(after stating the facts as above). In the case of International Nav. Co. v. Farr & Bailey Co., 181 U. S. 218, 21 Sup. Ct. 591, 45 L. Ed. 830, the Supreme Court, at page 226, 181 U. S., page 594, 21 Sup. Ct., 45 L. Ed. 830, say:
“Even if the loss occur through fault or error in management, the exemption cannot be availed of unless the vessel was seaworthy when she sailed, or due diligence to make her so had been exercised.”
*190The questions to be determined, therefore, are, first, was the ship seaworthy at the commencement of the voyage, and, second, if not seaworthy did her owners exercise due diligence to make her so ?
It is unnecessary to discuss the negligence of the engineers in leaving the sea cock open, for the reason that this fault is undisputed in the testimony and is practically conceded by the libelant. In the brief it is stated:
“If it is admitted, tliat it was a fault in management to allow the sea cock to remain open for five hours longer than usual, this error will not exonerate the shipowner, for he has failed to furnish a seaworthy ship.”
That the careless filling of the tank was a fault in the management of the vessel is manifest. The Glenochil (1896) Prob. Div. 10, 8 Asp. N. S. 218. For such faults the owner, if he has exercised due diligence to make the vessel seaworthy, is not responsible, under the third section of the Harter act.
The witnesses for the respondent agree that the pressure produced by the open sea cock was sufficient to blow out the packing no matter how carefully made. The libelant’s experts deny this, but admit that it would produce some additional pressure upon the top of the tank. One of them testified as follows:
“Tire pressure tliat could come from the sea would only be due to tbe ordinary floating of tbe ship, and that pressure is within the ordinary conditions for which a ship is constructed, and for which the tanks are constructed.”
No tests were made and no data appears in the record from which the amount of pressure can be estimated. In such circumstances the court must proceed upon the proof as made by the parties. We are not permitted 'to enter the domain of science and attempt an application of the principles of hydrostatics to the facts in hand even if competent to do so. We may, however, consider some propositions which appear to be elementary.
The top of the tank was about 17 feet below the level of the sea, and it is evident that if the manhole door and the sea cock had both been left open the water in the water-tight hold would in time have risen to the level outside. With the manhole door closed the pressure produced by this tendency of water to rise to its own level was resisted by the top of the tank, the door being a part thereof. The degree of force thus exerted is not shown, but it must have been considerable and it was a force that the door was not intended to resist.
Assuming that the cock had been closed when the water had risen to a point an inch from the top of the tank, it is apparent that, when the ship was on an even keel, no strain whatever would have been brought upon the joint. The rolling and pitching of the ship would have produced pressure in all respects similar to that which the joint successfully withstood when the ship was being painted at Java.
The contention seems to be well founded that the pressure last described might cause an imperfect joint to leak so that the water would percolate through the packing, but that it would not cause the entire packing to “blow out” as was the case with the joint in question. Such a result, it is argued, could only result from a much greater pressure than is produced by the normal conditions of navigation.
*191One of the libelant’s experts says:
“If the joint is effectively and properly made I do not think that any pressure coming on it would start the joint on a construction such as indicated by the description. The greater the pressure the tighter the joint will be.”
It would seem that this opinion leaves out of view the fact that the pressure upon the joint was equal in all directions. The lateral pressure upon the packing was as great as the vertical pressure upon the steel door itself. But, in any view, we are unable to accept the accuracy of the statement; it seems clear that sufficient pressure might be and, in fact, was brought against the packing to cause it to burst.
We have, then, a confessed act of negligence on the part of the engineers, sufficient to account for the damage to libelant’s cargo. All the witnesses who were present at the time, and are familiar with the facts, unite in saying that it was this fault which caused the leakage and that no ordinary joint could have withstood the pressure thus occasioned. The great weight of testimony is with the respondent upon this branch of the case, and the discussion might safely end at this point were it not for the libelant’s contention that the joint was unseaworthy and that this condition contributed, in connection with the fault of management, to produce the damage.
Did the owners exercise due diligence to make the Albion seaworthy when she sailed? Was the joint properly packed? The carpenter who packed it appears to be an intelligent and competent person and he testified that it was tight, strong and perfect in every particular. No one else' saw it and his testimony is wholly uncontradicted. After the joint was made it was tested by filling the tank and tipping the ship in various positions. The test showed no defect. This testimony is also uncontradicted.
The District Judge criticises this test for the reason that it does not satisfactorily show that the tank had been subjected to the amount of pressure which afterwards caused the leak. This seems to us hardly a fair criterion. The pressure which afterwards caused the leak was an abnormal and extraordinary pressure and one which the owners could not foresee and were not bound to guard against. If the manhole door were made in the usual way and withstood the ordinary tests it should be sufficient. The owners were not called upon to apply a strain to which, in a well-managed ship, it would never be subjected.
The accuracy of the tests at Java is also questioned by the libelant for various reasons, based upon assumptions not found in the record, but we are unable to accept these criticisms as well founded. It is true that all the details of the tests are not given, but if additional particulars were desired they could have been obtained on cross-examination. It is difficult to perceive what more the respondent could have done. No other test is suggested in the proofs. It is plain that no other test could be more effectual. If the tank withstood the pressure of ordinary filling nothing more was needed.
We do not feel justified in permitting the opinions of the experts to overthrow the direct and positive testimony that the joint was properly made and properly tested. If the tank had been filled in the usual *192way there would be unquestioned force in the expert’s theories, but when it appears that the treatment which the tank received might have destroyed a much better joint than the respondent was legally bound to furnish, the principal reason for accepting their opinions disappears. Having found a perfectly plain and adequate cause for the damage we are not required to resort to speculation and guesswork to find an additional cause. “The test of seaworthiness is whether the vessel is reasonably fit to carry the cargo which she has undertaken to transport.” The Silvia, 171 U. S. 462, 19 Sup. Ct. 7, 43 L. Ed. 241.
We think the Albion was reasonably fit to carry the libelant’s sugar, and that she would have carried it safely had not the gross carelessness of her officers permitted the influx of seawater.
The decree is reversed with costs and the cause is remanded to the District Court with instructions to dismiss the libel with costs.