GILBERT, Circuit Judge.
The Indrapura was a British steam vessel for the carriage of freight, and was of the type known as “water *713ballast” vessels. Water was supplied, to and discharged from her peak ballast tank by means of a pipe extending therefrom aft through the holds of the ship to the engine room, where it was fitted with a stopcock, and where the operation of filling or discharging was carried on. There were also means for filling the tank from the deck. The pipe, at the time of the voyage in question, which was in 1903, had been in service some two years. It was a 31/i> inch cast-iron pipe, % to % of an inch thick. It was put together in sections of some six to eight feet in length; the joints being connected by lead in order to ease the same with the roll of the ship. The pipe was laid on the floor of the vessel above the ceiling. It was boxed in a wooden case made of two-inch lumber to afford it protection against heavy cargo. It contained no valve either inside the tank or abaft the collision bulkhead. The vessel had been built five or six years previous to tlie voyage under a Lloyd specification, and she had an A-l rating and carried an A-l certificate, which had been issued on the inspection and survey made just prior to the voyage. Just before the voyage, the filling pipe had been tested by a pressure test of 300 pounds, according to the testimony of the ship’s engineer, and 400 pounds, as testified by the master. She left Yokohama March 12, 1903; her destination being Portland, Or. On her voyage she met with stormy weather, which continued for three or four days, during which she pitched and tossed heavily, and shipped water fore and aft. On March 23d her screw was racing all day, and on the following day the usual daily soundings disclosed that the peak ballast tank was leaking. The tank was pumped out, and, on arriving at Portland, it was found that the filling pipe had broken off between joints, letting water into hold No. 1, where the appellees’ gunnysacks were stored, causing injury thereto for which the libel in this case was filed.
The question presented to the District Court on the libel was whether the vessel was seaworthy as to cargo by reason of the fact that the filling pipe was not fitted with a valve either inside the peak ballast tank or stopcock abaft the collision bulkhead, or the fact that the filling pipe was not placed beneath the false bottom. Upon the evidence the court found the ship liable, and said:
“A reasonable precautionary measure would have been to fit the pipe with a valve to be operated by a. rod from the doelt or at some convenient point above the cargo. I'or the want of such a valve, I hold the vessel was unsea-worthy as to cargo in hold 2\To. 1.”
11 ] The question whether the vessel was unseaworthy in carrying the filling pipe through the hold or for want of means for excluding water from the pipe, either immediately within or immediately without the peak ballast tank, is one which should be considered, not only in the light of the evidence given by expert witnesses as to the usual and accepted construction of ballast tanks and their fittings, but: also in the light of' what would appear to be the obvious and prudent method of construction.
The expert testimony was more particularly directed to the question of the proper location of the filling pipe — whether it should be above or below the false bottom — than to the question of the necessity and *714feasibility of a valve' within or immediately without the tank. The expert testimony as usual was conflicting. For the appellee there were shipyard sujierintendents and engineers who testified that it is not proper marine construction to permit the drain pipe to run through the cargo; that it should be carried beneath the false bottom, so that in case of breakage the water would not reach the cargo; that, if it is placed above the false bottom, there should be a valve within or immediately without the peak ballast tank to shut off the water in the tank from the pipe. On the other hand, witnesses equally qualified testified that it was recognized as proper marine construction to place the pipe as it was placed in the Indrapura, and there was evidence that many freight ships were so constructed. One witness testified that it was highly improper to fit a valve forward of the collision bulkhead because of the difficulty of getting down to it in case it became obstructed, which he testified was likely to occur. Upon this conflicting testimony some difficulty is met in arriving at a conclusion. The fact must not be lost sight of that the seaworthiness of the ship Indrapura is to be measured by the standard of the time of the voyage in question. There is evidence that it is not unusual at that time to construct and locate the drain pipe as it was in the Indrapura. In Tidmarsh v. Washington Fire & Marine Ins. Co., 4 Mason, 439, 441, Fed. Cas. No. 14,024, Judge Story remarked that the standard of seaworthiness had been greatly raised within the last thirty jrears and in Burges v. Wickham, 3 B. & S. 693, Blackburn, J., said that the ‘‘standard of seaworthiness must rise with the improved knowledge of shipbuilding and navigation.” Judge Brown, in The Rover (D. C.) 33 Fed. 515, said:
‘'Scawortliiness does not require perfection in machinery more than anything else. Perfection is unattainable. Only a reasonable fitness for the service designed is required.”
And in The Lizzie Frank (D. C.) 31 Fed. 477, it was said:
“Where a vessel is constructed and equipped in the mode usual and customary with other vessels of like character, and in a mode approved by competent fudges and previous experience, then, in case of an accident happening by reason of a latent defect in the equipment and construction, there is no negligence on the part of the owner.”
In The Titania (D. C.) 19 Fed. 101, 102, Judge Brown said that the question of seaworthiness “is to be determined with reference to the customs and usages of the port or country from which the vessel sails, the existing state of knowledge and experience, and the judgment of prudent and competent persons versed in such matters.” But, while it is proper to consider evidence of the usual custom of shipowners and the usual method of construction of ships and their appliances, such evidence is not necessarily conclusive. It may be rejected altogether where the construction is obviously defective. Gilroy Sons & Co. v. Price & Co. (1893) A. C. 56.
[2] There is also to be taken into consideration the judgment of prudent and competent persons versed in such matters. In view of the testimony of such persons, which is found in the record here, to the .effect that the construction and maintenance of the filling pipe above ■the false bottom without a valve was faulty, and in consideration of *715the existing conditions and what appeared to tbe trial court to have been the obviously prudent thing to do, that court found that a valve in the pipe would have been but a reasonable precautionary measure. The case comes to us with this finding in favor of the appellees; and, although if the question were now first presented to us upon the evidence there might be serious doubt as to the proper answer thereto, we are not convinced, in view of the fact that such pipes in the holds of vessels have not infrequently been broken, as must have been well known to navigators, that common prudence did not demand the insertion of a valve to prevent injury to cargo in case of such breakage, or that the conclusion of the district judge should be set aside.
¡3 j Hut, irrespective of the question of the miseaworthiness of the ship on account of the location of the drain pipe or the absence of a valve therein at the bulkhead, we are of the opinion that the Indrapura was unseaworlhy as to the cargo in the hold in which the appellees’ goods -were carried for latent defects in the structure of or the protection to the pipe. Although the pipe was found to he broken immediately following a succession of stormy days, the evidence does not show that the weather was exceptionally stormy or more so than was ordinarily to be expected upon such a voyage. The pipe was inclosed in a wooden case constructed of stout planking two’inches in thickness rising ten inches above the false bottom, and measuring a foot across the top. On arriving at the vessel’s destination this casing was found to be intact, bearing no marks of strain or of collision with any article of the cargo. When it was removed, it was discovered that one of the lengths of the cast-iron pipe, eight feet in length, was broken in twain. This fracture, having occurred as it did, with the consequent injury to cargo, the burden was cast upon the shipowner to prove that the vessel was seaworthy when she started on her voyage, and that the pipe was properly constructed and attached and was securely in-cased. One witness testified that in his opinion if some of the hales of gunnysacks had got against the casing a heavy lurch of the vessel might have sprung it some, enough to break the pipe, “I say that it might, you know, I am not saying that it did.” Hut, if it is true that the pipe was broken in the manner so suggested, it was for the shipowner to show that the casing was reasonably sufficient for the purpose for which it was intended, and that it was securely fastened. No such evidence was offered. There was no testimony as to the manner in which the casing was fastened in place, and no proof that it was securely replaced after its removal at Yokohama. There can be no question that such a pipe could be. and ought to he, so securely protected by casing that no blow upon the casing from shifting cargo could so spring it as to break the pipe. From the fact that the pipe broke as it did it would seem clear, in the absence of proof to the contrary, that there must have been a defect in the pipe or that it was improperly mounted or placed, or that it was insecurely incased. If from either cause the pipe broke, the ship was unseaworthy as to the cargo. In The Glenmavis (D. C.) 69 Fed. 472, Butler, District Judge, in a similar case, said:
•‘I cannot: avoid the conclusion, however, that the easing was imperfect and unsafe at that time. Casing was essential to the safety of the pipe. *716Without it the latter would clearly have been insecure, and the ship have been subject to condemnation on that account. Any shifting of the cargo, such as might result from settling, or the motion of the ship in ordinary weather, would be likely to break it, if exposed. The sole object of the casing is to afford protection against such danger. Tt is necessary to this end, therefore, that the casing shall be very substantial, and be securely fastened in place.”
[4] There was in the bill of lading no exception of liability for loss 'or damage from latent defects of the ship, her machinery or appliances. The owners therefore are not entitled to the benefit of the Harter act. The Carib Prince, 170 U. S. 655, 18 Sup. Ct. 753, 42 L. Ed. 1181. In that case it was held that the Harter act did not exempt the vessel from liability for injury caused by a latent defect in the peak ballast tank consisting of a defective rivet which upon the voyage broke, leaving a hole through which water entered and injured the cargo, and that the ship was unseaworthy at the commencement of the voyage. In the Edwin I. Morrison, 153 U. S. 199, 210, 14 Sup. Ct. 823, 38 L. Ed. 688, the court approved the language of Mr. Justice Gray in his opinion in The Caledonia (C. C.) 43 Fed. 685, in which he said:
“In every contract for the carriage of goods by sea, unless otherwise expressly stipulated, there is a warranty on the part of the shipowner that the ship is seaworthy at the time of beginning her voyage, and that he has used his best efforts to make her seaworthy. The warrant}' is absolute that the ship is, or shall be, in fact seaworthy at that time, and does not depend on his knowledge or ignorance, his care or negligence.”
On the appeal from the decision in the case of The Caledonia, 157 U. S. 124, 15 Sup. Ct. 537, 39 L. Ed. 644, the court said:
“In our opinion the shipowner’s undertaking is not merely that he will do and has done his best to make the ship fit, but that the ship is really fit to undergo the perils of the sea and other incidental risks to which she must be exposed in the course of the voyage; and, this being so, that undertaking is not discharged because the want of fitness is the result of latent defects.”
And the court cited the case of The Glenfruin, 10 P. D. 103, where a steamship laden with cargo became disabled at sea in consequence of the breaking of her crank shaft resulting from a latent defect in the shaft, a flaw in the welding, which it was impossible to discover, and it was held that under his implied warranty of seaworthiness a shipowner contracts, not merely that he will do his best to make the ship reasonably fit, but that she shall really be reasonably fit for the voyage, and that as, when the Glenfruin started, the shaft was not reasonably fit for the voyage, she was unsea worthy. In the case of McFadden & Co. v. Blue Star Line (1905), 74 L. & J., K. B. 423, it was shown that, owing to the pressure of water, a joint in the tube communicating with the ballast tank which was insufficiently packed before the loading of a vessel was commenced, gave way and water flowed into the stoke holes, then into the bilges and then into a cargo hold. It was held that insufficient packing was itself a breach of warranty of seaworthiness.
The decree is affirmed.