ERVIN, Judge.

The Cassisis and their homeowner's insurer appeal from a summary judgment entered against them in a products liability action founded on theories of strict liability, negligence, and breach of an implied warranty. The only issue for our determination is whether the lower court correctly ordered summary judgment on the ground the Cassisis' proofs failed to show their damages were caused by a product in a defective condition at both the time of the accident and the time it was within the possession of the manufacturer or the retailer. Because we find genuine issues of material fact remain unresolved on that question, we reverse the summary judgment and remand the cause for further proceedings consistent with this opinion.

Mrs. Cassisi's deposition testimony was that she had purchased the alleged offending product, a clothes dryer, from the retailer, McDuff Appliances; that during the 19 months of its use, no maintenance work or repairs had ever been performed on it; and that it had always been normally operated. On the date of the accident, Mrs. Cassisi left her home with the dryer in operation; upon her return, she found the house ravaged by fire.

While appellants' expert, Clayton Morrison, a registered professional engineer, was unable to pinpoint a specific defect within the dryer (it had been badly damaged), it was his opinion that the fire had begun inside the dryer. When asked if a malfunction was inherent in the product, he responded:

My conclusion is that it was inherent . . . because I have concluded that the fire originated within the product. That is the source from which the fire pattern and the heat source emanates, is from the dryer itself. Now, you must therefore conclude, that it was due to some incorrect functioning of a part that was internal within the machine that caused the fire to start.

He also surmised that the malfunction was caused by an electrical short within the dryer. Mr. Morrison, however, was otherwise uncertain in his responses. He was unable to negate other possible causes of the fire, such as the possibility that flames from a fire originating outside the dryer during the machine's operation could have been drawn into its interior, causing the clothing to ignite; the possibility that the short could have occurred if either the circuit breakers were not functioning properly, or were not in existence; the possibility that the fire had been caused by a deterioration of the electrical wiring, such as its having been gnawed by rats; and, finally, the possibility that a defect could have occurred following the sale of the product to the Cassisis.

While appellants are not required to prove in a strict liability action that the manufacturer or retailer was negligent in the preparation or distribution of a product, they nevertheless have the burden, whether their case is founded in negligence, breach of an implied warranty, or strict liability, of establishing (1) that a defect was present in the product; (2) that it caused the injuries complained of; and (3) that it existed at the time the retailer or supplier parted possession with the product. 2 Frumer and Friedman, Products Liability, § 16A(4)(e)(i) at 3B-88, 89 (1980). Appellees argue that appellants' burden was not met because their evidence failed to pinpoint a defect inside the clothes dryer either at the time of the accident or at the time the product left the control of the manufacturer or supplier.

How a plaintiff meets his burden in a products liability case and thereby establishes a submissible case for jury consideration has been the source of frequent litigation. Since Section 402A of the Restatement of Torts (Second) has been adopted in Florida by West v. Caterpillar Tractor Co., Inc., 336 So.2d 80 (1976), that section should be carefully consulted in order to ascertain if the alleged offending product is one which may be subject to the Restatement's definition of defectiveness and, if so, to know what type of evidence must be presented to establish a submissible case of product defectiveness.

I. The Nature and Types of Product Defects

The black letter statement to Section 402A requires that a product be "in a defective condition unreasonably dangerous to the user or consumer or to his property. . . ." Thus, one who is injured while using a perfectly made axe or knife would have no right to a strict liability action against the manufacturer because the product that injured him was not defective. If the user is required to show only that the product was a factual cause in producing his injury, the manufacturer's liability to the injured party would be, as stated by West, supra at 90, and by Professor Wade, that of an insurer.

On first impression, the Restatement may seem to require proof that the product be both defective and unreasonably dangerous. That, however, is not the case. Section 402A defines defective condition by requiring the product to be "at the time it leaves the seller's hands in a condition not contemplated by the ultimate consumer, which would be unreasonably dangerous to him." See comment g. The words unreasonably dangerous are in turn defined as requiring the product to be "dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics." See comment i. It appears that the terms defective and unreasonably dangerous are redundant. Dean Prosser, the reporter for the Council, explains that the words "unreasonably dangerous" were added to foreclose the possibility that makers of products having the inherent potentiality for causing harm, such as drugs, whiskey, sugar, butter, etc., would become "automatically responsible for all the harm that such things do in the world." Prosser, Strict Liability to the Consumer in California, 18 Hastings L.J. 9, 23 (1966). Thus only "bad" whiskey, butter, etc., are subject to the Restatement's standard.

Concerned with the ambiguity in the standard, a number of courts have held it is unnecessary for a plaintiff to prove that the product be both defective and unreasonably dangerous. See, e.g., Cronin v. J.B.E. Olson Corp., 8 Cal.3d 121, 104 Cal.Rptr. 433, 501 P.2d 1153 (1972); Azzarello v. Black Brothers Co. Inc., 480 Pa. 547, 391 A.2d 1020 (1978); Pyatt v. Engel Equipment Inc., 17 Ill. App.3d 1070, 309 N.E.2d 225 (1974). Cronin, for example, rejected a defendant's instruction requesting that the plaintiff must establish not only that the product contained a defect which proximately caused his injuries, but also that such condition made the product unreasonably dangerous to the user or consumer. Cronin stated that the words "unreasonably dangerous" burden an injured plaintiff with an element which rings of negligence; that they are, when used in connection with the defective product terminology, susceptible to an interpretation which places a dual burden on the injured plaintiff to prove both that the product was defective and that it was unreasonably dangerous. 8 Cal.3d 121, 104 Cal.Rptr. 433, 501 P.2d at 1162. The court concluded that a plaintiff satisfies his burden in a strict liability action as to any alleged product defect if he proves there was a defect in the manufacture or design of the product and shows that the defect was the cause of his injuries. Id.

Notwithstanding the controversy by both academe and the courts over the proper test for determining a product's defective condition, the standard for all product defects under Section 402A is the same: Were the ordinary consumer's expectations frustrated by the product's failure to perform under the circumstances in which it failed? This standard works reasonably well as to those types of product defects characterized as defects resulting from manufacturing flaws — those caused by a miscarriage in the manufacturing process which produces an unintended result. Thus, when a new product suddenly and unexpectedly malfunctions during the course of its normal operation, causing injury, the standard is relatively easy to apply.

The consumer expectation standard, though adequate to identify unintended manufactured defects, is more difficult to apply as to the other two generally recognized types of product defects: (1) design defects — those which are due to design error because unforeseen hazards accompany normal use of the product created according to design, and (2) defects resulting from misinformation or inadequate warnings. As to the last two defects, the standard is said to be a very vague and imprecise one because the ordinary consumer cannot be said to have expectations as to safety regarding many features of complexly made products that are purchased, such as the risk of fire from the way gasoline tanks are installed in cars, or the magnitude of risks involved in vehicles overturning. Due to the difficulty in applying the consumer expectation standard to all types of product defects, many thoughtful commentators have suggested that it should be rejected, particularly as to those defects arising from design, in favor of a test that would weigh the utility of the design versus the magnitude of the inherent risk.

Heeding the call for a different test, some courts have explicitly rejected or modified the Restatement standard as to design defects. See Phipps v. General Motors Corp., 278 Md. 337, 363 A.2d 955 (1976); Barker v. Lull Engineering Co., supra, note 8; Caterpillar Tractor Co. v. Beck, 593 P.2d 871 (Alaska 1979). Barker qualified its prior holding in Cronin as to design defects only by applying to them the following two-pronged test:

First, a product may be found defective in design if the plaintiff establishes that the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner. [This part of the test accepts the Restatement standard.] Second, a product may alternatively be found defective in design if the plaintiff demonstrates that the product's design proximately caused his injury and the defendant fails to establish, in light of the relevant factors, that, on balance, the benefits of the challenged design outweigh the risk of danger inherent in such design.

573 P.2d at 455-56.

The most significant aspect of the Barker rule is its second prong. Once the plaintiff establishes a prima facie case showing that his injuries were caused by a product's design, the burden is shifted to the defendant to prove the design was not defective by presenting evidence of factors, such as the gravity of the danger posed by the challenged design, the feasibility of a safer design, the financial cost of the improved design, etc. 573 P.2d at 455.

Whether the Restatement standard should be augmented as in Barker is unnecessary for us to decide because we consider that appellants' evidence of a defect — if one existed — was attributable to a manufacturing flaw, and was not one of design. Appellants' expert indicated that the fire was caused by a malfunction in the clothes dryer. Such an occurrence differs either from the manufacturer's intended result or from other units of the same product line. See generally the discussion in Barker v. Lull Engineering Co., supra, 573 P.2d at 446. This type of defect is altogether dissimilar from the kind intentionally manufactured according to design that causes injury, i.e., the unguarded trencher in Auburn Machine Works Co., Inc. v. Jones, supra, note 9, whose dangers were obvious. Moreover, the alleged offending product before us is not one required to be excepted from the standard, such as one that is unavoidably unsafe. See note 4, supra. We are not required, therefore, to consider the balancing approach taken by other courts in determining whether a product was defectively designed. Our primary concern with the frustration of the ordinary consumer's expectations standard is its indiscriminate application to all types of design defects. The Restatement standard is particularly appropriate, however, to the facts before us. As stated in Phipps v. General Motors Corp., supra, 278 Md. 337, 363 A.2d at 959:

[T]here are those kinds of conditions which, whether caused by design or manufacture, can never be said to involve a reasonable risk. For example, the steering mechanism of a new automobile should not cause the car to swerve off the road, . . . the driveshaft . . . should not separate from the vehicle when it is driven in a normal manner, . . .; the brakes . . . should not suddenly fail . . .; and the accelerator of a new automobile should not stick without warning, . . . . [citations omitted]

Thus, evidence of the nature of an accident itself may, under certain circumstances, give rise to a reasonable inference that the product was defective because the circumstances of the product's failure may be such as to frustrate the ordinary consumer's expectations of its continued performance.

Our next task is to decide whether, given Mr. Morrison's testimony, and the evidence relating to the product's malfunction during normal operating conditions, the appellants established a submissible case for jury consideration.

II. The Sufficiency of Proof as to The Existence of a Manufacturing Flaw Within The Product

As previously noted, Section 402A places the burden on a plaintiff to establish that his injuries were caused by a product which was in a defective condition at the time it left the hands of the seller. Section 402A, however, is otherwise of little help in explaining how the injured plaintiff meets this burden. Many cases involving strict liability actions resolve the defect-causation issue by employing the same proof requirements common to negligence — such as by requiring the plaintiff to show that his injuries were proximately caused by the alleged defect. A plaintiff may of course establish a submissible case by either direct or circumstantial evidence, the latter frequently accomplished through the opinion testimony of an expert that the product was defective. See, e.g., Elmore v. American Motors Corp., 70 Cal.2d 578, 75 Cal.Rptr. 652, 451 P.2d 84 (1969); Bollmeier v. Ford Motor Co., 130 Ill. App.2d 844, 265 N.E.2d 212 (1970) (opinion given through answer to hypothetical question based upon expert's knowledge of similarly designed steering systems even though he had not examined the steering mechanism in question); Carey v. General Motors Corp., 387 N.E.2d 583 (Mass. 1979).

Difficulties arise, however, in those cases in which the plaintiff shows that there are several possible explanations for the accident, not all of which point to the product's defective condition. The answer frequently applied to such situations is to hold that since plaintiff's burden is to establish the presence of a defect at the time of the product's manufacture, his burden is not met — even though he presents evidence of the product's defective condition at the time of injury — without direct or circumstantial evidence showing the product was then defective. In effect, those decisions place on the plaintiff both the burden of producing evidence as well as the burden of persuasion as to that issue. To so hold, however, ignores the practical evidentiary problems which are common to products liability actions, and the policy considerations of Section 402A which were designed to assist the plaintiff in his quest for the jury's consideration of the issues alleged.

Appellees' primary argument is that appellants failed to establish a causal connection between their injuries and the alleged defect because they were unable to negate all alternative causes of the fire other than a defect within the dryer. There is authority for appellee's position, and it is exemplified most clearly by the New Jersey rule, the leading case being Jakubowski v. Minnesota Mining and Manufacturing, 42 N.J. 177, 183-84, 199 A.2d 826, 829-30 (1964). Accord, Scanlon v. General Motors Corp., 65 N.J. 582, 591, 326 A.2d 673 (1974). In Jakubowski, the court stated that when neither direct evidence is available to establish that a product is defective, or other evidence permitting an inference that a dangerous condition existed in it prior to sale, plaintiff is then required "to negate other causes of the failure for which the defendant would not be responsible, in order to make it reasonable to infer that a dangerous condition existed at the time defendant had control." Id. 42 N.J. at 183-84, 199 A.2d at 829-30.

If the Jakubowski rule were strictly applied to the facts in this case, it is clear that appellants could not recover since their circumstantial evidence of the product's defect did not negate other causes of the accident for which the defendants would not be responsible. There is, however, one type of proof of product defect, which, from a given set of circumstances, aids a plaintiff in meeting his burden by creating a legal inference that the product was defective both at the time of the injury and at the time it was within the control of the supplier. This inference arises from the occurrence of the accident itself, and the case most often cited as so holding is Greco v. Bucciconi Engineering Co., 283 F. Supp. 978 (W.D.Pa. 1967), aff'd, 407 F.2d 87 (3d Cir. 1969). We approve the Greco rule and apply it to the case now before us. That rule, applying Pennsylvania law, simply states that when a product malfunctions during normal operation, a legal inference, which is in effect a mirror reflection of the Restatement's standard of product defectiveness, arises, and the injured plaintiff thereby establishes a prima facie case for jury consideration.

The facts recited in Greco were that the defendant engineering company had manufactured a mechanical piler, designed to stack and transport steel at a mill where the plaintiff was employed. Part of the apparatus was a set of mechanical fingers which extended to support piles of steel. On one occasion, the fingers suddenly retracted, and a falling pile of steel sheets severed a portion of the plaintiff's right hand. At trial the plaintiff's evidence showed that the piler was only part of a production line which was run on a compressed-air system, and operated from a control panel, neither of which was manufactured by the defendant. Plaintiff presented no expert testimony as to proof of defect. The defendant argued that the case should never have been submitted to the jury because there was no evidence from which the jury could infer that a defect existed in the piler rather than in the compressed-air system or in the control panel.

In entering judgment on the jury's verdict in favor of plaintiff, and addressing itself to the defendant's contention that the verdict was speculative, the Greco court stated: "A finding that the malfunction was prompted by a defect in the control panel or air system would be equally `speculative'; no evidence in support of these theories having been adduced. A malfunction evidences a defect. [citations omitted] The plaintiff is not obliged to negate alternative grounds of causation." 283 F. Supp. at 984. In approving the lower court's judgment, the Third Circuit Court of Appeals held that a "`defective condition' within the meaning of Section 402A [can be established] by proving . . . the product functioned improperly in the absence of abnormal use and reasonable secondary causes." 407 F.2d at 89-90. And the inference was extended to the time of sale as well as the time of injury, the court observing that "there was . . . sufficient evidence to satisfy reasonable and well balanced minds that a defect existed at the time of sale." 407 F.2d at 91.

The Greco inference is somewhat analogous to the res ipsa loquitur inference applicable to negligence cases. While, as Prosser states, "[s]trictly speaking, since proof of negligence is not an issue, res ipsa loquitur has no application to strict liability; . . . the inferences which are the core of the doctrine remain, and are no less applicable." Prosser, Law of Torts, § 103 at 672-73 (4th ed. 1971). Both legal inferences are similar since they are based upon common sense assumptions that the occurrence of the accident is such that in the ordinary course of events it could not have happened — as to res ipsa, without the negligence of the person in control, Goodyear Tire Rubber v. Hughes Supply, Inc., 358 So.2d 1339, 1341 (Fla. 1978), and as to the other, without the product's defective condition. Additionally, both inferences are founded upon strong policy considerations that aid a plaintiff in meeting his burden of proof when direct proof of negligence or product defectiveness is wanting. Indeed, the case giving birth to the doctrine of res ipsa, Byrne v. Boadle, 2 H C 722, 159 Eng.Rep. 299 (Ex. 1863), was heavily influenced by the plaintiff's inability to extract an explanation from the defendant, as well as his incapacity to explain how the barrel which injured him fell from the second floor of Boadle's warehouse. Later cases applying the res ipsa doctrine have done so when the evidence of the true explanation of the accident is more accessible to the defendant than to the plaintiff. See Hughes v. Jolliffe, 50 Wn.2d 554, 313 P.2d 678 (1957); Appalachian Insurance Co. v. Knutson, 358 F.2d 679 (8th Cir. 1966); Wilson v. East St. Louis Interurban Water Co., 295 Ill. App. 603, 15 N.E.2d 599 (1938).

The application of the Greco inference as an aid to a plaintiff's establishing a prima facie case is perhaps most vividly portrayed by those cases in which the product was so badly damaged by a malfunction as to render impossible the plaintiff's ability to point with specificity the exact one of several potentially dangerous conditions which caused the accident. Thus, in Bombardi v. Pochel's Appliance T.V. Co., 9 Wn. App. 797, 515 P.2d 540 (1973), the court held that a submissible case was demonstrated despite evidence showing that the offending product, a television set, had been in use for three years before the fire and repaired several times, as well as the absence of any expert examination of the product due to its being "completely consumed in the fire." In so doing, the court applied the Greco inference, noting that once evidence of the malfunction was established, as well as of the product's normal use, a prima facie case was made out, particularly since the manufacturer failed to present any evidence indicating that the cause of the fire was some event or condition for which it was not responsible. The court justified its decision by relying on comment c to Section 402A, stating that the purpose of strict liability is to ensure that the costs of injuries resulting from defective products be borne by their makers who put them into the channels of trade rather than by injured persons who ordinarily are powerless to protect themselves. 515 P.2d at 545-46.

In Liberty Mutual Insurance Co. v. Sears, Roebuck Co., 35 Conn. 687, 406 A.2d 1254 (1979), the court relied upon Section 402A's definition of product defectiveness to uphold liability. There a fire occurred in an apartment near a television set. The owners testified that during the six months the set was in the household, it had received normal use. The plaintiffs' only expert witness was the chief of the local fire department who pinpointed the television set (described as a "mass of rubble") as the cause of the fire, but did not identify a specific defect. The court concluded that such proof, "together with the evidence that the set had in fact malfunctioned and the jury's knowledge from common experience that television sets, in normal use, do not self-ignite, was sufficient to establish the television set . . . was defective." 35 Conn. 687, 406 A.2d at 1257.

In the above two cases, the user's testimony as to the product's defective condition was corroborated by other witnesses. The Greco inference may arise independently of any corroboration, however, as the facts in McCann v. Atlas Supply Co., 325 F. Supp. 701 (W.D.Pa. 1971), demonstrate. There the plaintiff testified that he was driving his automobile at normal speeds when he heard a hissing sound like air escaping. His car and the trailer it was pulling suddenly left the roadway; both overturned and were destroyed by the ensuing fire. Because the tire was totally destroyed, the plaintiff relied solely on the circumstances of the accident to meet the burden of proof defect requirement of Section 402A. In response to the defendant's argument that the tire could have been damaged as a result of its striking objects in the road, the court admitted this possibility; but, relying upon Greco, stated that the plaintiff was not required to exclude every possible source of sudden deflation, and that the occurrence of the malfunction was circumstantial evidence of a defect, sufficient to carry the case to a jury. Id. at 703.

The inference, moreover, has been applied to cases in which the malfunction was such as to cause the product's disappearance. Thus, in Lindsay v. McDonnell Douglas Aircraft Corp., 460 F.2d 631 (8th Cir. 1972), an aircraft and its pilot were lost at sea. Just before the crash, a fisherman saw the plane on fire. The plaintiff was unable to negate the manufacturer's suggested cause of the crash as being due to pilot disorientation during a difficult maneuver and its hypothesis that the cause of the fire was due to the plane's venting fuel while descending in a stall with its afterburner on. In rejecting the defendant's arguments, the court stated the "proof required in a strict liability case must be realistically tailored to the circumstances which caused the form of action to be created." Id. at 639. It quoted with approval the rule proclaimed in MacDougall v. Ford Motor Co., 214 Pa. Super. 384, 257 A.2d 676, 679 (1969) that "`[p]roof of [a] specific defect in construction or design causing a mechanical malfunction is not an essential element . . .'", and that once a machine malfunctions, "`it obviously lacks fitness regardless of the cause of the malfunction.'" Id.

While all of the above cases involved products either lost or destroyed, the inference is not dependent solely upon such facts. The piler in Greco was not so badly damaged as to make expert examination impossible, although none was offered as evidence. Cases applying the Greco inference frequently involve evidentiary facts comprising both an expert's inspection of the product as well as proof of its malfunction, coupled with evidence of normal use. See, e.g., Winters v. Sears, Roebuck Co., 554 S.W.2d 565 (2d Mo. App. 1977); Knight v. Otis Elevator Co., 596 F.2d 84 (3d Cir. 1979); Farmer v. International Harvester Company, 97 Idaho 742, 553 P.2d 1306 (1976). At any event, the facts essential for the inference's application are simply proof of the malfunction during normal operation.

Without the Greco inference, all of the above cases could have been decided by directing a verdict in favor of the defendant. Apart from the evidence showing the occurrence of a malfunction during the product's normal operation, there was no proof offered that the particular alleged defective product was in fact defective. Moreover, there were a number of hypothetical causes of the accident for which the defendants would not have been responsible.

One commentator has suggested that a major consequence of the Greco inference is that it shifts the burden of producing evidence to the manufacturer once proof of the malfunction is established. Since the procedural effect of the Greco inference is identical to that of the res ipsa inference, it is more accurate to state that neither the burden of proof, nor any burden of producing evidence is cast upon the defendant "except in the very limited sense that if he fails to do so, he runs the risk that the jury may . . . find against him." Prosser, supra, § 40 at 229. Thus, the practical result of the inference is that if the manufacturer wishes to avoid a jury's consideration of the issues, it must offer evidence showing there are no genuine issues of material fact to be resolved by a jury — rather than suggest possible reasons for the product's malfunction.

Considering the complexities of products litigation, and again the policy factors of Section 402A, such a solution is the better reasoned one since it is the manufacturer — rather than the plaintiff — which presumably is in the better position to produce the technical evidence required for determining the existence of a product's defect. It is the one most familiar with the product, its expert's intimate knowledge of the product and its production process gives it a greater advantage over the average plaintiff in analyzing a product's failure. Additionally, the manufacturer should be able to trace the history of the product more easily than the plaintiff and to discover whether it was subjected to any abnormal handling.

Although we have observed the similarities between the two inferences, the important distinction between them is, of course, that of control. Before res ipsa may be applied, the injured plaintiff must prove the product causing his injury was under the exclusive control of the defendant. In a strict liability action, the fact that the defendant was not in control of the product when the injury occurred is obviously not a bar to the application of the product defectiveness inference — although it may be a factor for the jury to consider on the question whether the plaintiff's own conduct in fact contributed to his injuries. Thus, products such as television sets, electric stoves, clothes washers or dryers, which can be activated by human agency simply by turning a switch or pushing a button, but which are otherwise self-operating, could conceivably lessen a reasonable belief that the person's conduct was the cause of his injury. The user's relationship with such products is somewhat passive and custodial in nature, and does not normally involve hazardous conduct or active participation by him. On the other hand, the fact that a car leaves the road while under the control of its driver may not necessarily give rise to the same generous inference of defect.

Nevertheless, in a strict liability action, the factor of control, insofar as it relates to the time elapsed from sale to injury, is but one of several circumstances to be considered in determining whether the product was defective while it was within the control of the manufacturer or distributor. In addition to the product's age, other factors include the length of the product's use, the severity of its use, the state of its repair, its expected useful life, and whether it was subjected to any abnormal operations. Farmer v. International Harvester Co., 97 Idaho 742, 553 P.2d 1306, 1312 (1976). Ordinarily the weight of such evidence is for the trier of fact. See Sochanski v. Sears, Roebuck Co., 621 F.2d 67 (3d Cir. 1980); Liberty Mutual Insurance Co. v. Sears, Roebuck Co., supra. If, however, the product which malfunctions is shown to be so old, so frequently repaired, and subjected to such rugged use, its condition may be such as to negate any inference that it was defective at the time of its manufacture — despite evidence revealing the product's defective condition at the time of injury — because the ordinary consumer's expectations could not reasonably be said to have been frustrated by the product's failure under such circumstances. Cf. Kuisis v. Baldwin-Lima Hamilton Corp., 457 Pa. 321, 319 A.2d 914 (1974).

The Greco inference is particularly appropriate here. Combined with the expert's testimony indicating that a malfunction in the dryer was the cause of the fire, Mrs. Cassisi testified that the product had never been serviced or repaired, and that it had been normally operated during its 19 months' use. Given such evidence, it is immaterial that the plaintiffs failed to identify the specific cause of the malfunction since it is inferred that the malfunction itself, under such circumstances, is evidence of the product's defective condition at both the time of the injury and at the time of the sale.

Reversed as to both defendants and remanded for further proceedings consistent with this opinion.

SHIVERS and SHAW, JJ., concur.