The opinion of the court was delivered by

This action was brought in the Superior Court, Chancery Division, by the Passaic Junior Chamber of Commerce, Inc. and Charles D. Binns, Warren Binns, Harold Binns and Russell Binns, partners, trading as Binns Passaic Iron and Brass Foundry. The defendants are the Housing Authority, the Board of Commissioners, *385of forty pounds of air pressure thrown” into the bottle, which, according to the supervisor of production for the defendant, would “completely disintegrate” any bottle “defective for any reason.” The system of bottling was described by the same witness as “used in 95% of the bottling industry.” The witness did not, however, purport to bo a glass expert. Moreover, there does not seem to have been any procedure for testing filled bottles.

The motions for involuntary dismissal raised the question as to whether the doctrine of res ipsa loquitur was applicable in the circumstances of this case, there having been no direct evidence of negligence on the part of defendant in producing the bottled product which exploded or burst. Most of the early cases held that the rule of res ipsa was not applicable on the basis of proof of a single explosion, the incident occurring after the bottle passed from the control of the defendant manufacturer or bottler. See Annotation, 4 A. L. R. 1094 (1919). One of the early decisions, however, Payne v. Rome Coca Cola Bottling Co., 10 Ga. App. 762, 73 S. E. 1087 (Ct. App. 1912), held that where the plaintiff can “by direct or circumstantial evidence” (73 S. E., at page 1088) negative any inferences that such an explosion was due to mishandling by any custodian of the bottle between the time of its passage from possession by the bottler and the mishap there will survive a permissible inference that the manufacturer is to blame. Probably the majority of cases decided since the Payne case follow its broad rationale. The cases are collected in Annotation, 4 A. L. R. 2d 466 (1949); see Prosser, Law of Torts (2d ed. 1955), ■§ 42, pp. 203, 204; 2 Harper and James, Law of Torts (1956), § 19.5, p. 1077, n. 7 (“it is submitted that proof of the fact of the accident coupled with a showing that the bottle was subjected to neither extreme temperature nor negligent handling after it left the bottler’s control will constitute a prima facie case of negligence against the bottler”). Some jurisdictions follow the so-called North Carolina rule under which res ipsa does not apply unless there is evidence of the bursting of other bottles produced by the same defendant in substantially *386similar circumstances and reasonably proximate in time. Ashkenazi v. Nehi Bottling Co., 217 N. C. 552, 8 S. E. 2d 818 (Sup. Ct. 1940).

Although the courts of this State appear committed broadly to acceptance of the extended concept of res ipsa loquitur in this class of case, Taylor v. Berner, 7 N. J. Misc. 597, 146 A. 674 (Sup. Ct. 1929), reversed on the point of damages but affirmed on the applicability of res ipsa loquitur in 106 N. J. L. 469 (E. & A. 1930); MacPherson v. Canada Dry Ginger Ale, 129 N. J. L. 365 (Sup. Ct. 1943); Markowitz v. Liebert & Obert, 23 N. J. Misc. 281, 288, 43 A. 2d 794 (Sup. Ct. 1945); but see Dunn v. Hoffman Beverage Co., 126 N. J. L. 556 (E. & A. 1941); cf. Alston v. J. L. Prescott Co., 10 N. J. Super. 116, 120, 121, 124-126 (App. Div. 1950); and see Kramer v. R. M. Hollingshead Corp., 5 N. J. 386, 390 (1950), yet in none of the decided cases did the court have occasion to consider, as here we do, the precise nature of the function of appraisal by the trial court of the proofs purported to exclude inferences of causative connection between the conduct of others than the bottler-defendants and the damaging incident. How air-tight a factual case was required to be presented by the plaintiff negating any causal relationship between the conduct of others than the defendant and the explosion, in order to justify an affirmative answer to the question whether res ipsa loquitur was properly invoked here to aid the plaintiff, at the conclusion of his case ? Must the proofs on that issue be such that a jury could, under normal rules of sufficiency of proof to make a jury issue, decide it adversely to the hypothesis of such a relationship? Or is the criterion a finding by the court that the proofs are not of sufficient potency to destroy the continued subsistence of the probability that “in the ordinary course of things” the mishap could not have resulted unless there had been a lack of due care by the defendant? In other words, does the net balance of the probabilities from all the proofs at the conclusion of plaintiff’s case continue to preponderate for the hypothesis of defendant’s negligence and its proximate causal relationship with the accident?

*387In the Taylor case, supra, the question posed was not raised because no consideration at all was given to the possibility of negligence after delivery by the bottler; res ipsa loquitur was applied flatly on the basis of the assumption that an explosion of a bottle of vichy “does not ordinarily result” “with the exercise of reasonable care” by the bottler (7 N. J. Misc., at page 598).

In the Dunn case, supra, which is defendant’s principal reliance, the bottled sarsaparilla had been sold by the defendant bottler to a tavern and stored there for an unknown period of time prior to its sale to the plaintiff; it had then been transported in the trunk of a car some 30 miles, part way over dirt roads, before the plaintiff was injured when the bottle burst in his hand. There was no evidence as to the nature of the handling of the bottle in the tavern. The court held res ipsa loquitur not applicable. The court deemed that plaintiff had failed to discharge the obligation of showing, as part of his affirmative case, that there was due care in the handling of the bottle by all of its custodians after it left the control or management of defendant. Significantly, for purposes of our present inquiry, the court said (126 N. J. L., at page 559):

“At the conclusion of the case, when the motion for a direction of a verdict for defendant was made, it was just as reasonable to infer that while in the custody of Poole’s tavern the tensile strength of the bottle had been weakened by careless handling there, as it was to infer that the defendant had been negligent in the bottling.”

In other words, although not thus expressly articulated, the conclusion was that on the basis of the proofs presented the basic requirement of res ipsa loquitur no longer obtained, i. e., an apparent state of facts indicating that “in the ordinary course of things” the explosion would not have taken place if the defendant were exercising due care. Mumma v. Easton & Amboy Railroad Co., 73 N. J. L. 653, 658 (E. & A. 1906). The probabilities did not preponderate toward the hypothesis of defendant’s negligence as the cause of the explosion.

*388In the most recent of the New Jersey exploding beverage cases, MacPherson v. Canada Dry Ginger Ale, Inc., supra, the situation closely approximated that presently involved. The plaintiff was an employee in a confectionery store and was injured when a bottle already in a cooler exploded as she opened the cooler to put other bottles in. There was no evidence as to when the offending bottle was delivered, but, as in the present case, delivery to the store was by the defendant and “due care had been used at all times” in handling the bottles in the store. Bes ipsa loquitur was not mentioned, but the court relied upon Taylor v. Berner, supra. In affirming a judgment for the plaintiff the court said (129 N. J. L., at page 366): “The ease was one of fact. The trier of the facts was at liberty to accept the testimony of due care in subsequent handling and rest a verdict thereon.” While this language is unexceptionable insofar as it may refer to the ultimate finding of the facts supporting the judgment for plaintiff, it will not serve to explain the implicit determination as to the applicability of res ipsa loquitur.

A review of fundamentals reminds us that the determination of the applicability of that doctrine is a function ■of the court, not the jury. If the factual situation at the ■end of the plaintiff’s ease warrants its application, the court thereby determines that a prima facie ease has been presented by the plaintiff which permits, but does not require the jury to find for him. Wildauer v. Rudnevitz, 119 N. J. L. 471 (E. & A. 1938); Marzotto v. Gay Garment Co., 11 N. J. Super. 368, 375 (App. Div. 1951), affirmed 7 N. J. 116 (1951). Ordinarily, of course, the court may not determine the rule to be applicable unless it first finds that “the instrumentality or thing” involved in the injurious occurrence ■was under the control of the party charged with liability ■or his agents. Hamrah v. Clements, 3 N. J. 285, 289 (1949); Oelschlaeger v. Hahne & Co., 2 N. J. 490 (1949). In these ■exploding bottle cases, however, the adherence to res ipsa notwithstanding passage of the instrumentality from control ■of the defendant manufacturer or bottler prior to the accident is sometimes rationalized on the theory that control *389at the time of the act of negligence, rather than at the time of injury, is sufficient. Bradley v. Conway Springs Bottling Co., 154 Kan. 282, 118 P. 2d 601, 604 (Sup. Ct. 1941); Benkendorfer v. Garrett, 143 S. W. 2d 1020, 1022 (Tex. Civ. App. 1940); and see Kramer v. R. M. Hollingshead Corp., supra (5 N. J., at page 390); Alston v. J. L. Prescott Co., supra (10 N. J. Super., at page 125). But it has been aptly observed that this approach begs the question as to whether it was the bottler’s negligence that proximately caused the explosion. Case Comment, 36 Marquette L. Bev. 205, 206 (1952). It would appear preferable to discard the fiction of constructive control and rest the applicability of res ipsa loquitur in this type of case upon its underlying philosophy — • that it is simply a device for measuring the sufficiency of one kind of circumstantial evidence, where direct evidence of defendant’s negligence is typically inaccessible to plaintiff, to make out a prima facie case of negligence, Shaw v. Calgon, Inc., 35 N. J. Super. 319, 327 (App. Div. 1955)—and that the element of control in the defendant at the time of the accident is dispensable in this kind of case, if, in the light of all the circumstances reflected by the plaintiff’s proofs, including the strength of his negation of causative relationship between the conduct of the plaintiff and others than the defendant and the harmful event, it still appears probable “in the ordinary course of things” that the accident (explosion) was the proximate result of defendant’s negligence. 2 Harper and James, op. cit., supra, § 19.7, p. 1087; Prosser, op. cit., supra (§ 42, p. 205).

It should now be clear that of the two alternative criteria posed earlier in this opinion for determining the sufficiency of the plaintiff’s case to withstand a motion for dismissal in this kind of action, absent any evidence of defendant’s negligence other than the event of the explosion of the bottle, the second is the appropriate one. It is for the court itself to determine, on the basis of its own judgment “as to the general experience of mankind” (cf. Cicero v. Nelson Transportation Co., Inc., 129 N. J. L. 493, 495 (Sup. Ct. 1943), whether the probabilities still preponderate for the *390hypothesis of defendant’s negligence as the cause of the explosion after weighing the force of plaintiff’s evidentiary exclusion of other possible factors, including his own conduct. If they do, res ipsa loquitur applies, entirely apart from the circumstance that it will ultimately be for the jury to consider all the facts and to determine whether the parties have sustained the burden of proof normally allocable to them, respectively, as to such matters as negligence, contributory negligence and proximate cause (including the incidence of responsible intervening cause). See 2 Harper and James, op. cit., supra, § 19.8, p. 1093.

"We thus arrive at the question as to whether the trial court correctly determined that res ipsa loquitur precluded a dismissal at the end of plaintiff’s case. Factually the situation is closely comparable with that presented in the MacPherson case, supra. It is stronger in the sense that here the offending bottle was shown not to have been in the luncheonette more than 11 days, and probably only three days. In the MacPherson case there was no evidence as to when the bottle was delivered to the store. It is strongly inferable from the proofs that the exploding bottle was not touched from the time the case was delivered by defendant until the plaintiff carried it into the service room. The proofs tend to negate any untoward handling of the bottle at that time. Contrary to the factual conclusion by the court in the Bunn case, supra, as to the particular situation there manifested, it does not here appear from the proofs, as they stood at the conclusion of plaintiff’s case, to be as probable that the explosion was the result of mishandling of the bottle by others than defendant, or of other causes not defendant’s responsibility, as that the responsible and proximate cause was lack of due care by defendant in putting out the product. Res ipsa loquitur precluded dismissal at the end of plaintiff’s case.

We do not reach a more favorable conclusion upon defendant’s contentions that there should have been a dismissal at the end of the entire case, or the grant of a new trial on the thesis that the jury verdict was against the *391weight of the evidence. The weight to be accorded defendant’s detailed proofs concerning its exercise of care in bottling the beverage was for the jury. MacPherson v. Canada Dry Ginger Ale, Inc., supra (129 N. J. L., at page 366). As tort duty is measured by the scope of the risk which negligent conduct foreseeably entails, 2 Harper and James, op. cit., supra, § 18.2, p. 1018, it is fair to assess the duty of a producer of bottled carbonated beverages which have the capacity for sudden explosion and serious hurt to bystanders in terms of such reasonable care as takes account of the kind of handling and the range of atmospheric change to which bottles of such beverages are commonly and foreseeably subjected in ordinary commercial experience. Cf. Harpell v. Public Service Coordinated Transport, 20 N. J. 309, 315-317 (1956). “The public * * * does not expect to handle beverage bottles with all the tenderness of a new father for his first born child.” Dingwall, “Exploding Bottles,” 11 NACCA L. J. 158, 170 (1953).

Affirmed.