The§e are five suits in admiralty, brought by owners of portions of the cargo of the steamer River Meander, to recover for damage to the cargo caused by sea water on a voyage from Smyrna to New York. The steamship was a single screw steel steamer built in 1906, 373 feet long, 50 feet broadband 24.9 depth of hold. Her hold was divided into five compartments separated by water-tight bulkheads. She was purchased by the American Levant Line in 1912 for ¿39,400 and was turned over to her purchasers about August 31, 1912. She was at that time in' a drydock on the Tyne. A surveyor made a favorable report of her condition to the purchasers at the time of her purchase.
The steamer sailed August 30, 1912, from the Tyne with a cargo of coal and delivered her cargo at the Island of Elba. She then proceeded to Machri, where she took on a part of her cargo for her voyage to New York, and then to Smyrna, where she completed her cargo. The merchandise in question in these suits was taken on at Smyrna. She sailed from Smyrna, stopped at Algiers for coal, passed Gibraltar on October 18th, and arrived at New York on November 5th. The steamer had fairly good weather until she got about four days west of Gibraltar. From that time she encountered exceptionally heavy weather, consisting of a substantially continuous series of full gales, until she passed Nantucket Lightship November 4th.
After loading at New York, she started on a return voyage to England, and, when a few days out, a leak developed, and water entered in such quantities that the crew abandoned the steamer and left her to founder in mid-ocean.
The part of 'the ship in which the cargo was stowed, for damage to which this suit was brought, was in the No. 3 hold on the port side. In that hold there was below the main deck a section called .the “between-decks.” Below the between-decks was a deep tank, sometimes used for water ballast and sometimes for the carrying of cargo. On this voyage it was filled with bales of Turkish tobacco. It is the next compartment aft of the engine room and is separated from the forward part of the No. 3 hold by a water-tight bulkhead extending upward to the between-decks. The between-decks above is also water tight, and is fitted with two doors or covers on the floor, through which cargo, when carried in the deep tank, is taken into and out of the tank. These doors or covers are of iron. They stand on coamings 9 inches high and can be screwed on with rubber washers so as to make them entirely *933water tight. They are so screwed on when the deep tank contains water ballast, but on this voyage, after the deep tank was filled with' bales of .tobacco, the doors or covers of these entrances into this deep tank were simply laid on the coamings. Then the between-decks above was filled with figs in wooden cases.
There was a large ventilator of the usual cowl type leading to the forward part of the between-decks about two feet from the after side of the bulkhead of the bridge deck structure, and in a position about opposite the port forward corner of the main deck hatch. The top of the cowl of this ventilator was 8 feet above the main deck. The open-, ing of the cowl was 2 feet 8 inches in diameter, and the size of the main ventilator was 16 inches in diameter. There was also another and smaller ventilator leading through the between-decks into the forward part of the deep tank. The top of this ventilator was six feet above the main deck. It was a little nearer the center of the ship than the large ventilator. Its cowl was 1 foot 10 inches in diameter, and the main shaft of this ventilator was 12 inches in diameter. This small ventilator was about a foot aft of the bulkhead of the bridge deck structure.
The steamer had a full equipment of gear for closing the ventilators in bad weather. Canvas' covers were provided to be placed over the mouth of the cowls and lashed fast with a cord, very much as a housewife ties a paper over the mouth of a tumbler of jam. All the ventilators on the ship- were so covered as soon as the weather became heavy. Wooden plugs were also provided, to be inserted in the main pipes of the ventilators after the cowls were taken off, for use in very heavy weather.
On the morning of October 26th about 6:15, these canvas covers on the two ventilators leading into the No. 3 between-decks and deep tank were found to be off the ventilators. The assumption is that they were blown off by the wind or washed off by seas. One of the officers testified that they were in their place and in'proper condition that morning when he went off watch a little after 2 a. m. It therefore appears that those canvas covers were off those ventilators not more than four hours that night, and of course may have been off any less time. The officer who discovered that they were off immediately put new covers on. The ventilators were constructed in the usual way. A solid iron pipe was riveted to the deck, coming up a certain distance above the deck, and having an iron band around it a portion of the distance from the deck, upon which was fitted the upper part, or cowl, of the ventilator, very much as one piece of a stovepipe is put upon another. By this arrangement the cowls of the ventilators can be turned in any direction. The evidence is that both these ventilators were turned that night so that the cowls turned toward the port quarter of the ship, away from the prevailing northwest wind.
On the forenoon of the 26th, the cowls of these two ventilators were removed and a wooden plug inserted in each of the iron pipes of the ventilators. These plugs were covered with tarpaulins, and in this way the ventilators were made absolutely water tight. These plugs were provided for the vessel in casé of very heavy weather, and apr *934parently were not used in any of the other ventilators, which during the continuous gales of that voyage were simply protected by the canvas mouth covers.
The officers testify that they had no idea at the time that any water to any serious extent had gotten down those ventilators. Soundings were taken whenever possible at regular intervals during the voyage to see whether there was any water in the holds. There were some days during this very severe weather on which soundings could not be taken, and on those occasions the pumps were set to work and pumped any water out. No unusual amount of water was discovered in the ship until the 3d of November, eight days after the ventilator covers blew off. On the morning of biovember 3d the soundings showed 4 feet 6 inches of water in the No. 3 port bilge. This was pumped out, but water continued to accumulate there in large quantities, and on the morning of the 5th, when the steamer reached New York, there was found to be two feet of water in that bilge.
There was a leak also in the port side of the No. 1 hold, beginning about October 25th, and continuing substantially all the way across the Atlantic. The pumps kept the water out, but it collected in that part of the ship, apparently at the rate of about a foot in 12 hours, substantially through the voyage. It seems to be admitted by all the par-. ties that the water under the No. 1 hold could not have affected the cargo in the No. 3 hold because of the intermediate water-tight bulkheads ; but the amount of water which entered the No. 1 hold, as well as that in the No. 3 hold, may be propeidy taken into consideration on the question of the seaworthiness of the ship.
When the hatches were opened and the cargo taken out in New York, it was found that a portion of the cargo in the between-decks and in the deep tank on the port side of No. 3 was damaged by sea water. A drawing has been made of the damage, which is marked “Claimant’s Exhibit 3” of November 25, 1912,. and which, according to the evidence, correctly shows the part of the cargo, which was injured. Under the small ventilator a small portion of the tobacco in the deep tank was wet. This wet tobacco did not extend down to the floor of the deep tank, and was situated immediately under the ventilator. Under the large ventilator opening into the between-decks, cases of figs were wet, and this wet portion extended down directly from a point each side of the ventilator entrance, widening somewhat as it went down until' a point about 10 or 12 inches above the bottom of the between-decks, and at that point all the cases were wet clear across the bottom of the between-decks. The only part of the figs in the between-decks which were wet were those under the ventilator and those on the floor. The rest were dry. In the deep tank immediately under the door or covering under the large ventilator opening into the tank the bales of tobacco had sunk down, leaving- an open space; then, from the top of the bales under the covering down to a point near the bottom of the deep tank, the bales were wet, sodden, and steaming, and all the bales at the bottom of the deep tank were also wet and steaming. All the rest of the tobacco in the deep tank was entirely dry, except the small portion under the small ventilatof which has been already described.
*935After the vessel had been unloaded, there was evidence of a leak in three frame spaces on the starboard side of the ship at the forward end of the between-decks about opposite where the injured part of the cargo had been. At that place it was found that the iron was corroded, and there were signs that water had been running there. The stringer plate under the deck was slightly bent down, and a wooden chock was displaced which had been fastened by cement, and the angle iron on the bulkhead had four rivets missing. Some of the witnesses say that the water stains there showed that a good deal of water had been coming down there for quite awhile. Others say it showed only a slight trickling. There was an icebox and a coal bunker on the deck above this place. They were removed and a test to ascertain whether there was a leak there, made by building a cofferdam and putting water to the depth of about a foot on the deck; but no evidence of any leak there was discovered by that test. Some of the libelants’ witnesses claim from the signs of water on these three forward' frame spaces that there had been a leak there which had admitted a considerable amount of water while the ship was laboring in the heavy weather, and that the reason that no leak appeared when tested in New York was that the ship was lying still. But I cannot believe that sufficient water came into the steamer at that place to afford any explanation of the damage to the cargo which is the subject of these suits.
The claimant’s theory is that all the water which damaged this cargo came down the two ventilators. Claimant’s counsel argued on the hearing that it was not necessary to infer that it came down the two ventilators during the period of not exceeding four hours when the ventilator covers were off, but that the ventilator covers may have been taken off at other times during the voyage. In my opinion the evidence affords no basis for the inference that any water got down.those ventilators except for the brief period when the covers were off on the morning of October 26th. There is no evidence that the covers were off those ventilators before October 26th at any time when the weather was stormy enough to cause water to accumulate on deck in such a mass as to pass down the ventilators, and the evidence is uncontradicted that on the forenoon of the 26th the cowls were taken off, the wooden plugs inserted, and both ventilators made absolutely water tight, and that that condition continued substantially during the rest of the Voyage. The theory, therefore, that the damage to the cargo was caused by sea water coming down through those ventilators, must start with the assumption that the only time during which that could have occurred was during that part of the period between about 2 a. m. and 6 a. m. on the morning of October 26th, when the covers were off.
It is estimated that the entire amount of water which was ultimately found in the between-decks and deep tank amounted to about 50 tons. The claimant’s theory is that this water entered through these ventilators; that it went substantially straight down through the cases of figs, and then straight out on the- floor of the between-decks, until it reached a point higher than the coaming of the door into the deep tank, and that when it reached that point it soaked through under the loose cover over the coaming into the deep tank; that when it got into the *936tobacco it was absorbed by the tobacco to the point of saturation, and then passed out into the bilges; and' that that is the explanation of the fact that on November 3d there was found 4%. feet of water in the bilge of No. 3. This theory seems to me simply incredible. In the first place, if we are to judge by the situation of the damaged cargo, the amount of water that went down the small ventilator was slight; it only wet the tobacco immediately under that ventilator for a portion of' the way through the deep tank, so that the great bulk of the water must have gone' down through the large ventilator. The cases of figs were not so stowed as to completely fill the between-decks; there was the usual space of from six inchés to a foot at the top of these cases of figs.
If the water came through this ventilator it is hardly possible to imagine it as coming down in a steady stream, no matter how heavy the seas were that were breaking on the ship. Even if it came down in a steady stream, it could not, it seems to me, have simply wet the cases immediately under the ventilator, but would have dashed off in every direction over the whole of the cargo. If the water came down intermittently, it nevertheless must have come at times in such quantities as to produce the same result, if it be assumed that as much as 50 tons came down in four hours. Moreover, the evidence is specific that the cowls of these ventilators were turned that night towards the port quarter away from the prevailing wind. How could a large quantity of water get down there in so short a time? No seafaring men were produced who ever heard of such a case. I have no doubt that the damage to that portion of the figs in the between-decks immediately under the large ventilator was caused by water coming down through this ventilator; but I think the fact that the water only damaged those cases immediately under each ventilator shows that the water which came in. was mostly spray or small quantities of water ’entering from time to tíme. I do not believe that much of the water in the bottom of the between-decks came in through the ventilators, or that much of the water which got into the deep tank was originally held in the bottom of the between-decks and from there drained through the door over the deep tank. I think that most of the injury to the tobacco was caused by the four feet six inches of water found in the port side of No. 3 hold on the 3d of November, and I think that that water got into the ship at that point in some manner which is not explained, and did not get into it through the ventilators.
[ 1 ] The claimant’s position is that, the water having entered through the ventilators, the damage was caused either by a peril of the sea or by mismanagement in the navigation of the steamer, and that, in either case, the claimants are exonerated from liability by .the provisions of the Harter Act. But the Harter Act only exonerates a shipowner from responsibility for damages or loss resulting from faults or errors in navigation or in the management of the vessel or from dangers of the sea, in case the owner has exercised due diligence to make the vessel in all respects seaworthy. So far as the damage to the tobacco in the deep tank is concerned, I think the burden of showing that the damage arose from one of the excepted causes was upon the carrier, that the *937evidence leaves the efficient cause of the damage wholly unexplained, and that therefore the doubt as to the cause of the entrance of the sea water must be resolved against the carrier. The Folmina, 212 U. S. 354, 29 Sup. Ct. 363, 53 L. Ed. 546, 15 Ann. Cas. 748. So far as the damage caused by water entering through the ventilators is concerned, I think it is clear that such damage was caused either by a peril of the sea by which the canvas coverings were either blown off or washed off in the very heavy weather on the morning of October 26th, or through fault in the navigation of the vessel by the omission to take the cowls of the ventilators off earlier and put the wooden plugs in the ventilators.
But if the vessel was not seaworthy when she started upon the voyage, then the provisions of the Harter Act do not apply to- the case at all. In the first place, there can be no presumption of seaworthiness in such a case as this. The Wildcroft, 201 U. S. 378, 26 Sup. Ct. 467, 50 L. Ed. 794. It must be affirmatively proved by the shipowner. Now there is no evidence in this case that the vessel was seaworthy when she began her voyage. The only evidence relied on to show that she was seaworthy at that time is a report made to the purchasers by a surveyor before the ship sailed from the Tyne. The surveyor was not called as a witness, and his report is mere hearsay. No evidence has been given as to the actual condition of the ship before she sailed from the Tyne. Nor has any evidence been given of her condition before she sailed from Smyrna. It may be in this case that the Tyne is to be taken as the place from which the voyage commenced, but the cargo 'taken out to the Mediterranean was completely unloaded and a new cargo taken in for New York; and; if Smyrna is to be regarded as the place from which the voyage began, then there is no evidence at all of the condition of the ship before she sailed from Smyrna. Moreover, on the voyage to New York the serious and substantially constant leak in the No. 1 port hold, which began shortly after leaving Gibraltar, shows, in my opinion, that the ship was not seaworthy during the voyage. It is said that that leak had nothing to do with the damage to the cargo, which I think is true; but the requirement of the Harter Act is that the owner must provide a ship that is in all respects seaworthy before he can take advantage of the exemptions from liability required by that act. Moreover, in my opinion, the water which was found in the port side of the No. 3 hold on the morning of November 3d showed that the ship was not seaworthy at that time. I believe that it came in in some way entirely unexplained. But whatever explanation may be made of its getting in, the fact is that it got there, and that after it was pumped out it came in again at that place in large quantities. The ship was not drydocked at New York. There is a good deal of evidence to the effect that she was examined in New York as well as she could be in the water when not put in the drydock, and that she'appeared to be in good condition; but it seems to me that the leaks which had occurred on the voyage to New York would have led any prudent shipowner to have had her drydocked here before her return voyage to Europe. When the fact is considered that when she was four or five days out on her return voyage, after a day or two *938of heavy weather she sprung a leak on the port side of the ship about at the place where the damage to the cargo occurred, and that although all the pumps were put in operation the water gained upon them so rapidly that at the end of about 36 hours the ship was abandoned in a sinking condition, it seems to me an inevitable conclusion that this ship was not seaworthy during the voyage to New York. In my opinion, if a ship is shown to be unseaworthy during a voyage, the inference may be drawn, in the absence of any explanation to the contrary, that-"she was unseaworthy when she started. Cargo owners usually cannot prove unseaworthiness at the time a voyage begins.
[2] It is the duty of the shipowner to know that his ship is seaworthy before the voyage begins, and if he does know it he can prove it. If a vessel proves to be .unseaworthy during a voyage, the burden should be on the shipowner to prove affirmatively that she was seaworthy at the time the voyage began.
[3] No such affirmative evidence has been given in this case; and my conclusion upon the whole case is that the River Meander was not seaworthy when the voyage to New York began, and that the claimants are not entitled, in respect to any of the damage which was caused by sea water on this voyage, to exemption from liability by any of the provisions in the bill of lading or by the terms of the Harter Act.
The libelants are entitled to a decree in each case as demanded in the libel. If the damages in each case are not agreed upon, there should be the usual reference to fix the amount.