(dissenting). In People v. Rubin (284 N. Y. 392), this court upheld the defendant’s conviction for illegal parking solely on proof that he was the owner of the automobile, and I perceive no reasonable or sound basis for not applying the principle of that case to the one before us. Proof of ownership of a vehicle, the court there decided, constitutes ‘ ‘ sufficient basis for an inference of personal conduct ”, even though no statute so provides and justifies the conclusion that it was “ the owner * * * [who] controlled the car and personally violated the regulation ” (284 N. Y., at pp. 396-397).
Whether we use the label “ presumption ” or “ inference,” it would seem indisputable that stare decisis compels affirmance. The similarities between the Rubin case and the present one are unusually striking. Just as in Rubin, so here, the trial judge relied upon the inference that the owner was the operator at the time in question, in order to bridge the gap between proof of offense and identity of offender. As here, the inference was not aided by any statute, but resulted solely from the court’s use of “ a common process of reasoning ” (284 N. Y., at p. 396); it was also in the field of ‘ ‘ traffic ’ ’ infractions, and the penalty was almost identical. There, too, the effect of the inference was to raise prima facie proof of guilt, which the defendant could have refuted by offering evidence of an alibi, through the testimony of others if he chose not to take the stand. And, finally, as in this case, the conviction likewise depended upon a finding that it was the defendant himself, not his car, that broke the law.1
Without a dissenting voice, this court held that it was permissible to find — given the offense and the fact of ownership — that it was “ the owner * * * [who] controlled the car and personally violated the regulation ” (284 N. Y., at p. 397).
We should not now lightly disregard the Rubin decision, with all its parallels of fact and reason. Surely, we should not do so, merely because in the present case the car was moving and *403attended at the time of infraction, whereas Rubin’s was at a standstill and driverless. Nor may we disregard that earlier case, because there are in this state more licensed drivers than cars. It is, nevertheless, as likely that the owner "will be at the wheel of his car when it speeds as when it parks.
If, as the majority opinion declares, it is “ not unreasonable ” to charge the owner of a parked vehicle for “ an illegal storage * * * in a public street ” (opinion, p. 401), then, it necessarily follows, it is equally reasonable to charge the owner of a speeding vehicle for an illegal journey upon such street. In neither case is it proper to relate guilt to mere ownership of the offending vehicle, but in neither case is that done. Here, as there, guilt is predicated upon the inference, always rebuttable, that the owner was the operator.
Wholeheartedly agreeing with the court’s insistence upon “ consistency” (opinion, p. 400), I favor affirmance upon the clear and compelling authority of the Rubin decision. I am more impressed by the need for present fidelity to its settled rule and rationale than I am by a concern that some day a vehicular manslaughter case, or other serious case, may induce infidelity to today’s decision.
However, quite apart from the Rubin case, both principle and common sense support affirmance of this conviction. The fact that the legislature has not seen fit to enact the presumption into a statute is of no moment. Nonstatutory presumptions are as deeply imbedded in the fabric of our law as their statutory counterparts. Mention need but be made of the presumption of guilt of crime from proof of defendant’s possession of its fruits (People v. Galbo, 218 N. Y. 283, 290; Knickerbocker v. People, 43 N. Y. 177); of due delivery of a letter from proof of its mailing (News Syndicate Co. v. Gatti Paper Stock Corp., 256 N. Y. 211); and of legitimacy from proof of marriage (Matter of Findlay, 253 N. Y. 1), to recall a few in common use in trials. Indeed, the very presumption here in question has long been recognized and applied in civil cases. (See, e.g., Potts v. Pardee, 220 N. Y. 431, 433; People v. Rubin, supra, 284 N. Y. 392, 396.)
The nature and origin of such presumptions have been well explained by the court in People ex rel. Woronoff v. Mallon, 222 N. Y. 456, 464-465: ‘ ‘ Presumptions are constantly arising from the proof of facts which move juries or courts to a determination. *404Because these presumptions are not classified into statute law does not weaken their force in the particular case. ‘ Presumptive evidence and the presumptions or proofs to which it gives rise are not indebted for their probative force to any rules of positive law; but juries, in inferring one fact from others which have been established, do nothing more than apply, under the sanction of the law, a process of reasoning, the force of which rests on experience and observation, and such inferences are presumptions of facts. ’ (Justice v. Lang, 52 N. Y. 323, 329.) ”
Rules of presumption are, of course, nothing more or less than rules of evidence. (See, e.g., People v. Nelson, 234 App. Div. 481, affd. 260 N. Y. 559.) As such, they are subjects as appropriate for the court as for the legislature. Their validity depends, not upon which body or tribunal proclaims them, but upon how well they square with traditional guarantees of due process. As the United States Supreme Court declared in Tot v. United States (319 U. S. 463, 467), in invalidating a statutory presumption aimed at possession of weapons by ex-convicts, “ The rules of evidence, however, are established not alone by the courts but by the legislature. * * * The section under consideration is such legislation. But the due process clauses of the Fifth and Fourteenth Amendments set limits upon the power of Congress or that of a state legislature to make the proof of one fact or group of facts evidence of the existence of the ultimate fact on which guilt is predicated.” It is, therefore, clear that legislative action in this field would not alter the question.
The test of the validity of a presumption has often been formulated. It requires that the fact proved, that is, the fact from which the inference proceeds, be rationally connected to the fact inferred. We have said, quite recently and in a serious criminal case, that a presumption is good or bad depending " upon whether, based on life and life’s experiences, a rational connection between the fact proved and the ultimate fact presumed may be said to exist.” (People v. Terra, 303 N. Y. 332, 335.)
The inference is, of course, rebuttable, and, to satisfy the demands of due process, it is not essential that the inferred fact follow as an inevitable conclusion from the proposition which *405has been proved. Here, for example, the circumstance that it is not always the owner of a speeding car who is at the wheel does not ipso facto invalidate the presumption. The basic test is whether common experience supports the probability, not the certainty, that, if the first fact is true, the second is also true. To the Supreme Court, speaking in the Tot case, we are indebted for this analysis of the use of presumptions in criminal cases (supra, 319 U. S., at pp. 466-467):
“ An indictment charges the defendant with action or failure to act contrary to the law’s command. It does not constitute proof of the commission of the offense. Proof of some sort on thé part of the prosecutor is requisite to a finding of guilt; it may consist of testimony of those who witnessed the defendant’s conduct. Although the Government may be unable to produce testimony of eye witnesses to the conduct on which guilt depends, this does not mean that it cannot produce proof sufficient to support a verdict. The jury is permitted to infer from one fact the existence of another essential to guilt, if reason and experience support the inference. In many circumstances courts hold that proof of the first fact furnishes a basis for inference of the existence of the second.”
I cannot say that the courts below were in error in holding that the defendant’s ownership of the car “ furnishes a basis for inference ”, supported by “ reason and experience ”, that he was driving. And, if that cannot be said, this court may not void the inference, merely because some assign to it a somewhat lesser degree of probability than others have selected.
Beyond that, in assessing the validity of a rule of presumption, we should not wholly lose sight of the practical ends of justice. Speeding on the highway — carrying with it, as it does, an ever-growing toll of death and injury — presents not only a serious police problem, but a grave social problem as well. Scientific devices are at hand to meliorate the menace, but they will be made useless if too heavy insistence is placed upon the need of chase and capture to sustain each conviction. It is impossible to blink the fact that there are simply not enough police officers to patrol the roads efficiently without the assistance of mechanical aids, such as the phototraffic camera here *406employed. It is a great pity that the advantages to be derived from its use are to be denied to this state by judicial action.
I do not mean to suggest that these practical considerations would be sufficient to justify an inference otherwise impermissible, or that a presumption would become just and proper merely because it will facilitate a conviction. The balance between the rights of the individual and the interests of society may at times be a delicate one, but here the court, in my view, is going far beyond the necessities of the case. The body of the offense, speeding, has been proved, and only the question of identity remains. It does not strike me as harsh, or as imposing any but the lightest burden, to require a defendant in a case such as this to offer evidence upon a fact peculiarly within *his own knowledge, particularly since it may be done through the testimony of others. (See, e.g., 8 Wigmore on Evidence [3d ed., 1940], § 2273, pp. 426-427, 431.)
What the court, through Judge Seabs, wrote in 1940 in People v. Rubin (supra, 284 N. Y. 392, 396-397) — and it constitutes an impeccable summary of the good sense of the rule for which I speak — might well have been written for the very case before us:
“ To rule that this inference may not be drawn from the established facts would be to deny to the trier of the facts the right to use a common process of reasoning. (Justice v. Lang, 52 N. Y. 323.) Ownership of a vehicle in civil cases has long been recognized as prima facie proof that it was being operated by or for the owner. [Oases cited.] Here, ownership has been held a sufficient basis for an inference of personal conduct. If he was not in control he could easily have produced a witness or witnesses to show it. [Cases cited.] We find it competent under the circumstances to conclude from the proof that the owner of the car controlled the car and personally violated the regulation. [Cases cited.] ”
The conviction should be affirmed.
Fboessel, Van Voobhis and Btjbke, JJ., concur with Desmond, J.; Fuld, J., dissents in an opinion in which Conway, Ch. J., and Dye, J., concur.
Judgments reversed, etc.
. The regulation involved in the Bubin ease was in no sense an “in rem” provision, directed against the car itself, but was aimed at the individual who actually parked it; thus, the regulation provided, in so many words, that “No person shall park a vehicle ” for more than one hour in certain streets. (New York City Traffic Regulations, § 85a; see, also, People v. Rubin, supra, 284 N. Y. 392, 394.)