Nos. 84, 85, Dockets 74-1629, 74-1632.United States Court of Appeals, Second Circuit.Argued October 9, 1975.
Decided November 26, 1975.
Alexander S. Kritzalis, New York City (Burlingham, Underwood
Lord, Hervey C. Allen, and John S. Rogers, New York City, of counsel), for plaintiff-appellant.
Donald M. Kennedy and Thomas D. Toy, New York City (Donovan, Donovan, Maloof Walsh, Francis J. McCaffrey, and Hill, Rivkins, Carey, Loesberg O’Brien, New York City, of counsel), for defendants-appellees.
Appeal from the United States District Court for the Southern District of New York.
Before KAUFMAN, Chief Judge, and FRIENDLY and SMITH, Circuit Judges.
 Hellenic Lines Limited (Hellenic), owner of the vessel Hellenic Sailor, appeals from the dismissal by Judge Griesa, in the District Court for the Southern District of New York, of two consolidated complaints seeking general average contributions of approximately $135,000 from marine insurers of cargo; the sum represents a share of the extra expenses incurred in preserving the ship and cargo and completing the voyage after a fracture of the vessel’s crankshaft which occurred in the Red Sea on December 24, 1967, some 36 days after she left New York, her last loading port, and as she was approaching Jeddah, her first port of discharge.
 There is no dispute over the governing rule of law. The bills of lading contained New Jason clauses, the effect of which, in light of the Carriage of Goods by Sea Act [COGSA], 46 U.S.C. § 1300 et seq., is that “the ship is entitled to recover general average expenses in just exactly the same cases as those in which it would have been immunized against cargo claims, had cargo been damaged.” Gilmore Black, The Law of Admiralty, at 267 (2d ed. 1975). COGSA requires that “The carrier shall be bound, before and at the beginning of the voyage, to exercise
due diligence to — (a) Make the ship seaworthy,” 46 U.S.C. § 1303(1), and provides that “Whenever loss or damage has resulted from unseaworthiness, the burden of proving the exercise of due diligence shall be on the carrier . . .,” 46 U.S.C. § 1304(1). That the damage in this case resulted from unseaworthiness is not contested.
 Recognizing that decision turned on the resolution of extremely technical issues of fact, Judge Griesa, who obviously had paid exceedingly close attention as the evidence came in, dictated an admirable opinion from the bench, to which we can add little. He found that the fracture of the No. 2 after side rod crankshaft web, just in front of the No. 3 bearing, began as a fatigue crack followed by a sudden rupture of the rest of the web; that the crack and rupture were the result of a bending stress on the web; and that the probable cause of the bending stress was the flexing of the nearby large crankshaft web resulting from the crankshaft’s being out of alignment. We do not understand the correctness of these findings to be seriously disputed; in any event they surely are not clearly erroneous. Hellenic’s attack is directed rather at the judge’s conclusion that it had failed to demonstrate the exercise of due diligence with respect to the non-alignment of the crankshaft.
 The court placed primary, although by no means sole, reliance on two sets of circumstances. The first was this: On October 11, 1967, when the Hellenic Sailor was making an earlier stop in New York, a wire gauge reading of crankshaft alignment was taken by Golten Marine, Inc., a concededly competent marine repair firm. This showed sags of 32/1000ths of an inch at No. 4 main bearing, 42/1000ths of an inch at No. 3, and 32/1000ths of an inch at No. 2. At the time, the vessel was lightly loaded, with a draft of 12’6″ forward and 22’10” aft. When Charles Allan, Hellenic’s marine superintendent, returned from a vacation on which he had gone shortly after October 11, he wrote a letter, dated November 6, to Marcos Evangelou, the Hellenic Sailor’s chief engineer. This enclosed a copy of the readings, asked Evangelou to note the position of the crankshaft “with Nos. 2, 3 and 4 below the `0′ line,” and added that this had occurred while the ship was in a comparatively light condition. Allan concluded: “We will take further readings at the next available opportunity with the ship in both loaded and light conditions.” In a deposition Allan testified that his concern was that the readings showed a deflection somewhat greater than it should have been and that the difference between the draft fore and aft might have made the readings unreliable.
 Evangelou testified that on November 13 he took shaft alignment readings at Philadelphia when the vessel was much more fully, and more evenly, loaded, with a draft of 24’6″ forward and 28’3″ aft. After Evangelou had unsuccessfully sought to identify a document as showing these readings, Hellenic advised the court that it did not contend that the document reflected readings taken at Philadelphia. The Philadelphia readings were not produced, and no explanation for their non-production was made.
 The second set of circumstances was this: There was evidence that what is known as a web deflection test is also useful in detecting inordinate stress on the webs. Allan testified in his deposition that web deflection tests are “essential”; that the chief engineers on the
Hellenic Line’s vessels took such tests almost every voyage; that the results would be recorded and kept on the vessels; and that he would not be informed unless there was something abnormal. Beyond that the manufacturer of the vessel had sent a letter to the American Bureau of Shipping on September 14, 1960, with a copy to Hellenic which was produced from its files, enclosing a copy of “recommended deflections of crank webs,” and Doxford
Sons, Ltd., the original designer of the type of engine used on the vessel, had engaged in extensive research and published web deflection standards for their crankshafts. The record contained web deflection readings made on the Hellenic Sailor on October 31, 1960, November 23, 1960, and sometime during September, 1962, but none thereafter.
 Hellenic does not argue that it had ordered web deflection tests to be taken during the voyage and the Chief Engineer failed to make them. Rather it contends, based on some of the evidence but contrary to the opinion of its marine superintendent, that the wire gauge test and the web deflection test are duplicative, and that therefore there was no need to take a series of web deflection tests. Judge Griesa found that “the plaintiff has failed to sustain its burden of proof in connection with its assertion that it did not need to take web deflection readings,” and that, accordingly, it had not “satisfactorily explained the absence of any specific information regarding such readings . . . following September 1962.” We are not persuaded that we should disturb these determinations. However, even if we were to adopt the plaintiff’s contention, and were to say that the absence of web deflection readings could be justified by the presence of wire gauge test results, this would not carry the day for the plaintiff in light of the conclusions with respect to the Philadelphia readings.
 As to these the judge was justified in concluding that one of two things occurred — either they were never made or they were made and had become unavailable. If no tests were made, the Chief Engineer had failed to do what Hellenic’s own marine superintendent had directed, and the vessel departed New York, her last loading port, without these having been made. If they were made, as Evangelou testified, the judge was justified in inferring from their unexplained non-production that they were unfavorable. As Mr. Justice Story elegantly observed, such non-production is “a very awakening circumstance, calculated to excite the vigilance, and justify the suspicions of the court,”The Pizarro, 2 Wheat. (15 U.S.) 227, 241, 4 L.Ed. 226 (1817). We recently summarized the developed law on this subject i Tupman Thurlow Co., Inc. v. S. S. Cap Castillo, 490 F.2d 302, 308 (2 Cir. 1974), and need not repeat that analysis. See also 2 Wigmore, Evidence § 291(3) (3d ed. 1940). Although in the Cap Castillo case we “hesitate[d] to conclude that the non-production can stand alone as the determinative factor . . .,” id., here we have, in addition to the failure to produce web deflection readings, Allan’s own concern about the New York wire gauge measurements and other evidence relating to the unfavorable sag trend which was shown in successive wire gauge measurements of the crankshaft.
 Hellenic seeks to answer all this on the basis of expert testimony that a wire gauge reading at Philadelphia, or a web deflection test during the voyage, would have disclosed nothing alarming since telescopic readings taken at Port Sudan in early February, 1968, after the crankshaft fracture and an effort at temporary repairs, showed sags considerably less than the New York October 11, 1967 readings. The judge rejected this on the basis of Allan’s deposition testimony that the post-fracture readings would not be indicative of pre-fracture conditions. While Hellenic mounts a valiant effort on this point, we are unconvinced that the judge’s conclusion was clearly erroneous or indeed was erroneous at all.