[1] This is an action to recover damage to 1,078 packages of dry salted hides shipped on the steamer Citta di *523Palermo at Genoa, Italy, in December, 1910, for New York. The hides were stowed in the forepeak, in the space below the between-decks. On the voyage the forepeak filled with salt water and the hides were practically destroyed. Upon arrival in New York the forepeak was pumped out and the cargo removed by divers. The steamer was then placed on dry dock, and it was found upon examination that on the starboard side, about on the 13-foot mark and forward of the collision bulkhead, the points of two rivets in the middle of one of the plates were off. It is conceded that the water which caused the damage entered through these rivet holes.
The libelant alleges that the steamer was uuseaworthy at the outset ox the voyage, in that these rivets were defective, and because the forepeak was not. properly fitted for the carriage of cargo, having a defective pump, no sounding pipe, and no dunnage.
The Citta di Palermo was about 13 years old. She was rated in the highest class in both the English and Italian Lloyd’s. In March, 1910. she had undergone her No. 3 survey at Genoa, and after general repair she retained her rating. The hides in question were take aboard on December 15, 1910, and on the following day the steamer left Genoa for Leghorn, where she arrived on December 17th. After taking on additional cargo she left on December 25th for Oran, and thence to Lisbon, where she arrived on January 3, 1911. At Lisbon 19 bales of cork were stowed in the forepeak on lop of the hides. Proceeding fronii Lisbon, very severe weather was encountered on January 10th, and it prevailed with little interruption for 10 days. The weather conditions were described in emphatic terms by the officers, few of whom, however, appear to have had long experience. Most of the damage sustained by the vessel was done to,the shelter deck, which formed no part of the vessel’s original structure. The indications are that the weather, although severe and long continued, was not extraordinary for the Atlantic in midwinter. On January 19th it was observed that the steamer was going by the head. Soundings disclosed that the forepeak was flooded. Efforts were thereupon made to pump the water out, but, these efforts proving ineffectual, the forepeak was closed up, and in this condition the steamer arrived in New York on January 27th.
The claimant introduced the depositions of sixteen witnesses bearing upon the shipment of cargo and the voyage, namely: From the vessel : Zauucchy, master; Rossi, first officerGavignin, second officer; Giovanelli, chief engineer; and Mezzano, carpenter. Prom Genoa: Bruno, the owner’s superintendent; Gozzini, a clerk; Cremonini, naval expert for the Genoa Chamber of Commerce; and five stevedores who participated in loading the Citta di Palermo at Genoa; also three stevedores who participated in loading the additional cargo at Lisbon. The substance of this evidence may be stated thus: The proof concerning the No. 3 survey at Genoa in March or April, 1910, is indirect. The survey is characterized by some of the deponents as most rigorous, but no witness testified with respect to1 what was actually done by the surveyors, none of whom was produced. Bruno, the owner’s superintendent, and the only witness who says he was present at the survey, *524states that general repairs were then made, and he mentions particular repairs in the forepeak as including one deck plate, the pump, sounding pipe, and chain locker. Giovanelli, the chief engineer, who says he was on board during the survey, testified that the pump was tested at that time and worked all right. No mention was made by any witness. of any inspection or repair of plates or rivets.
Bruno, the vessels’ officers, and the stevedores who put on cargo at Genoa, all testify génerally that the forepeak hold was dry and in good condition when the hides were stowed. The second officer and three stevedores who stowed the bales of cork in the forepeak at Lisbon say that there was no odor or other indication that water had entered at that time. They are corroborated by the carpenter, Mez-zano, who joined the ship at Leghorn, who says he sounded the fore-peak regularly each morning and evening from tire date of departure from Leghorn, and found no water until the morning of January 19th. Mezzano asserts that he discovered water in the forepeak upon his regular morning sounding at 7 o’clock, that it was not discovered that the vessel was down by the head in consequence of this leak until after he made his 'sounding, and that he inf ormed the first officer of the situation. This is not exactly in agreement with the testimony of the first officer, who says he was with the carpenter when he made soundings on the evening of January 18th, between 4 and 4:30 o’clock, and at 7 o’clock on the morning of the 19th. Nor does it accord with the entry in the log under date of January 18th and 19th:
“There was noticed an abnormal leaning to head of the vessel, and a consequent loss in the speed of the vessel, and,, on trying to ascertain the cause, it was found that the forepeak was completely filled in by sea water.”
At all events, the hand pump in the forepeak was immediately started. The chamber of this pipe consisted of three sections, having therefore two joints, and at each joint there was a flange. The chief officer and the chief engineer testified to the condition of this pump at Genoa.. The chief officer said he “had it examined” and that it was in good condition. He did not assert that he actually saw it, much less that he made any examination of it. Although the chief engineer stated that he actually examined the pump, he did not say that he examined the chamber in the space between decks, nor did he assert that he made any test of the pump. Both these officers, as well as Capt. Zanucchy, testified that when the pump was started on January 19th it was in good working order, and that no repairs were made on it during the voyage. When asked about the cement that appeared around the flanges, the captain said that this had been done by the carpenter, from a quarter to a half hour after the pumping was begun, because they “thought that the water did not come with sufficient force.” The chief officer and the chief engineer said the cement was put on as a precaution only. — the former, so as to prevent any air coming through the joints; tire latter, so that they would not have to watch the flanges. All three agree, however, that after the flanges were cemented tire pump worked just as it had before, neither better nor worse. They were unable to malee headway on tire water with the hand pump. Connections were then made with a steam pump, but as *525the tube had to be put through the hatch, and the forepeak was filled with cargo, this, too, proved ineffectual, and after four days’ unavailing work the forepeak was closed.
The witnesses produced at the trial took up the story from the steamer’s arrival at New York. Among them were the libelant’s witnesses Ritchie and Evald, marine surveyors who examined the vessel here, a diver named Anderson, who assisted in the discharge of the vessel, and Bagger, an expert on hand pumps. The claimant produced, among others Cioe, the stevedore who discharged the forepeak; and a diver named Greer who assisted in this work; Nevada, a plumber who did some work on the vessel; and five surveyors who examined the vessel here, namely, Farrar, Murray, White, Gillet, and Ross.
After the forepeak had been pumped out and the cargo removed, the vessel was placed on drydock, and while there was examined by the surveyors. Surveyors Evald and Ritchie, who testified on behalf of the libelant, examined the rivets in question from the scaffolding alongside, which enable them to get within two or three feet of the rivets. They testified that the points of these two rivets were corroded and eaten away the depth of the countersink in the plate. On the inside the heads of these rivets were, tight up against the frame, and the part oí the rivets extending from the head to the point where the frame came against the plate was there. This is established as a fact by the statement of the diver Greer, who tried to stop the leak by wedging a shingle in between the frame and the plate. He was unable to cover the lióles because the rivets projected out beyond the frame. Hence their conclusion was that the water that came through these rivet holes worked in between the frame and the plate. It is agreed that none of the surrounding plates or rivets was disturbed, and both Evald and Ritchie testified that the cement between the frames near this plate was not even cracked.
Of the surveyors who testified on behalf of the claimant, Farrar, Ross, and Murray inspected these rivets from the bottom of the dock, eight feel below them, and on the outside, about 4 o’clock on a February afternoon. They state that the rivets were broken off between the plate and the frame, for they saw the hole or countersink in the plate. White testified to the same effect. He also examined them from the inside, and says Ke could see the light come through. But lie admits, that he did not examine into the matter very carefully. — “I saw that the rivets were broken off and for my purpose that was sufficient.” White, Farrar, and Ross all deny that there was corrosion around the rivet holes in the plate. The claimant failed to produce any other witness who had examined the rivets at any closer range than the surveyors mentioned-- — conspicuously, the workmen who replaced the rivets.
The conclusions drawn by the claimant’s witness vary. The chief engineer, Giovauelli, thought it must have been caused by ‘‘striking some floating wreckage or something, which might have loosened the rivets.” First Officer Rossi testified to the same effect, and he added that the paint on the vessel’s side was scraped as if she had struck something. Second Officer Gavignin thought the damage was due. to *526straining or collision — probably the latter. Capt. Zanucchy appears to have taken the same view so far as his confused testimony is intelligible at all. The surveyors ascribed it to the panting of the vessel in the heavy seas, although admitting that the breaking of two rivets in the middle of a plate, unaccompanied by any damage to the surrounding rivets, was unusual.
[2] The burden is upon the claimant to show the cause of the damage. The Folmina, 212 U. S. 354, 29 Sup. Ct. 363, 53 L. Ed. 546, 15 Ann. Cas. 748. The vessel encountered very héavy weather, but this fact alone is not sufficient to support a finding that the damage was due to perils of the sea. The Edwin L. Morrison, 153 U. S. 199, 14 Sup. Ct. 823, 38 L. Ed. 688. As the claimant’s witness Earrar admitted, nobody can say that the damage was done by heavy weather. What Earrar and his fellow surveyors say is that the damage could have been caused by heavy weather. Upon all the evidence it cannot be said that the claimant has proved that the breaking of the rivets was due to the straining of the vessel at sea. Proof of heavy weather, if coupled with proof that the rivets were in good condition at the beginning of the voyage, might warrant such a conclusion. But mere proof of heavy weather, coupled with the proof herein submitted as to the condition of the rivets at the end of the voyage, is insufficient to support such a finding.
[3] The witness Ross was recalled for the purpose of proving that there may have been a latent defect in the rivets due to crystallization or unfairness of the.holes. The object was to bring the case within the provision of the bill of lading exempting the owner of the vessel from loss or damage caused by' “unseaworthiness of the ship at the commencement or at any period of the voyage arising from any latent defect in hull.” Assuming the exemption to be valid, the proof does not bring this case within its terms. Such an exemption must be strictly construed and cannot avail- the claimant unless the evidence shows that the 'defect, if any, was latent at the beginning of the voyage. There is no proof in this case that any one had made an actual visual inspection of these rivets since the vessel was launched. There is no evidence that the rivets were inspected or tested in the No. 3 survey nine months before, and there is no suggestion that they were inspected or tested at any later time. The only evidence that the condition found at New York did not exist at Genoa is th'e fact that no water entered the forepeak prior to July 19, 1911. But this does not prove that the defect in the rivets, if any, could not have been discovered by an inspection at Genoa. It is to be observed, that the bill of lading does not contain the common provision substituting for an absolute warranty of seaworthiness the mere obligation to use due diligence to make the ship seaworthy. No question of diligence is involved in this case. The question here is, not whether the defect in the rivets would have been discovered in the exercise of due diligence, but whether it could be discovered at all. It cannot be maintained that, so long ás water did not enter tire forepeak, the supposed defect in the rivets was not discoverable by inspection. In view of the libelant’s testimony, it is quite possible that the countersink of the *527rivets may have been wholly or partly corroded at the beginning of the vo) age, although the shank of the rivets still remained in place and prevented the entry of water. Claimant’s witness Gillett admitted that if a rivet was a little cold at the time it was hammered, or if in any other way water got in behind the countersink, the rivet would naturally corrode. It appears from Ross’ testimony that such a condition is not infrequently met with; indeed, his testimony discloses very clearly the necessity of inspection and test of rivets to detect defects, it is immaterial for the present purpose whether such tests are usual. The material question is whether, if used, they would have disclosed the defect. It is worthy of note, however, that the claimant offered no evidence to show what are the usual and proper methods of testing rivets, or what kind of defects would be disclosed by such tests.
Turning, now, to the issue with respect to the pump, White and Farrar admit that it was cracked and had been patched with cement. The libelant’s witnesses testified without contradiction that the upper flange was broken and that both flanges were patched, the upper with cement, the lower with white lead and canvas. The fact that these repairs were made in the chamber located in the hetween-deck from 15 to 30 minutes after pumping began confirms Capt. Zanucchy’s admission that the water did not come with sufficient force. The claimant’s expert White admitted that the crack would impair the efficiency of the pump. Therefore, in view of the uniform testimony of the crew that the pump worked no better after it was repaired, it would appear that the pump was not efficient either before or after being repaired. The significance of this appears from the proof that the pump, had it been in good condition, could have taken care of the leak. The,estimates of the various witnesses with respect to the amount of water that the two rivet holes would admit naturally varied in accordance with their assumption of the distance separating the iJiell plating from the frame of the vessel, and as to whether the rivets were broken off flush with the frame or extended to or into the plate. But, assuming conditions favorable to the greatest inflow of water, it appears from the evidence that 30 gallons of water per minute was the maximum amount of water that could enter through the tW'O rivet holes in question, while the capacity of the pump was not less than 45 gallons per minute.
Did this defective condition of the pump exist at the beginning of the voyage? The only evidence hearing upon the condition of the pump at Genoa is the testimony of the chief officer and the chief engineer, The chief officer stated that he had “had it examined” and that it was in good condition. It does not appear that he actually saw it at that time, much less that he made any examination of it. It is true that the chief engineer states that he inspected the pump at Genoa; but he does not say that he actually examined the chamber, which was in the dark space between decks, nor does it appear that he made any test of the pump at that time. Moreover, both these witnesses stoutly maintain that the pump was in good order when started after the heavy weather; indeed, none of the steamer’s witnesses admitted that the pump was cracked, even at New York. Their testimony *528therefore does not establish that it was not cracked at Genoa. If the pump chamber had been cracked during the • voyage, it .would seem an easy matter to prove; but there is no, testimony in the record to that effect. There is y.ti fact no evidence that it could have occurred on the voyage. It is not suggested that any other piping or machinery below decks was broken on this voyage. In the absence of any evidence, it cannot be assumed that the pump in the forepeak was broken by heavy weather.
In view of the foregoing conclusions, it is unnecessary to examine in detail the evidence bearing upon the issues with respect to cargo battens and sounding pipe. . Four witnesses testified positively that there were no battens around the sides of the forepeak. The fair inference from all the testimony is that, if any wooden battens were used, they were not secured to the frames of the ship, but were merely loose boards, which would be likely to shift as soon as heavy weather was encountered. This has a direct bearing upon the alleged breaking of the sounding pipe.
There is serious conflict in the testimony relating to the question whether or not the forepeak was fitted with a sounding pipe, and it is difficult to reach a definite conclusion in this matter. The necessity of having a sounding pipe is shown by the fact that soundings taken through the pump would reach only to the elbow in the chamber, nearly 4^4 feet above the bottom of the ship. Several of the steamer’s officers testified positively that there was a sounding pipe in the fore-peak, that it was in good condition at Genoa, and that it was used daily until after the discovery of the leak on January 19th. Their testimony is, however, conflicting and contradictory. Capt. Zanucchy first testified that nothing happened to the sounding pipe on.the voyage. Afterward he conceded that it had been broken the day after the leak was discovered, in which he. is supported by the first and second officers. But the carpenter, Mezzano, who did the sounding, contradicts the officers by stating that the sounding pipe was broken two days before the steamer arrived in New York, which would make it six days after the leak was discovered., Ross and Greer saw no sounding pipe, nor were White and Farrar positive that they saw one. On the other hand, Evald and Ritchie testify that they looked for a sounding pipe and that there was none. But in view of the testimony of the plumber, Nevada, who says he removed it, the evidence as a whole preponderates in favor of the claimant.
My, conclusion is that the claimant has failed to- prove that tire rivets in question were in good condition at the beginning of tire voyage, or that their defective condition was latent. The claimant has also failed fir prove that the hand pump, which was found in bad condition at the end of the voyage, was in good condition when the voyage began. The libelant is therefore entitled to a decree.