There is some question, upon the proofs, as to whether the cargo of the Squires was insured solely by a certain certificate of insurance, or by a more elaborate form of policy to which the certificate referred. This question need not be discussed, since the result would be the same, whichever way.it were answered. If the sinking of the boat were caused by impalement on an unknown obstruction alongside the dock, the insurance company would be liable. If it were not so caused, the evidence warrants no other conclusion than that the accident happened because of the boat’s unseaworthiness, and in that case the insurer would not be responsible.

Except as his testimony is corroborated or corrected as to time, the sole witness to the accident is the captain of the canal boat. According to his narrative, loading began at 9 a. m., and was finished about 3 p. m. The superintendent and assistant superintendent make it later, which seems more in accord with the captain’s further statement that after she was loaded, about 5 p. m., he shifted her position, hauling down nearer the mouth of the slip, so as to be more conveniently placed when the tug should come. The tugman had' told him the tide would not be right for going over the bar at Kings-bridge until the morning. The entire cargo (55 tons) was placed on-deck. There is evidence that the Squires had carried more than that weight safely, when distributed, part -on deck and part below. She lay bow in, and, when the loading was completed, had a list to port; i. e., inshore or towards the dock. Calculating from the tide tables, it appears to have been low water at Hell Gate about 4 p. m. Some of the witnesses testified to a half hour’s difference between Hell Gate and Mott Haven Canal, and the District Judge found it was low water at the latter place at 4:3o p. m. Appellant criticises this finding, and contends that it should fix the time an hour later. In the view we take of the case, the additional hour of ebb is immaterial. During loading the boat was- pumped twice—at 5 p. m., and apparently again at 6 p. m. The most water she had in her at any time until 6 p. m. (not an unusual amount of leaking for her) was four inches on one side, and a half inch on the other, at the bow, and, at the stern, two to three inches on one side, and nothing at the other. The greatest depth was on the side which listed towards the dock. About 6:30 p. m. the captain went ashore for supper, returning about 7, or half past, when he found the same amount of water,- and pumped again.. He then went ashore to get some oil for his lamp, and some time after his return, while he was pumping, he felt her list the other way,. *863He thereupon took his lamp, went down the stern hatch, and walked up toward the middle, and heard the water pouring in. He fixes this at about io p. m. or later. The testimony of other witnesses seems to indicate it was not long after 8 p. m. He got a line ashore, made it fast to a freight car, hurried to the house of the watchman on the dock, tried unsuccessfully to telephone for a steam pump, and then went off to get some men to help unload. When he got back, shortly before midnight, he found the boat had sunk.

The theory of the libelant is that the causé of the accident was a hidden obstruction. About a month after the accident it was discovered that there were two short stumps of small piles alongside of the dock where the boat had been berthed. The top of one of them had been broken off and splintered so that, if opportunity afforded, it could penetrate a boat’s bottom. Of course, it needs more testimony than this to warrant a finding that this broken pile actually penetrated the side or bottom of the canal boat; and, in our opinion, libelant has failed to establish the necessary connection between the two. As to the wound alleged to have been received by the boat, the evidence is unpersuasive. It is as follows. About three days after the accident, the captain took her out and tried to beach her himself, but could not. So he hired a tug to shove her on. That did not suit. So, after she had lain on the beach about ten days, he got the tug to haul her out again, and “put her up higher, so it would be better—get her higher up.” No survey of her was made till a month after the accident, and it disclosed no penetrating wound, such as would be made by a sharp-pointed pile. The entire testimony as to survey is as follows: The captain says that he took the two surveyors—McCarthy for himself, and Onderdonlc for insurance company—up there in a rowboat; that the boat was lying in the mud, and the condition of the tide such that they could not see whereabouts the leak was—could feel it under water with a pole, at the turn of the bilge. The water passing in and out had washed a hole in the mud. Witness had noticed the place about a week before, when the tide was lower. The bilge plank was torn off. McCarthy says he went with the captain and Onderdonk in a rowboat at low water, and found that on the port side, somewhere about amidships, her bilge plank was gone. “There was a. round bilge one center plank, and one each side of it, and the center plank was gone.” Onderdonk confirms this testimony, and adds that she also had a big hole on the port side nearest the stern, but that seems to have been caused after the accident by the tug putting her on the beach. Manifestly, the boat suffered hard usage after the accident—pushed ashore and pulled off again, and lying meanwhile unwatched and exposed to contact with whatever floating stuff might be thrown against her. If there were nothing else in the case, it would be an unsatisfactory conclusion that would find a causal relation between the sunken pile and the abandoned wreck of a. month later. But there are other points in the case. As was said before, the District Judge reached the conclusion that it was low water at 4:30. Appellant contends that the Harlem river tides are somewhat erratic; that a strong wind will delay or advance them considerably (according to the captain’s *864story, there was no strong wind the day his boat sank)'; that the ship •canal has produced changes. But upon all the evidence in the case, we cannot find any warrant for holding that the District Judge was more than an hour out of the way in his calculations, or for fixing low water any later than 5:3o p. m. It is evident from the testimony that the captain’s recollection as to time is quite inaccurate. Appellant’s counsel proves by documentary evidence that he is some two hours out of the way as to time he discovered the water coming in. He is the only one who fixes the time of loading as early as 3 to 4 p. m. Of the other witnesses called by libelant, the superintendent says that the feed was all on by 3 o’clock, or half past 3, but that the hay was not put on until the last thing, and that was along toward night. The assistant superintendent says that the loading was finished between 5 and 6; nearer 6 than anything; about the time witness was leaving for the day. It seems to be fairly established that it must have been about half-past 5 o’clock when the captain changed the boat’s position half her own length. He himself says it was as late as 5 p. m. It is evident that when he moved her she was not impaled on a pile, and the few minutes left before the rise of the tide began could not have brought her so far down as to cause her penetration by a pile over which she was floated at 5 :3o p. m.

.It also appears that the exact location of these piles (there were two of them) has been fixed by a survey with measurements made by a diver. Their tops extended to about four inches below low-water mark, and about one inch in front of the line of the outer faces of the fender piles, and they measured about five inches across. When we take into consideration the fact that the boat was not moved to the vicinity of the pile ,until 5 p. m. or later, that the alleged wound was in the turn of the "bilge—a point somewhat further from the dock than the line of fender piles against which the boat’s side rested—■ and that continuously, during the entire time subsequently to loading, and before she was found to be leaking dangerously, the boat lay with a list towards the dock, thus throwing her bilge plank still further away from it, we are unable to reach the conclusion that the ■sunken pile pierced the bilge plank, as appellant contends. There is a more satisfactory explanation of the accident to be found in the record. The boat was a very old one—weak and patched up so as barely to navigate the still waters of the upper Harlem. The owner bought her for $150. He had recently, as he testifies, “had her calked and fixed up good to her 4 or 5 ft. mark, about light-water line.” He says that, with the load she had on her, the “water line would be about 3 ft. 3 or 4 inches—maybe a little more, maybe a little less.” There is not much margin between a 4-foot mark and one 3 feet 4 inches—“maybe a little more.” Although the load was not an excessive one, it was all on deck, and the boat was undoubtedly tender. It is conceded that she had a list—how much, we do not know, but it would take very little to bring her port side down till the water reached above the line where she had been “calked and fixed up good”; and, once she began to leak seriously, so old a boat would be likely to develop quickly new elements of weakness. Upon the proofs, we are not satisfied that the hidden obstruction caused the *865injury, nor that the boat sank from any cause other than her own inherent weakness.

The decree of the District Court is affirmed, with costs.