KNAPPEN, Circuit Judge.
[1] On May 18, 1915, a coal company shipped on board the steamer Viking, then lying at Buffalo, N. Y., a cargo of coal, in good order and condition, for transportation to Sault Ste. Marie, Mich., at an agreed rate of freight, to be delivered to the consignee .in like good order and without delay, .dangers of navigation only excepted. The Viking left Buffalo during the afternoon of May 18th, arrived at Toledo in the late afternoon of the next day, and moored at a dock in that harbor, where she remained until she sank at 5:30 a. m. on May 21st. The cargo was insured by'the Providence-Washington Insurance Company, which paid the damage, took an assignment of the shipper’s cause of action, and was thereby subrogated to the shipper’s right of action. Federal Insurance Co. v. Detroit, etc., Ins. Co. (C. C. A. 6) 202 Fed. 648, 651, 121 C. C. A. 58. This libel was then filed by the Insurance Company alleging two causes of action: First, the failure to deliver the cargo in good order; and second, a wrongful deviation in the voyage to Sault Ste. Marie, Mich., by going to Toledo.
[2] As to the first cause of action, the answer alleged that the sinking was caused by the negligence or mismanagement of a member of the crew, and that claimants were accordingly exonerated by virtue of the Harter Act (Act Feb. 3, 1893, c. 105, 27 Stat. 445; U. S. Comp. St. 1916, § 8031), which relieves the vessel and her owners in case due diligence has been exercised to make the vessel seaworthy and properly manned, equipped, and supplied; as to the second cause of action, that the proceeding to Toledo was for the purpose of picking up the steamer’s regular consort, in accordance with the usage of the trade and the custom on the Great Lakes. The District Judge found neither of these defenses sustained, and held the steamer liable accordingly.
*803The question of seaworthiness arises out of this situation: The Viking was a freight boat of the single-deck type; i. e., she had but one main deck. When loaded, she sat low in the water and had but a small amount of freeboard. She was equipped with a 2J/t>-inch siphon for discharging water from the bilge, suction being created by the introduction of steam at a connection near the lower'end of the siphon, whose outboard end was at the 14 foot 3 inch draft line. There was no valve in the siphon, except a plug cock (which could be operated only by a large monkey wrench) located in the bilge, possibly 2 feet above the keel and 10 or 12 feet below the working platform of the engine (the engine room was on the main deck), and was reachable by a descent of six or seven steps. There was no way of telling whether the stop cock was open, except by the direction of a certain mark from the plug cock. When1 the vessel left Buffalo, she was drawing 14 feet forward, and 13 feet 11 inches aft, and the freeboard end of the siphon was thus above the water line. On the 20th, she coaled at Toledo, thus increasing her draft to 14 feet 7 inches, and making the discharge end of the siphon about 3 inches under water. When the boat was raised, the stop cock was found open, and it is clear from the record that the sinking was due solely to this condition of the siphon.
The District Judge, in rejecting the defense of seaworthiness, said that, in his judgment the boat—
•‘was not seaworthy to be loaded more than 14 £e,et. It seemed to the court ou the trial, and time has not remoyod this impression, that, if this boat were to bo so deeply loaded, the most ordinary precaution to make her seaworthy would liave been to have placed a cock near the point of discharge of the siphon where it would be readily accessible. Without such an obvious and important, yet inexpensive, addition to her equipment, she cannot be said 10 have started from Buffalo in a seaworthy condition when loaded so deeply. We do not believe that the Harter Act should lie so- construed as to pass to 1he cargo owner responsibility lor carelessness on the part of the crew when that carelessness operates on a failure of ordinary prudence when furnishing' equipment. ‘To send this vessel on a voyage with her [outboard] siphon end under water, and with no safeguard save a cock so inaccessibly 1 seated and awkwardly operated as in this case, is without excuse under any reasonable construction of the Harter Act.”
[3] This conclusion meets our approval. We do not think it open to the criticism that the District Judge supposed the boat left Buffalo with the discharge end of the siphon under water. The opinion was a mere summary memorandum, reciting no facts. Presumably, the court had in mind the situation after coaling at Toledo, after which the major part of the voyage was to occur, and which coaling was contemplated as a necessary part of the voyage. This being so, the situation is no different than if the coaling had been at Buffalo. The Southwark, 191 U. S. 1, 7, 8, 24 Sup. Ct. 1, 48 L. Ed. 65; The R. P. Fitzgerald (C. C. A. 6) 212 Fed. 678, 683, 129 C. C. A. 214.
[4] While there was testimony lending to show that such a siphon construction was seaworthy, it is not persuasive. The record impresses us otherwise, and that the more usual method was to have a globe valve on the inboard side of the vessel, reachable from the working platform, and which would plainly show whether it was open or closed. True, the *804testimony indicates that a vessel is not rendered unseaworthy by the mere fact that the discharge end of the siphon is allowed to be under water, , provided the stop cock therein, wherever located, is kept closed. The unseaworthiness as respects equipment lies in the fact that, if the valve is not kept closed and the outer end of the siphon remains under water, the ship is bound to sink sooner or later, if that condition continues long enough—coupled with tire fact of inaccessibility of the stop cock and the practical impossibility of closing it after a few feet of water have entered the hold, and the greater danger that the open condition of the valve will be left unnoticed, if located where it cannot readily be seen and operated by the engine room crew. The fact that the boat passed official and other inspections is not conclusive of the fact of seaworthiness. The R. P. Fitzgerald, supra, 212 Fed. at pages 686, 687, 129 C. C. A. 214, and cases there cited.
[5] Claimant contends, however, that this condition of the siphon, even if amounting to unseaworthiness, did not contribute to the sinking, for the reason that the engineer testified, without dispute, that when he arrived on deck, after being called by the watchman (all the remainder of the crew, as well as the master, being turned in for the night), the ship had so far settled that the water was pouring in through the gangways, portholes, and over the deck; and it is accordingly argued that, even had the valve been located near the discharge end of the siphon, its closing at this juncture would not have saved the vessel from sinking to the bottom. The watchman, who discovered the settling and called the engineer and master, was not produced (nor was any other member of the crew), and the defense of noncontribution assumes not only that the watchman called the engineer and master as soon as possible after he discovered the sinking condition, but that, when so discovered, the boat had settled to the point of letting the water through the gangways and portholes and over the deck, in the face of the natural improbability that a watchman actually on duty would haye failed to notice that the boat was settling until the water was coming in freely to the extent stated. But if these assumptions are to be indulged there remains the greater probability that a visibly open valve, accessibly located, would have been discovered by the engine force long before water began to come in to an extent making complete sinking inevitable.
[6-8] The defense of noncontribution also assumes that the open valve was due to negligent operation. The vessel being unseaworthy, fo say the very least, liability cannot be escaped without the clearest showing that the disaster was bound to have happened despite the unseaworthiness, and this burden we think not clearly sustained. As párt of the evidence of seaworthiness, the claimant presented the testimony of the engineer, who answered in the affirmative this question:
“Beaving Buffalo on this occasion, or before leaving Buffalo, can you state whether or not the stop valve, the safety valve, in the siphon pipe in the Yiking was closed?”
We assume that the witness meant that the stop cock was closed before and at the time the ship left Buffalo, and we assume that he thought *805such was the case. The view which the District Court took of this evidence does not appear; but, while it is not directly disputed, it is not convincing of the fact. In substance, the engineer’s testimony, taken as a whole, comes merely to this: That the last use which the engineer knew had been made of the valve was during the discharge of a cargo of corn, which immediately preceded the loading of the coal in question. There is no testimony of the actual closing of the valve, or that it was looked after or even thought of before the ship left Buffalo, or at any time after it was used during the discharge of the corn cargo until the ship was found to he sinking at the Toledo dock. The engineer knew of no use had of the siphon (or occasion for such use) after the corn cargo had been unloaded. There had been no occasion to use the siphon at Toledo. The fires were put out in the middle of the forenoon of the 20th, and the siphon could not have been operated after that time. The engineer knew of nobody who had opened the stop cock; on the contrary, he testified that none of the crew would admit any knowledge of it. The natural inference from the engineer’s testimony is that he had no affirmative recollection of closing the valve, but assumes that it was closed because it plainly should have been. The performance of that duty cannot be left to presumption. The Wildcroft, 201 U. S. 378, 388, 26 Sup. Ct. 467, 50 L. Ed. 794. As the discharge end of the siphon was out of water during the voyage from Buffalo to Toledo, and, indeed, until the coaling was nearly completed on the 20th, water would, so far as appears, normally not have come in from the sea in such an amount as to be discoverable even in the hold.
[9] It is not claimed that the valve was purposely left open on the boat’s departure from Buffalo, and thus that the failure to close it later was due to negligence in operation. The burden of showing that the valve was closed when the boat left Buffalo thus rested on the ship (International Nav. Co. v. Farr, 181 U. S. 218, 222, et seq., 21 Sup. Ct. 591, 45 L. Ed. 830), and we think that burden not sustained. The ship being unseaworthy, it is unnecessary to consider the question of deviation.
The decree of the District Court is affirmed.