dissenting. The essential questions in this case are whether plaintiffs introduced sufficient credible evidence from which it could be inferred that a defect existed' in the alignment of the transmission and the neutral safety switch, and whether that defect existed at the time the automobile left the hands of the defendant. The plaintiffs’ evidence, simplified, is basically that the 17-month-old car, which had previously exhibited no malfunction, behaved in an unusual manner which contributed to an accident; that no servicing had apparently been done on the transmission or the ignition before the accident; and that after the accident the car behaved in an unusual manner, in that it could be started, and perhaps run, when the transmission w,as japimed pp/j the ppinjtpr incjicatpd ttyat i[, yras ip Driye, *219No direct physical evidence was presented that any part was defective, and, in fact, there was evidence that the safety switch itself was properly aligned and that the. transmission worked properly as indicated. The plaintiffs’ case is that from this evidence one can infer a defect in the car, in that the car conld be started in Drive, and that this defect existed at the time of manufacture.
In products liability cases, proof that a defect existed is often difficult and complex. Frequently the product in dispute will have been destroyed, beyond any possibility of analysis, or be so complex that a plaintiff would have a greater difficulty in determining the presence of a defect than would the manufacturer. In most cases, proof of the defect must necessarily be by circumstantial evidence and inference. No general rule can adequately apply to the wide range of such cases, each involving a different mixture of fact and inference, but fundamental to any such case is that some defect must be proved. As Prosser states, in Strict Liability to the Consumer in California, 18 Hastings L. J. 9, 52-54:
‘ ‘ The mere fact of an accident, as where an automobile goes into the ditch, does not make out a case that the product was defective; nor does the fact that it is found in a defective condition after the event, when it appears equally likely that it was caused by the accident'itself. The addition of other facts, tending to show that the defect existed before the accident, may make out a case, and so may expert testimony. So likewise may proof that, other .similar products made by the defendant met with similar misfortunes, or the elimination of other causes by satisfactory evidence. In addition, there are some accidents, as where a beverage bottle explodes or even breaks while it is being handled normally, as to which there is human experience that they do not ordinarily occur without a defect. As in cases of res ipsa loquitur, the experience will give rise to the inference,'and it may be sufficient to sustain the plaintiff’s burden of proof,” '
*220Although plaintiff’s evidence may be sufficient to permit an inference that something was wrong with the car, that alone is not sufficient to establish a defect, except perhaps in cases, analogous to res ipsa loquitur, in which ordinary human experience tells us that the event could not happen without a defect. The instant case is not such a ease, for driver error, failure of some part, accidental or unwitting damage to the car, and other possibilities do provide other explanations. See Gast v. Sears (1974), 39 Ohio St. 2d 29, 313 N. E. 2d 831. In this case, the fact that something went wrong is not sufficient to support an inference that a defect caused the accident.
The particular defect that plaintiff alleges is that the indicator and the transmission gear linkages were both malaligned in a similar fashion, relative to the neutral start switch, so that the car as manufactured could start in Drive rather than, as intended, only in Park and Neutral. There are various ways in which that particular fact could be proved, by means of several types of evidence. Keeton, Manufacturer’s Liability: The Meaning of Defect in the Manufacture and Design of Products, 20 Syracuse L. Rev. 559, for example, suggests five ways by which the particular fact could be proved. (See, also, Rheingold, Proof of Defect in Product Liability Cases, 38 Tenn. L. Rev. 325.)
(1) Plaintiff might introduce evidence by an expert based upon an examination of the product in question following the happening of the damaging event. Expert evidence would be di.re.ct evidence of an identifiable defect. In the instant case, two expert witnesses testified, and neither was able to point out an identifiable defect.2 Both testified that the car could be started in an indicated Drive position, but neither identified a cause for that based upon their examination. The only explanation, offered by one of the experts, was that the pointer was probably damaged.
*221(2) There may simply be evidence of a damaging event occurring in the course of or following use of a product, whether by the testimony of the user or otherwise. This may be sufficient in the case where, as a matter of common knowledge, a defect is the probable cause. As already indicated, the instant case is not one involving such common knowledge.
(3) A plaintiff may produce both evidence of a damaging event occurring in the course of or following the use of a product and expert evidence that the most likely probable cause was attributable to a defect in the product being used at the time.
In the instant case, the only expert who was qualified to state such an opinion, the company expert called as a witness by the plaintiff, was not asked to state whether a defect was the probable cause, and in fact made clear in his testimony that he believed it probable that there was no defect and that the apparent starting of the car in Drive was probably caused by damage from the accident to the indicator. He stated only that it was possible for various components to be malaligned as plaintiffs’ theory required.
(4) In addition to evidence of an accident and the probable cause of such accident, evidence could be introduced to negate the existence of “probable causes” not attributable to the maker.
This type of evidence was not introduced in the instant case, except with respect to the issue of maintenance on the car in the 17 months after delivery.
(5) In some cases, the physical evidence of the actual condition of the product after the accident would be such that a layman could infer that it was defective.
No such physical evidence was introduced in the instant case.
All of these forms of proof relate to whether, after something has gone wrong, that event can properly be attributed to a defect. The facts of the individual case will *222be crucial in determining whether such an influence is permissible. In State Auto Mutual Ins. Co. v. Chrysler Corp. (1973), 36 Ohio St. 2d 151, 304 N. E. 2d 891, for example, a new one-half ton pickup truck had been used for 77 days; the brakes failed because of a loss of brake fluid from a hole, with fibers blown outward, in a brake line which was original equipment and had not been tampered with; and the actual brake hose was unavailable, due to no fault of the plaintiff. The issue in that case was specifically whether the hole was caused by a defect or by an accidental impact of some kind, and the circumstances of that case were held sufficient to raise a jury issue. As the court stated, at page 161:
“Obviously, the most exact method of ascertaining a defect in the hose would be to subject the hose to scientific tests, under pressures applicable in an ordinary breaking situation. In the absence of any direct evidence that the hose introduced by appellee Chrysler Corporation is the hose in question, this cannot be done.”
In the instant case, the most exact method of determining whether there was in fact a malalignment of the transmission parts would have been actual examination of those parts, which were under the control of plaintiffs for several months after the accident. The examination actually made was incomplete. No identifiable defect was found; nor was the testimony of the mechanics who repaired the automobile offered. Failing such examination, even by a non-expert, plaintiffs could have introduced expert testimony that the probable cause of the accident was a defect. This was not done, and the expert testimony was only to the effect that a malalignment was possible, but, according to one expert, was not probable. The sum of the evidence is thus only that something unusual happened in the car, and that a possible explanation of that happening is a defect. I believe plaintiffs could have and should have done more in order to make a case for the jury, for the essential link of actual proof, between the accident and any possible explanation, is missing.
*223The same difficulty arises with regard to the issue of whether the claimed defect existed at the time the car left the hands of the defendant. Plaintiffs introduced evidence that the transmission had not, been serviced or tampered with, and that the family’s method of starting the car could have permitted a defect to remain undiscovered. But this, again, is only evidence of the possibility of a defect. Here, the' car was 17 months old, there was no expert testimony that the claimed defect was one which would probably have existed at the time the car was manufactured, and common experience does not permit any such inference. In tracing the defect in the product into the hands of the defendant, “* * * [t]here is first of all the question of lapse of time and long continued use. Tins in itself will never prevent recovery where there is satisfactory proof of an original defect; but where there is no such definite evidence, and it is only a matter of inference from the fact that something broke or gave way, the continued use usually prevents the inference that more probably than not the product was defective when it was sold. The seller certainly does not undertake to provide a product that will never wear out.” Prosser, supra, at 54. In the instant case, plaintiffs’ negative evidence indicates the possibility that the claimed defect could have existed at the time of manufacture; but this possibility remains only a theory, for there is a lack of any positive evidence which would overcome the inference arising from the long-continued use of the car.
At the center of plaintiffs’ case, which is made up of evidence of an unusual event and a possible explanation thereof, there should be some positive proof that the possible malalignment was something more than a theory. Such proof was not presented and remains a matter of speculation. For that reason, I agree with the trial judge that plaintiffs failed to submit sufficient evidence from which it could be inferred that a defect existed in the Friedman car at the time it left the hands of the defendant.
One expert was the company witness called by the plaintiffs. The second was a metallurgist called by plaintiffs, who was not, however, qualified as an expert in automotive mechanics.