BURNETT, Judge.

This is a toxic tort case. James Earl has alleged that his lungs were injured when he was exposed at work to vapors emitted from a plastic film manufactured by the Cryovac division of W.R. Grace Company. He filed a worker's compensation claim against his employer and filed this tort action against Cryovac. The worker's compensation claim was settled. In this case, the district court entered summary judgment against Earl, holding that he had failed to establish any causal connection between his injury and Cryovac's product. Today, we vacate the summary judgment and remand the case.

In Part I of our opinion, we discuss the elements of a toxic tort action, the requirement of proximate cause, and the use of expert testimony to establish causation. In Part II, we focus on the issue of causation in the present case, summarizing the evidentiary facts and the opinions of expert witnesses. In Part III, we enunciate the standards governing summary judgments and explain why a summary judgment should not have been entered in this case.

I

When a plaintiff brings an action against the manufacturer of a product, seeking damages for negligence or for strict liability in tort, he carries the burden of showing (1) that he has suffered an injury, (2) that the product was defective or unsafe when it left the control of the manufacturer, and (3) that the plaintiff's injury was proximately caused by the product. E.g., Henderson v. Cominco American, Inc., 95 Idaho 690, 518 P.2d 873 (1973). If the product is alleged to be unsafe because it is toxic, the causation issue turns upon two subsidiary questions: (a) Did the product, or a substance in the product, have the capacity to cause the type of harm claimed by the plaintiff? (b) Was the plaintiff's exposure sufficient to produce a toxic effect? Farber, Toxic Causation 71 MINN.L.REV. 1219 (1987).

Because toxic torts typically involve a period of latency between exposure and manifestation of injury, the outcome of the litigation often will turn upon the issue of causation. This issue may be addressed by general or particular evidence. General evidence, derived from research in medicine, chemistry or other disciplines of science, may establish the toxic potential of a substance under certain conditions of exposure. Particular evidence, arising from diagnosis and treatment of the plaintiff's ailment, may prove that an exposure has occurred and may demonstrate a manifestation of the product's toxic potential.

A

Both types of evidence, general and particular, are probative as to the legal requirement of proximate cause. Under Idaho law, proximate cause, in the sense of cause in fact, embraces two closely related elements:

First, an event is the cause in fact of a succeeding event only if the succeeding event would not have occurred "but for" the prior event. Thus, an act or omission is not the cause in fact of ensuing damage if the damage likely would have occurred anyway. The second element is a requirement that the first event be a "substantial factor" in producing the succeeding event. [Citation omitted.] Thus, a defendant's conduct is the cause in fact of an event only if it was a material element and a substantial factor in bringing it about. [Citation omitted.]

Edmark Motors, Inc. v. Twin Cities Toyota, Inc., 111 Idaho 846, 849, 727 P.2d 1274, 1277 (Ct.App. 1986) (quoting Challis Irrigation Co. v. State, 107 Idaho 338, 343, 689 P.2d 230, 235 (Ct.App. 1984)). However, a proximate cause need not be the sole or primary cause in fact. It may be concurrent with other causes which, in combination, cause the harm. Idaho Jury Instructions (IDJI) No. 230; Fouche v. Chrysler Motors Corp., 107 Idaho 701, 692 P.2d 345 (1984).

When a case goes to trial, the existence of proximate cause, like any other required element of the plaintiff's case, must be established by a preponderance of the evidence. The trier of fact must be persuaded that the plaintiff's claim of causation "is more probably true than not true." IDJI No. 112-1. By employing a probability standard, the law strikes a balance between plaintiffs' and defendants' rights. It avoids compelling a plaintiff to meet the virtually impossible burden of proving causation with certainty in order to obtain compensation for an injury. It also avoids compelling a defendant to pay damages when his connection with the plaintiff's injury is nothing more than a mere possibility. This balance reflects a value judgment based on our society's intuitive sense of civil justice. See generally Calabresi, Concerning Cause and the Law of Torts, 43 U.CHI.L.REV. 69 (1975).

When doctors and scientists evaluate causation, however, they do not strike a value-based balance. In their work, they apply standards of greater or lesser rigor than probability. See generally Nesson, Agent Orange Meets the Blue Bus: Fact-Finding at the Frontier of Knowledge, 66 B.U.L.REV. 521 (1986) (hereafter cited as Fact-Finding at the Frontier of Knowledge). In scientific research, where the replication of an observed event is the ultimate test of truth, the usual standard of causation is a high degree of certainty. Conversely, in diagnosis and treatment of a specific patient, where the objective is to find a cure or to prevent further harm, a doctor may ascribe causal significance to a possibility that falls short of a probability.

Accordingly, when the courts apply medical and scientific evidence to a question of causation, they must interpret the evidence carefully in light of the applicable standard. They may not assume that a causal relationship is probable merely because a physician deems it significant in his diagnosis and treatment of a patient's condition. Neither may they assume that a causal relationship is improbable merely because it has not been documented in a body of research literature where a high degree of certainty is demanded. These distinctions are particularly important in a toxic tort case where, as here, the issue of causation is framed by the expert opinions of scientists and treating physicians.

B

Causation is a question of fact for the jury to resolve. E.g., Nelson v. Northern Leasing co., 104 Idaho 185, 657 P.2d 482 (1983). However, the admissibility of expert testimony on that subject is a question for the trial judge to decide. I.R.E. 104. The judge's decision rests on three criteria. First, the witness must be qualified as an expert who possesses "scientific, technical, or other specialized knowledge [that] will assist the trier of fact to understand the evidence or to determine a fact in issue. . . ." I.R.E. 702. Second, if the qualified witness intends to give testimony containing an opinion or inference, and such testimony will be based on facts not otherwise in evidence or within the expert's personal knowledge, then the facts must be "of a type reasonably relied upon by experts in the particular field. . . ." I.R.E. 703. Third, if the testimony is relevant and admissible under the foregoing criteria, it may nevertheless be excluded if its probative value would be "substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time or needless presentation of cumulative evidence." I.R.E. 403.

In many toxic tort cases (including, as we shall see, the present case), the controversy over expert testimony arises from the second of these criteria. The question is whether the experts' opinions are based upon facts known to them or otherwise admitted in evidence — and, if not, whether the facts are "of a type reasonably relied upon by experts in the particular field," as required by Rule 703. In addressing these questions, a trial judge must take care not to allow his decision on the admissibility of expert testimony to be influenced by his perception of whether the testimony would be persuasive to a jury at trial. The weight given to expert testimony is to be determined by the jury. E.g., State v. Hopkins, 113 Idaho 679, 747 P.2d 88 (Ct.App. 1987). In evaluating the facts upon which an expert bases an opinion, the judge must not "infringe upon the fact-finder's role in assessing the weight of the expert testimony." In re Agent Orange Product Liability Litigation, 611 F. Supp. 1223, 1244-45 (E.D.N.Y. 1985).

It is particularly important in toxic tort cases, and in other litigation where highly technical issues are presented, that the judge not exclude expert testimony by second-guessing the facts upon which the experts choose to rely. "Judges, after all, are lay persons, no matter how well-read they are in science." Fact-Finding at the Frontier of Knowledge, supra, at 531. If an expert is qualified to render an opinion, based on his expertise in the subject, then he must be accorded substantial deference in the selection of data upon which he chooses to base his opinion. In re Japanese Electronic Products, 723 F.2d 238 (3d Cir. 1983), reversed on other grounds sub. nom. Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986); see generally Rheingold, The Basis of Medical Testimony, 15 VAND.L.REV. 473 (1962). With this constraint in mind, we now turn to the evidence adduced during the summary judgment proceedings in the present case.

II

On the causation issue, the plaintiff's case rested largely upon the testimony of three qualified expert witnesses. The first to be deposed was Charles E. Reed, M.D., the plaintiff's primary treating physician. Dr. Reed was, and is, board-certified in internal medicine and in the subspecialty of pulmonary disease. He examined the plaintiff on many occasions, ran numerous laboratory and clinical tests, and devised a treatment program that ultimately alleviated many of the plaintiff's symptoms.

Dr. Reed obtained a patient history which paralleled the deposition testimony of the plaintiff himself. The plaintiff worked in the meatpacking industry for approximately twenty-five years, of which twenty years had been spent in "slaughter" rooms and the most recent five years had been spent in a "packing" room. Until moving to the packing room, the plaintiff had not suffered any severe breathing disorders. At one time, he had been a smoker of cigarettes, but he had discontinued this habit ten years before seeing Dr. Reed. Approximately seven years before he saw Dr. Reed, the plaintiff was hospitalized briefly for shortness of breath; however, no diagnosis was made, and he suffered no severe symptoms until he began work in the packing room. In Dr. Reed's opinion, this prior history was unremarkable.

The plaintiff consulted Dr. Reed, upon reference from another physician, when he experienced a worsening problem of chest tightness and shortness of breath. He noticed that his symptoms would temporarily subside during weekends and vacations, but would resume when he returned to work in the packing room. Dr. Reed learned that the packing room contained a machine that wrapped meat in plastic bags and then heated the plastic in water at or near the boiling temperature. The hot water caused the plastic to shrink tightly around the meat; accordingly, the machine was known as a "shrink tunnel." Because the packing room was kept at a low temperature, the boiling water in the shrink tunnel produced a heavy fog. Some employees, who later submitted affidavits in this case, reported that they could smell plastic, and occasionally burning plastic, in the air.

Dr. Reed also learned that the plastic material used in the packing room was a shrinkable thermoplastic "barrier bag" manufactured by Cryovac. Dr. Reed found medical literature documenting the existence of "meatwrapper's asthma" or "meatwrapper's syndrome," a chronic lung disease observed in employees of meatpacking plants and butcher shops where plastic bags were cut by a thermal process known as a "hot wire." The disease, which produces symptoms similar to those experienced by the plaintiff in this case, has been attributed in the literature to the release of vapors during the heating of the plastic material. As noted by Dr. Reed, the researchers are satisfied that a causal connection exists between the heated plastic and the disease, although the precise chemical (or group of chemicals) responsible for the etiology of the disease has not been identified.

In consultation with Dr. Reed, the plaintiff quit his job in the packing room at the meatpacking plant. During the course of treatment prescribed by Dr. Reed, his symptoms gradually receded. When asked to characterize the plaintiff's condition at the time depositions were taken in this case, Dr. Reed stated that the plaintiff suffered from chronic obstructive pulmonary disease. He observed that "some of the reversible component of his disease [has] completely cleared leaving a residual component of chronic obstruction."

Upon these facts, Dr. Reed formed an opinion, expressed in his deposition, that the plaintiff's disease was caused or exacerbated by exposure in the work place to vapors emitted from the plastic bags as they were heated in the shrink tunnel. Dr. Reed described this causal connection not merely as a possibility but as a "high probability." He elaborated as follows:

James Earl had no prior history of disease. James Earl worked with a material that is known to cause lung disease. James Earl developed lung disease. James Earl left the employment and was treated for a period of time and became clinically well. . . . [W]e are left with an assumption, a high probable cause, that his exposure at the work place caused or greatly exacerbated the lung problem.

Dr. Reed suspected that polyvinyl chloride, a chemical commonly found in plastic films or bags, was a significant factor in the plaintiff's disease. However, he did not rest his opinion upon the existence of any single chemical. Rather, he said, "I would expect it to be a combination of components from the results of the reading I have done. . . . [T]here are a number of polyvinyls [in the plastic bag] and it would appear that any one or perhaps all of them can cause this kind of difficulty in their breakdown products."

Dr. Reed's opinion was challenged by Cryovac's able counsel, partly because it did not specify the particular component(s) of the plastiv vapors which caused the plaintiff's disease, and partly because the temperature in the shrink tunnel process was less than half the usual temperature of the "hot wire" implicated in the "meatwrapper's asthma" cases. On each point, however, Dr. Reed's opinion was corroborated by the deposition of Allen S. Todd, an industrial hygienist with more than twenty years of experience in toxicology, including the toxicology of plastics. Having become familiar with the Cryovac barrier bag and with the shrink tunnel process, Mr. Todd stated in a deposition: "I think scientifically there will be vapors emitted under the conditions of use under [sic] the barrier bag." He added that the vapors would "range from the monomers to oxidative by-products of some of the lower molecular components [and] perhaps your stabilizers or your plasticizers or other materials which are components of the ultimate products and could vary from time to time." When asked whether the vapors could be harmful, he replied:

I would expect many of the compounds we are talking about to have impact on the lungs. Okay. We know some of the monomers do. I would suspect some of the stabilizer intermediates or oxygenative products to similarly impact upon the lungs.

Although Mr. Todd suspected that vinyl acetate might play an important part in the plaintiff's disease process, he — like Dr. Reed — emphasized the synergistic effect of multiple chemicals rather than narrowing his opinion to a single component.

Mr. Todd also concurred with Dr. Reed that the "meatwrapper's asthma" literature was instructive. He acknowledged that a similar body of literature did not exist with respect to shrink tunnels, but he stated, "[T]hat doesn't mean anything. It just means nobody specifically has published on it." He noted that when a "hot wire" is changed to a "cold wire" (one which is heated to a lower temperature), the concentration of airborne contaminants is reduced, but not eliminated. He further noted that a "hot wire" process results in both pyrolysis and degradation. He stated that degradation would also occur at shrink tunnel temperatures. He elaborated:

[T]here are some common denominators with either process . . . scientifically it would make sense that you could do it in a shrink tunnel and in fact you have a second factor and that is even though your temperatures are lower, your residence time and contact is probably far greater.

Based upon his experience in the toxicology of plastics, and upon information furnished to him regarding the plaintiff's occupational and medical history, Mr. Todd concluded:

[The plaintiff's disease] is work oriented. And I think the orientation is related to the contaminants generated in his last occupation and it really [is] as far as you can go at this point in time. . . . What I am saying is within, with a reasonable scientific certainty, with all the information available, yes, it does appear to be a cause and effect relationship.

The last expert witness to be deposed was Jiri E. Kresta, Ph.D., a research professor at the University of Detroit with more than twenty years of experience in chemical technology, particularly in polymers. Dr. Kresta arranged for a laboratory experiment in which a sample of a Cryovac plastic bag was heated to the temperature of boiling water and the liberated vapors were collected. Upon analysis, Dr. Kresta found that the vapors contained no vinyl acetate or polyvinyl chloride as such, although certain breakdown products — acetic acid and chlorinated alkanes — were discovered. A host of other chemicals also appeared, including a vaporized plasticizer identified by Dr. Kresta as diphenyl octyl phosphate. During his deposition, Dr. Kresta read from product literature issued by Monsanto, a manufacturer of diphenyl octyl phosphate under the trade name "Santicizer 141." "According to the manufacturer," Dr. Kresta stated, "this chemical 'can cause irritation. Elevated temperatures may cause release of vapors which are harmful if inhaled.'"

Although Dr. Kresta characterized his experiment as "qualitative" rather than "quantitative," he said that the diphenyl octyl phosphate was a "major component" of the vapor. He further remarked that all of the substances discovered were present in quantities probably equaling or exceeding one part per million. Substances of a lesser concentration were not likely to be detected with the equipment and methodology employed in the experiment. Dr. Kresta's observations in this regard were pertinent in light of prior testimony of Dr. Reed, who acknowledged that he could not postulate an "average threshold" chemical exposure that would produce harm to lung tissue, but that the minimum physiological "insult" level was unlikely to be less than 0.7 parts per million in ordinary persons. It could be lower in sensitized individuals. Dr. Kresta also noted, as had Mr. Todd, that if vapors were liberated by a process of hydrolysis — that is, by chemical interaction between the plastic and hot water — the concentration of effluent could be increased. Moreover, as Dr. Reed stated in his deposition, contaminants carried in water vapor, such as the steam surrounding a shrink tunnel, could be expected to make more harmful contact with lung tissue than contaminants borne in dry air.

III

Despite the foregoing facts and expert opinions, the district court entered summary judgment in favor of Cryovac. The court held that the plaintiff had failed to establish a genuine issue of material fact regarding causation. With due respect to the capable district judge, we disagree.

The standards governing the entry and review of a summary judgment are well settled. Controverted facts are viewed in favor of the party resisting the motion for summary judgment. Where, as here, a jury has been requested, the non-moving party is also entitled to the benefit of every reasonable inference that can be drawn from the evidentiary facts. Anderson v. Ethington, 103 Idaho 658, 651 P.2d 923 (1982). Thus, the burden of a plaintiff, when faced with a motion for summary judgment, is not to persuade the judge that an issue will be decided in his favor at a trial. Rather, "he simply must present sufficient materials to show that there is a triable issue." 6 J. MOORE, W. TAGGART, J. WICKER, MOORE'S FEDERAL PRACTICE § 56.11(3), at p. 56-243 (2d ed. 1988) (emphasis original).

A "triable issue" exists whenever reasonable minds could disagree as to the material facts or the inferences to be drawn from those facts. E.g., Petricevich v. Salmon River Canal Co., 92 Idaho 865, 452 P.2d 362 (1969); Snake River Equipment Co. v. Christensen, 107 Idaho 541, 691 P.2d 787 (Ct.App. 1984). Therefore, although the plaintiff carries the ultimate burden at trial of proving causation to a standard of probability, the court in a summary judgment proceeding does not weigh the evidence for probability. The court determines only whether the evidence frames an issue upon which reasonable minds could disagree. Beyond this threshold of reasonableness, weighing the evidence is a task reserved to the jury, who will have a first-hand opportunity to consider conflicting evidence and to observe the cross-examination of witnesses. See generally Schwarzer, Summary Judgment Under the Federal Rules: Defining Genuine Issues of Material Fact, 99 FEDERAL RULES DECISIONS 465 (1984).

Here, the judge acknowledged that the elements of a plaintiff's case could be established with circumstantial evidence. Farmer v. International Harvester Co., 97 Idaho 742, 553 P.2d 1306 (1976). Indeed, our Supreme Court held in Farmer that a plaintiff may use circumstantial evidence to show that a product is unsafe, without identifying and proving a specific defect in the product. Accord Curtis v. DeAtley, 104 Idaho 787, 663 P.2d 1089 (1983). A circumstantial showing that the product is unsafe may be sufficient of itself, or it may be supplemented by expert opinion testimony. Fouche v. Chrysler Motors Corp., 107 Idaho 701, 692 P.2d 345 (1984); Edwards v. Conchemco, Inc., supra note 5. In the present case, the plaintiff presented both circumstantial evidence and expert opinion testimony on the issue of causation.

In his memorandum decision, the district judge made only passing reference to the circumstantial evidence. He focused instead upon a perceived inadequacy of the expert opinion testimony. Although the judge did not expressly hold the opinions of Dr. Reed and Mr. Todd to be inadmissible under I.R.E. 703, he attacked those opinions as being unsupported by the underlying facts of the case. We acknowledge, of course, that the wholly unsupported conclusions of experts are not sufficient, by themselves, to create a genuine issue of fact. In re Agent Orange Product Liability Litigation, 611 F. Supp. 1285 (E.D.N.Y. 1985) (a decision related to the other Agent Orange case cited supra). However, as mentioned earlier in this opinion, a judge should avoid weighing the evidence on summary judgment, and should accord substantial deference to a qualified expert's selection of the facts upon which he bases his opinion. Those facts, as we have noted, may be within the expert's knowledge; they may be reflected elsewhere in the record; or they may come from outside the record if they are of a type upon which experts in the field reasonably would rely. In our view, the opinions of Dr. Reed and Mr. Todd meet this standard.

The real thrust of the district judge's decision — and of Cryovac's argument on appeal — appears not to be that the experts relied upon improper facts; rather, it is that they relied upon facts insufficient to support conclusions that the plastic vapors had a toxic potential and that the plaintiff was sufficiently exposed to suffer harm. This is not an attack upon the experts' qualifications or upon the admissibility of their testimony; it is an attack upon the content of the testimony itself. Such an attack is inappropriate in summary judgment proceedings, except to the very limited extent of determining whether the experts' opinions, coupled with the circumstantial evidence, are sufficient to frame an issue upon which reasonable minds could disagree.

We believe this narrow test has been satisfied here. On the question of whether fumes from the heated plastic were potentially toxic, the opinions of Dr. Reed and Mr. Todd were consistent with the literature on "meatwrapper's asthma" and with Dr. Kresta's finding of at least one harmful chemical in the effluent produced at shrink tunnel temperatures. Regarding the sufficiency of the plaintiff's exposure, the opinions of Dr. Reed and Mr. Todd were consistent with Dr. Kresta's finding of effluents in measurable quantities higher than the lowest likely "insult" level for an ordinary person.

Moreover, with respect to both toxicity and sufficiency of exposure, the expert opinions were supported by the particularized evidence of the plaintiff's experience while working near the shrink tunnel in the packing room. As noted in this opinion and in Dr. Reed's deposition, the plaintiff suffered a worsening of symptoms while on the job and an improvement of symptoms when he was not working. During a prior twenty-year period, when the plaintiff worked at locations in the meatpacking plant other than the packing room, his medical history was, in Dr. Reed's view, unremarkable. These facts were considered significant by Dr. Reed and Mr. Todd in reaching their respective conclusions that it was not merely possible, but was highly probable, that the plaintiff's disease was caused or exacerbated by the heated plastic.

We do not consider it fatal to the plaintiff's case that the etiology of his disease has not been traced to a discrete component or set of components within the heated plastic vapor. As explained by our Supreme Court in Farmer v. International Harvester Co., supra, the plaintiff need only show that the product is unsafe; he need not identify and prove the specific defects which render it unsafe. The same approach is reflected in the cases cited at footnote 2, where victims of "meatwrapper's asthma" have been allowed to recover despite scientific uncertainty as to the precise etiological link between their disease and specific chemical(s) in the heated plastic vapors. See also Robinson v. SAIF Corp., 78 Or.App. 581, 717 P.2d 1202 (1986) (allowing recovery of worker's compensation benefits for chemical exposure, based on treating physicians' opinions, although exact etiology of disease was undetermined). Moreover, the question of legal liability does not turn upon the isolation of the offending substance(s) where each suspected substance emanates from the defendant's product. Cf. Basko v. Sterling Drug, Inc., 416 F.2d 417 (2nd Cir. 1969) (plaintiff not required to show which of two drugs caused his injury where both drugs were produced by the same manufacturer).

Finally, we note that the plaintiff's claim in a toxic tort case does not fail merely because the circumstantial evidence and the expert opinions are unsupported by animal or epidemiological studies confirming the existence of a cause-and-effect relationship. As noted by the United States Court of Appeals for the District of Columbia:

As long as the basic methodology employed to reach such a conclusion is sound, such as the use of tissue samples, standard tests, and patient examination, products liability law does not preclude recovery until a "statistically significant" number of people have been injured or until science has had the time and resources to complete sophisticated laboratory studies of the chemical.

Ferebee v. Chevron Chemical Co., 736 F.2d 1529, 1536 (D.C. Cir. 1984), cert. denied, 469 U.S. 1062, 105 S.Ct. 545, 83 L.Ed.2d 432 (1984). Accord City of Greenville v. W.R. Grace Co., 827 F.2d 975 (4th Cir. 1987); Wells v. Ortho Pharmaceutical Corp., 788 F.2d 741 (11th Cir. 1986).

Upon the record before us, viewing the facts and legitimate inferences in best light for the plaintiff, we conclude that reasonable minds could fins, or at least could disagree regarding, a causal connection between the plaintiff's disease and Cryovac's product. Therefore, a genuine issue of material fact exists. The summary judgment is vacated and the case is remanded. Costs to the plaintiff (appellant), James Earl. No attorney fees on appeal.

WALTERS, C.J., and SWANSTROM, J., concur.