LUMBARD, Circuit Judge:
On February 7, 1970, the ANTONIO DEMADES, a 700 foot cargo ship owned by the defendant, Anthony Shipping Co., Ltd., then under lump-sum charter to the Hugo Neu Corporation, and carrying over 25,000 tons of steel scrap, sank in the North Pacific on its way to Japan. The cargo owner, Yawata Iron & Steel Co., Ltd., brought suit in the Southern District on February 3, 1971, claiming the shipowner was responsible for loss of the vessel and Yawata’s steel scrap valued at |1,458,014.58 The determination of the cause or causes of the sinking in the February storm faces many difficulties of proof largely because the master and the first mate and the ship’s logs were lost along with eight members of the crew. As the court cannot say that the cargo owner has established its case by a fair preponderance of the evidence, judgment must be for the shipowner.
The ANTONIO DEMADES sailed from Boston for the Panama Canal on January 6, 1970. The crossing of the Atlantic Ocean and Caribbean Sea was relatively uneventful (except for one period of bad whether) and the ship ar- ‘ rived at Cristobal in the Panama Canal Zone on January 13, 1970. The ANTONIO DEMADES was reprovisioned and then transited the Canal and sailed from Balboa for Japan on January 14. After the ship crossed the International Date Line on February 1-2, 1970, the weather worsened. According to Second Mate Gregos (the only surviving deck officer), in the early morning hours of February 6 the wind was about force 7 on
the Beaufort Scale,
and the vessel reduced its speed. At about 1400 hours on February 6, the McGregor steel hatch cover on the No. 1 cargo hold (the forwardmost hold) failed, and the hold flooded. At that time the ship was sailing into the storm and the wave action on the forward part of the ship was such that the Master had to reverse course and reduce the vessel’s speed by one half in order to be able to lead a party forward to examine the hatch cover and No. 1 hold. It was discovered that two sections of the hatch cover had been twisted open and had been thrown on the deck. The hold was filled with water and the hatch cover could not be reclosed. Three hours later the Master turned the ship back into the wind and resumed his course towards Japan. About this time the ship’s crew began pumping water out of No. 2 and No. 3 cargo holds and No. 2 and No. 3 double bottoms. Ten hours later (0500 hours on February 7 the Master ordered the crew to stand by to abandon ship and ordered S.O.S. signals sent. At 0815 hours the order was given to abandon ship. Thereafter the vessel gradually went down by the head and sank. Twenty members of the crew were saved by a ship responding to the S.O.S., but all of the ship’s records were lost.
I.
The applicable law is contained in the Carriage of Goods by Sea Act (Cogsa). 46 U.S.C. §§ 1300-15. Under Cogsa plaintiff established a prima facie case by showing that the scrap was loaded on board the ANTONIO DE-MADES and that it was not delivered. After such a showing the burden is on the carrier “to bring itself within an excepted cause [under Cogsa] or to prove it exercised due diligence to avoid and prevent the harm.” Lekas & Drivas, Inc. v. Goulandris, 306 F.2d 426, 429 (2d Cir. 1962). If the carrier establishes that it falls within such an exception under 46 U.S.C. § 1304(2), the cargo owner must then establish that the vessel was unseaworthy and that the unseaworthiness was at least a concurrent cause of the loss. If the cargo owner establishes that, the carrier can still avoid liability if it shows that it exercised due diligence in an attempt to make the ship seaworthy. 46 U.S.C. § 1304(1); In re Grace Line Inc., 517 F. 2d 404 at 406 (2d Cir. May 19, 1975); Director General v. S.S. Maru, 459 F.2d 1370 (2d Cir.),
cert, denied,
409 U.S. 1115, 93 S.Ct. 898, 34 L.Ed.2d 699 (1972); J. Gerber & Co. v. S.S. SABINE HOWALDT, 437 F.2d 580, 588 (2d Cir. 1971); G. Gilmore & C. Black, Law of Admiralty § 3-43, at 183-85 (2d ed. 1975).
II.
Defendant urges that two statutory exceptions absolve it of liability.
First, it suggests that the storm encountered by the ANTONIO DEMADES in the North Pacific was so severe that it constituted a peril of the sea. See 46 U.S.C. § 1304(2) (c). Second, it argues that the loss was due to an act by, or the neglect of, the master in the navigation or man
agement
of the ship. See 46 U.S.C. § 1304(2)(a).
A. PERIL OF THE SEA
The Court does not believe that the storm encountered by the ANTONIO DEMADES was a peril of the sea. The evidence indicated that storms such as the one involved here were common occurrences during the month of February in this area of the North Pacific. Indeed, at trial defendant’s expert meteorologist testified that this storm was not even the worst storm of that month. Of course, the fact that such a storm should have been anticipated does not mean that the storm cannot have been a peril of the sea. However, in this case the court does not believe that the winds and sea encountered by the ANTONIO DEMADES were of such magnitude to constitute a peril of the sea.
Each side called an expert meteorologist who prepared exhibits showing the wind, sea, and swell conditions in the area of the North Pacific where the ANTONIO DEMADES sank. These exhibits were based on weather data that consisted largely of readings taken by other ships in the general vicinity of the ANTONIO DEMADES. According to custom, these readings were taken every six hours. Thus, the weather data most relevant to the conditions faced by the ANTONIO DEMADES when the hatch gave way at 1400 hours on February 6, 1970, were the readings taken at 1100 hours and 1700 hours on that date.
The two experts (Robert Raguso for the plaintiff and William Kaciak for the defendant) offered conflicting versions of the weather faced by the ANTONIO DEMADES at both 1100 hours and 1700 hours. Much of this difference can be explained by two factors. First, the two experts placed the ship in different positions. Kaciak assumed that the ship was in the same location (i. e., the S.O.S. position) at 1100, 1700, 2300 hours on February 6 and 0500 hours on February 7 while Raguso assumed that the ship gradually reached the S.O.S. position from points to the southeast. Second, Kaciak erred when he plotted the position of the ship. His exhibits indicate that he meant to place the ship at 33°15'N, 157°30'E, but close examination of the exhibits he prepared shows that he mistakenly placed the ship about 20 minutes to the north of that location.
In Raguso’s opinion, at 1100 hours the ANTONIO DEMADES was encountering sustained winds from the west northwest of 40-45 knots (Beaufort force 8-9) and seas of 15 feet from the same direction. Kaciak thought that the ship encountered winds of 50 knots (force 10) and seas of 16 feet. If the ANTONIO DEMADES had been correctly plotted on Kaciak’s chart, it appears that the ship would be experiencing sustained winds of about 47-48 knots (highest force 9 — lowest force 10) and seas of about 15 feet. At 1700 hours Raguso thought that the winds were at 40 knots (force 8) and that the seas were at 15 feet. Kaciak testified that the winds were at 50 knots and the seas at 26 feet. Kaciak’s chart (corrected for the plotting error) shows winds of 47-48 knots and seas of 19 feet. Second Officer Gregos stated in his deposition that the wind was blowing at about force 9-10 while he was on the bridge from 1200 hours to 1600 hours.
While no one will ever know exactly where the ship was at 1100 and 1700 hours on May 6, the court finds that the positions assumed by Raguso are more accurate than those assumed by Kaciak. It seems unlikely that the ship made no headway in the 18 hours preceding the sending of the S.O.S. While it is true that the ship reversed course between 1400 and 1700 hours on February 6, Gregos testified that the ship’s speed during that period was about 30-50 RPMs while its speed in the other direction prior to 1400 hours and subsequent to 1700 hours was 80 RPMs. Thus, the court finds that Raguso’s estimate of
the weather faced by the ANTONIO DEMADES to be more accurate, and concludes that during the period of time when the No. 1 hatch failed the ANTONIO DEMADES did not encounter sustained winds in excess of the upper extent of force 9. It should be noted that even if defendant’s evidence was accepted, the winds would only be at the lowest levels of force 10. Moreover, the seas found by both Raguso (15 feet) and Kaciak (15-19 feet) are within the range of normal expectation for force 9 winds.
The question therefore becomes whether force 9 winds constitute a peril of the sea. I do not think that they do in the circumstances of this case. While it is probably impossible to define precisely the term “peril of the sea,” in J. Gerber & Co. v. S.S. SABINE HOWALDT, 437 F.2d 580, 594-97 (2d Cir. 1971), Judge Anderson, after a comprehensive review of the cases that have attempted to define and apply the phrase, concluded that “[tjhere are, however, few eases in which the winds are force 9 or below (i. e., 54 land miles per hour or 47 knots) in which there has been found to have been a peril of the sea.” 437 F. 2d at 596.
The winds encountered by the ANTONIO DEMADES were at most on the order of 47-48 knots. Under Judge Anderson’s analysis these winds would not be such as to constitute a peril of the sea in the usual case.
This result seems especially appropriate since none of the special circumstances discussed in the SABINE HOWALDT are present in this case. Thus, there is no reason to make it one of the rare exceptions to the general rule. The evidence indicated that the ANTONIO DEMADES did not encounter cross seas. (A cross sea occurs when the swell and sea are coming in different directions by 45° or more). At most times the seas and swell encountered by the ANTONIO DEMADES came from the same direction and during the short period of time in which there was some divergence in direction, it was only about
22yz°.
There was no indication that the ship was being buffetted about unduly in the storm for a long period of time. Indeed, the hatch failure occured soon after the storm commenced. It appears that the ship was able to steer a steady course into the storm. Moreover, it does not appear that the wave that broke the hatch cover was abnormally large.
There were many other ships in this storm, some in areas where the storm was worse, yet none of them sank.
Thus, the court concludes that the factors and cases discussed in the SABINE HOWALDT opinion support rejection of defendant’s peril-of-the-sea defense. Defendant did not satisfy its burden of
establishing such a defense by a preponderance of the evidence, especially since its principal witness made erroneous assumptions concerning the location of the ship and incorrectly plotted the ship when he performed his analysis.
B. ACT
OR
NEGLIGENCE OF THE MASTER
Defendant next claims that it should be absolved of any liability because, it asserts, the sinking of the ship was due to the negligence of the master in deciding to turn back into the teeth of the storm at 1700 hours on February 6 instead of continuing downwind until the storm subsided.
While it is impossible to know what would have happened had the master continued downwind, all of the evidence at trial indicated that the ship would have survived longer. When the hatch broke, it was impossible for the crew to inspect the damage because the ship was heading into the storm and the front of the ship was literally awash. However, when the master changed course and turned downwind into the storm, he was not only able to lead a party forward to inspect the damage, but they remained forward for an hour or two. Thus, the direction the ship was going clearly had a considerable impact on the extent to which the front of the ship was exposed to the sea and to direct wave action.
It is clear that the ship would not have sunk if only No. 1 hold filled with water. The crucial question thus becomes — what effect on progressive flooding would going into the wind as opposed to sailing downwind have had. As indicated later, the court finds the opinion of defendant’s naval architect Ganley as the most credible explanation of how the ship sunk. In his view, two of the more important factors in the progressive flooding were the direct wave action on the front of the ship which probably broke off and unplugged ventilator pipes and the flooding of No. 2 cargo hold through leaks created by the pressures placed on the No. 1-No. 2 bulkhead by the water in No. 1 hold. Had the ship been going downwind the direct wave action would have been much less (as evidenced by the fact that the master’s party was able to go forward to inspect the hatch only when the ship was heading downwind). Any progressive flooding that occurred because water entered the holds through openings in the deck (e. g., unplugged vents) would have been much less — both because it would be less likely that such vents would be broken off or become unplugged and because it would be less likely that water would enter those that had been unplugged or broken since some of them were located fairly high above deck level. ■
In addition, the action of the waves in rocking the ship would tend to spill some of the water out of No. 1 hold. Since the three-foot high coaming would offer more protection to that hold when the waves were striking the ship from the rear then when they were coming directly over the bow, less water would probably have remained in No. 1 hold. Also, to the extent that No. 1 hold was contributing to the flooding of other areas of the ship, there might be less flooding since No. 1 hold would have less water in it thereby reducing the pressure and the source for the flooding. Perhaps the difference in the rate of flooding would not have been great with respect to that caused by No. 1 hold, but as to any water entering through the ventilator pipes, etc. it seems that it would have made a considerable difference.
In any event, both experts, Ganley and Gilbert, agreed that the progressive flooding would have occurred more slowly if the ship had headed downwind. This difference probably would have been crucial. The log of the rescue ship the M.S. CHILE MARU indicated that after the ANTONIO DEMADES was abandoned the weather improved (e. g., the winds died down to force 7). Plaintiff’s meteorologist, Raguso, indicated that by 1100 hours on February 8, the storm center had moved far to the northeast and that it was weakening rapidly. Kaciak testified that the storm
lasted a little more than 24 hours after the hatch cover failed.
The ANTONIO DEMADES was not abandoned until 18 hours after the hatch cover failed. It appears that for most of that period of time (i. e., until the S. O.S. signal was sent), the master thought that the flooding situation was under control. If the master had remained on his downwind course, the flooding situation probably would have been brought under control. Had the ship continued downwind, both experts agree that it would have survived longer (since there would have been less progressive flooding). Thus, the longer she remained afloat the less additional water she would have taken and the more likely that her pumping efforts would have been successful and that repairs could have been attempted. The court concludes that Ganley was correct in saying that there would have been a good chance that the ship would not have sunk had the master not decided to sail back into the force of the storm.
The court concludes the casualty probably would not have occurred but for the error of the master in turning back into the storm. Thus, defendant met its burden of establishing that it fell within an excepted clause of Cogsa (i. e., 46 U.S.C. § 1304(2)(a)). Consequently, it becomes plaintiff’s burden to establish that the ship was unseaworthy.
III.
Plaintiff urges that the ship was unseaworthy for four reasons. First, it suggests that the ship had insufficient bunkers (fuel) to make the voyage from Panama to Japan; second, it asserts that the McGregor hatch cover on No. 1 hold had been improperly altered, thereby contributing to its failure; third, it argues that the occurrence of progressive flooding in No. 2 and No. 3 holds indicates that the ship was structurally unsound and that it had not been properly repaired after a prior casualty; and fourth, it claims that the ANTONIO DEMADES was overloaded. The court concludes that the plaintiff has failed to establish any one of these claims of unseaworthiness.
A. Insufficient Bunkers.
Plaintiff suggests that the ANTONIO DEMADES was unseaworthy because it did not have sufficient fuel for a trip to Japan. Although it never made the argument explicit, it appears that plaintiff hoped by this contention to negate the negligence-of-the-master defense by arguing that the master had no choice but to continue on to Japan because he was short of fuel. Examination of the evidence does not support such a conclusion. According to the ship’s agent’s report, the ANTONIO DEMADES left Cristobal on the Atlantic side of the Panama Canal with approximately 1006 tons of intermediate fuel oil (IFO) (which is used in the main engines) and 124 tons of disesel fuel (which is used in the generators) on board. During her short layover in the Canal Zone and her transit of the Canal, the ship probably used, at most, 14 tons of IFO and 2 tons of diesel. Thus, on leaving the Pacific side of the Canal the ship had approximately 992 tons of IFO and 122 tons of diesel available for its voyage. Some of this fuel may have been unusable since it could not be gotten out of the fuel tanks, but according to defendant’s expert master mariner, Capt. Fertig, this amount could be kept at a minimum by applying heat to the tanks which would prevent the fuel from forming a coating on the tank walls.
The distance from Panama to Japan by the great circle (shortest) route is 7,680 miles. The shortest distance to Japan from Panama which would avoid the winter zone (where the ship would be more severely restricted in the amount of cargo she could legally carry) was 7,850 miles. However, it appears that the actual route taken by the ship was not the shortest route, but' was rather one which covered 8,075 miles.
The ANTONIO DEMADES had a maximum speed of 15 knots and maximum fuel consumption of 35 tons of IFO and
iy2
tons of diesel per day. It appears from the three weekly reports sent by the ship to its owners during the crossing of the Pacific that the average speed of the ship was 13.63 knots and that the average daily consumption of fuel was 34.66 tons.
Since the Master was undoubtedly aware of the route the vessel would take and of the fact that the vessel had not performed and would not perform at maximum capacity, the relevant figures for the purpose of determining whether the bunkers were sufficient at the time the ship left Balboa, are a distance of 8075 miles, a speed of 13.63 knots, and fuel consumption rate of 34.66 tons per day.
At an average speed of 13.63 knots the ANTONIO DEMADES would take 24.68 days to go the 8,075 miles from Panama to Japan and would consume 855.4 tons of IFO and 37.0 tons of diesel fuel. If this amount is increased by 20% to allow for the normal reserve that should be carried, the ANTONIO DEMADES should have left Panama with 1026.5 tons of IFO and 44.4 tons of diesel fuel. Thus, the ship’s store of IFO was 34.5 tons short of what the Master should have taken on board at Panama. However, the ship had 77.6 more tons of diesel fuel than was required for the voyage. Since the diesel fuel could be used in the main engines with little difficulty (it could be mixed in with the IFO in gradually increasing proportions), the court concludes that the ANTONIO DEMADES left Panama with an excess of 43.1 tons of fuel over the 20% reserve that plaintiff’s expert mariner, Capt. Patterson testified was required by customary industry practice. The ship was not unseaworthy because of insufficient bunkers; it had a reserve fuel supply of about 25% when it sailed from Balboa.
In any event, the sufficiency of the fuel supply was not a factor in causing the loss of the ship. At the time of the casualty the ship had virtually all its reserve left and it was only 3 days from Japan.
B. McGregor Hatch Cover on No. 1 Hold.
The ANTONIO DEMADES had been stranded for eight days on Quita Suena Bank in the Caribbean in October-November 1968. In December, she entered drydock in Osaka, Japan, for inspection and repairs. The surveyor at the shipyard indicated that the hatch covers and the top edge of hatch coamings of the holds were deformed. The repair bills show that one piece of the No. 1 hold hatch cover had been removed to the shop, faired, and refitted, and that the coamings had been repaired. Although the existence of such deformation of deck fittings indicated that the ship had been subjected to great stresses when it was grounded, there was no evidence to suggest that the repairs were inadequate. Significantly, if the repairs were done properly, plaintiff’s expert naval architect, John Gilbert, testified that the hatch covers and coamings would have been restored to their original strength. The independent surveyor in Japan indicated in his reports and in his deposition that the repairs were done properly.
Gilbert testified that the repairs would have been much more costly in the United States. While this explains one cause of the demise of the American shipbuilding industry, it is not evidence of inadequate repairs in the aggregate, let alone inadequate repairs to the hatch covers and coamings.
There was testimony to the effect that overall deformation in the ship’s hull probably occurred as a result of the grounding and that a transit would have been necessary and should have been used to determine if there was such hull deformation. It does not appear that a transit was used. However, this seems insignificant with respect to the seaworthiness of the hatch covers. While deformation of the hull might have caused the hatch covers to buckle at the time of the grounding, there was no evidence that the mere existence of deformation in the hull would cause any weakness in hatch covers that had been refitted subsequent to the deformation.
Thus, the court concludes that plaintiff failed to carry its burden in attempting to prove that the hatch cover on No. 1 hold was inadequately repaired so that the vessel was unseaworthy.
Plaintiff also suggests that the vessel was unseaworthy because the hatch cover on No. 1 hold was altered in violation of American Bureau of Shipping rules that prohibit unauthorized structural modifications in a vessel. The plans of the ANTONIO DEMADES indicate that there were four sections in the hatch cover of No. 1 hold and five sections in the covers for the other six holds. The shipyard’s repair bill indicates there were only four sections in the No. 1 hold hatch cover, and there is no evidence that the cover was ever modified. Two of the surviving crewmen made statements that sections 4 and 5 of the hatch cover failed. Plaintiff urges the court to conclude from this that an unauthorized structural change was made in the vessel. The court finds, however, that the two men were confused and that it is likely that the sections of the hatch cover that failed — two end sections, which would have been numbers 4 and 5 on the other six hatch covers — were actually the third and fourth sections of the four-section hatch cover.
The court concludes that although the failure of the No. 1 hold hatch cover was unexplained, plaintiff has not established that it was due to improper repairs of prior damage or unauthorized structural modification.
C. Progressive Flooding.
The ANTONIO DEMADES would not have sunk if only the No. 1 hold had filled with water. It was because of the progressive flooding of Nos. 2 and 3
holds, Nos. 2 and 3 double bottoms and other areas that the ship eventually went down. . Plaintiff claims that the fact of progressive flooding shows that the vessel was unseaworthy because it was structurally unsound. First, plaintiff suggests that the failure to test the ship with a transit when it was repaired in 1968 meant that locked-in stresses caused by hull deformation might have been present and that these stresses might have so weakened the bulkheads between the holds that the ship should be considered unseaworthy. Second, plaintiff also contends that there might have been small hidden leaks in the bulkheads or double bottoms which could have been found by a hose test but which would not have been found by a visual inspection.
According to plaintiff’s naval architect, Gilbert, many of the bulkheads and the double bottoms in the forward area of the ship were repaired. Omachi, the surveyor in Japan attending the repairs, testified in his deposition that hose tests were performed on the double bottoms and on the bulkheads between No. 1 and No. 2 holds as the repairs progressed.
In any event, Gilbert only speculated that damage to the bulkheads and the double bottoms might be a possible source of progressive flooding. He admitted that proper repairs would restore the watertight integrity of these sections and there is no evidence that the repairs were inadequate. The surveyor, on the other hand, testified that they were adequate.
Even if it is assumed that no hose tests were conducted, the leaks that might be found in the bulkheads would be mere pinpricks since the bulkheads were visually inspected subsequent to the repairs by surveyors hired by cargo owners (who, of course, would be looking very carefully for any preexisting damage to the ship so as to prevent their employer from being blamed for it). I do not think that the existence of minute bulkhead leaks would establish that the vessel was hnseaworthy. There was no evidence that the existence of pinprick-size holes would measurably reduce the strength of the bulkheads, and, in any event, it would seem that the pumps could easily handle the small amounts of water which would accumulate from such leakage.
Similarly the fact that a transit was not used does not establish that there was unrepaired hull deformation. However, even if there were locked-in stresses at the time the ship left the shipyard, there was testimony by defendant’s expert naval architect, Ganly, that these stresses would have been immediately noticeable or would have worked themselves out over time as the ship sailed the seas. At trial Ganly evidenced considerable familiarity with the concept of locked-in stresses and stated that the literature in the field supported his view. Gilbert, on the other hand, only testified that locked-in stresses could have caused the progressive flooding. He did not provide any evidence to show that locked-in stresses were ever present in the ship, let alone present at the time of sinking. Thus, it is concluded that there is no evidence that there were any locked-in stresses present in the ANTONIO DEMADES which caused, or partially caused, the casualty.
In all likelihood defendant’s expert Ganly accurately described the way in which the ship sank. With No. 1 hold filled with water, the deck structures on the front part of the ship would be subjected to forces to which they were not normally exposed and which they were not designed to withstand. It is likely that the exposed parts of the ventilator pipes to the below deck areas would be swept away. Although some of the vents were plugged it is quite possible that prolonged direct wave action could dislodge the plugs. Thus, other areas in the forward area of the ship would start to fill (even if the bulkheads were watertight). In addition, as the ship pitched in the storm the sloshing of tons of water (and perhaps some scrap steel) back and forth in the No. 1 hold would subject the bulkhead to forces that it was not designed to withstand and might well result in some water getting into the adjacent hold.
The burden is on the plaintiff to show that unrepaired structural defects caused the vessel to be unseaworthy and in the court’s opinion, it has failed to establish this. It is more likely that the ship sank as described by Ganly, whose description of the vessel’s sinking did not depend on unproven structural defects, but which assumed that the ship was sound.
D. Overloading.
Finally, plaintiffs claim that the ship was unseaworthy because it was overloaded in violation of the International Loadline Convention and that the overloading was a contributory cause of the casualty.
See The Smith Voyager, 439 F.2d 109, 113 (2d Cir. 1971). At the outset the court notes that there was no incentive on the part of the shipowner to overload the ship as the charter provided for a lump-sum payment and not a sum dependent on the amount of cargo carried.
Determination of this claim is complicated by substantial inconsistencies in the measured drafts of the vessel that were taken at various times in her voyage Four measured drafts are before the court. According to the draft statements prepared by the Master and surveyor at Boston the ship had a mean corrected draft of 35 feet 6% inches.
The draft statement of the surveyor appears to have been prepared with considerable care. It is far the most complete evidence of the ship’s draft and the court finds that it can be taken as an accurate record of the ship’s draft when it left Boston.
The second draft figure is provided in the Ship’s Information and Quarantine Declaration filed with the Panama Canal authorities on arrival in Cristobal and
prior to bunkering. That statement indicates an uncorrected draft of 34 feet six inches forward and 35 feet three inches aft. Plaintiff’s expert Gilbert used these figures and calculated that the corrected draft was 35 feet two and six-tenths inches. Gilbert also started with the Boston draft and adjusted it in light of the ship’s expected consumption of fuel and water between Boston and Panama, in order to calculate on arrival draft based on the Boston draft. He found that his calculated arrival draft was basically equivalent to the measured arrival draft. Based on the amount of fuel and water loaded at Panama, Gilbert calculated that the ANTONIO DE-MADES was not overloaded when she left Panama, that she was five inches below her marks on entering the summer zone and that she was 3.265 inches over her marks at the time of the casualty.
The third recorded draft is contained in the Ship’s Condition Report prepared by the Panama Canal pilot. This report, if correct, would have indicated that the ship was one and a half feet over her tropical marks when leaving Panama and more than a foot over her marks at the time of the casualty. Gilbert indicated, however, that he did not consider this report because it was so far out of line with the other two. The court agrees with him and finds that it does not accurately reflect the ship’s draft.
Finally, there is a letter from the Port Captain at Cristobal (received after trial but offered sight unseen by defendant and received into evidence) which states that a recorded (uncorrected) draft taken when the ANTONIO DEMADES was at Cristobal shows a draft of 35 feet six inches forward and 37 feet aft. Plaintiff argues that if the same sag correction is used as was used at Boston then the corrected mean draft was 36 feet 6% inches or 5% inches over Gilbert’s calculated draft. Plaintiff argues that this 5% inches should be added to Gilbert’s calculations so that the ship was 2 inches below her marks both when she left Panama and at the time of the casualty. The court disagrees.
It is necessary to choose between the careful draft calculation made at Boston which (according to Gilbert) was consistent with the arrival draft at Cristobal and the Port Captain’s letter which does not even attach a copy of the draft measurement to which it refers. Gilbert’s calculations of the draft are more believable. He started with the arrival
draft which was consistent with the Boston draft and he considered what weight the ship took on at Cristobal. If one were to accept the plaintiff’s interpretation of draft in the Port Captain’s letter, it would be necessary to account for an additional 500 tons of cargo or fuel (the additional weight represented by a difference of 5% inches in draft) and there is no explanation offered. Moreover, there are several reasons why the plaintiff’s interpretation of the Port Captain’s draft may be inaccurate. First, it assumes that the sag correction used in Boston should be used in Cristobal even though the fuel taken on at Cristobal was loaded mainly in the fore and aft sections of the ship thus tending to reduce sag. Second, the draft was taken in Cristobal Harbor and that harbor has a lower salinity than regular sea water because of the fresh water released when the Panama Canal locks are emptied into it. Fresh water is less buoyant than salt water and a correction must be made in order to adjust draft measurements made in fresh water to those permitted by the load line certificate issued under the International Load Line Convention.
In sum, the Boston departure and Cristobal arrival drafts seem to be accurate. The draft referred to in the Port Captain’s letter is inconsistent with those drafts and with the amount of fuel and water actually taken on board at Cristobal. Thus, the court credits the Boston draft and the arrival draft as accurate and considers the other two drafts to be inaccurate. The court concludes, as did Gilbert and Patterson (both plaintiff’s experts), that the ship was not overloaded when she left Panama and was not overloaded when she sank. The ANTONIO DEMADES was not unseaworthy because of overloading.
Since the defendant has established that it falls within an exception under Cogsa, 46 U.S.C. § 1304(2) (a), since the plaintiff has failed to establish that the ship was ■ unseaworthy, and since in any event the defendant exercised due diligence to provide a seaworthy ship, the complaint is dismissed.
This opinion constitutes the court’s findings of fact and conclusions of law. Rule 52(a), F.R.Civ.P.
So ordered.