MEMORANDUM AND ORDER

NEAHER, District Judge.

This is a diversity negligence action against an automobile manufacturer, arising out of an unusual accident in New York, that came to trial before the undersigned in the fall of 1973. At the close of the evidence, after three weeks of trial, the court reserved decision on cross-motions for a directed verdict and charged the jury of 12 sitting in the case. The jury disagreed and a mistrial was declared. The case has remained pending before the court on the Chrysler defendants' motions for judgment in accordance with their prior motions, notwithstanding the absence of a verdict. Rule 50(b), F.R.Civ.P.

Before discussing the merits of those motions, it may be worthwhile to observe how poorly the federal unanimous jury rule in this diversity case served the declared goal of a ‘ just, speedy, and inexpensive determination of every action,’ Rule 1, F.R.Civ.P., by giving the plaintiff an advantage he would not have enjoyed had he brought his case in the State court.

After a lengthy deliberation the jury reported, via a note, that the ‘ Jury is Deadlocked.’ Court Exhibit 3. In response, the court gave the jury the modified Allen charge, and the jury resumed deliberations. Some time later, another note was received, confirming the continuing deadlock:

‘ The vote at the present time is 11 to 1 in favor of defendant ‘ Chrysler’ .' Court Exh. 4.

The court's response was again to require the jury to deliberate further, and to tell them that ‘ the verdict must be unanimous.’ Trial Record (T.R.) 79-E (October 26, 1973). Finally, the following note was sent to the court:

‘ Jury's Verdict ‘ 11 to 1 in favor of Chrysler Corp. Deadlock, no possibility of change.’ Court Exh. 7.

With that report, and a lack of positive response to the court's face-to-face inquiry of the jury as to whether there was any hope that further deliberation might alter the situation, T.R. 103-E (October 26, 1973), the court declared a mistrial and dismissed the jury.

Thereafter the parties submitted briefs on the still-outstanding motions for a directed verdict. In addition, the court requested the parties to brief the following issue:

‘ Whether, in a diversity civil action originally brought in a federal district court in the State of New York, the principles of Erie R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938), in light of the principles enunciated in Williams v. Florida, 399 U.S. 78, 90 S.Ct. 1893, 26 L.Ed.2d 446 (1970), Johnson v. Louisiana, 406 U.S. 356, 92 S.Ct. 1620, 32 L.Ed.2d 152 (1972), and Apodaca v. Oregon, 406 U.S. 404, 92 S.Ct. 1628, 32 L.Ed.2d 184 (1972), and other relevant authorities, require the federal court to receive a verdict of not less than five-sixths of the jurors constituting a jury as provided by the New York Civil Practice Law and Rules, § 4113(a), as authorized by Article 1, § 2, N.Y.Constitution.’ Memorandum Order of October 31, 1973, at p. 2.

The court has concluded for the reasons expressed in part I below that the jury's communication that eleven of the twelve jurors had agreed on a verdict does not constitute a legal verdict in this case. But as part II, infra, explains the court is satisfied that judgment should be entered for the Chrysler defendants dismissing the complaint on the merits.

I.

The court requested the attorneys to brief the jury unanimity issue because of a recent series of cases in the Supreme Court which, when read together, substantially, if not entirely, erode the conceptual underpinnings of what were previously thought to be essential attributes of trial by jury, as preserved by the Sixth and Seventh Amendments to the Constitution, in the federal courts.

Starting with Williams v. Florida, 399 U.S. 78, 90 S.Ct. 1893, 26 L.Ed.2d 446 (1970), the Court held that a 12-man panel is not a necessary ingredient of trial by jury, and a State criminal trial by a jury of six did not violate the Sixth Amendment rights of the convicted defendant ‘ as applied to the States through the Fourteenth.’ Id. at 86, 90 S.Ct. at 1898. In Johnson v. Louisiana, 406 U.S. 356, 92 S.Ct. 1620, 32 L.Ed.2d 152 (1972), the Court held that criminal jury unanimity was not a requisite of due process of law. Id. at 359. That same day the Court decided Apodaca v. Oregon, 406 U.S. 404, 92 S.Ct. 1628, 32 L.Ed.2d 184 (1972), five Justices upholding defendants' State court convictions by non-unanimous (10-2 and 11-1) criminal juries. Four of the majority felt that the Sixth Amendment, as made applicable to the States through the Fourteenth Amendment, did not require that the jury's vote be unanimous. Id. at 407-10, 92 S.Ct. 1628. The Court's reasoning was that

‘ the relevant constitutional history casts considerable doubt on the easy assumption . . . that if a given feature existed in a jury at common law in 1789, then it was necessarily preserved in the Constitution’, id. at 408-09, 92 S.Ct. at 1631 (footnote omitted).

Finally, in Colgrove v. Battin, 413 U.S. 149, 93 S.Ct. 2448, 37 L.Ed.2d 522 (1973), the Court decided the question reserved in Williams v. Florida, supra, 399 U.S. at 92 n. 30, 90 S.Ct. 1893, whether

" additional references to the ‘ common law’ that occur in the Seventh Amendment might support a different interpretation' with respect to jury trial in civil cases.' 413 U.S. at 152, 93 S.Ct. at 2450.

In concluding that they do not, the Court examined the historical underpinnings of the Seventh Amendment. The Court's conclusion was indeed a sweeping one:

‘ [B]y referring to the ‘ common law,’ the Framers of the Seventh Amendment were concerned with preserving the right of trial by jury in civil cases where it existed at common law, rather than the various incidents of trial by jury. In short, what was said in Williams with respect to the criminal jury is equally applicable here: constitutional history reveals no intention on the part of the Framers ‘ to equate the constitutional and common-law characteristics of the jury.’ ' Id. at 155-56, 93 S.Ct. at 2452 (footnote omitted) (emphasis in original).

The Court went on to hold that because in Williams it had ‘ rejected the notion that ‘ the reliability of the jury as a factfinder . . . [is] a function of size,’ ' it concluded that a 12-man civil panel was not ‘ a substantive aspect of the right of trial by jury.’ Id. at 157, 93 S.Ct. at 2453.

With the way thus cleared, the Court then concluded that the Seventh Amendment guarantee of a jury trial in federal civil cases was satisfied by a six-man jury. Id. at 158-60, 93 S.Ct. 2448. Interestingly, the Court also noted that a local district court rule to that effect was not inconsistent with Rule 48, F.R.Civ.P., holding that Rule 48 could not be read as implying ‘ a direction to impanel a jury of 12 in the absence of a stipulation of the parties for a lesser number.’ Id. at 163, 93 S.Ct. at 2456.

In light of these authorities, and especially the broad reasoning under which the Court's conclusions were reached, it is difficult to see how there remains, except through historical custom and usage, any Seventh Amendment requirement of unanimity in a civil case. Older cases all relied on the assumption we are now told was erroneous— namely, that the various incidents of a jury trial were preserved by the Constitution. Indeed, it would be anomalous to require a stricter standard of consensus in a federal civil trial than in a State criminal trial to which the Fourteenth Amendment has incorporated Sixth Amendment guarantees under Duncan v. Louisiana, 391 U.S. 145, 88 S.Ct. 1444, 20 L.Ed.2d 491 (1968). See Cooley v. Strickland Transportation Company, 459 F.2d 779, 781 (5 Cir. 1972); 9 Wright & Miller, Federal Practice & Procedure § 2492 at 482.

It was with that background that the court directed the parties to brief the Erie issue in this case, for it seemed that the federal interest in a unanimous verdict in a diversity case was far less clear than it was once thought to be. On the other hand, the unambiguous requirement in New York law is that agreement among at least ten of the twelve jurors would have resulted in a verdict in this case, had this case been tried in the courts of this State. Art. 1, § 2, N.Y.Const.; CPLR § 4113(a). The State interests behind such a rule are clear— chief among them to reduce the waste of judicial time and resources of the court, the parties and their counsel by reducing the frequency of hung juries. See Apodaca v. Oregon, supra, 406 U.S. at 411, 92 S.Ct. at 1633 (‘ Requiring unanimity would obviously produce hung juries in some situations where nonunanimous juries will convict or acquit’ ); Weinstein-Korn-Miller, supra, § 4113.02 (‘ Modification of the common-law rule [in New York] was stimulated by the costly consequences of jury disagreements, its tendency to encourage compromise verdicts, and the delays caused by the crowded docket on the jury calendar. . . . As predicted, the rate of jury disagreements declined [following enactment of the modification] from slightly over five per cent to approximately three-and-one-half per cent . . .’ ); see generally Recent Statutes, 37 Colum.L.Rev. 1235 (1937).

With the demise of Rule 48, F.R.Civ.P., in Colgrove, supra, it can no longer be thought that there is a federal procedural rule dictating unanimity in federal civil trials. Thus, the Erie question becomes much more complex, for there is no longer any inflexible federal rule to which the court undoubtedly would have been obliged to defer under Hanna v. Plumer, 380 U.S. 460, 471, 85 S.Ct. 1136, 14 L.Ed.2d 8 (1965). Instead the question is one that appears to be controlled by the principles enunciated by the Court in Byrd v. Blue Ridge Electric Cooperative, Inc., 356 U.S. 525, 78 S.Ct. 893, 2 L.Ed.2d 953 (1958), to which might be added Szanty v. Beech Aircraft Corp., 349 F.2d 60, 63-64 (4 Cir. 1965), and Miller v. Davis, 507 F.2d 308, 315 (6 Cir. 1974). In fact, some courts have already had occasion to reexamine, in light of the above recent developments, the Erie question with respect to the size of the jury to be afforded in diversity cases. See Palmer v. Ford Motor Corporation, 498 F.2d 952 (10 Cir. 1974); Wilson v. Nooter Corporation, 475 F.2d 497, 502-04 (1 Cir. 1973). Both Palmer and Wilson upheld use of a local district court rule requiring six-man juries in those cases.

Despite these portents, the court is satisfied that it would not be fair to apply the New York rule retroactively to this case, even if convinced of its applicability. The fact is that all parties and the court proceeded to and through the trial on the implicit assumption that the verdict, if any, had to be unanimous. This was evident in the court's instructions both at the close of the case and during actual deliberations. In short, it was, and is, the law of the case. See 1B Moore's Federal Practice ¶ 0.404[1] at 401-03. On reflection of the matter, it is simply too late in the day for a jury verdict to be salvaged from the proceedings in this case. We therefore turn to the question of whether the proof warranted jury consideration in the first instance.

II.

In determining whether a motion for a directed verdict should be granted, it makes no difference whether the court applies the New York rule, Calvert v. Katy Taxi, Inc., 413 F.2d 841, 846 (2 Cir. 1969), or a more general federal formulation, O'Connor v. Pennsylvania R.R. Co., 308 F.2d 911, 914-15 (2 Cir. 1962), for the standards are in general accord. Noonan v. Midland Capital Corporation, 453 F.2d 459, 461 (2 Cir. 1972). The evidence must be viewed most favorably to the party against whom the motion is made, and that party is also given the benefit of all reasonable inferences from the evidence. Unfavorable evidence, to be considered, must be uncontradicted and unimpeached. Bigelow v. Agway, Inc., 506 F.2d 551, 554 (2 Cir. 1973). The motion will be granted only if

‘ (1) there is a complete absence of probative evidence to support a verdict for the nonmovant or (2) the evidence is so strongly and overwhelmingly in favor of the movant that reasonable and fair minded men in the exercise of impartial judgment could not arrive at a verdict against him.’

Noonan v. Midland Capital Corporation, supra, 453 F.2d at 461. See also Bernardi v. Rederi A/B Saturnus, 512 F.2d 660 (2 Cir. 1975). A mere scintilla of evidence is insufficient to warrant jury consideration. Epoch Producing Corp. v. Killiam Shows, Inc., 522 F.2d 737, 742-743 (2 Cir. 1975).

The accident out of which this litigation arose occurred on August 22, 1967 in a parking lot operated by plaintiff in Bronx County, New York. It resulted in plaintiff's sustaining serious injuries when he was pinned between his recently acquired 1960 Valiant automobile and another vehicle, a panel truck parked in the lot. The Valiant had been manufactured by defendant Chrysler Corporation (‘ Chrysler’ ) and contained what is commonly referred to as a ‘ push-button’ transmission. The theory of plaintiff's case was that due to a design defect in the push-button transmission, it was possible for the car to move forward under its own power even though the neutral transmission button was depressed. Plaintiff referred to this asserted condition as a ‘ fake neutral.’ His claim was that at the time the accident occurred the neutral button was depressed, yet the car moved and struck him as he was standing in front of it. Chrysler's position, in defense, was that if the car had started in neutral, it was not mechanically possible for it to have moved forward in neutral, since a change into drive gear was required to produce the hydraulic flow that effected the link between engine power and driveshaft.

In order for plaintiff to succeed it was necessary to establish that (1) the push-button transmission mechanism was defective when manufactured; (2) Chrysler failed to exercise the requisite due care in designing, manufacturing or testing the push-button transmission mechanism; (3) the defect was the proximate cause of his injuries; and (4) he himself was free from contributory negligence. For the reasons which follow, the court is of the opinion that plaintiff adduced insufficient proof as to (1) and (2) to merit submission of those issues to the jury and, accordingly, that Chrysler is entitled to a directed verdict in its favor. At this point a brief recapitulation of the evidence is in order.

Plaintiff's somewhat confused version of the events preceding the accident was as follows: the Valiant was parked in his lot about eight feet behind a panel truck. After several unsuccessful attempts, he started the car, ‘ in park,’ but noticed that the motor was running irregularly and smoke was coming out from under the car. He set the emergency brake, by depressing a lever to the floor with his foot, turned off the motor and got out of the car. He observed gasoline dropping to the ground and forming a pool. He saw the co-defendant Alfred A. Diaz approaching and requested his assistance.

According to Diaz, on August 22, 1967 at about 5:00 p.m. he went to plaintiff's parking lot to retrieve his own automobile. When he arrived at the parking lot, plaintiff asked him to help adjust the carburetor on the 1960 Valiant. Diaz started the car and then got out to open the hood. He was unable to open it and requested plaintiff to do so, which plaintiff did. Diaz, in his effort to adjust the carburetor, pulled a rod connecting the carburetor to the gas pedal, causing the motor to race very fast and the car suddenly to lurch forward, pinning the plaintiff, who had been standing directly in front of the vehicle, against the panel truck. Diaz, who had been standing alongside the Valiant, was apparently uninjured.

Plaintiff's attempt to show that the push-button transmission in the Valiant was defective rested solely on the testimony of his expert witness, Elliot M. Frimmer. Frimmer stated that during his 27-year career he had worked on tens of thousands of transmissions, including hundreds of 1960 Valiant transmissions and that the latter were often unreliable in that even when the neutral button was depressed the car might move forward if the gas pedal were pushed. He explained that this occurred because pressing the neutral button on the dashboard did not always ensure that the gears actually moved completely into neutral position; and that when the engine was raced, accomplished by either depressing the gas pedal or manually operating the gas feed, sufficient pressure built up to move the gear into drive and consequently propel the car forward.

This malfunction occurred, in his opinion, because the flexible cable which moved the detent plate in the transmission would stretch when the driver, attempting to release the oft-struck parking pawl, would force it. He also stated that a spring-activated ball, which was supposed to lodge itself in the designated detent position, did not work properly because Chrysler used a weak spring. The detent plate, which is serrated, and described by Frimmer as consisting of ‘ hills and valleys,’ controls the flow of transmission fluid. The movement of the transmission fluid, in turn, causes the gear designated by the driver to engage. When the neutral button is pushed, if the detent plate does not properly align, as a result of a stretched cable and weak spring, ‘ [i]t [the ball] could get stuck on the top of the hill and stay there.’ Then ‘ if the engine would race up sufficient hydraulic pressure could accumulate to force the car to go into drive. Even though the button was in neutral.’ This defect could manifest itself at any time from ‘ 0 to 100,000 miles' and was not the result of normal wear and tear.

Frimmer further stated that he had received many such complaints. He said that about one month after the accident, he tested plaintiff's 1960 Valiant and that one out of five times it moved forward when ‘ given gas' in neutral. He admitted that he did not examine the actual mechanism in the car .

Chrysler attempted to refute Frimmer's testimony through its own experts, who testified in substance that not only was his ‘ fake neutral’ theory contrary to the physical facts but that an examination of plaintiff's 1960 Valiant years after the accident revealed that the push-button mechanism was operating perfectly.

Jeremiah T. Ball, Assistant Chief Engineer of Mechanical Engineering at Chrysler, testified that if the parking cable in the car were to stretch, as Frimmer testified it did, a built-in ‘ fail-safe’ device would prevent the driver from getting the car out of park. He also stated he had measured the diameters of the cables in plaintiff's 1960 Valiant, although not at every point, and that they were of normal width, indicating no stretching.

Ball also controverted Frimmer's weak spring theory in several ways. First, he stated that the ball and spring, which ride in the detent, are not necessary to the proper alignment of the detent plate but are just an additional safety device. He explained that a slide mechanism actually rotates the detent plate and the steel ball simply engages the already positioned detent plate and locks it into place.

Furthermore, he testified that from 1960 to 1964 Chrysler produced approximately 460,000 cars with push-button transmissions involving the same cable, detent, ball and spring mechanism and this was the only car he ever heard of starting in neutral and then going forward in neutral.

In November 1970, Isaac Stewart, a private consulting engineer, examined plaintiff's Valiant and found the push-button unit unbolted from the dashboard and hanging back down from its cables toward the floor; the push-button box assembly itself was missing. No functional testing was done at that time.

On January 15, 1973, he had the car raised on a lift, the transmission pan dropped and the fluid drained. Using a dental mirror he inspected the detent plate and ball, ascertaining that it was in the neutral position. The neutral slide in the shift box, which was hanging in back of the dashboard, was depressed. He then had a mechanic who was working with him depress various slides while he, from underneath the car, observed the detent plate and the ball. He testified:

‘ At the same time I observed the movement of the manual valve as the detent plate moved it. I observed that when I asked the mechanic to press the specific button or specific tang that the reaction that occured down below was consistent with my request. ‘ . . . I asked the mechanic to press the D button or the drive tang and observed that the detent plate assumed the D position. I then asked him to press another one— the N or whichever one it was at the time— I don't remember the specific sequence at this time— and the detent plate then moved to the position that I requested. I used all five combinations.’ Tr. 929.

The slide box was then placed in its normal position behind the dashboard and the test run again and successfully so.

Finally, Stewart testified that the ball never got caught ‘ on a hill’ between drive and neutral and that for it to do so would be impossible since there was not enough room on the point of the ‘ hill’ for the ball to sit.

Additionally, Dr. Clifford Becker, who had driven the 1960 Valiant for seven years and then sold it to plaintiff, testified that the car never moved forward when started in neutral.

It is apparent that the evidence tending to establish that the transmission in plaintiff's 1960 Valiant was not in fact defective and that Frimmer's ‘ fake neutral’ theory was inconsistent with the actual operation of the gear control mechanism was overwhelming. Frimmer admitted that he never made any examination of the operative parts. In fact, the only evidence of a defect is Frimmer's testimony that he tested the car one month after the accident and that it moved forward in neutral one out of five times. At no time did he suggest that the forward movement he observed was at all comparable to the sudden and powerful surge forward which pinned the plaintiff.

To be weighed against this is Ball's testimony that he measured the diameters of the cables in the 1960 Valiant and they had not stretched (Frimmer had testified that a stretching of the cables was the reason for misalignment of the detent). He also stated that the detent was essentially slide-operated and would correctly position itself even absent the spring-operated ball (Frimmer had attributed the ‘ fake neutral’ situation in part to the ball being pushed by a weakened spring).

Additionally, Stewart testified that he actually tested the push-button mechanism in plaintiff's car and found it to be completely operative. He also stated that it was impossible for the ball to rest on top of one of the detent positions because they came to a point and provided inadequate room for the ball to sit on (Frimmer's ‘ fake neutral’ theory had the ball resting on the detent point separating drive from neutral).

Chrysler's expert testimony was essentially uncontroverted.

In sum, the court concludes that the evidence is so strong and overwhelming that reasonable and fair-minded jurors in the exercise of impartial judgment could not conclude that the push-button transmission in plaintiff's Valiant was in fact defective in design or manufacture and that consequently plaintiff has failed to establish a requisite element of his cause of action.

While there was, as discussed above, overwhelming evidence that the 1960 Valiant's transmission contained no defect, there is a more basic reason why liability cannot attach in this case. It is well settled that to avoid liability in negligence Chrysler must undertake to use reasonable care in its manufacturing process. MacPherson v. Buick Motor Company, 217 N.Y. 382, 394-95, 111 N.E. 1050 (1916). Reasonable care in these circumstances necessitates

‘ such inspections and tests during the course of manufacture and after the article is completed as the manufacturer should recognize as reasonably necessary to secure the production of a safe article.’ Smith v. Peerless Glass Company, 259 N.Y. 292, 296, 181 N.E. 576, 578 (1932).

Here, Ball's uncontroverted testimony was that tests conducted by Chrysler in 1956 on the same detent, cable, spring and ball, which formed the operative section of the 1960 Valiant push-button transmission, showed that the device operated properly 50,000 times ‘ even when the buttons were pushed by twice the pressure normally exerted by one's finger.’ Ball also testified that the push-button mechanism on each car was physically tested after the car came off the assembly line.

There was absolutely no evidence tending to show that this testing was inadequate. Kross v. Kelsey Hayes Company, 29 A.D.2d 901, 287 N.Y.S.2d 926, 927 (3rd Dep't 1968).

Even assuming that there was a latent defect in the push-button transmission caused by a stretching of the cable and consequent misalignment of gears (aided by a weak spring), there was no evidence that this later condition could have been ‘ reasonably apprehended by the manufacturer at the time of manufacture of the subject vehicle.’ McNally v. Chrysler Motors Corporation, 55 Misc.2d 128, 284 N.Y.S.2d 761, 764 (Sup.Ct.1967). This is to say that Chrysler does not have to manufacture perfect products but only must, in their manufacture, ‘ exercise a reasonable degree of care commensurate with all the circumstances.’ Bernstein v. Remington Arms Company, Inc., 16 A.D.2d 694, 227 N.Y.S.2d 802, 804 (2d Dep't 1962).

Plaintiff, in his cross-examination of Ball, elicited testimony to the effect that Chrysler, as reflected in its trouble diagnosis charts produced for its service people (Plaintiff's Exh. 4), was aware that its cars sometimes drove in neutral. However, this is entirely consistent with Ball's testimony that while it was possible for the detent ‘ to be off’ an entire gear, plaintiff's own testimony that prior to the accident the car operated properly in reverse and started in neutral established that his car could not have been off one entire gear. Frimmer also testified that one month after the accident all gears except neutral operated properly. Plaintiff introduced no evidence linking the problem alluded to in Chrysler's service manual to his ‘ fake neutral’ theory, and for a jury to infer that it somehow did relate to that theory would require an exercise in pure speculation.

Similarly, Frimmer's testimony that in 1960 or 1961 he had a meeting with Chrysler engineers at which they discussed a tool, designed by him, the purpose of which was to better synchronize the gear-shifting process, was never directly linked to Frimmer's ‘ fake neutral’ theory and could well have related to some other transmission problem. It would be improper to allow the jury to impute to Chrysler knowledge of plaintiff's specific ‘ fake neutral’ claim based on this general discussion. ‘ An inference will be upheld only if application of common experience and logic to the underlying evidence will support it.’ Epoch, supra, 522 F.2d at 744.

In sum, there was no proof in this case that Chrysler was aware or should have been aware of any defect or potential defect in the push-button transmission mechanism in 1960 Valiants as that claimed here. Nor was there any proof that Chrysler had received any complaints similar to plaintiff's. And finally actual inspection of the parts themselves and their mechanical operation revealed no defect. In a nutshell, there was no proof from which the jury could infer that it was reasonably foreseeable to Chrysler that injury might result from a defect in its push-button transmission and any verdict to the contrary could only result from impermissible speculation.

For the foregoing reasons, Chrysler's motion for a directed verdict is granted.

So ordered.

On the other hand, the jury's note, coupled with the colloquy in open court, might well be deemed a theoretically sufficient verdict. Certainly there was no doubt about where everyone stood and any requirement of a publicly-stated verdict seems to have been met, at least in spirit, if not technically so. And the declaration of a mistrial can be revoked. See Grace Line v. Motley, 439 F.2d 1028, 1033 (2 Cir. 1971). Finally, plaintiff's authorities also point out that the right to demand a jury poll is waived if not asserted before the jury's discharge.

In this case it is clear, however, that substantially all of the evidence that might have been adduced on the issue of breach of warranty was placed in the record in an effort to establish negligently defective design and manufacture of the Valiant. If anything, more detailed proof was required on the latter issue than would have been on the former. That being the case, the court sees no need to reconsider its decision on the statute of limitations question previously decided. The evidence presented would not have supported a verdict for plaintiff under either theory, so no injustice results from such an approach.