This libel contains two counts, for damage to separate consignments of rubber which were shipped at Singapore on board the steamship Mursa, destined for San Francisco. Both consignments arrived damaged in the same manner, by contact with crude oil and salt water, and, since their good condition at the time of delivery is not contested, the sole issue is whether or not respondents are liable for the losses sustained.

All of the rubber was stowed in the after part of the aftermost lower hold, where it was placed on both sides of the shaft alley. Its stowage was careful and proper, and no complaint is made upon that basis; but on arrival at San Francisco the lower tier of eases was found to have been stained by water and oil to heights varying between 2 and 10 inches. A coating had been formed upon their surface, which was so thick that it could be scaled off by band, and the combination of oil and water had penetrated far into the rubber which they contained. Inasmuch as the dunnage, together with its supporting timbers and flooring, had elevated the lower tier by almost 6 inches from the steel floor of the hold, it is apparent that at some time during the voyage there was a mixture of liquid present which had a depth of at least 8 inches. This the movement of the ship splashed and threw to a considerably greater height, even staining cases above those in the lower tier.

There is no conflict as to where it came from; it escaped from the shaft alley through defective rivets and improperly laid caulking of the angle bars which fasten the walls of that passage to the floor of the hold, and through two leaks, one on each side of the shaft alley, had found access to the cargo. Water evidently had made its way into- the alley through a leak in the stern gland, or had dripped from leaking bearings, and had washed oil, of which there was a considerable; quantity loose there, into the hold.

The bills of lading exempt both owner and operator from liability for "unseaworthiness, even existing at the time of shipment or sailing on the voyage> provided the owner or operator has exercised due diligence to make the vessel seaworthy,” as well as from "latent defects in hull, machinery, or appurtenances.” They also provide that the ships on which such goods are carried are not warranted seaworthy as to “any defect or deficiency not discoverable by due diligence by the owners.” The question thus is one of the exercise of due diligence, and if it can be answered in the affirmative, although at the inception of the voyage the Mursa actually was unseaworthy, respondents must be exonerated.

The libelants submit, and without conflict the witnesses testified, that both divisions between the shaft alley and the hold should have been water tight. Respondents claim *423 that they exercised due diligence to make them so, and attribute the leales which were found (as counsel for the libelants so aptly-put it) “to the stereotyped storm, to the tired wings of which earners with damaged cargo endeavor without fail to shift the burden of their liability.” Respondents also maintain that any fault of which they may have been guilty was committed in the navigation or management of the vessel, through failure to sound the well of the hold in question, which is excused by section 3 of the Harter Act (Comp. St. § 8031), and that in any event they are entitled to the benefit of libelants’ insurance.

The Mursa was built for the Shipping Board in 1920. During construction, she was examined by an inspector, who at the time was inspecting five vessels a week. He caused the after peak tank and double bottom No. 7 to be subjected to hydrostatic tests, by being filled with water, and found both to be water tight. lie omitted, however, to cause the same test to be made in the case of the shaft alley. Instead, he directed a man to play a hose against the outside of the tunnel, while another, using a flash light, followed the discharge on the inside to look for traces of leakage; of which none were found. The fact is significant that, when the damage had been discovered in San Francisco, ill order to ascertain its source, marine surveyors caused the shaft alley to be partially flooded with water, with the result that the defects immediately became apparent.

Where preof of a delivery in good order to a carrier has been made, with that of a redelivery in bad order, the carrier prima facie is liable. Respondents, therefore, are called upon to explain this loss. The R. P. Fitzgerald (C. C. A. 6) 212 F. 678, 684 , 129 C. C. A. 214 , certiorari denied 234 U. S. 757 , 34 S. Ct. 675 , 58 L. Ed. 1579 ; The Folmina, 212 U. S. 354, 362 , 29 S. Ct. 363 , 53 L. Ed. 546 , 15 Ann. Cas. 748; The Rosalia (C. C. A. 2) 264 F. 285, 288 . Their first contention is' that due diligence was exercised to make the vessel seaworthy; in other words, that the “'hose’ test, which was made on September 10, 1920, while iho Mursa was still on the ways and unlaunehed, was adequate and sufficient. With this I am compelled to disagree.

The voyage with which we are concerned began on June 1, 1923, almost three years after the test referred to. It was the maiden voyage, for throughout the intervening period the vessel had been laid up at Benicia, inside the harbor, and almost entirely idle. Although the tanks were retested just before she left San Francisco, the “hose” test of the shaft alley was not repeated, and on the basis of that test, made at the time stated, she was declared fit for sea and ready for cargo. That the test by flooding would have been entirely practical is indicated by the fact that it actually was made after the damage had occurred. Respondents’ witnesses testified that the “hose” test was that customarily used. It seems clear, however, that the principal, if not the only, reason for applying it, in preference to the former test, was that it would impose considerable weight upon the ways supporting the vessel, and that in consequence its construction would be interfered with.

No reason whatever was shown for the failure to make the- test by flooding during the 2% years after she had left the ways, which elapsed before the voyage began, or for that to renew the test by hosing. The latter, as has been shown, consisted in running water from a hose, under a pressure of 90 to 100 pounds, over the entire outer surface of the tunnel ease, with a man inside to watch for traces of the ingress of water, guided only by sound, in perfect darkness except for the illumination of a flash light, and required to follow all rivets, plate and hounding bar edges, breaks, and connections.

A water test, say respondents, is never made on the inside of a shaft alley, for the “obvious” reason that it would not try the water-tightness of the top of the tunnel. Conceding that fact to be true, it does not follow that a test which is sufficient for that purpose is adequate for the purpose of finding leaks elsewhere situated. The leaks in this case not only would have been (and were) exposed by flooding the shaft alley, but were by no means likely to be discovered by the use of a hose, particularly at high pressure. The requirement of duo diligence under the Harter Act, of course, is not satisfied by diligence in the procuring of surveyors’ certificates. The R. P. Fitzgerald (C. C. A. 6) 212 F. 678, 687 , 129 C. C. A. 214 , and eases cited. I therefore hold that the respondents failed in their duty duly to exert themselves to- make the vessel - seaworthy prior to the commencement of the voyage.

In avoidance of this position, the argument next is made that, even if proper tests had been made at or prior to that time, the defects would not have been disclosed by them, because they did not exist at the out *424 set, but developed en route. This is somewhat at variance with respondents’ contention on the first branch of the ease, that “experience has proved that, after the rivets and caulking edges are once sound, they should remain that way for a considerable length of time.” It also is at variance with the contentions of the libelants, 'who argue that, since the vessel in general arrived in good shape, the fact of damage in only one place'on the ship indicates unseaworthiness, and not an injury resulting from a peril of the sea. The Asuarca (D. C.) 291 F. 73 , 75, affirmed (C. C. A. 2) 291 F. 77 ; The Rosalia (C. C. A.) 264 F. 285, 288 .

As was said in the ease last cited: “It is true that the strain of long-continued storm may injure some one piece of apparatus, good of its kind, while leaving the rest of the ship substantially unhurt. * * * But this is very rare; and * • * the assertion belongs to that class of unusual claims, which ‘ought to be sustained by evidence correspondingly convincing.’ ” Here the only evidence offered is that no water or oil was discovered in the hold where the rubber was stored, either at Manila or at Singapore, on the voyage out from San Francisco, and libelants are asked to account for the fact. Even assuming that an explanation is necessary, clearly the burden is on respondents to explain, and that cannot be done by asking questions. They, and not the libelants, were present during the voyage, and their affirmative burden cannot be so discharged.

Furthermore, both in answer and brief, respondents assert that the leak on the port side of the alley was caused, not by a defective rivet, but by the improper manner in which the caulking edge was laid up against the steel of the tunnel wall. It is scarcely conceivable that a storm could have damaged the caulking edge without affecting the rivets in it. This defect, therefore, did not develop after the vessel’s departure, but existed previous to it. As for the other, the weather and seas which were encountered appear not to have been unusually rough. In consequence of these facts I am unable to believe that a peril of the seas caused or in any way contributed to the damáge.

Respondents’ third contention is that at best the loss is attributable to fault in the navigation or management of the ship, through failure to sound the bilges and well of this hold. An obvious answer is that the proximate cause was unseaworthiness (The Christel Vinnen, 30 Com. Cas. 32, reversing 16 Asp. M. C. 292); and, second, that it is a condition precedent to the application of section 3 of the Harter Act that due diligence shall have been exercised to make the vessel seaworthy (The R. P. Fitzgerald [C. C. A. 6] 212 F. 678, 682, 683 , 129 C. C. A. 214 , and cases cited; Compagnie Maritime Francaise v. Meyer (C. C. A. 9) 248 F. 881, 884 , 160 C. C. A. 639 ; The Willdomino (C. C. A. 3) 300 F. 5 , 1924 A. M. C. 889). Since I have held that due diligence was not exercised here, this contention, also, is of no avail.

The question of benefit of insurance was not discussed by respondent in its brief. The omission appears to have been well advised, for the law is now settled that a carrier cannot have the benefit of any loan made by an insurance company to a cargo owner. The Turret Crown (C. C. A. 2) 297 F. 766, 779 ,1924 A. M. C. 253, certiorari denied 264 U. S. 591 , 44 S. Ct. 403 , 68 L. Ed. 865 . The fact that the recovery now had will inure to the benefit of an insurance company 'is immaterial, for this suit is, in point of law, that of the shippers, and it is solely with their rights that we are concerned.

From these considerations, it follows that a decree must be entered for the libelants, in the amount of their damage, for interest, and for costs. The matter is referred to Commissioner Krull for the purpose of ascertaining the amount of damage, who will take testimony, prepare findings, and recommend conclusions, subject to the approval of the court.