $3,867 for prepaid freight, wharfage, and advance charges must be deducted from the amount the petitioner should be required to pay in order to secure limitation of liability. As to an item, however, of $7,770 passage money, it was held:
“As the passage money in question was prepaid under an express agreement that the owner of the ship should not refund it, notwithstanding a failure to deliver passengers at the places of destination, we think it clear that it must be regarded as earned.”
The court further held that the sum thus directed to be paid in should not be diminished by the amount expended by the owner in forwarding passengers to their destination, nor by the amount of certain sums voluntarily given to some of the passengers. The case of The Scotland, 105 U. S. 24, 26 L. Ed. 1001, was distinguished because it disclosed no such agreement; while in the case of The [436] Corona “the amount prepaid for passage was, by the express stipulation of the parties, made absolute and unconditional, and should, in our opinion, be regarded as earned.”
We fully concur in these views expressed by the Court of Appeals in the Ninth Circuit, and find in the case at bar a similar agreement both as to freight and passage money. The bills of lading all contained the clause: “11. Also that the freight prepaid will not be returned, goods lost or not lost.” The first clause in the passage ticket reads: “Passage money to be paid in full before the departure of the steamer and at all events belongs to the company.” And in the sixth clause it is provided that: “The company will not be liable fpr loss or damage occasioned by accidents, fire, perils of the sea, or unforeseen circumstances, or by barratry, fault or negligence of the captain, pilot, sailors, members of the crew or passengers.” We are of the opinion, therefore, that the petitioner should pay over to the trustee 100,703.08 francs for prepaid passage money and 12,716.43 francs for prepaid freight, New York to Havre, aggregating 113,419.57 francs, with interest from the date of the disaster.
. 2. As to the freight and passage money collected for the crossing from Havre to New York. The statute uses the phrase “freight for the voyage,” and claimants contend that the voyage in question was the round trip from Havre to New York and back again to Havre. Very many authorities are cited by both sides, and the District Court has 'discussed the question at some length. As was to be expected, an examination of the citations shows that the word “voyage,” like so many other words, is somewhat elastic. The meaning to be given to it in any particular case is largely dependent upon the facts of that case. We are dealing now with the word as used in a statute which provides that the shipowner’s liability for loss of freight, etc., on a particular voyage may be limited when he gives up the earnings of the same voyage. The voyage which exposes the property to risk is the voyage the earnings of which are to be paid in. As was said in The Main, 152 U. S. 123, 14 Sup. Ct. 486, 38 L. Ed. 381, “The real object of the act was to limit the liability of vessel owners to their interest in the adventure”; i. e., to the “adventure” in which the persons or property transported was put at risk. In determining precisely what such adventure is under this statute, we concur with the District Judge in the conclusion that the controlling circumstances are not to be found in the shipowner’s agreements with individual shippers, nor in the length of time for which a crew may be hired or the ship provisioned; nor is it important what nomenclature may be adopted in the shipowner’s logbooks or in the daily talk of its officers, nor how it keeps its accounts, nor how often the ship is inspected. The fundamental question seems to be this: Considering .the merchandise and passengers which are shipped as a whole, when does the ship reach a port where such merchandise and passengers are no longer any part of them, exposed to the risks of transport by that ship? Now, it may very well be that an empty ship will take on a cargo, complete or partial, in a particular port, and may visit in [437] succession other ports, in some or all of which it discharges more or less of its original- cargo, taking on other cargo in its place, which it carries to one or other of the ports it subsequently touches at, until finally it reaches a port of complete discharge. Whether the whole or some particular portion of that continuous transit should, in the case of disaster between two of the ports touched at, be considered the “voyage” of the statute, may present interesting and difficult questions; but in our opinion the case at bar is free from difficulty. L,a Bourgogne is one of a fleet of steamers which are operated solely as an “ocean ferry” between the ports of Havre and New York. Each trip or crossing of each steamer is by itself a complete maritime transaction. A steamer lying at .Havre, empty except for her own equipment and stores, takes on board a cargo under contract with shippers to deliver it in New York. No merchandise is shipped to be carried by that steamer to New York and back again to Havre. In like manner she takes aboard her complement of passengers under contract to take them all to New York, and to land them there. When she sails from Havre she exposes this cargo and these passengers to the risks and peril of that crossing; but when she reaches New York she discharges all her freight and lands all her passengers, and becomes herself an empty ship again, ready to take on other cargo and other passengers for another - crossing. When cargo and passengers are safely landed in New York, all their risks of the crossing are over; the adventure as to them and for which they paid freight and passage money is fully terminated; and it seems unreasonable to hold that the money thus paid by them should be made applicable as a fund for the relief of other cargoes and other passengers, who subsequently leave a place of safety in New York to be embarked independently in another maritime adventure from New York to Havre. Passage tickets are often issued covering a trip out and back, but that means only back by the same ferry, some other steamer, or the same steamer on some subsequent crossing, will usually carry them. And even if in some instances a few individuals may return by the next trip of the same steamer, that does not alter the situation. They do not remain on the ship. They are landed in a place of safety ashore. Their connection with the ship is terminated, and the adventure on which they originally embarked is at end. We concur with the District Judge in the conclusion that the freight and passage money for the voyage from Havre to New .York need not be paid over to the trustee.
3. As to the subsidy received from the French government. The contract under which this was paid has been put in evidence. It shows that the contractor (the petitioner here) undertook to operate a weekly steamship line from Havre to New York, 52 voyages going and returning, each 3rear; that it agreed to employ new or suitable steamers in number always sufficient to insure a complete performance of the service. The character, size, speed, equipment, etc., of the steamers are carefully provided for. The contractor engages to transport gratuitously all of the mails upon the line from Havre to New York. (Presumably this covers mails [438] from New York to Havre, although the language is not entirely clear.) The contractor also engages to transport gratuitously all gold, silver, and copper coins for the use of the state. The loss of one of the ships shall not be an excusé for interruption of service. The contractor shall only be authorized to temporarily replace the steamer lost by a steamer accepted by the minister, and shall be allowed a delay of 30 months to place a steamer in service fulfilling all of the conditions of the specifications. In the event that by reason of accident befalling one of the vessels the voyage commenced cannot be completed, the postal agent on board shall be authorized, if it can be done, to secure the transport of the mail by the first French or foreign steamer bound for their place of destination. The expense of this special transportation shall be retained from the payment of the subsidy attaching to the interrupted voyage. If the mails cannot be sent forward, “the passage not accomplished under the conditions of the present article shall occasion a proportionate reduction from the subsidy.” Various penalties are provided for, and also premiums for speed in excess of the requirement of the contract, and there are provisions as to the disposition of the ships in case of war. The compensation for the whole service is fixed at 5,480,000 francs, payable in equal monthly installments. We are of the opinion that this subsidy cannot be considered as freight paid for transport of the mails, Vb2 of the whole being arbitrarily apportioned to each round voyage. It must be treated as an entirety. It is the lump sum which the French government has seen fit to pay in order to stimulate the creation and maintenance of a fleet of steamers of a certain class, no old steamer being allowed in the 'service unless it “has been Frenchified before May 22, 1883,” and every new steamer to be built in France for the promotion of French industry; all sailing under the French flag for the promotion of French commerce. In one sense it may be said t-hat by making each double crossing the steamer making it earned a part of the whole sum, but what part it so earned we cannot say. The compensation covers not only the crossing steamers, but also those held in reserve to take ■the place of any which may break down, and, as to all of them, '“in every extraordinary political circumstance, even outside of the case of maritime war,” the right secured to the government to “purchase or take for freight one or more steamers,” the details of whose speed, capacity, and construction it has arranged for to suit itself. It seems impossible to say what fraction of the 5,480,000 francs is to be considered the compensation to any single steamer for undertaking to transport the mails on a single trip from New York to Havre.
4. The claims for damages by reason of loss of life. The disposition of these claims is already settled by decision of this court. “The territorial sovereignty of a state extends to a vessel of the state when it is upon the high seas, the vessel being deemed a part of the territory of the state to which it belongs; and it follows that a state statute which creates a liability or authorizes a recovery for the consequences of a tortious act operates as efficiently [439] upon a vessel of the state when the vessel is beyond its boundaries as it does when it is physically within the state.” International Nav. Co. v. Lindstrom, 123 Fed. 475, 60 C. C. A. 649. The Bourgogne upon the high seas was a part of the territory of France, and the record shows conclusively — indeed, the proposition is not disputed — that, if the Bourgogne were in fault for the collision (the French courts held she was not), the claims for damages by reason of loss of life could have been maintained in the French courts against the fund paid in, in limitation of liability. The decree appealed from should be modified accordingly.
5. The alleged faults of the Bourgogne. The English court, upon evidence mainly from the Cromartyshire, supplemented by three witnesses from the Bourgogne, and partly also upon calculations based on the injuries apparently sustained, found that the steamer was going at “full speed reduced.” The French court found that she was going “at twelve knots, about.” The District Judge has reached the conclusion that the speed was about 10 knots. A careful examination of all the testimony produced here has satisfied us that, although there may have been a reduction, she was certainly not going any slower, and probably was going faster, than 10 knots. It is unnecessary to rehearse the evidence. The statement in the opinion below is sufficient indication of the grounds for this conclusion. The character and extent of the wound received by the Bourgogne are suggestive of a high speed on her part. Undoubtedly the fog was exceedingly dense; that fact is uncontradicted; and the steamer had not “reduced her speed to such a rate as would enable her to stop in time to avoid collision after an approaching vessel came in sight, provided such approaching vessel were herself going at the moderate speed required by law.” The Chattahoochee, 173 U. S. 540, 19 Sup. Ct. 491, 43 L. Ed. 801. We are emphatically of the opinion that such a speed under the circumstances was excessive, and, since it probably prevented an earlier fog-horn blast being heard from tile Cromartyshire, it cannot be held not to have been a proximate cause of the collision. We express no opinion as to any fault of the Cromartyshire. We are not trying her cause, and none of her witnesses (except the captain) have been examined in this cause. Indeed, there is some question raised as to the admissibility of his examination. Moreover, with allowance of the death claims, even one-half of the damages found in this proceeding will greatly exceed the sum transferred to the trustee in limitation of liability.
6. As to the contention that the disaster occurred with the “privity or knowledge of the owner.” None of the officers of the defendant were on the Bourgogne, nor did any of them direct, interfere with, or approve her navigation. There was undoubtedly no direct privity or knowledge of this disaster on the high seas, thousands of miles from the home office of the defendant. The contention, however, is that petitioner never made proper regulations to govern its masters as to navigation of its vessels in a fog; and that it encouraged, or sanctioned, or knowingly tolerated a persistent violation of the rules of navigation (in that particular) on the part of [440] its officers. As we have seen, the line is operated under a contract with the French government, which imposed certain penalties for delays and stoppages “not justified by the circumstance of vis major,” and provided for the presence on each steamer of a so-called “agent of Postes.” The petitioner issued to all its captains a set of rules and regulations called “Reglement du Service a Bord.” Article No. 395 of such rules reads as follows:
“In conformity with the rules of international regulations having for object the prevention of collisions — ‘All vessels under steam which approach each other so that there may be risk of collision, must diminish their speed, or stop and go backwards, if necessary. All vessels under steam must, during foggy weather, preserve a moderate speed.’ The captain under these circumstances must diminish the speed of his engines and, in agreement with the agent of Postes, the captain must make known by proees verbal the delays which such manoeuvres may have occasioned.”
It is argued that this rule is insufficient, because it would be a sufficient compliance with it if, in a dense fog, a captain diminished speed somewhat from the maximum, although the diminution was not sufficient to bring the steamer down to “moderate speed.” This is not a fair construction of article 395. Possibly, if the international rule were not quoted; there might be room for argument that the article lacked explicitness; but with the full text of the international rule incorporated as it is, no captain could be in doubt that the diminution he was directed to make should be such as would conform his speed to the navigation which the international rule prescribed. Plowever positive and precise a rule may be, it will not avail the person who has prescribed it, if he encourages, or sanctions, or knowingly tolerates its violation. It is contended that petitioner did thus sanction or knowingly tolerate excessive speed, and claimants seek to establish that proposition by proof that the steamers of this line so frequently violated the rule of “moderate speed in a fog” that it must be assumed that petitioner’s officers knew of such violation, and, by not taking any steps to put a stop to such a reprehensible practice, must be held to have tolerated or encouraged it. There may be a widespread impression that the various Atlantic liners, which seek for and obtain the patronage of the traveling public by means of the quick' passages they make, are able to make such passages only by running at excessive speed in a fog. Nevertheless the court cannot take judicial notice that any particular line thus constantly violates the law, nor can it hold the owners of the line in damages upon the theory that they are persistent lawbreakers without proof of the fact. The evidence produced to show infractions of the “moderate speed” rule so frequent as to bring knowledge thereof home to the officers of petitioner is as follpws: During the 11 years preceding the loss of the Bourgogne there appear to have been three occasions when steamers of petitioner’s line were in collision with other vessels in a fog, and were held by the court which investigated the accidents to have been going at excessive speed: Da Bretagne with the Tellus, May 7, 1887; La Normandie with the Charlotte Webb, May 18, 1889 (58 Fed. 427); La Touraine with the Sully, September 10, 1894. These three detached occurrences are not especially [441] important, and they lose all persuasiveness when it is seen that subsequent to the last of them there are only two reported occasions when steamers of petitioner’s line were in collision with other vessels in a fog, and in both cases the courts held petitioner’s vessels to be going at moderate speed, and free from fault. La Bourgogne with the Ailsa, February 29, 1896, 86 Fed. 475, 30 C. C. A. 203; La Champagne with the Ville de Rio, July 4, 1898. Claimants called six witnesses who had made voyages on the company’s steamers, with the following results: Peters was in the employ of petitioner from 1886 to 1889 as cook, deck steward, and second steward, and made about 88 voyages on the Gascogne, Bourgogne, and Navarre. He testified that about six or seven voyages out of ten there were periods during which the weather was foggy; that the steamers never slowed down; that in mid-ocean they went full speed, but in the English Channel the speed was now and then reduced; that it was the same on the Hamburg Line and the Red Star Line, on which he had served. He had always seen them in mid-ocean, when it was foggy, going full speed to make the time. Levanoux, a flour merchant, had crossed frequently by the French line since the fast steamers were built in 1886 till 1893, his last trip in 1893 being made on the Bourgogne. On that trip there was a fog, lasting how long he could not say, but he didn’t notice any slacking of the speed, which he had noticed before the fog was full speed, and doesn’t recollect any diminution indicated in the posted daily runs. His recollection is that the same was true as to the daily records on other trips. He had noticed delay in arrival on this side occasioned by fog, but as to arrivals at Havre he “could not answer so very well.” There were some variations in time of arrival. His observations were made in a casual way. Baruch, an importer of embroideries, had crossed on steamers of this line most every year since 1895. He testified that during those voyages there were periods of fog, sometimes three, sometimes four, days, and that there was no slowing down. Riker, an importer of millinery goods, had made two trips a year since 1889; had noticed periods of fog, which might last from 12 to 24 or 36 hours, but never noticed any lessening of speed during such fogs; nor could he see that it made any difference in the daily runs. Steiner, who is engaged in business in New York, had made six voyages by the same line at intervals between 1885 and 1900. The only time he recollects fog was during a passage by L’Aquitaine in 1900. One day was quite clear, and the next they met a very thick fog, but at the end of the day had made as many knots as the day before, when it was clear. Burras, a buyer of flowers and feathers, had crossed by the line twice a year for 16 or 18 years, and never noticed any change of speed or in the announcement of daily run during periods of fog: Of the witnesses called by petitioner, Felsenberg, a commission merchant, had made three voyages by the same line — one by the Bourgogne (her last voyage out before she foundered) — and noticed they slackened speed during fog when they came to the Banks, and they had about two days’ thick fog. As to the daily runs, he re[442] membered that when the weather was dark or foggy the number of knots was not so great. On the trip by the Bourgogne they reached Havre Sunday morning, only a few hours late. Pagnon, a silk importer, who had crossed by petitioner’s steamers more than 60 times, testified:
“I have seen very often — not to say always — a reduction of speed in case of dense fog. Sometimes I counted myself the revolutions of the engine. In the old time, when they had side-wheelers, it was very easy to count. The revolutions were less numerous. At that time I remember pretty well that when there was a reduction of 20 per cent, we considered it fair. With the screw steamers very often I have seen, and I can be a witness to it, a sensible reduction; 15 to 20 revolutions, as far as I can remember. I think I have seen more too. I have seen those steamers stop in case of fog. * * * Q. How about reduction of speed, if you have observed any, on the Banks or about the Banks, or in mid-ocean? A. Tes, very often. I have seen reduction under the same circumstances. I have seen diminution of speed, coming nearly to a stoppage, in a place which is called by the French navigators the ‘Banquereau’ — the small banks, which are north of the large banks. I have seen a steamer there stop entirely on account of the fog, and when the fog disappeared then we saw a few fishermen there. * * * In fact, I don't think I have seen any part of the ocean where they did not diminish in speed in case of severe fog. And sometimes it happened to me to spend twelve days in a journey which we could have made in that steamer in nine days, because constantly we were going at reduced speed, except when there was a little opening. « * * Q. What are the limits that you give to a sensible reduction? A. I think I said also 20 per cent, and more, but usually, I think, 20 per cent. At 70 revolutions 20 per cent, would make it 56, and sometimes it went lower. Q. Are you prepared to swear that you can say that there has been reduction of speed more than 20 per cent, in fog? A. Sometimes, sometimes; yes. I went so far as to state that I have seen the boat stop. Q. I wasn’t asking about stopping. I was asking about reductions of speed of the engines. A. I have seen it go very slow; that is all I can swear, very slow; say not half; very seldom. I have seen them go carefully, feeling their way, and anxious and desirous to avoid any mishap. Q. These cases have been, as I understand you, in the English Channel, or approaching the lightship? A. And also in the Banks, where they often would have a great many fogs, you know. Everywhere where there is fog.”
Benson, agent for a silk house, had crossed by these steamers more than 60 times, had several times looked at the indicators in dense fog, and found that the speed had been reduced; the indicator was very often to half speed, or even lower than that, down to what they call slow; and he remembered that during fogs there was a decided difference in the day’s run. Metz crossed in the Bourgogne in May, 1898. She was then commanded by Capt. Deloncle, who went down with her. He noticed that “she was going at a slow rate during a fog of about an hour’s duration.” Two of the captains testified that they always slowed down in a fog, increasing the speed whenever the fog lightened up, as, in their experience, it frequently did.
As the District Judge remarked: “The determination of the question as to what is to be done in all the varying stages between a light haze and a dense fog rests upon a great variety of circumstances and conditions.” There is no fixed speed prescribed by the international regulations. It should be varied according to the distance at which objects can be observed by the navigators and lookouts, who are better situated to appreciate the density of the fog [443] than are observers from .the promenade deck, or the door of some saloon. The estimates as to speed given by passengers who are not experts are generally unsatisfactory. Upon the proof as it stands we cannot find that the petitioner’s officers knowingly tolerated or encouraged the running of its steamers at excessive speed in fogs, or were negligent in failing to enforce the rules. Certainly they used due diligence in securing officers of experience and ability. We concur in the conclusion that the disaster was “done, occasioned, or incurred without the privity or knowledge of the owners.”
7. It is contended that the ship was in actual violation of statutory rules as to equipment (life boats, floats, rafts, life preservers, boat-disengaging apparatus, etc.). We do not think it necessary to add anything to the opinion of the District Judge on this branch of the case. Counsel for one of the claimants criticises the statement that it was admitted in the District Court that section 4493, Rev. St. U. S. [U. S. Comp. St. 1901, p. 3058], is not applicable. Whether or not there was such an admission is unimportant, since section 4400 [U. S. Comp. St. 1901, p. 3015] expressly declares that “vessels of other countries” (which the Bourgogne was) shall not be subject to the provisions of section 4493.
8. In view of the findings above set forth, there was no error in overruling the exception to so much of the commissioner’s report as found that the S. S. White Dental Company had no claim.
Various minor questions as to admissibility of evidence, etc., need not be discussed here. We concur with the District Court in the disposition made of them.
The decree is reversed, and the cause remitted to the District Court for modification in accordance with this opinion. Inasmuch as both sides appealed and neither prevailed as to all the points argued, there should be no costs of this appeal.