GARRECHT, Circuit Judge.

An original libel and two amended libels were brought by the Los Angeles Soap Company, one of the appellees, to recover for shortage and contamination of a consignment of cocoanut oil shipped, in September, 1927, in the deep tanks of the steamship West Cajoot, owned by the appellant, from Manila to Los Angeles.

The appellant filed two libels to recover general average contribution alleged to have been due from the appellees, as a result of expenditures following the stranding of the West Cajoot off the coast of Japan.

Since most of the testimony was by deposition, the case, as the appellant itself points out, is before this court de novo, "both in theory and in fact."

On September 23, 1927, the West Cajoot left Manila with 1,223,583 pounds of cocoanut oil consigned to the appellee soap company. In the early morning of October 2, 1927, the vessel struck some obstruction in Van Diemen Straits, off the Japanese coast. Her engines were immediately put astern, and the ship floated a few minutes later. The West Cajoot's course was then altered for Kobe, Japan, where the vessel arrived on the evening of October 3, 1927. Repairs to the vessel were made on dry dock.

On October 5 it was discovered that the cocoanut oil in the port deep tank had been contaminated by fuel oil, and that some oil had leaked down into the No. 3 port double bottom fuel tank, which was directly underneath the port deep tank. The cocoanut oil was not removed from the deep tanks while the ship was at Kobe.

The West Cajoot left Kobe early on November 1, 1927. The next morning the vessel reached Yokohama, where she took on 1,984 barrels of fuel oil. The ship left Yokohama on the evening of November 2, arriving at San Francisco on November 22, 1927. She was dry-docked at that port for further examination, and left for Los Angeles on November 29, arriving at her destination on the morning of December 1.

The appellees' suit and the appellant's two suits were consolidated for trial. The court below found that the West Cajoot was in a seaworthy condition when she sailed from Manila, but that she "was unseaworthy on and after her arrival at Kobe by reason of the leak between the tanks above described, and remained so until the discharge of the cargo at the port of Los Angeles." The court also found that all of the cocoanut oil in the port deep tank was contaminated with fuel oil when the cargo arrived at Los Angeles, and that there was, moreover, a shortage of 90,620 pounds. In a memorandum opinion, the court stated that the appellant was liable for the "damage done to the cocoanut oil occurring after the arrival of the ship at Kobe," and added that "this quantity includes all of the oil in the tank affected except about ten tons."

The District Judge also found that the stranding of the vessel was the result either of "errors in navigation or in the management" of the West Cajoot. This finding is assigned as error, but, in our view of the case, it is immaterial.

In an interlocutory decree, the lower court ordered that the appellee soap company recover from the appellant all damages caused by the contamination of the cocoanut oil in the No. 3 port deep tank, except the 10 tons above referred to, together with all damages resulting from the shortage. The court also decreed that the appellant recover from both of the appellees general average contribution for the stranding of the steamship, and referred the cause to the United States commissioner to compute the amounts due to the respective parties.

The appellant is here attacking, first, so much of the interlocutory decree as holds that the appellee soap company is entitled to recover damages for the contamination of the cocoanut oil, except about 10 tons, and for the shortage of 90,620 pounds; and, second, so much of the decree as holds that the appellant should recover on its libels in general average for damage suffered only prior to the vessel's arrival at Kobe.

In their cross-appeal, the appellees assign as errors, inter alia, the lower court's failure to find that the West Cajoot was unseaworthy upon sailing from Manila and the court's allowance to the appellant of recovery in general average for such damages as were directly attributable to the stranding.

The cross-assignments of error assailing the seaworthiness of the West Cajoot are bottomed upon the alleged incompetency of the second assistant engineer, M.P. Appel. It is also asserted that the lower court erred in failing to find that, by permitting the West Cajoot to sail from Manila with Appel as an officer, the appellant, knowing Appel's incompetency, failed to exercise due diligence to make the vessel seaworthy in all respects.

The principal testimony as to Appel's alleged incompetency is found in the deposition of Paul Frazier Blinn, who was chief engineer of the West Cajoot on the voyage in question.

On cross-examination, Blinn's testimony as to Appel's mentality was as follows:

"A. I asked my second assistant where he was pumping his oil from, to burn in the ship, and he told me, `Number three.' I says, `No, there's no oil in number three; how can you pump from number three?' He said, `Yes, the port tank is full.'

"Q. Hadn't you already instructed him to discontinue pumping from number three? A. Yes.

"Q. How did he happen to be pumping from number three if you had instructed him not to? A. Well, he found there was oil in there. I always told the engineers to drag and get all the oil out of the tank they can — after they are empty there is a certain amount that will run down. And this particular man I had, he was a man that wasn't of sound mind, and a man that couldn't remember anything.

"Q. What was his name? A. Martin Appel.

"Q. Was he the man that you had to promote to the job of assistant engineer because of one of the engineers leaving you at Manila? A. No. No, I wouldn't promote him, because he was not a responsible man for the position.

"Q. Did you know he was of unsound mind when you departed from Manila? A. Yes.

"Q. And you think that was why he didn't obey your orders? Yes."

During the same cross-examination, the proctor for the appellee read the following excerpt from the engineer's log for September 21, 1927, entered while the vessel was still at Manila: "M.J. Peralta, Junior, First Assistant Engineer, paid off to transfer to S.S. Crissfield as chief engineer. W. Parke refused to be raised to second assistant engineer. All engine gang threatened to walk off ship if M.P. Appel was raised to first assistant. Unable to get engineers on shore. Under these conditions I was forced to raise J.L. Mone, water tender, to first assistant engineer, Mr. Mone having in his possession discharges to certify that he has held a chief engineer's position on foreign ships." (Italics our own.)

Blinn stated that the foregoing entries were correct. His testimony then continued:

"Q. Who were the vessel's agents at Manila? A. Swayne Hoyt."

"Q. Did you report to them this trouble with Mr. Appel? A. Yes.

"Q. Did you tell them that he was of unsound mind? A. Yes.

"Q. What did they have to say to that? A. Well, they didn't say anything — to get along with him until we got back to 'Frisco. * * *

"Q. Did you get rid of Mr. Appel as soon as you arrived in San Francisco? A. Yes. Just the minute we paid off I got rid of him."

On recross-examination Blinn stead-fastly adhered to his testimony regarding Appel's incompetency:

"Q. Do you think Mr. Appel was a man of intelligence? A. No."

"Q. Do you think he was a competent man? A. No, he wasn't competent at all."

The appellant seeks to avoid the effect of the foregoing evidence by adducing a lengthy stipulation as to certain matters to which Chief Engineer Blinn would have testified had he been recalled to the stand. The substance of the stipulation is that: "In filling the settling tanks from No. 3 double bottom tanks, Second Assistant Engineer Appel was acting in accordance with the regulations of the vessel and did not disobey any instructions of the Chief Engineer when he filled the settling tanks from No. 3 double bottom tank at any time prior to the discovery of the leakage from the deep tanks into the double bottom tanks." The stipulation sets forth that Blinn would explain his change of testimony as to the filling of the settling tanks by the fact that at the time he gave his original deposition there were available to him only the "smooth engine room log books" instead of the "rough log."

The appellant also seeks to explain away Blinn's testimony by referring to the deposition of Henry F. Gelhaus, the marine superintendent of Swayne Hoyt, of San Francisco, a firm that operated the West Cajoot for the appellant, and were the vessel's agents at Manila. Gelhaus thus testified as to the "temperamental qualities of Mr. Appel" which might have given Chief Engineer Blinn the "impression" that "there was anything wrong with Mr. Appel mentally": "Men aboard ship are something like a family. They get quarrelsome over nothing, and it is very easy for a junior officer to criticize a senior officer, and also a senior officer to criticize a junior officer for personal reasons. They come with petty complaints and say the man is crazy, or he is a hot-head, or various expressions like that, and when they are boiled down, on investigation we find there is no sense in them in fact."

Finally, the appellant refers to the fact that it presented Appel to the court below, and argues that his "frank, intelligent testimony and engaging appearance completely demonstrate the man's qualities." It is also contended that "the Court's finding that the West Cajoot was seaworthy on sailing from Manila necessarily implies its finding that Appel impressed the Court as being competent." Reference is likewise made to the fact that the District Judge commented favorably upon the neatness of Appel's entries in the rough engine room logbook.

It is to be observed that Blinn's alleged opinion that Appel was of unsound mind was evidently based upon the latter's asserted "forgetfulness." At any rate, Blinn specified no other evidence of his subordinate's asserted mental weakness. The chief engineer explained his statement regarding Appel's mentality as follows: "Q. When you say he is not of sound mind, just what do you mean? A. Well, his peculiar actions on the ship; and he had a tendency to forget things."

Nowhere in Blinn's testimony do we find any specific reference to Appel's "peculiar actions." As to the latter's alleged forgetfulness, the stipulation to which we already have referred shows that, as to the filling of the settling tanks from the No. 3 double bottom tanks, it was Blinn's recollection, and not Appel's, that was "confused."

It is well settled that a lay witness is not competent to give his general impression as to the sanity of another person without specifying the data upon which his opinion is based. See Craig v. United States (C.C.A.9) 81 F.2d 816, 829-831, and the cases there cited.

Therefore, considering the present record as a whole on the subject of Appel's competency, including the evidence of his long employment record both before and after his service on the West Cajoot, we find no reason for disturbing the finding of the court below that the West Cajoot was in a seaworthy condition when she left Manila.

Much is said in the briefs concerning the cause of the stranding. The appellant contends that the mishap occurred "on an uncharted pinnacle," while the appellee charges that it "resulted from negligence in navigation."

In our view of the case, it is immaterial which theory is adopted, since, if the vessel is seaworthy, the Harter Act (46 U.S.C.A. §§ 190-195) exempts the shipowner from liability in either case. Section 3 of that statute (46 U.S.C.A. § 192) reads as follows: "If the owner of any vessel transporting merchandise or property to or from any port in the United States of America shall exercise due diligence to make the said vessel in all respects seaworthy and properly manned, equipped, and supplied, neither the vessel, her owner or owners, agent, or charterers, shall become or be held responsible for damage or loss resulting from faults or errors in navigation or in the management of said vessel nor shall the vessel, her owner or owners, charterers, agent, or master be held liable for losses arising from dangers of the sea or other navigable waters, acts of God, or public enemies, or the inherent defect, quality, or vice of the thing carried, or from insufficiency of package, or seizure under legal process, or for loss resulting from any act or omission of the shipper or owner of the goods, his agent or representative, or from saving or attempting to save life or property at sea, or from any deviation in rendering such service."

The Harter Act, however, places upon the shipowner another burden, which is altogether independent either of seaworthiness or of careful navigation. As the appellant itself concedes, "the shipowner remains * * * still liable for negligence in the custody, care and proper delivery of the cargo." The pertinent provision of the Harter Act is section 1 (46 U.S.C.A. § 190), which reads as follows: "It shall not be lawful for the manager, agent, master, or owner of any vessel transporting merchandise or property from or between ports of the United States and foreign ports to insert in any bill of lading or shipping document any clause, covenant, or agreement whereby it, he, or they shall be relieved from liability for loss or damage arising from negligence, fault, or failure in proper loading, stowage, custody, care, or proper delivery of any and all lawful merchandise or property committed to its or their charge. Any and all words or clauses of such import inserted in bills of lading or shipping receipts shall be null and void and of no effect." See, also, section 2 of the same act (46 U.S.C.A. § 191).

The foregoing rule is expressly recognized in the bill of lading under which the cocoanut oil in question was shipped. The document contains the following provision: "It is expressly agreed that if shipowners shall have exercised due diligence to make the vessel herein mentioned seaworthy and properly manned, equipped, and supplied, carrier shall not be or be held liable for any loss of or damage to cocoanut or other vegetable oil carried in bulk which shall be the result of contamination, discoloration, leakage, seepage, rust, effects due to carrying said oil in bulk, effects of steam coils and/or their connection, or the result of any other causes or circumstances of any nature, kind, or character unless it be first proven that such loss or damage was caused by or resulted from carrier's neglect or fault or failure in proper loading, stowage, custody, care or proper delivery of said oil."

When a stranding results from a cause excepted by the bill of lading or by the Harter Act, the carrier nevertheless must exercise reasonable care in preserving the cargo from further damage, even though such cargo may have been partly damaged, as here, by the stranding itself as an immediate cause.

In The Portsmouth, 9 Wall. (76 U.S.) 682, 687, 19 L.Ed. 754, the court said: "Were it necessary, it would be easy to show that the conduct of the master after the vessel was stranded was entirely unjustifiable. It was his duty even then to take all possible care of the cargo. * * * Losses arising from dangers of navigation, within the meaning of the exception in the bill of lading, are such only as happen in spite of the best human exertions, which cannot be prevented by human skill and prudence. But in this case no effort was made to save the cargo. The salt was not thrown overboard until after the arrival of the tug. The fog had then lifted. The wind and the sea had subsided. It is evident the salt might then have been saved, if it could not have been removed before."

And in The Willdomino (C.C.A.3) 300 F. 5, 16, affirmed 272 U.S. 718, 47 S. Ct. 261, 71 L.Ed. 491, the following language was used: "It was the duty of the carrier to exercise reasonable care, diligence, and activity in preserving from further damage any portion of the cargo that had been damaged by the stranding so far as it could be done consistently with the carrier's duty to the owners of other portions of the cargo. [Authorities cited.]" See, also, Bixby v. Deemar (C.C.A. 5) 54 F. 718, 720; McNeil Higgins Co. v. Old Dominion S.S. Co. (C.C.A.7) 235 F. 854, 856.

The burden of showing that the damage to cargo arose from one of the excepted causes is on the shipowner. The Folmina, 212 U.S. 354, 363, 29 S.Ct. 363, 53 L.Ed. 546, 15 Ann.Cas. 748; The Rosalia (C.C.A.2) 264 F. 285, 288.

If, however, the shipowner brings himself within the exceptions of the Harter Act or the bill of lading, the burden rests upon the shipper to establish negligence on the ship's part. The Florinda (C.C.A.2) 31 F.2d 262, 264, certiorari denied Saitta v. The Florinda, 279 U.S. 874, 49 S.Ct. 514, 73 L.Ed. 1009.

In the instant case, Kobe was the port of refuge, at which the West Cajoot went on dry dock for repairs. At Kobe, the vessel was boarded by the agent of her operators, Swayne Hoyt, Inc., and by the port superintendent and other officials of the United States Shipping Board. Those officials, who unquestionably can be regarded as the responsible representatives of the shipowner, held a conference and "discussed ways and means of handling the damage to the ship as well as the cargo."

Under such circumstances, if the West Cajoot sailed from Kobe in an unseaworthy condition, the fault is attributable to the shipowner, and cannot be regarded as a fault of navigation or management, which would come under the exceptions of the Harter Act.

In May v. Hamburg, etc., Aktiengesellschaft (The Isis), 290 U.S. 333, 344, 54 S. Ct. 162, 164, 78 L.Ed. 348, Mr. Justice Cardozo said: "If the master of the Isis had acted on his own responsibility at Bremen in sending the vessel on, the fault would have been negligence in management, or so we may assume. But that is not what happened. The owner intervened by its marine superintendent, who was sent from Hamburg to Bremen to inspect the disabled vessel. * * * He consulted with the master and others. The decision in the end was his. This he tells us very frankly. If reasonable diligence would have shown that the vessel was unseaworthy when he sent her on her way, there was something more than an error in navigation or management on the part of master or of crew. There was a failure by an owner to fulfill the condition on which immunity depended." See, also, The Guildhall (C.C.A.2) 64 F. 867, 868, 869.

If the West Cajoot left Kobe with a leak between her port deep tank, containing cocoanut oil, and her port double bottom tank, containing fuel oil, she was unseaworthy for the carriage of cocoanut oil. "A tank from which none of the liquids on earth except cocoanut oil are likely to escape is unseaworthy for the carriage of cocoanut oil." The Arakan (D.C.) 11 F.2d 791, 793.

Finally, if the loss to cargo is caused jointly by an excepted cause and by the shipowner's negligence in the custody of the consignment, the shipowner has the burden of proving what damage is attributable to the excepted cause and what loss has been the result of his own negligence.

In Schnell v. The Vallescura, 293 U.S. 296, 305, 306, 55 S.Ct. 194, 197, 79 L.Ed. 373, Mr. Justice Stone said:

"Here the stipulation was for exemption from liability for a particular kind of injury, decay. But the decay of a perishable cargo is not a cause; it is an effect. It may be the result of a number of causes, for some of which, such as the inherent defects of the cargo, or, under the contract, sea peril making it impossible to ventilate properly, the carrier is not liable. For others, such as negligent stowage, or failure to care for the cargo properly during the voyage, he is liable. The stipulation thus did not add to the causes of injury from which the carrier could claim immunity. It could not relieve him from liability for want of diligence in the stowage or care of the cargo.

"It is unnecessary for us to consider whether the effect of the clause is to relieve the carrier from the necessity, in the first instance, of offering evidence of due diligence in caring for a cargo received in good condition, and delivered in a state of decay. [Cases cited.] For here want of diligence in providing proper ventilation is established and it is found that the failure to ventilate has caused the damage. It is enough that the clause plainly cannot be taken to relieve the vessel from bringing itself within the exception from liability for damage by sea peril where the shipper has carried the burden of showing that the decay is due either to sea peril, in that bad weather prevented ventilation, or to the vessel's negligence. Where the state of the proof is such as to show that the damage is due either to an excepted peril or to the carrier's negligent care of the cargo, it is for him to bring himself within the exception or to show that he has not been negligent. [Case cited.]

"Similarly, the carrier must bear the entire loss where it appears that the injury to cargo is due either to sea peril or negligent stowage, or both, and he fails to show what damage is attributable to sea peril. [Many cases cited.]"

See, also, The Medea (C.C.A.9) 179 F. 781, 786.

The foregoing principles govern the appellant's responsibility for the custody of the cocoanut oil after the West Cajoot reached Kobe, and also the appellant's duty to make the vessel seaworthy before leaving that port. It next becomes necessary to apply those principles to the facts disclosed by the apostles.

At the outset, it is conceded that the contamination of the cocoanut oil was caused by a leak or "aperture" between the port deep tank, containing the appellee's cargo, and the port double bottom tank, containing fuel oil. It is likewise conceded that this aperture was not repaired at Kobe, despite the fact that the vessel was there exactly four weeks, most of which time was spent on dry dock. It is also undisputed that the cocoanut oil was not removed from the deep tanks until the West Cajoot discharged its cargo at its destination, Wilmington, Cal.

Two defenses are offered by the appellant for its failure to remove the cocoanut oil from the deep tanks and repair the aperture between the port deep tank and No. 3 port double bottom tank:

"A. The contamination of the cocoanut oil had already occurred when the vessel reached Kobe and did not increase after it was discovered on October 5th.

"B. There were no facilities at Kobe for stowing, refining or otherwise handling the cocoanut oil so as to decrease the damage."

We will first consider the contention that the contamination of the cocoanut oil did not increase after October 5.

In the first place, it is argued that the aperture was "insignificant" and "minute in the extreme." Elsewhere in its brief, however, the appellant points out that "the amount of fuel oil needed to contaminate cocoanut oil is astonishingly small." In view of this latter statement, which accords with the facts, the appellant's contention that the leak was "insignificant" reminds one of Mercutio's grim comment regarding his mortal wound: "No, 't is not so deep as a well, nor so wide as a churchdoor; but 't is enough; 't will serve."

Again, it is urged that both the shortage and the contamination had already occurred when the vessel reached Kobe, and did not increase after the contamination was discovered on October 5.

If the appellant's theory on this point is correct, it is excused from liability for either the shortage or the contamination; because in that event the loss and the shortage would be due entirely to an excepted cause — the stranding — regardless of whether or not the stranding was caused by a danger of the sea or by errors in navigation or management.

If, on the other hand, the appellant's contention is not borne out by the facts, and it is found that only a part of the loss was caused as a result of the stranding, and the rest was the result of negligence in the care of the cargo, the shipowner is exempted from liability only for the loss attributable to the stranding.

The burden of bringing itself within the exception of the Harter Act by showing that the loss and damage were due in whole or in part to the stranding rests upon the appellant. If it establishes that all of the shortage and contamination were the result of the stranding, the appellant is exempt from liability. If, on the other hand, the appellant can show only that part of the loss was attributable to the stranding, and the appellee sustains its burden of proving that the appellant has been guilty of negligence in the care and custody of the cargo, the further burden of attributing the proper proportion of the entire damage to stranding on the one hand and its own negligence on the other rests upon the appellant.

The appellant contends that both the shortage and the contamination of the cocoanut oil already had entirely occurred when the vessel reached Kobe, and did not increase after the contamination was discovered on October 5. This contention is bottomed upon several grounds:

First, that some of the fuel oil in the No. 3 double bottom tanks was drawn off on October 3, when the settling tanks were filled; that, after the discovery of the contamination on October 5, no additional fuel oil was pumped from the No. 3 double bottom fuel tanks; and that, when the cargo was discharged at Los Angeles in the early part of December, there was found a 9-inch vacuum in the No. 3 port double bottom tank, which was below the port deep tank and into which the cocoanut oil had leaked. From all this it is argued that the leak had already been plugged up when the fuel oil was drawn off from the double bottom tanks on October 3.

Second, that John H. Bower, a chemist for the United States Bureau of Standards, had analyzed samples of the cocoanut oil taken at Kobe and "compared the results of his analysis" with an analysis by expert chemists employed by the appellees, who examined the cocoanut oil when it was discharged at Los Angeles. Bower testified that "both analyses indicate the same thing, only slight contamination," and that the report of the appellees' chemists did not indicate "any increase in contamination over the analysis" made by him.

Without entering upon a detailed analysis of the evidence, we will discuss these two grounds seriatim.

In support of its contention that "at some time between the stranding on October 2nd, and prior to the discovery of the contamination [on October 5], the aperture had become sealed," the appellant points to the fact that, when the cargo was discharged, there was found a 9-inch vacuum in the port double bottom tank, which was below the port deep tank, and into which the cocoanut oil had leaked.

From this fact it is argued that all of the shortage must have occurred between October 2 and October 5, and that therefore the appellees' "contention that the shipowner was derelict at Kobe in the care and custody of the cargo can relate only to the `delivered, contaminated' cargo; it can, in no way, impose liability for the previous shortage."

In support of its theory that no pumping of oil could have occurred after October 5, the appellant points to the testimony of Chief Engineer Blinn and Second Assistant Engineer Appel, to the effect that on that date "all of the pipes by means of which fuel oil could be pumped from the No. 3 double bottom tanks were disconnected and `blanked off'." To make sure that the valves could not be opened, they were fastened together with wire. The appellant also relies upon the testimony of two witnesses for the appellee — Capt. Joe Jory and Capt. W.E. Heppell. Captain Jory stated that, whenever the vacuum occurred, it was caused by the fact that the cocoanut oil had congealed and stopped the leak. He further said that such an empty space would indicate that fuel oil had been pumped out "at some time." Captain Heppel testified to the same effect. Both experts said that they could think of no other explanation for the vacuum. Neither expert, of course, attempted to fix the time of the last pumping from the double bottom tank, and the questions asked of them on that subject omitted the time element.

The appellant's theory as to no leakage and no pumping from the port deep tank after October 5 overlooks one stubborn fact, positively testified to by the chief engineer, one of the appellant's principal witnesses. Blinn swore that the No. 3 double bottom tank, containing the fuel oil, was full when the vessel sailed from Kobe on November 1. In view of his testimony, it is clear that the vacuum of 9 inches in that tank must have occurred some time during the trip from Kobe to Los Angeles, and not at some time prior to October 5, as the appellant contends.

Whether this much-discussed vacuum was created by a leak in the double bottom tank or by pumping off some of the fuel oil during the trip across the Pacific to Los Angeles, via San Francisco, is not known. Nor is it encumbent upon the appellees to account for that space in the double bottom tank. In The Medea, supra, this court said:

"The reason why the burden of proof is placed upon the carrier is that he or his servants know, or at least ought to know, the circumstances connected with the loss or damage to the cargo, while the owner of the lost or damaged goods has no such knowledge, and if he were required to furnish such proof it would operate as a denial of justice. [Cases cited.] * * *

"These cases settle the rule beyond controversy that with respect to the liability of a common carrier for loss or damage to goods while in his possession the question as to the burden of proof is not one of pleading, but of primary liability which the carrier must meet according to the tenor of his contract."

In connection with the possibility of a leak in the double bottom fuel tank, it should be observed that the record shows there was a leak from the starboard deep tank, probably into the port deep tank. Regarding this leak, Chief Officer Sorensen testified as follows:

"Q. According to these measurements, then, and according to the outages between October 6th and October 12, the port tank gained four inches of oil? A. Yes.

"Q. And during the same period the starboard tank lost about 3½ inches of oil? A. 3½ inches of oil.

"Q. Have you any explanation of those figures? A. No, the only explanation I can give is there was a leak somewhere where it ran in from the starboard side into the port side, a leaky rivet in the bulkhead, for instance, or something like that."

Later on in his testimony, Sorensen suggested an alternative explanation for the admitted leakage:

"Q. That showed that some oil had been lost from the starboard deep tank? A. Yes.

"Q. How did you say you accounted for that? A. The only way would be it leaked out of there.

"Q. How could it leak out of there? A. Through various ways, for instance like leaky rivets in the bulkhead.

"Q. Where would it leak to? A. If the leaky rivets was in the bulkhead, that is the amidships bulkhead, it would leak in to the port side; if the leaky rivet was in the shell of the ship it would leak over the side."

The appellant does not seem to have paid any attention to this leakage between the two deep tanks, and its proctors ignore the subject in their brief. For a better understanding of the magnitude of the shortage from this starboard deep tank, we append the following table:

Pounds Total cocoanut oil shipped ............ 1,223,583 Each deep tank carried approximately one-half of the cocoanut oil ......... 611,791½ Starboard Deep Tank Oil shipped ........ 611,791½ pounds Oil recovered ...... 553,943 _______ Shortage from starboard deep tank ..... 57,848½ Port Deep Tank Oil shipped ........ 611,791½ pounds Oil recovered ...... 574,670 _______ Shortage from port deep tank .......... 37,121½ _________ Excess of shortage in starboard deep tank over shortage in port deep tank 20,727 pounds

From the foregoing, it will be seen that, although the leakage into the double bottom fuel tank was only from the port deep tank, the shortage from the port deep tank was only three-fifths of that from the starboard deep tank.

A careful study of the record has convinced us that the appellant has failed to sustain its burden of proving that all of the shortage of cocoanut oil had occurred prior to the discovery of the contamination, on October 5.

We turn next to the question of whether or not all of the contamination had occurred prior to the discovery of some leakage between the double bottom tanks and the deep tanks on October 5.

Here again the appellant relies upon the mysterious 9-inch vacuum in the port fuel tank discovered when the West Cajoot arrived at Los Angeles early in December; and here again the same answer can be made to its contention. Chief Engineer Blinn, one of the appellant's chief witnesses, testified that the No. 3. double bottom fuel tank was full when the West Cajoot left Kobe on November 1. Hence the 9-inch vacuum in that tank must have occurred some time during the eastward trip across the Pacific, and not prior to October 5.

It is still necessary, however, to consider the testimony of Bower, the chemist of the Bureau of Standards. It had been agreed between counsel that samples of the oil taken at Kobe should be sent to the Bureau of Standards for analysis. Bower testified that his analysis of the oil taken out at Kobe and an analysis of the cocoanut oil made by Curtis Tompkins, Limited, chemists employed by the appellees, both indicated "the same thing — only slight contamination," and that the Curtis Tompkins report indicated no increased taint.

From this testimony the appellant argues that there was no increase in contamination after October 5, and that "any failure of the shipowner to take action at Kobe becomes moot, as it resulted in no increase of the damage to the cargo." The testimony of Bower, however, does not establish that neglect at Kobe did not increase the contamination of the cocoanut oil. In the first place, it should be observed that only one sample from each tank was analyzed by the government chemist. Yet many samples, from different levels of each tank, were taken after the contamination was discovered on October 5. First, on October 6, at the direction of Samuel Reid, the appellant's cargo surveyor at Kobe, Chief Engineer Blinn took samples from the No. 3 deep tanks. On October 12 Blinn again took samples of the cocoanut oil. The chief engineer testified that the samples were taken "at different levels — the bottom, center, and the top of the tank." He kept one sample aboard the ship, and sent the rest to A.C. Watson, agent at Kobe for Swayne Hoyt, operators of the West Cajoot. Watson sent two of the samples to Swayne Hoyt at San Francisco, from where they were forwarded to the United States P. I. Agency at New York, that agency being the claims department of the Shipping Board. By that agency the two bottles were mailed to the Admiralty Counsel of the Shipping Board, which in turn mailed them to the Bureau of Standards.

Surveyor Reid himself took samples of the cocoanut oil in the deep tanks on October 4, or "a day after the vessel arrived." He testified that a sample taken from the lowest part of the deep tank on the port side was found to be discolored.

Reid also stated that at Kobe samples of the cocoanut oil were taken from different levels of the deep tanks. On that subject, his testimony continued:

"A sample taken from the lowest part in the deep tank on the port was found to be discolored. The oil in all the other parts was normal. Samples were taken from the upper, the middle, and the lowest stratums respectively, on the port and the starboard of the vessel. Six bottles of samples were taken in all. The only contamination was found to be at the lowest part on the ship's port. * * *

"The contamination in the oil existed from the bottom of the tank to a depth of some six inches and the upper portion in the tank was clear in general."

In this connection, it should constantly be borne in mind that, when the cocoanut oil in the port deep tank was discharged at Los Angeles two months later, it was not "clear in general," but "contaminated."

The question of the levels from which the various samples were taken, therefore, is important in determining the extent of the contamination from October 6, when Blinn's first samples were taken, to the early part of December, when the cocoanut oil was discharged from the port deep tank and was found to be entirely contaminated. Yet there is nothing either in the stipulation that samples of the oil should be sent to the Bureau of Standards or in Bower's own testimony or report that specifies the levels from which they were taken. According to Watson's own testimony, not all the samples taken were sent by him to the United States for analysis: "There were samples forwarded to Swayne Hoyt at San Francisco; there were samples put aboard the vessel; one set I believe was given to the Chief Engineer and one to the captain, and the surveyor retained a set."

Similarly, the record is silent as to whether the samples analyzed by Bower were taken on October 6 or October 12.

Chief Engineer Blinn and Surveyor Reid were not the only ones who took samples of the cocoanut oil in the deep tanks of the West Cajoot at Kobe. Chief Officer Sorensen also took samples, from different levels, on October 6, in Capt. Reid's presence.

Reid testified that "all the samples were handed to A.C. Watson, the vessel's representative." The record is clear that Watson sent only two of the samples to the United States for analysis. The question presses upon us as we study the apostles: Which of these many samples did Bower have before him when he made his analysis, so much relied upon by the appellant?

As a matter of fact, there is considerable evidence in the record — including testimony by the appellant's own responsible officials — tending to establish the fact that the contamination of cocoanut oil increased during the period between October 5 and December 1-2, when the oil in the port deep tank was discharged.

We have already referred to Surveyor Reid's testimony to the effect that, at Kobe, "the only contamination was found to be at the lowest part on the ship's port."

Watson admitted that he knew that the contamination would increase:

"Q. But did you seriously think, Mr. Watson, that all of the damage to this cocoanut oil had already been done between the time of the stranding and the time of this vessel's arriving in Kobe? A. No, I did not.

"Q. Didn't you know that the damage was going to continue to go on as long as the two different classes of oil were left in the two tanks adjoining each other? A. I did."

Elmer Ord Slater, a chemical engineer with 23 years' experience — including experience in analyzing cocoanut oil — gave the following testimony:

"Q. Why, in your opinion, should it have been taken out immediately upon discovering the contamination? A. For the reason that when oil is held in storage, that is, when cocoanut oil is held in storage and it is still liquid, and fuel oil has gotten into it, the fuel oil is miscible with the cocoanut oil.

"The Court: Is what? A. It is miscible, that is, it will mix readily with it. I mean by that, it differentiates between getting water into the oil, which would stay on the bottom and not have a tendency to work up into the oil. There are several factors that would tend to make the fuel oil work up into the cargo of cocoanut oil so long as it is liquid."

Slater testified that, even if all of the cocoanut oil in the port deep tank was "more or less contaminated by fuel oil," he "would still consider it advisable to take the oil out of the tank."

Slater's testimony as to the spread of contamination resulting from failure to remove the cocoanut oil was corroborated by Harvey C. Bennett and William F. Beedle, experienced chemists, and by Capt. Leonard Arthur Waters, shipmaster and marine surveyor, who had inspected the West Cajoot several times; Capt. W.E. Heppell, shipmaster and marine surveyor, who had handled cocoanut oil cargoes; and Capt. Joe Jory, shipmaster and surveyor, who likewise had had experience with cocoanut oil, and whom this court has declared to be "a navigator of impressive seamanly qualifications." California Hawaiian Sugar Refining Corporation v. Rideout (C.C.A.) 53 F.2d 322, 326.

It should be observed that Bennett, Beedle, Slater, and Waters testified in open court, and that therefore the District Judge was in a position to note their manner on the witness stand. Apparently the court below was impressed, as we are, by the positive testimony of those experts, to the effect that the contamination increased after October 5; for the court found that "there was a failure on the part of respondent in the proper care and custody of said cargo in not removing and saving it at Kobe in order to avoid further damage."

Accordingly we hold that the appellant has failed to sustain its burden of showing that all the loss was occasioned by a cause coming within the exceptions of the Harter Act, to wit, by the stranding. In other words, it has failed to establish its first defense; namely, that the contamination of the cocoanut oil had already occurred when the vessel reached Kobe, and did not increase after the taint was discovered on October 5.

Before the appellees can recover for the damage, however, they, in turn, must sustain their burden of proving that the appellant was negligent in the care and custody of the cargo at Kobe. A consideration of this charge of negligence involves an inquiry into the merits of the appellant's second defense — that there were no facilities at Kobe for stowing, refining, or otherwise handling the cocoanut oil so as to decrease the damage.

At the outset, the appellant insists that the testimony of the appellees' witnesses on the question of what should have been done at Kobe "should be absolutely excluded, as none of these witnesses were in any way qualified to testify on this point." The appellant's chief attack upon the qualifications of these disinterested experts is that they had been at Kobe either not at all, or not long enough, or too long ago.

We do not agree with the appellant's implication that a seasoned seaman, an experienced surveyor, or an expert chemist required a long-time residence in Kobe, continuing almost up to the date of the trial, before he would be qualified to testify as to what could have been done at Kobe with the cocoanut oil in the two deep tanks. A shipping expert need not be one of the "oldest inhabitants" of the port concerning whose facilities he is asked to testify.

Here again, however, it is not necessary to depend upon the testimony of the appellees' experts; for the appellant is convicted of negligence out of the mouths of its own witnesses.

In the first place, the master, the chief officer, and the chief engineer of the West Cajoot were not consulted about the advisability of removing the cocoanut oil from the No. 3 port deep tank.

Capt. Thorsen testified that he did not remember any one's discussing the subject with him. Chief Officer Sorensen said that his advice was not asked as to whether it would be wise to remove the cocoanut oil. Chief Engineer Blinn, who personally took soundings of the double bottom fuel tank, and took samples of the cocoanut oil, frankly admitted that he "never gave that a thought," and that he was not consulted.

Nor was counsel taken with Bunichi Nagamatsu, surveyor for the Teikoku Kaiji Kyokai, the representative in Japan of the American Bureau of Shipping. Nagamatsu testified:

"After the completion of the repairs to the S.S. West Cajoot, I considered that she was seaworthy to return to the United States, and I never gave any thought about a leakage from the deep tank to the double bottom tank. Consequently, I never had thought of whether or not the vessel was seaworthy to carry cocoanut oil in her deep tank. * * *

"The damaged portion was repaired in accordance with my recommendation, and I did not discover whether or not there was a leak from the deep tank to the double bottom tank. Consequently, I was never requested by the customer to pump out the cocoanut oil from the deep tank into a temporary storage and make repair of the leak."

The survey report made by Nagamatsu, in which he declares the West Cajoot seaworthy "so far as the damaged parts are concerned," is dated October 27, 1927. Yet on October 3 Watson "suspected" that there was a leak from the deep tank into the double bottom fuel tank. On October 4 Surveyor Reid found discoloration in the cocoanut oil. On October 5 Blinn actually discovered contamination. "Several days" after the arrival at Kobe, Watson "discovered" that what he had long suspected was true; namely, that the cocoanut oil was contaminated. He immediately notified R.M. Johnson, assistant director for the Orient of the United States Shipping Board Merchant Fleet Corporation, who in 1927 was district engineer for the Orient of that same corporation, and also notified Thornton, the general agent for the Shipping Board. On October 14 Watson cabled the news of the contamination to Sawyne Hoyt at San Francisco.

This mass of evidence leaves no doubt in our mind that, by the middle of October, 1927, at the latest, every responsible officer of the Shipping Board at Kobe knew that there was a leak between the port deep tank and the port double bottom tank. Yet none of those representatives of the United States felt it necessary to notify the surveyor of the leak, but all permitted him to issue a certificate of seaworthiness in ignorance of it!

Nagamatsu was the appellant's own witness. We do not doubt that he told the truth when he testified that he did not "discover" or give "any thought" to a leak, and that "the customer" — namely, the United States — never requested him to pump out the cocoanut oil and repair a leak. In a word, Nagamatsu was not notified that a leak existed.

We turn, finally, to the question of whether or not the appellant's officials exercised due diligence in endeavoring to obtain facilities at Kobe for stowing, refining, or otherwise handling the cocoanut oil so as to decrease the damage. That question is involved in the appellant's second defense; namely, that there were no such facilities.

Kobe is the largest port in the Orient, though it is not a point of shipment of cocoanut oil. A priori, it is quite unlikely that a port of that size would not have facilities for the discharge of approximately 260 long tons of cocoanut oil.

Indeed, Agent Watson admitted that Kobe surpassed even the great port of Yokohama in general facilities.

If the appellant failed to use due diligence in the care and custody of the cargo after the stranding, it is liable for all damage attributable to such lack of diligence.

The fact of the matter seems to be that the responsible officials of the appellant made only half-hearted efforts to obtain facilities at Kobe. Surveyor Reid admitted that "at that time there were many vessels sailing from Kobe to the Pacific ports, besides the S.S. West Cajoot, but I do not remember the names of those vessels nor have I made any investigation thereof." He also admitted that, "theoretically speaking, it would be possible to transfer the cocoanut oil from one vessel to another vessel," but added that it would be "disadvantageous to do so from a business standpoint"; "that practically it could not be done," etc. He testified that he thought cocoanut oil "could be transferred to the vessel Achilles, but I never made an inquiry."

Agent Watson admitted that there were Shipping Board vessels "coming in and going out of Kobe all the time" — "vessels like the West Cajoot." Watson likewise admitted that he gave no "consideration to transferring this shipment of cocoanut oil from the West Cajoot to any other Shipping Board vessel."

In view of the fact that transfer of the cocoanut oil from one Shipping Board vessel to another would have reduced the expense of transshipment to the minimum, Watson's evident embarrassment when the question of such transshipment was pressed is quite understandable:

"Q. Did you give any consideration to transferring this shipment of cocoanut oil from the West Cajoot to any other Shipping Board vessel? A. Well, I do not believe that — I do not know what to say, only to say that there are no homeward bound vessels calling there or there were not at that time. There may have been one occasionally that would go to North Japan, but I do not think there was anything home bound that would have a deep tank available.

"Q. Did you make any inquiries as to that situation at the time? A. No, I did not.

"Q. You don't really know whether there was a vessel available or not? A. No."

Similarly, Watson made no inquiries as to whether tank cars could be used, and said that he believed that he had never seen a tank car in Japan. Surveyor Reid, however, testified that at or near Kobe there were tank cars available for the storage of cocoanut oil, but that "every one of them was dirty and unfit," and "could not be rented." Reid, however, does not say how he reached the conclusion that the cars could not be rented, but on the contrary adds that it was not his duty to make the investigation concerning storage facilities at Kobe.

Johnson, the appellant's district engineer at Kobe, testified that there were storage tanks for fuel oil at Noda, which is a part of Kobe, but that he was informed by Watson that he had made inquiry from the Rising Sun Oil Company and found that there were none available for storage of cocoanut oil.

Watson himself under oath repeatedly denied that he had made inquiries concerning the Noda tanks or any shore tanks whatsoever.

Indeed, the only inquiries regarding storage facilities at Kobe that Watson seems to have made related to lighters or barges. These inquiries were made before he had discovered that the cocoanut oil was contaminated, and were not renewed after such discovery. Watson's reason at that time for wishing to have the cocoanut oil discharged was to enable repairs to be made to the vessel and also to lighten her before dry-docking.

So far we have considered only the appellant's own testimony as indicating a lack of energy and resourcefulness on the part of its responsible officers at Kobe. In addition, there is a mass of impressive evidence given by five of the appellees' experts — Bacon, Stanley, Heppell, Jory, and Waters — to the effect that the cocoanut oil in the deep tanks could have been removed at Kobe into barges, another vessel, a shore tank, tank cars, drums, or the West Cajoot's own forepeak and afterpeak tanks.

It would unduly lengthen an already lengthy opinion to analyze this expert testimony in detail. Suffice it to say, neither the professional standing nor the motives of these specialists have been impugned, and their positive evidence alone amply supports the lower court's finding that: "This uncontaminated cocoanut oil could have been removed and saved at Kobe, and there was a failure on the part of respondent [the appellant] in the proper care and custody of said cargo in not removing and saving it at Kobe in order to avoid further damage."

On the question of whether or not lighter tanks, railroad tanks, or shore tanks available at Kobe were "dirty and unfit," we will permit ourselves a brief quotation from the testimony of Capt. Waters, a veteran surveyor and shipmaster, who had been in Kobe the year after the stranding, discharging a cargo consisting principally of oils in drums, and who testified in open court that he knew that tank cars, oil tanks, lighters, and drums were to be found in Kobe during a period covering a number of years. Capt. Waters graphically described how he himself had cleaned tanks for the reception of cocoanut oil: "A. There are various methods of cleaning them [tanks]. I have cleaned a number of tanks personally. I have supervised the cleaning of tanks and helped to clean them out, and in the particular case, taking cocoanut oil, the usual practice is to steam the tank down for a number of hours, to cut the oil off the sides of the tanks, wash down with heavy hose and hot water, and a lye or a caustic is used. After that it is pumped out and washed down, dried. Sometimes it is washed out with fresh water and then they take cocoanut meal, wipe the tank down with cocoanut meal, and sometimes they use the oil itself to wipe the tank out. Then this is done under survey and when the surveyor approves the tank fit to take the oil, it is taken aboard and taken into the tank."

Capt. Waters' testimony is too detailed and positive to admit of error. He was either telling the exact truth or he was intentionally falsifying. The latter alternative is untenable.

Nor does the appellant's criticism of Capt. Waters' qualifications, based upon alleged insufficient residence in Kobe, apply here. We do not believe that the appellant will contend that oil tanks at Kobe are harder to clean than are oil tanks elsewhere!

We next must consider whether the West Cajoot was seaworthy when she sailed from Kobe on November 1 after responsible representatives of the shipowner had taken supervision of her repairs.

Capt. Jory testified that he would not have given the vessel a seaworthy certificate, in view of the fact that there was a leak between her port deep tank and her port double bottom tank. Such a leak he said, was contrary to rules laid down by Lloyd's, the American Bureau, and the Bureau Veritas. Apart from any rules, however, "a vessel which sailed from Kobe on a trans-Pacific voyage, with a leak of unknown dimensions in one of her tank tops," was not in a seaworthy condition, he said.

Jory's testimony was corroborated by that of Capt. Heppell, Stanley, and Capt. Waters.

Furthermore, the leak between the starboard deep tank and the port deep tank rendered it even more advisable to remove the cocoanut oil from the port deep tank, according to the testimony of Capt. Jory and Capt. Waters.

We believe that the evidence adduced by the appellant fails to sustain the shipowner's burden of proving that the West Cajoot was seaworthy when she sailed from Kobe. The Isis, supra, 290 U.S. 333, at pages 346, 347, 54 S.Ct. 162, 78 L.Ed. 348.

Since, therefore, under the teaching of the Vallescura Case, supra, the appellant has failed to show what proportion of the damage resulted from an excepted cause, and what proportion was caused by its own negligence — such negligence having been sufficiently proved by the appellees — the appellant is liable for the entire shortage and contamination of the cocoanut oil, except the ten tons that the court found had been contaminated when the vessel reached Kobe. This latter contamination involved the bottom six inches in the port deep tank.

In their cross-appeal, the appellees assert that the lower court erred in failing to hold that the West Cajoot's call at Yokohama for fuel oil was a deviation. This alleged error is neither specified nor argued in the appellees' brief, and we must assume that it has been abandoned. See Craig v. United States, supra, 81 F.2d 816, at page 831. At any rate, we do not believe that the claim of deviation has merit, since the bill of lading gave the vessel liberty "to call at any port or ports in or out of or beyond the customary or advertised route for the purpose of bunkering," etc.

Both sides attack the lower court's decree with regard to the allowance for general average. The appellant complains that the court erred in granting general average only for damage suffered prior to the arrival of the vessel at Kobe, and in not holding that the appellant should recover the amount "found to be due under the general average statement prepared in this case." The two libels filed by the appellant allege that "the contributory value of said shipment of cocoanut oil is * * * $70,581.49, and the amount of the proportionate contribution payable by said shipment is $2,588.92." In its brief, the appellant asserts that "the other 52 interests shown in the general average statement or their underwriters have paid their proportionate amounts."

The appellees, on the other hand, object to the allowance of any general average.

The interlocutory decree grants the appellant "general average contribution directly resulting from the stranding," and refers the cause to "the United States Commissioner to ascertain and compute the amounts due the respective parties." The memorandum of decision "allows recovery * * * for general average for damage suffered prior to the arrival of the vessel at Kobe by reason of the stranding."

The appellant complains that the court's memorandum eliminates "usual general average expenses at the port of refuge, which are considerable, including crew's wages, fuel consumed," etc.

The bill of lading in the instant case contained the customary "Jason Clause," as follows: "If the owner shall have exercised due diligence to make the vessel in all respects seaworthy and properly manned, equipped and supplied, it is hereby agreed that in case of danger, damage, or disaster resulting from faults or errors in navigation, or in the management of the vessel, or from any latent or other defects in the vessel, her machinery or appurtenances or from unseaworthiness, whether existing at the time of shipment, or at the beginning of the voyage (provided the latent or other defect or the unseaworthiness was not discoverable by the exercise of due diligence), the shippers, consignees, and/or owners of the cargo shall nevertheless pay salvage and any special charges incurred in respect of the cargo, and shall contribute with the shipowners in general average to the payment of any sacrifices, losses, or expenses of a general average nature that may be made or incurred for the common benefit or to relieve the adventure from any common peril."

Under the teachings of The Jason, 225 U.S. 32, 55, 56, 32 S.Ct. 560, 564, 56 L.Ed. 969, the foregoing provision is valid, and entitles the shipowner "to contribution under the circumstances stated." See, also, Aktieselkabet Cuzco v. The Sucarseco, 294 U.S. 394, 402, 403, 55 S.Ct. 467, 79 L.Ed. 942.

We believe that the language of the interlocutory decree here appealed from, "general average contribution directly resulting from the stranding," should be interpreted as covering "the payment of any sacrifices, losses, or expenses of a general average nature * * * made or incurred for the common benefit or to relieve the adventure from any common peril." Any expenditure at Kobe, necessitated by the stranding as a proximate cause, including crew's wages and fuel consumed at the port of refuge, is subject to general average contribution by the appellees. So construing the interlocutory decree on the subject of general average, we give it our approval.

Accordingly, the decree is affirmed both as to the award of damages to the appellees for shortage and contamination of the cocoanut oil and as to the appellees' liability for general average contribution.

Decree affirmed.