McLAUGHLIN, Circuit Judge.

Nathan Prashker was a young man, twenty-seven years old, unmarried, president of a family corporation, and the holder of a private pilot's license with some 215 hours flying time to his credit. On October 5, 1953 he undertook to fly from Allegheny County Airport near Pittsburgh to Cleveland, about 130 miles distant, with his cousin, a young man who was the father of several children. The visual flight plan filed with the Allegheny tower by Prashker called for a flight of approximately 42 minutes at 3,000 feet in a Beechcraft Bonanza, N 2168 D, recently acquired by the family corporation and in which Prashker had logged some forty hours of flying time.

At the take-off point the ceiling was estimated as 1800 feet, visibility four miles with light rain and haze. At the Greater Pittsburgh Airport conditions were better with broken clouds at 3,000 feet, overcast at 9,000 feet, visibility 12 miles with light rain. At Cleveland the ceiling was 15,000 feet with broken clouds below that, visibility 12 miles. There is no indication other than that he was in the habit of doing so that the weather conditions actually were checked by Nathan Prashker.

N 2168 D with Nathan Prashker at the controls departed Allegheny County Airport at 12:09 P.M. headed in a northwest by west direction with instructions to use caution in passing through the Greater Pittsburgh area which had a lot of instrument traffic. At 12:22 the pilot called Allegheny tower by radio, advising that he was over the City of Pittsburgh at 2,300 feet, was unable to maintain visual references, and that he was into and climbing up through the overcast. Immediately after the tower had advised that their last report for the tops of the overcast had been 3,800 feet, Pittsburgh tower, who apparently had heard Prashker report, called Allegheny tower to advise that a jet reported the cloud tops at 33,000 feet. This information was relayed to N 2168 D and the pilot was then asked for his approximate position. Prashker replied tensely that he didn't know. He was asked if he wanted to return to the airport and he nervously said that he did but that he didn't know his position and that he was incapable of flying on instruments.

Immediately Allegheny tower asked Pittsburgh tower to take a direction-finding reading on N 2168 D's transmissions and the pilot was told to give a short count. The tower heard him give the count and his identification. Pittsburgh tower then told them to have the pilot steer 325 degrees magnetic; this was immediately relayed to the pilot with the information that it would take him to Greater Pittsburgh Airport and that he should break out "in the vicinity of the river". The acknowledgement came as a broken and poor transmission. Allegheny tower repeated the steer and advised that the plane switch radio frequencies to work the Greater Pittsburgh Airport, but there was no acknowledgement. Repeated calls in the next minute were not answered.

Then at 12:27 an outside telephone call was received from a woman reporting that she and a neighbor had just seen an aircraft go behind a hill and that they thought it had crashed. Eight minutes later a man called to report that he and his co-workers had seen a wing fall off a plane and that the plane had crashed in Pangburn Hollow, near Elizabeth, Pennsylvania.

The crash scene was in a hilly, heavily wooded area near but out of sight of the Monongahela River, about 8 miles south-southeast of the Allegheny County Airport. The left wing was detached from the main body of wreckage and lay about 250 yards back along the flight path. The left stabilizer and part of the rudder were missing.

These actions were brought in Delaware against Beech Aircraft Corporation as the manufacturer of the Bonanza and against Atlantic Aviation Corporation as the seller of the aircraft, both of them Delaware corporations, by the executrix of the estate of Nathan Prashker and by Choice Embroidery and Lace Company, the Prashker family corporation, incorporated in New York, which owned the aircraft. Jurisdiction is based on diversity of citizenship.

The claims against Atlantic are for breach of the implied warranty of merchantability and for negligent failure to warn of known dangerous propensities of the aircraft. Four allegations of negligence are made against Beech: (1) that the airplane was improperly designed; (2) that after receiving notice of the improper design it failed to make changes; (3) that there was a failure to include necessary safety devices; and (4) that there was a failure to warn of the prospective dangers. Additionally there is an allegation sounding in deceit that it was a conscious policy of Beech to disseminate false information as to the strength and safety of the Bonanza. There was further a claim against Beech for breach of warranty.

At the conclusion of a three week presentation of the plaintiffs' evidence, the trial court directed a verdict in favor of the defendants on the grounds that Nathan Prashker had incontrovertibly been contributorily negligent, thereby barring recovery on any claims of negligence, that there was no privity between Beech and the plaintiffs to sustain recovery on breach of warranty under the applicable law of Kansas, and that though there was an implied warranty of merchantability between Atlantic and the plaintiffs there could be no recovery thereon when the damage to the plaintiffs had been proximately caused by their own representative's contributory negligence.

It is a truism that on a motion by the defendant for a directed verdict at the close of the plaintiff's case, the evidence and all the inferences reasonably to be drawn therefrom must be viewed in the light most favorable to the plaintiff.

We consider first the matter of Nathan Prashker's contributory negligence. Because the harm occurred in Pennsylvania, that is the place of the alleged tort. The United States District Court of Delaware in a diversity action posing a conflict of laws question looks to the conflicts rule of Delaware to determine the law to be applied. Klaxon Co. v. Stentor Electric Mfg. Co., 1941, 313 U.S. 487, 61 S.Ct. 1020, 85 L.Ed. 1477. Delaware follows the usual course of looking to the law of the place of the alleged tort to determine the substantive rights of the parties arising therefrom. Pack v. Beech Aircraft Co., Del. 1957, 132 A.2d 54; Restatement, Conflict of Laws, § 384. Thus, the law of Pennsylvania was applied by the District Court and will be looked to by us.

The Civil Air Regulations, which were put into evidence in this case, prohibit flying under instrument conditions by a pilot without an instrument rating and without recent practice. A pilot without instrument qualifications must adhere to the visual flight rules. These prohibit flying when visibility is less than three miles or when the ceiling is less than 1,000 feet. Additionally they require that a visibility of three miles be maintained at all times while in flight and that the plane approach clouds no closer than 2,000 feet horizontally, 500 vertically under, and 1,000 feet vertically over. The Regulations do contain a provision, however, which permits departure from the other Regulations in event of emergency. We mention the regulations of the Civil Aeronautics Board in order to establish impartial and authoritative criteria for judging the question presented to us.

The evidence shows that the necessity of prohibitions on flight under instrument conditions by unqualified pilots is dictated by the extreme likelihood of loss of control over the aircraft once visual points of reference become unavailable. This likelihood arises out of a vertigo induced by the movements of the aircraft in which the pilot becomes disoriented in space to the point where he has difficulty in distinguishing between right and left and up and down. Normal sense perceptions are unreliable to the point of being actively misleading. The plane can quickly go out of control in such circumstances. Even if the pilot can recognize what maneuver the out-of-control plane is performing his tendency is to over-correct. The consequence is for the plane again to go out of control, or if the correction is too violent resulting in an attempted maneuver for which the plane is not designed, to tear the plane apart in mid-air by reason of the extreme forces exerted on it.

Nathan Prashker knew, or certainly should have known, about and understood these and other factors involved in flight safety. Much of the literature put into his hands by the defendants at the time the Bonanza was purchased and afterwards emphasized the necessity for maintaining positive control over the aircraft at all times; it emphasized, too, the extreme danger of losing that control when instrument flight conditions were encountered by a pilot insufficiently skilled in coping with the difficulties peculiar to instrument flying. Nathan Prashker knew, too, that his craft was a faster, aerodynamically "cleaner" airplane than had been the three he had owned previously at various times, and that such an airplane could rapidly get into serious trouble if not skillfully operated. A fast, "clean" airplane is more difficult to fly, both on visual flight rules and on instruments because it responds to control quickly and easily. The control exercised must, therefore, be skilled. In fact, Nathan Prashker doubtless had the skill requisite to handle the Bonanza so long as he remained within the limitations imposed by the visual flight rules and stayed out of instrument conditions.

The facts, however, demonstrate that Nathan Prashker, wittingly or not, made a number of decisions involving bad judgment which set the stage for the ensuing tragedy. To begin with he elected to take off from the Allegheny County Airport in marginal weather conditions, judged by the visual flight rules. He thus assumed the risk of getting into poorer flight conditions, even though he undoubtedly hoped and may even have assumed that they would improve. It is inferable from the time-intervals involved that he shortly did encounter worsening conditions and that he turned back toward Pittsburgh. Over the city he made the grievous mistake of electing to go into the overcast in an attempt to climb through it.

There was elicited on cross-examination from a part-time employee of Atlantic's, who was also an airlines pilot of some 10,000 hours experience, testimony that he had never encountered a weather condition which he could not have avoided either by going around, by turning back, or presumably by staying on the ground in the first place. It is apparent, then, that what are alleged to be defects in the Bonanza only become significant because Prashker had gotten himself into a situation in which causes beyond his ability to control inexorably commenced to operate.

A decision based on Prashker's contributory negligence alone would depend on a preliminary holding that the evidence would not support an inference that bad weather may have closed in on him even though he was exercising due care for his own safety. We incline to the view that no such inference is available, but concede that the question may be arguable. Therefore, and because the case has been so thoroughly and well presented to us, we prefer to rely also on an alternative ground for decision. We think that no negligence can be imputed to the defendants.

Plaintiffs asserted that the Bonanza had been negligently designed; they have attempted to spell out three specific defects. The first of these alleged defects is that the Bonanza model involved is subject to a lateral oscillation of some two or three degrees on either side of a norm in rough-air flight, and that this oscillation induces vertigo whenever visual references become unavailable. But there was undisputed evidence that any aircraft under similar flight conditions is subject to some lateral oscillation. The conclusion is that, although the existence of oscillations may explain how vertigo is induced, vertigo cannot be avoided except by avoiding the loss of a visual reference; in instrument flight the reference is supplied by the instruments which in their use require a degree of skill amounting to an art.

Another alleged defect was that the plane tended to assume a nose-down attitude and once in that position accelerated dangerously. But the testimony was that the nose of the Bonanza does not drop of its own accord. Any airplane certificated by the C.A.A. as was the Bonanza must be statically stable. The acceleration once the plane was in a nose-down attitude was not extreme for an aircraft of clean design; again this particular aircraft required a measure of skill in flying it, and Nathan Prashker apparently had that skill so long as he remained within the limitations of the visual flight rules as he was required to do. Plaintiffs allege that the Bonanza would have been a safer plane had it been equipped with dive brakes. Beech had in fact considered the use of dive-brakes but had abandoned the idea largely because much the same result could be achieved by extending the landing gear. It is doubtful in the extreme that Nathan Prashker would have thought to use the dive brakes even had they been installed because the landing gear was still retracted after the plane had crashed.

The third alleged defect of which plaintiffs complain is that stick-loads were dangerously light. They are important in preventing the pilot, when under stress, from violently moving the controls, thereby imposing such stress on the plane structure in its attempt to respond that it disintegrates. It appeared, however, that the Bonanza's stick-loads exceeded those imposed by the military for their aircraft designed to perform more violent maneuvers than is the Bonanza.

In short all these alleged defects were either nonexistent or present to an unavoidable extent known to Nathan Prashker. Heeding the fundamental warning against an unqualified pilot's getting into instrument conditions is the surest means to avoid unintentionally exceeding the inherent restrictions in the design of any aircraft. That Nathan Prashker could have violated that tenet without negligence on his part does not, of course, make the defendants liable absent negligence on their part. Even if it can be said that there is a duty to warn of the dangers in exceeding the inherent restrictions in any aircraft, that duty was adequately discharged by the warnings to avoid instrument flight conditions which would be the only anticipatable way the restrictions could be unintentionally exceeded by a pilot.

We turn now to a consideration of the claims for breach of warranty. Again the conflicts rule of Delaware is looked to in order to determine what law shall be applied to judging these claims arising out of contract. No reason is perceived why Delaware would refuse to apply the law that the parties have stipulated to be the governing law by reason of the contract of sale of the airplane having been entered there by Beech as seller and Atlantic as buyer. Cf. Wilmington Trust Co. v. Wilmington Trust Co., 1942, 26 Del. Ch. 397, 24 A.2d 309, 139 A.L.R. 1117. The law of Kansas will, therefore, be applied to determining the breach of whatever warranties accompanied the sale.

We think, on the basis of the foregoing discussion dealing with the physical limitations on airplane design, that the plaintiffs' evidence concerning breach of warranty was insufficient to go to the jury. The evidence fails to demonstrate any breach of either an express or implied warranty by Beech. We therefore pass without deciding the question of whether the law of Kansas would require privity between Beech and the plaintiffs.

As for the claimed breach of the implied warranty of merchantability between Choice and Atlantic, we apply, through the Delaware rule of conflicts, the law of New Jersey where the sale occurred, both for construing the contract, Harris v. New York Life Insurance Co., 1943, 27 Del. Ch. 170, 33 A.2d 154; Lams v. F.H. Smith Co., 1935, 6 W.W.Harr. 477, 36 Del. 477, 178 A. 651, 105 A.L.R. 646, and for ascertaining damages arising from any breach, Canadian Industrial Alcohol Co. v. Nelson, 1936, 8 W.W.Harr. 26, 38 Del. 26, 188 A. 39.

The measure of damages which Choice asserts is the value of the airplane, and not simply the difference in value between the aircraft at the time it was sold and the value of a similar plane free from the alleged defects making the plane unmerchantable. The measure of damages for a breach of warranty is, under New Jersey law, the loss directly and naturally resulting in the ordinary course of events from such breach. DeRose v. Hunter Lindsay Corp., App. Div. 1956, 41 N.J. Super. 178, 124 A.2d 349; Somerville Container Sales v. General Metal Corp., 39 N.J. Super. 348, 120 A.2d 866, amended App. Div. 1956, 39 N.J. Super. 562, 121 A.2d 746. The inescapable inference is that the damage complained of must stem from the breach of warranty and not from some other cause. Thus, it is clear that if the damage to the plane was proximately caused by the negligence of Choice's pilot, Choice cannot recover. Cf. Heath v. Channel Lumber Co., Inc., App. Div. 195 3, 25 N.J. Super. 6, 95 A.2d 425. But since we have not rested our decision as to the negligence issues solely on Nathan Prashker's contributory negligence, neither do we do so on this point; again, the plaintiff's evidence fails to prove that the plane was not reasonably fit for flying.

One point remains for discussion and decision. Plaintiffs complain of the trial court's ruling which prevented them from putting into evidence the reports of forty-five fatal accidents involving structural failure in flight of Bonanzas both on the issue of cause and on the issue of notice. The court permitted the introduction of only seven of the accident reports limited solely to the issue of notice of the alleged defects. The plaintiffs also offered evidence, which was rejected on both issues, of government statistics known to Beech that over a six year period prior to the accident in question the Bonanza had a rate of structural failure in flight approximately fifteen times the rate for all other single engine civil aircraft.

The first issue confronting the trial court upon the offer of the reports was whether evidence of accidents to one model of the Bonanza could be relevant as to the cause of accidents in other models. The court was justified in putting the plaintiffs to their proof that the five different Bonanza models involved in the series of accidents were similar. The court ultimately decided that the models were more like each other than like any other airplane.

At the same time the court was justified in excluding evidence as to accidents to the first model. The sum of the changes made in the design of the first model through four subsequent models permitted the trial court to regard the first and fifth planes as dissimilar aircraft. Evidence of accidents in the first model would not have been sufficiently material to the cause of accidents in the fifth model to require its admission. The two most significant changes from the preceding models occurred between the first and the second and between the third and fourth. The first of these changes involved increasing the limit design load factor from 3.8 G's to 4.4 G's which moved the plane from the C.A.A.'s normal category to the utility category for certification purposes. At the same time gross weight was increased. These changes improved the structural strength of the airplane. The second significant change was to the dihedral and area of the tail which affected the plane's handling characteristics.

The second issue confronting the trial court, once it had initially determined what models for purposes of the case were identical and which reports could therefore be relevant and material, was to decide whether the accident reports were indeed relevant on the issues for which they were offered. The court found that on the issue of notice of possible defects the reports were relevant, but that as to the cause of this accident they were irrelevant. Similarly the statistics were ruled irrelevant.

The determinations of irrelevancy were proper. Each accident report involved an instance of structural failure in flight where the airplane was, as plaintiffs assert, "pulled apart by loads in excess of its stress limitations." And yet the stress limitations exceeded those which the Civil Aeronautics Administration required as sufficient to assure flight safety. Showing that the planes came apart is not enough because of the compromise required with structural strength in designing an aircraft which will get off the ground. To be relevant on the issue of causation attributable to a design defect the accident reports would have had to show at least that the planes involved had not gone out of control prior to their coming apart, or if they had gone out of control, that it had been without fault by the pilot. The seven reports in evidence — and they may be assumed best to present the plaintiffs' case — invariably involve pilots unqualified to fly under instrument conditions who nonetheless were in such conditions or, in a few instances involving the unadmitted reports, pilots qualified for instrument flight who were in turbulent weather when the accidents occurred. To hold the aircraft responsible on such evidence would be utterly to disregard the factor of human fallibility known inevitably to occur in such circumstances and would be patently unjustified. The reports, therefore, do not tend to prove the existence of a defect in design of the Bonanza. The cases cited by plaintiffs on the admissibility of prior occurrences as evidencing existence of a dangerous condition are consequently inapposite.

The statistic as to rate of structural failure of the Bonanza is, on its face, startling. But it involves the same accidents which were the subjects of the reports above discussed. In view of what amounts to an explanation of each accident which contributed to the ultimate statistic, that statistic itself has no relevance as notice to Beech of the existence of a defect. As to the issue of causation by a design defect, the statistic tends to prove nothing more than the individual accident reports. It is extremely dubious, in fact, whether statistics involving probabilities as to causation have any significance at all when applied to a particular incident.

Even were it assumed that the accident reports and statistics prove the Bonanza to be accident-prone under conditions of turbulence in instrument flight, probably the warning to stay out of such conditions and emphasizing the consequences of ignoring it would be enough. By "accident-prone" is meant that the plane has some characteristics which may, when a pilot loses visual reference, call upon him for a much greater degree of skill than theretofore he had been required to exercise in controlling the airplane. If he has the skill and can make the adjustment quickly, he is safe. If the characteristics of the plane take it out of control before he adjusts, he is in serious trouble. If it is possible to get into turbulent instrument flight conditions despite the exercise of due care, then the pilot can be said, in view of the warning, to have assumed the risk of the accident proneness every time he flies subject to the possibility that he may encounter turbulent instrument conditions.

The limitation on introducing the evidence offered of other accidents for the purpose of showing notice of defects to Beech was, at the least, a proper exercise of the trial judge's discretion in conducting the course of the trial.

The judgment is affirmed.