The carrier relies upon three exceptions in the bill of lading—first, against “perils of the sea”; second, against “unseaworthiness”; third, against “collision.”
Section 3 of the Harter Act (Comp. St. § 8031) provides that the ship shall be exempt from liability for “perils of the sea” only if due diligence has been used to make her seaworthy. Therefore, if “collision,” without fault of the ship, be a “peril of the sea,” the ship must show due diligence. In the view I take of the evidence, it will not be necessary to decide that question, since I think that the scow has shown due diligence.
Section 2 of the Harter Act (Comp. St. § 8030) forbids the insertion of any exception in a bill of lading relieving the ship of due diligence to make herself seaworthy. The question here is whether an unconditional exception against unseaworthiness is within that section. This question, too, I find it unnecessary to decide, because I think that the ship has shown that the loss was wholly occasioned by a collision and that she was seaworthy at the outset of the voyage. By “collision” I include the possibility that the plank may have been broken by the jolting of the barges against each other without the actual impact of another vessel, though that seems to me by far the most likely explanation. The ship, of course, has the burden of showing that the loss was due to collision, and the evidence is all circumstantial. The case, as I view it, is resolved into the inferences to be drawn from what I regard as substantially undisputed facts, and turns upon the sufficiency of those facts as a basis for reliable deduction. I proceed to state why I think them adequate.
[1.2] When raised, the barge showed a break in the fourth strake down on the starboard quarter 10 or IS feet from the stern. If submerged, this would quickly have sunk her, and it was attributed as the cause of the loss by substantially all the witnesses. If actually at the water line, it could easily have been caused by any piece of wood caught between the barge and the Haverstraw, had the barges been squeezed together. It was not at all necessary that this piece of wood should be caught on end; a small, square, floating block would have been enough. The No. 11, a new barge outside the barge in question, was during the same night injured on opposite sides, her planks being cracked, as though she had been squeezed. It did not appear whether these injuries were in that part of the No. 11 which lay opposite the break in the barge .in question, but they may have been so.
The depth of the barge was 9 feet and the number of strakes either 9 or 10, their width either 10 or 12 inches. Bagger was not sure of *277the width, but thought them 10 inches; Devlin says.“about 10 inches”; Johnson, the carpenter, who would be most likely to know, says 12 inches. If he is right, as I am inclined to believe, the bottom of the plank, which was split for its whole width, was 4 feet below the deck. Both the witnesses, who saw the barge when fully loaded, say that she had a freeboard of 3% feet. If they are right, two things follow: The break could have been caused as suggested, and1 could not have occurred very long before the barge foundered. She was in fact fully loaded at 8 a. m., and did not sink until 5 o’clock on the next morning. If the break was made before she left Brooklyn, it must certainly have been well above the water line, a point to which I shall presently recur.
If we assume that the barge filled for some other reason than the break, we must suppose that she leaked. All the evidence contradicts that supposition. She had been on the dry dock within 10 months; she bore no other injury which would naturally have filled her. After she was raised, Bagger searched her seams and found them tight. He reported her as seaworthy, except for' the break. Johnson, Crear, and Gordon, disregarding Devlin, all had examined her shortly before she left Brooklyn and found' her sound. Devlin’s declaration to Jeffcott throws some doubt upon his testimony, but even so does not indicate unseaworthiness. In the face of this proof, the most plausible theory for the libelant is, not that she was generally leaky, but that the break had occurred before she started, and had not been discovered, and that she filled on the way over or while lying at either pier. The possibility that she did so fill I pass for the moment. At present I need only notice the improbability that such a break should have escaped the observation of four men, all going through her hold, although it must have been very apparent, and was necessarily at about the level of their chests as they went over her with a light. Of course, they may have all been untruthful; but, if so, I had no means of observing it at the trial.
But, supposing that they overlooked it, the question arises whether she could have lived as she did for so long. The libelant insists that the break was a foot above the water line. She lived for over 20 hours, during which time she was towed around from the East to the North River and lay at piers, subjected to the wash of passing steamers and to such pitchings as these might give her. Besides, even in those waters, the surface is not quite like a mill pond. Is it not a strong supposition that she would not have continually taken in water through that period ? How much we cannot tell, nor can we be sure that she might not have endured; but the probabilities seem to be against It.
Again, the libelant urges that she careened to port; but that proves nothing. Her midships partition was not a water-tight bulkhead. Their main reliance is .therefore upon Capt. Jeffcott’s calculation of what must have been her freeboard, which he finds to have been at least 5 feet. That is based upon the assumption that when light she drew 18 inches, and that her cargo set her down not more than 2 feet 6 inches. Hence the break was almost a foot or more above water, and could not have been caused by the squeezing of the barges together. This suggestion in the first place contradicts the testimony of *278both witnesses who observed the barge when loaded. Devlin says 3%• feet; Stein (certainly disinterested) says 3% to 3% feet. Crear says that before the loading was completed she had between 4 and 5 feet freeboard.
The estimate is that she drew 18 inches when light, and with 6 inches-of water in her bottom, but is at most merely a guess. She had a deck and a house covering the whole of her, except at the extreme ends. No one can possibly say what her draft would be without experience. Moreover, her rake was 7 feet at each end and her length over all 91-feet, so that her bottom was only 77 feet long. If sunk 18 inches in the water, she measured on the water line, as I make it, about 79 feet, and, if sunk to 4 feet, 83 feet. Her average length was say 82 feet. If there was 200 tons of cargo on her, as some of the testimony justifies one in supposing, she would be sunk about 2 feet 9 inches, and the break would have been only 9 inches above water. All this is purely speculative; but, even so, if she had a list of only one-half of one degree by the stern and one degree to starboard, the break would' have been substantially at the water line. If she were one degree by the stem and mathematically trimmed thwartships, the break would have been only an inch or so above water. In short, to be effective as an answer, the hypothesis presupposes conditions which are not realized in practice. If the break were at or within an inch of the water line, it would have been under water half the time, and the chance of her living for nearly 24 hours would be greatly lessened. When a barge is well trimmed, no one expects that she will bear the test of a spirit level. Hence, as it appears to me, we have no right to say that, even if her cargo, if strictly trimmed, would not have put her down to the-break all over, she was not so down at the place where the break actually was.
It appears to me, therefore, to be fairly proved by a preponderancéof evidence that the -break was caused by the same violence which squeezed in the sides of the No. 11, and that it was at the water line, or nearly enough above it to be the result of some floating object between-the barge and the Haverstraw. The most obvious suggestion is that a tug had come in during the night and collided with all three, but I need not commit myself to that supposition. It is enough that everyone assumed after the injury that the break caused her to sink, and that the weight of proof is that the break was not there when she left the East River pier.
As in all such cases, demonstration.is not possible; but the barge need not exclude all reasonable doubts. The probabilities seem to me to weigh enough in her favor to exonerate her from liability, and the-libel will be dismissed, with costs.