This case presents a libel in personam for damages by fire to the steamship “Dupont” belonging to the State of Maryland. The respondent is the Spedden Shipbuilding Company of Baltimore, which, at the time of the fire on March 17, 1934, was engaged under written contract with the State in performing certain work on the boat. At the time of the fire the boat was on the Marine Railway in the shipyards of the respondent, but the interior of the boat was in charge of employes of the State and not of the respondent. The work that was being done by the Spedden Company was the placing and welding of certain steel doubler plates on the starboard and port sides of the boat near the water line. The amount of damage done by the fire is shown to have been $24,378.24, and the libellant claims this amount with accumulated interest aggregating in all $27,046.55. The State has collected all of its loss with the exception of $250 (not covered by the policy! from the Federal Insurance Company, for whose benefit by subrogation the suit is thus principally brought.
The work to be done was covered by a simple contract in the form of a letter dated March 8, 1934, from the president of the Spedden Company to the Conservation Department of the State of Maryland, Mr. Swepson Earle, Commissioner. In substance it recited an agreement on the part of the Spedden Company for $3,800, to “fit y¿' doubling plate over inside strake on water line * * * on each side for a distance of 52 feet” *194at designated places along the water line; same to be electric welded. “Plates to be formed to present shape of hull and to have red lead freely used between plates and then drawn up hard before welding is commenced.” The agreement was evidently verbally accepted by the State and a few days later the boat was placed on the marine railway and the work of welding the doubler plates, which had previously been fitted to the hull of the boat on both sides, was begun on Thursday, March 15, continued throughout Friday the 16th, and was being further continued on Saturday, March 17th when, a little after ten o’clock in the morning, the fire was first discovered.
It should be said that the Conservation Department of the State of Maryland, whose activities it is well known relate very largely to the fish, crab and oyster industries in the Chesapeake Bay, has two boats, one the Dupont, and the other the McLean. Customarily the Dupont is used in summer and the McLean in winter. During the winter season the Dupont is customarily moored at the Spedden plant and it had been so moored during the winter season of 1933-34 prior to the repair work above mentioned. Shortly before ttndertaking the repairs it was discovered that a. leak had developed in the forward portion of the hull caused probably by the rather severe ice of that season which occasioned the necessity for certain repairs. As a result of determination to make these, it was also decided to add the doubler plates as covered by the contract mentioned. The repairs caused by the damage by ice were also apparently going forward at the same time and were covered by a separate contract which, however, is in nowise involved in this case. For some time prior to the fire of March 17, the steamer McLean had also been anchored temporarily at least near the Dupont; and the general custom was that one crew served both steamers, the main force, of course, being on the McLean, with some few members of the crew thereof giving necessary attention to the Dupont. One or more members of the crew slept at night on the Dupont to give necessary watchman service. Captain A. S. Creighton was in general charge of both boats. It appears that he was absent, however, from both boats at the time of the fire and for some few days prior thereto. He had delegated direct supervision over the Dupont to Capt. Samuel F. Creighton who was also absent at the time of the fire, having left Baltimore on Thursday to spend the week-end at his home on the Eastern Shore. Before leaving he had notified the Second Officer of the joint crew of the two boats, one Guy Andrews, to give necessary supervisory attention to the Dupont. The electric welding of the plates was, of course, being done on the outside of the steamer and the interior of the steamer was under the charge and control of the State employes at the time of the fire. On the Friday afternoon prior to the fire, Mr. Andrews had been personally present on the Dupont giving it supervisory attention and when he left about 4 p. m., while the welding work was still in progress on the outside, he instructed a seaman, one Dawson, to remain in the after cabin near' where the welding work was being done on the starboard side until the welders ceased work for the day. Andrews then went to the McLean for his supper and when the welders quit work for the day about five o’clock, Dawson also left the steamer, one Phillips, however, employed as an assistant machinist, staying on board for the night as watchman. It appears that Andrews had expected to resume personal supervision of the Dupont the next morning but did not do so because he was instructed by the First Officer of the McLean to go with it to a nearby point for coaling. It resulted that no member of the crew was present on the Dupont at the time of the fire except Phillips who was engaged in some repairs to the machinery in the engine room of the Dupont at the time of the discovery of the fire.
The size, deck arrangement, location of the machinery and of the doubler plates on the starboard side of the Dupont are all clearly depicted in a blueprint of the ship offered in evidence as Libellant’s Exhibit No. 1. From this it is shown that the Dupont was a steam vessel with length of 154.5 feet, breadth of 23.6 feet, depth of 12.8 feet; gross tons 303 and net tons 206. She had two decks, a main deck and a berth deck. On the berth deck aft in what are referred to as'the owner’s quarters, there were several staterooms, three on each side, being separated by a two-foot six inch passageway. On the starboard side there was a linen locker, five feet four inches wide, the door of which opened at the bottom of a stair which led *195from the berth deck up to the main deck into the main cabin there situate. There was a porthole in the linen locker which was the fifth on the starboard side counting from the stern of the ship. The cabin above on the main deck had a door leading to the deck which apparently had been locked by Phillips, the watchman, on his rounds of inspection the night before the fire, and was still locked upon the discovery of the fire. No inspection had been made of the after part of the ship on either deck on Saturday morning prior to the discovery of the fire.
The fire evidently originated in or near the linen locker on the berth deck aft."* On that morning the two welders who, with their welding equipment had been furnished to the Spedden Company by the United Electric Welding Company, which had been impleaded into the case under admiralty rule 56, had started work at 7.30 o’clock. It is clearly established by the testimony that the welding work, the whole of which on the starboard side aft extended from a point forward of the linen locker to a point considerably aft of the locker, had been completed on Friday afternoon to a point some four or five feet aft of the linen locker. The welding on the starboard side immediately outside the linen locker had been done on Friday. At the time of the outbreak of the fire the welding was being done some ten or twelve feet aft of the linen locker.
The welders were working on a staging on the outside of the ship and the porthole in the linen locker was some little distance above their heads so that they either could not or did not look into the linen locker while the welding work was progressing either on Friday or Saturday morning. When the Dupont was put up in winter quarters a framework had been built on her main deck which was completely covered over with canvas. The fire on Saturday morning was first discovered by some one on an adjacent ship or on the shore who called the attention of one of the welders to smoke coming somewhere from the ship in the after portion. About the same time the watchman, Phillips, came out of the engine room and noticed the smoke. He procured a fire extinguisher and went aft on the main deck to the cabin door which he found locked. Upon unlocking it he was met with a heavy burst of flame which forced him back, and he then sounded the fire alarm which in due course brought the City Fire Department, and the fire was extinguished doubtless as promptly as possible, but very extensive damage had been done, although the area of the fire was limited to the linen locker and the adjoining stairway and the expensive mahogany woodwork in the cabin above and its somewhat luxurious furniture and appointments. It is said the steamer although purchased a few years previously by the State at a cost of only $30,000, had been built about twenty years previously at a cost of approximately $900,000. Its interior evidently embraced very handsome and expensive woodwork and fittings.
It will have been noted that at the time of the fire the Dupont was to some extent in charge of the Spedden Company but also in part in charge of the State employes. The work of the Spedden Company was being done entirely on the outside of the boat and the interior was under the supervision of and in charge of Phillips actually, and nominally at least in charge of the State employes consisting of the crew of the nearby McLean. Therefore we do not have a case where the whole control and care of the boat at the time of the damage had been entrusted to a bailee for work to be done. It was a ca£e of divided control and supervision. And it has already been noted that the place where the fire broke out was apparently within the exclusive control of the boat owner with access thereto, under lock and key, in charge of the owner’s agents. Nevertheless it is the contention of the libellant here that the fire was caused by the electric welding work on the outside of the hull and that the bailee for work to be done, the Sped-den Company, failed to exercise ordinary care in the performance of the work which caused the fire; that is to say, that the bailee’s negligence caused the fire with the resulting damage.
The course of the trial was that the libellant offered rather briefly testimony which it may be assumed constituted a prima facie case. The Spedden Company then called some sixteen witnesses who gave in considerable detail testimony, the general purpose and effect of which was to explain fully the process of electric welding and to demonstrate to the extent possible that there was no reasonable probability that the fire was due to the *196work, and further that it had exercised ordinary diligence and care in doing the particular class of work. The libellant then offered some testimony in rebuttal. Comprehensively considered the testimony as a whole involved the production and examination of apparently all the witnesses to the facts, including both of the workmen doing the electric welding, the president and superintendent of the Spedden Company, and other of its employes, and the respondent also called as witnesses many of the State’s employes constituting some of the members of the joint crew of the Dupont and the McLean. The respondent also offered the testimony of a Mr. Boetcher, an assistant engineer of the Consolidated Gas Electric Light and Power Company of Baltimore, as an expert familiar with the process of electric welding, under whose supervision a number of detailed experiments had been conducted, under conditions said to be similar to those of the particular work, to demonstrate that the fire in this case could not have been caused by the welding process.
Under these conditions we are not in this case concerned with the legal problem of the burden of proof as a procedural matter, that is to say, in the sense that the term “burden of proof” is used to describe the duty of going forward with the evidence during the progress of the trial, because in this case both parties have in turn gone fully forward as far as the available testimony permitted. It follows that we are here concerned with the burden of proof (which was much discussed in the argument of the case) only in its primary meaning which describes the duty resting on the libellant in this case to establish by a preponderance of evidence the propositions which are essential to the successful maintenance of its action. See Kohlsaat v. Parkersburg & Marietta Sand Co. (C.C.A.4) 266 F. 283, 284, 11 A.L.R. 686; Jefferson Standard Life Ins. Co. v. Clemmer (C.C.A.4) 79 F.(2d) 724; Southern R. Co. v. Prescott, 240 U.S. 632, 641, 36 S.Ct. 469, 60 L.Ed. 836. In my opinion it therefore results that the libellant, to maintain the respondent’s liability in this case, must establish by the preponderance of the evidence that (1) the respondent, the Spedden Company, failed to exercise ordinary care as bailee in doing the electric welding work and (2) the fire was caused as the proximate result of its negligence. This very subject-matter was very fully considered by the Circuit Court of Appeals for this Fourth Circuit in Newport News Shipbuilding & Dry Dock Co. v. United States, 34 F.(2d) 100, 103, 104, in Judge Northcott’s opinion. Therein are reviewed a number of prior decisions of the same general legal character, in some of which, as in the particular case then under consideration in admiralty, a libellant sought to establish liability on a shipyard for fire damage to a ship. In that particular case the court held the liability was established by reason of the fact that the shipowner made out a prima facie .case which the shipyard failed to rebut by affirmative proof on its part as to the circumstances and conditions of the origin of the fire. In that case the fire occurred in a stateroom when only employes of the shipyard were present, doing plumbing work which involved the operation of a lighted open flame furnace without a helper, in a stateroom recently painted. The shipyard did not call as witnesses any of its employes who were present at the origin of the fire. The contract in the case under which the work was being done also affirmatively called for a high measure of diligence and care in the performance of the work. The facts in that case are very different indeed from those now under consideration.
In a later case in this Circuit, Thompson v. Chance Marine Construction Co., 45 F.(2d) 584, 585, in affirming a decision by Judge Coleman in this court in a somewhat similar case in admiralty, the Circuit Court of Appeals in a per curiam opinion said:
“We hold the proper rule to be that the burden is upon the bailee to prove absence of negligence upon his part, but we are of the opinion that this rule may not be extended so far as to cast upon the bailee the burden of proving the origin of the thing that caused „the damage. It may well be, and in fact often is, the case that the origin of a fire is not known and is not susceptible of proof, and the bailee sufficiently carries the burden imposed on him by proving absence of his own negligence or that of his employes.”
After appropriate reference to ■ the Newport News Case, supra, the opinion continued:
“But in no instance do we find the bailee held responsible where he had made full disclosure of all facts within his *197knowledge or within the knowledge of those under his control, and has proveii to the satisfaction of the court absence of negligence on his part.”
It will be noted that in these two cases the court speaks of the burden of proof as to absence of negligence being on the bailee, but as I read the cases the expression is there used in the procedural sense of going forward with the evidence or as equivalent to the “burden of explanation.” I do not understand that what was there said on this subject was intended to be a departure from what had been expressly previously ruled in Kohlsaat v. Parkersburg &' Marietta Sand Co., supra, cited with apparent approval in both cases, and which it may be noted had only recently theretofore been cited with approval by the Supreme Court in the Malcolm Baxter, Jr., 277 U.S. 323, 334, 48 S.Ct. 516, 72 L. Ed. 901, in connection with Southern R. Co. v. Prescott, 240 U.S. 632, 640, 36 S.Ct. 469, 60 L.Ed. 836, supra. See also in this connection the annotation to the Kohlsaat Case, 11 A.L.R. 690; and the discussion of the Newport News Case in 71 A.L.R. 777. See, also, in relation to the same general subject matter of the burden of proof in such a case, National Sugar Refining Co. v. Tietjen & Lang Dry Dock Co. (D.C.N.Y.) 33 F. (2d) 354, affirmed (C.C.A.) 33 F.(2d) 356 (involving a factual situation generally similar to the instant case) ; and Warren & Arthur Smadbeck, Inc., v. Heling Contracting Corp. (C.C.A.2) 50 F. (2d) 99, 101; Vol. V, Wigmore on Evidence (2d Ed.) § 2508; Beck v. Wilkins-Ricks Co., 179 N.C. 231, 102 S.E. 313, 9 A.L.R. 559; 6 C.J. 1158, 1160.
Applying these principles to. the facts of this case, and looking at the effect of the whole testimony in the case, I find that the libellant has failed to establish by a preponderance of the evidence the proposition of fact which it is essential for it to successfully maintain in order to establish the liability of the respondent in this case.
In the first place, as to the respondent’s alleged negligence, the weight of the evidence seems to me to support the view that the respondent exercised at least ordinary care in the performance of the electric welding work. The positive testimony on this point is practically limited to that of the respondent’s president and superintendent, supported to some extent at least by the testimony of the workmen who were doing the actual welding. This testimony was to the effect that work of this character on exterior doubler plates on the ship’s hull was not generally known to be attended with the danger of causing a fire to any interior part or contents of the ship, and that in fact these witnesses knew of no fire which had been caused by such work under comparable conditions. The distinction between welding work done by electric power and that done by the oxy-acetylene blow-torch is emphasized. In the latter there is much more danger of causing fire from the nature of the process. But the only fire danger from electric welding would lie in the super-heating of the steel plates which are being welded, and the presence on the inside of the plates of some very combustible matter in immediate contact therewith. The existence of such conditions would be very unusual, as the inside of the plates (called the skin of the ship) is ordinarily separated from the interior woodwork by an air space of one or more inches, caused by the presence of the angle-irons of the hull construction. Where electric welding is being done in proximity to gas or oil tanks or known highly combustible or inflammable material, it is customary and ordinary prudence to make a careful inspection before the work is begun. But no such conditions were known or anticipated in this case, and in fact did not exist. The work was not of a new kind or character to the Spedden Company which has been in business for a great many years in Baltimore, and has done similar work on ships over a period of twenty years or more in the past without any fire damage ever having occurred. And this general experience is supported by the particular experimental tests made by the respondent including those conducted by the expert witness, Boetcher, the general conclusion from which would seem to be that the temperature engendered in the steel plates on the inside of the ship in this case could not have been more than about 400 degrees, and even this temperature could not have been maintained in any one place for more than a few minutes. This testimony as to the known dangers attending such work or the absence thereof, and the general experience of people engaged in this class of work, has not been contradicted by any positive testimony pro*198duced by the libellant. Notably no witnesses from shipyards engaged in similar work were called as witnesses in this case. On this issue the argument and contention of counsel for the libellant is based purely on what it puts forward as common senses in such cases. It calls attention to the fact that there was a fire and urges that it should be found that the fire was caused by the welding, and because a fire was so caused, as they contend, it should be inferred that it was due to absence of common precaution. It is true, I think, that the prima facie case in this respect was made out by libellant’s witnesses but it has been rebutted by the evidence submitted by the respondent. At least I cannot find that on the whole testimony the libellant has established by a preponderance of the evidence that the respondent failed to exercise ordinary care, which was all that was required from it in this case. International Mer. Marine S. S. Co. v. W. & A. Fletcher (C.C.A.2) 296 F. 855. Counsel for the libellant contend that the burden of persuasion as to absence of negligence was on the respondent, on the whole testimony in the case, but even if this were so (and my view as already stated is to the contrary) nevertheless I find the preponderance of the evidence affirmatively establishes that ordinary care was exercised in this case by the respondent.
The specific fault attributed to the respondent is its failure to warn the libellant’s agents of the danger of fire from the welding, so that a careful watch would be maintained on the interior of the ship during the welding operation. But this assumes that the danger was known, or should have been known from experience —a fact negatived by the testimony in this case. And it may not be without significance that the libellant’s agents did maintain a watch on the interior of the ship near the welding operation during Friday (though stated not to have been for the particular purpose) when, as we shall hereafter see, the fire must have originated if it was caused by the welding; but neglected to do so on Saturday when the fire was discovered.
Hindsight is, of course, better than foresight, and it may very well be that the experience gained by this case, involving as it does the possibility that the fire may have been caused by the welding operations, should require a greater measure of care to be exercised by the respondent in the future in doing this work, including a more detailed and careful inspection of the interior of the ship, than was given in this particular case. But in the absence of affirmative testimony that such greater care is customarily given by others in the performance of similar work, I cannot reach the conclusion on the testimony in this case that there was failure to exercise ordinary care commensurate with the knowledge and information of the » respondent with regard to the dangers or lack of them involved in the welding work.
Nor am I able to find, from a preponderance of the evidence, that the fire was in fact caused by the welding operations. It is true that it may possibly have been so caused in this case but to affirmatively find that it was so caused involves finding as facts a number of things which are highly speculative and conjectural. To make this clear requires some further detailed statement as 'to the electric welding process. This consists in applying to the small crevice between the superimposed doubler steel plates and the adjoining exterior plates or strakes of the hull of the ship, a metal electrode which, under energy of electric power, melts both itself and the surface of the two steel plates so that a complete fusion is made and the two plates are welded together. Before the heating begins the added plates are very carefully fitted on the side of the ship and then drawn tight so that the edges of the plates (called butts) are brought very closely together in a close or tight fit which, however, may be as much as one-eighth of an inch apart. The end of the metal electrode which, of course, is like a long thin pencil, is, however, wider than one-eighth of an inch, and could not well be inserted into the interstices between the plates so that it can come into direct contact with the lower plate or strake on which the doubler plate is being superimposed. In practice, the edges of the doubler plates are partially filed away or bevelled off to receive the end of the electrode which is inserted into the widened crevice made by the bevel, and deposits itself as a bead and simultaneously fuses together the edges or butts of the adjoining steel plates. The plates are one-quarter of an inch thick, and it will have been noticed that the doubler plates are superimposed on an existing lower plate *199also one-quarter of an inch thick. It is the contention of the libellant’s witnesses in this case that the welding fused together not only the edges of the doubler plates with the contiguous external plate already existing on the ship, but that the doubler plate was also fused to the underlying existing steel plate and that to accomplish this latter the electrode had to and did penetrate to the exterior surface of the inner plate. This, however, is disputed as a fact by the respondent’s witnesses, and as theirs is the only affirmative testimony as to what actually was done on this particular ship, I am unable to find as a fact that the libellant’s contention, which is based on opinion rather than fact, is established by a preponderance of the evidence. The importance of the disputed fact has relation to the ultimate question of what degree of heat was engendered on the inner side of the inner plate. Libellant’s witnesses contended as a matter of expert opinion that these temperatures must have been possibly 2000 degrees or at least 1200 degrees; while the respondent’s contention is that the temperature would not have been over 400 degrees. The controversy is whether the inside plates, forming what is called the “skin of the ship,’’ were heated at any time during the welding process to red or glowing heat which occurs at a temperature of about 1200degrees. The libellant’s expert, Mr. Haight, a marine surveyor who surveyed the ship after the fire on March 23, carefully examined the skin of the ship as exposed in what had been the linen locker, and testified that he found in several places the white paint coating on the skin of the ship along the line of some of the welding had been burned off, exposing the surface of the metal which, when scraped with a knife, showed a blue color which in his opinion indicated that the metal had been made red hot by the welding process. On photographs taken after the fire showing the interior condition of the linen locker he pointed out at least two places which he said indicated the conditions above mentioned. He gave it as his opinion that the fire was caused by this superheating óf the skin of the ship, in contact with some very combustible material such as feather pillows. And there was testimony to the effect that the linen room contained woolen blankets, linen sheets and feather pillows. It appeared also that in the linen locker against the steel sides of the ship there was built in a wooden cupboard but at one place between the top of the cupboard and an extra upper shelf, the skin of the ship was visible although it had been painted with several coats of white paint. That is to say, while most of the wooden inner structure of the ship was built inside of the steel hull, two or three inches away from the skin of the ship, nevertheless in the linen locker there was a space of several feet of the skin of the ship exposed and open to contact with things that might have been placed upon the shelving. Mr. Haight also testified that he found the same conditions as to paint and metal with regard to apparent overheating visible in some of the other staterooms when the wooden work was removed, which however had not caused fire.
But with regard to the actual condition in the linen locker, there was no testimony that feather pillows had in fact been placed on the shelf where they could or would have come in contact with the particular spot on the skin of the ship which had been as alleged so superheated. It is true that Mr. Haight said he found some feathers on one of the shelves six days after the fire, but there was explanatory testimony as to this to the effect that much of the contents of the linen locker had been thrown out of the porthole after the fire by the firemen. Clearly there was no positive testimony that any particular inflammable material was on this shelving at the time of the fire.
Again it is to be noted, in relation to actual combustion from the superheated metal that the process of electric welding on the outside of the ship was not continuous in a particular spot for a considerable period of time. The testimony was that a foot of electric welding, horizontally or vertically, can be done in about four minutes. It would, therefore, appear that no particular spot of the size indicated in Mr. Haight’s testimony, which he said must have been superheated, would have remained superheated for any great length of time continuously.
We, therefore, have the conditions necessary to gratify the facts assumed by the libellant’s expert to involve the following: (1) that the inner surface of the metal was superheated to at least 1200 degrees; (2) that this occurred in a particular spot where there were feathers or some equally highly combustible material in imme*200diate contact; (3) that it continued long enough to actually cause combustion, and (4) that this combustion originated some time Friday before 5 P. M., and did not manifest itself for at least 17 hours thereafter, to wit, 10 o’clock the next morning. The weight of the evidence is against (1), and the existence of (2) and (3) are wholly conjectural.
The fourth of these required facts grows out of the consideration already stated, that there was no welding work done outside of the linen locker on Saturday morning. And it is conceded by libellant’s counsel that if the fire was due to the welding it must have originated some time Friday before 5 P. M. It is true that the libellant called a fire expert, the Chief of the Baltimore Salvage Corps, who testified that cases of smouldering fires where there was not free access to air as in a small room, more or less sealed from the air, have been known to exist for two or three days before being discovered. But such cases are exceptional and there is no positive testimony in this case that this condition did occur here.
The respondent offered testimony tending to show that the fire, which admittedly occurred in or near the linen locker, might well have been caused by defective electric wiring resulting in a short circuit and a consequent fire. Considerable testimony was given as to the nature of the electrical wiring throughout the Dupont. And it may be safely said that the testimony very clearly showed that the wiring was done in what is now an obsolete way which would not be approved under any conditions for house wiring and much less for electrical wiring on a ship. There was also positive testimony that there was a short circuit in the electrical wiring in the ceiling of the linen locker, after the fire, and that the ceiling was made of cardboard. But whether this short circuit was the cause of, or was caused by, the fire, is speculative. It is, however, I think, unnecessary to pursue this line of inquiry in detail. It is sufficient to say that the described conditions were such that it is possible the fire did occur from defective electrical wiring but there is no more certáinty that this was the cause than that the cause was the electric welding. Both are speculative and conjectural. In this connection reference again may appropriately be made to the above quotation from Thompson v. Chance Marine Const. Co. (C.C.A.) 45 F.(2d) 584, 585. Here, as there, the burden is not on the bailee to positively show the true origin of the fire which often, as in this case, is very difficult to determine. There were other minor facts and circumstances covered by the testimony bearing on the possible origin of the fire, some of which seemed to point to the welding operations as the probable cause, others with probably equally probative force indicating the defective electrical wiring as the cause. The photographs of the conditions in the linen locker and the adjoining hallway with respect to the electrical fixtures after the fire and the course of the burning of the fire, lead to different and varying speculative conclusions as to the exact place of origin and cause of the fire. It is unnecessary to further analyze these considerations which in the ultimate analysis are all purely speculative.
Considering 'all the testimony in the case and the reasonable inferences therefrom, I find as ultimate facts that the libellant has failed to establish by a preponderance of the evidence either the alleged negligence of the Spedden Company or the cause of the fire as attributable thereto. On the contrary I find the respondent has successfully carried the burden, if on it, to show absence of negligence on its part. For these reasons I reach the conclusion of law that the libel must be dismissed with costs allowed to the respondents.
I should add that the admiralty jurisdiction in this particular case was previously established by Judge Coleman after hearing the exceptive allegations of the respondent with regard thereto. I have, of course, accepted and acted on that view of the jurisdiction in the case.