Plaintiff Henry Hickerson lost his wife and home in a fire. He sued Defendants alleging a motorized wheelchair or scooter they manufactured and sold to him, a Pride Jet 3 Power Chair ("PowerChair"), was defective and caused the fire. Mr. Hickerson designated a fire causation expert witness who offered an opinion regarding the location of the point of origin of the fire and identified the PowerChair as a possible cause of the fire. The expert based these conclusions on burn patterns and damage and on the facts that the remains of the PowerChair were sitting at the center of the area of origin and that no other appliance in the area of origin contained batteries or was connected to an external power source at the time of the fire. Defendants moved to exclude Mr. Hickerson's expert witness on the basis that he was not an expert in the engineering or manufacturing of motorized chairs or scooters and was therefore unqualified to identify a specific defect in the PowerChair. Defendants also moved for summary judgment. The district court granted in part and denied in part Defendants' motion to exclude the expert witness and granted Defendants' motion for summary judgment. We reverse and remand.

The fire consumed Mr. Hickerson's home in the early morning hours of October 22, 2003. Mr. Hickerson escaped from the house, but his wife did not. That same day, two investigators from the office of the State Fire Marshal examined the site and interviewed the firefighters and other people at the scene. David Owens, one of the State's investigators, found the remnants of the PowerChair in an area of a living room that he identified as the area of the most intense heat and a likely point of origin for the fire. He examined the PowerChair and found that a battery cable had shorted to the metal of the chair. He could not determine whether the shorted cable was a cause or a result of the fire. He also identified the remains of a tower-type space heater near the remains of the PowerChair and identified shorted wires in the home's electric system. Again, he could not determine if the shorted wires were a cause or a result of the fire. Mr. Owens ultimately concluded that an electrical engineer should examine the PowerChair. He also concluded that the cause of the fire could not be determined due to the amount of damage to the scene.

Mr. Hickerson's insurance company hired a consulting firm to investigate the fire. The investigator from the consulting firm was William L. Schoffstall, and Mr. Hickerson designated Mr. Schoffstall as an expert witness regarding fire origin and causation. Mr. Schoffstall was an experienced firefighter and a certified and experienced fire scene investigator. Mr. Schoffstall arrived at the site of the remains of the Hickerson home on October 23, 2003, the day after the fire. When he arrived, family members were already sorting through the remains and were in the process of removing items from the scene.1 The family members complied with a request to exit the fire scene until after the investigation was complete.

In a written report, Mr. Schoffstall described his examination of the interior and exterior of the home. He eliminated various rooms as possible locations for the origin of the fire based on observations of the relative degrees of damage, the pattern and direction of damage, and the type of damage, i.e., fire, smoke, or heat. He noted an oval burn pattern in an area of the living room that he identified as the area of the highest heat and the likely point of origin of the fire (the same area identified by State Fire Marshal investigator David Owens). Mr. Schoffstall also noted that Mr. Hickerson had said the area of the oval burn pattern was the area from which he noticed high heat as he escaped from the house. Mr. Hickerson commented only as to heat because he could not see through the smoke when escaping the home.

Mr. Schoffstall found no signs of faulting with the home's circuit breaker box, but noted that some of the breakers appeared to have been "tampered with" after the fire. Having found no signs of faulting, he eliminated the breaker box as a possible point of origin or cause. He also examined the furnace, the washer and dryer, the oven/stove unit, the dishwasher, and some of the house's wiring. He found no signs of faulting and eliminated these items as possible causes. He did note fire damage to some of the house's wiring, but observed no evidence that the wiring was a cause or point of origin for the fire. There is no suggestion that the damaged wiring ran through the area of the oval burn pattern.

Mr. Schoffstall examined and sifted through ashes and debris at the point of origin and identified the remains of the PowerChair. Mr. Schoffstall found and preserved as evidence several parts of the PowerChair including wires, battery parts, parts of the seat, the base of the PowerChair, and an item identified possibly as the charger. He examined the wiring of the PowerChair and determined that it showed signs of faulting.

In the written report, Mr. Schoffstall concluded as follows:

In the professional judgment of this investigator, this fire is classified as undetermined. The fire was found to have started in the living room area near the west end of the structure. The only source of ignition in this area was the scooter and with the damage found to the wires of the unit, it could not be ruled out as the source of the fire.

In an affidavit that he prepared later, Mr. Schoffstall further stated:

5. The only items found in the area of origin were the remnants of Plaintiff's scooter, including: the base of the scooter, batteries and possible charger, parts of the seat, and some wiring.

6. That [sic] the remnants of Plaintiff's scooter, identified in Statement No. 5, were the only possible sources of ignition found in the area of origin. All of the items found in the area of origin, identified in Statement No. 5, were preserved for investigation and available to Defendants for inspection.

At the time of the fire, the PowerChair was approximately six months old. Mr. Hickerson and his wife had purchased the PowerChair new from Defendant The Scooter Store-Kansas City LLC. Defendant Pride Mobility Products Corporation had manufactured the PowerChair. Mr. Hickerson stated in a deposition that the space heater identified by David Owens was not attached to a power source at the time of the fire. It is undisputed that the space heater contained no batteries or other independent, on-board power source. Mr. Hickerson stated in an affidavit that he had placed insulation in the crawlspace under his home, observed the floor joists, and observed that no wiring ran through the floor at the alleged area of origin.

Defendants' experts were Francis Burns, a fire investigator, and Thomas J. Bazjek, an electrical engineer. These experts were not afforded an opportunity to examine the fire scene, as Defendants did not receive notice of the fire until after cleanup had occurred at the site. The preserved remains of the PowerChair were available for inspection by Defendants at a firm in Tennessee, Diversified Product Inspections, Inc., and, in fact, Mr. Bazjek participated in a group examination of the remains. Mr. Burns and Mr. Bajzek also examined the photographs of the fire scene taken by Mr. Schoffstall, discovery materials related to the litigation, and official reports from the State Fire Marshal and the local fire department.

Mr. Bazjek stated that when he examined the PowerChair remains, the battery cable was no longer fused to the frame of the chair, as described by the State Fire Marshal. However, there was evidence of electrical activity at the corresponding location of the cable. He also determined that certain pieces of debris collected by Mr. Schoffstall were not part of the PowerChair, but rather, were an electric plate and electric heating element "consistent with the type found in portable space heaters." Mr. Bazjek ultimately concluded that there was "no evidence of a defect or failure in any component [of the] PowerChair that could have caused the fire." He further stated that the damage to the PowerChair was consistent with "attack by an external fire," and he could not rule out two floor lamps, a table lamp, a heating pad, a television, or a VCR (all purportedly in the living room), or the house's wiring as possible sources of the fire.

Expert Francis Burns concluded that nothing from the group examination of the remains of the PowerChair indicated that the PowerChair was a cause rather than a victim of the fire. Like Mr. Bajzek, he identified additional "energized" appliances in the living room, including the two floor lamps, the table lamp, the television and VCR, and the heating pad. In addition, he noted that Mr. Hickerson was a smoker and had been smoking in the house on the night of the fire. There was no suggestion that Mr. Hickerson had been smoking in the area identified as the origin of the fire, and Mr. Hickerson stated in his deposition that he had not smoked in the room of origin on the night of the fire. There was also no suggestion that any appliances or devices other than the PowerChair and the space heater were located in the area of the living room identified as the origin of the fire.

Defendants moved to exclude Mr. Schoffstall's testimony and moved for summary judgment. The district court granted in part the motion to exclude. The court found that Mr. Schoffstall was not qualified as an expert regarding the manufacture or failure of electric scooters, but that he was qualified as an expert regarding examination of fire scenes and identification of points of fire origin. The court ultimately stated that Mr. Schoffstall could not testify that Defendants' PowerChair was the cause of the fire, but could testify "only to his report and observations he made at the scene of the fire."

The court went on to hold that, without testimony regarding the existence of a defect in the PowerChair, Mr. Hickerson could not establish a necessary element of his product liability claim against Defendants. Accordingly, the court granted summary judgment in Defendants' favor on all counts. Mr. Hickerson appeals, alleging that under Missouri law, expert testimony regarding a specific defect is not necessary if circumstantial evidence is sufficiently strong to permit the trier of fact to infer a defect and causation. Under this theory, Mr. Hickerson alleges that Mr. Schoffstall's testimony suffices to place the origin of the fire at the location of the PowerChair and to eliminate other possible causes at that location with a sufficient degree of certainty to permit a jury to infer a fatal product defect. Defendants argue that Plaintiff's expert is unqualified, his conclusions rely upon speculation, and that without a product or manufacturing expert, any inference of a defect in the present case would be unreasonable and impermissibly speculative.

We review the district court's grant of summary judgment de novo, viewing the evidence and testimony in a light most favorable to the non-moving party and drawing all reasonable inferences of fact in that party's favor. Martin v. E-Z Mart Stores, Inc., 464 F.3d 827, 829 (8th Cir. 2006). We review for abuse of discretion the district court's ruling regarding the admissibility of expert testimony under Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993). General Elec. Co. v. Joiner, 522 U.S. 136, 139, 118 S. Ct. 512, 139 L. Ed. 2d 508 (1997).

Defendants oppose the use of Mr. Schoffstall as an expert witness on two grounds. First, they argue he is not qualified as an electrical engineer or manufacturing expert and therefore not qualified to offer an opinion as to the existence of a defect in the PowerChair. Second, they characterize his conclusion that the PowerChair was the only possible source of ignition as too speculative to satisfy the reliability requirements of Daubert. Regarding the first point of contention, Mr. Hickerson does not offer Mr. Schoffstall as an expert in the fields of electrical or mechanical design or engineering or to identify any specific, alleged defect in the PowerChair. In fact, nothing in Mr. Schoffstall's written report or affidavit purports to describe the nature of any alleged defect or suggest an alternate design that might have been superior. Mr. Hickerson offers Mr. Schoffstall only as a fire cause and origin expert and relies entirely upon a res ipsa-type theory of an implied product defect. Because Mr. Hickerson is not attempting to prove a specific defect and Mr. Schoffstall is not being offered as an engineering or product design expert, and does not seek to testify as to these issues, Defendants' argument in this regard is without merit.

Regarding the second point, Mr. Schoffstall has extensive experience as a firefighter and fire investigator, and he is a certified fire investigator. This experience demonstrates that he is qualified in the area of expertise for which he has been designated. In fact, Defendants do not seriously contend that Mr. Schoffstall is unqualified to conduct a fire scene investigation and analysis to determine a point of origin. Rather, Defendants contend that his conclusions regarding causation and origin are unsupported by reliable methodology.

We disagree. The methodology he used to generate his opinion is sound. He examined burn patterns, examined heat, fire, and smoke damage, considered this evidence in light of testimony regarding the fire, and identified a point of origin. He then considered as possible causes of the fire those devices that contained or were connected to a power source and that were located at the identified point of origin. He eliminated as possible sources those devices that were not in the area of origin or that were not connected to a power source and contained no internal power source. We can find nothing unreliable in this accepted and tested methodology. See, e.g., Weisgram v. Marley Co., 169 F.3d 514, 519 (8th Cir. 1999) ("Now, as a qualified expert in fire investigation, Freeman was free to testify—as he did—that the burn and smoke patterns and other physical evidence indicated that, in his opinion, the fire started in the entryway and radiated to the sofa.").

There is no suggestion that the space heater, found near the PowerChair, was plugged in at the time of the fire, and Mr. Hickerson stated in his affidavit that it was not plugged in. Defendants' experts identified other appliances in the living room that may have been connected to power sources at the time of the fire, but there is no evidence that tends to place these other powered appliances specifically in the area of origin. Defendants' experts also emphasized the fact that Mr. Hickerson was a smoker and that he admitted smoking on the night of the fire. As already noted, however, Mr. Hickerson stated that he only smoked in a different room on the night of the fire. Taking the evidence in a light most favorable to the plaintiff, as we must, it does not appear that Mr. Schoffstall failed to eliminate any powered device in the area of origin or any other suggested ignition source as a possible cause. Based on the identification of a point of origin and the elimination of other possible causes, it is permissible for Mr. Schoffstall to testify as to the point of origin and to explain that he inferred through process of elimination that the PowerChair was the cause of the fire.2

It is well-established under Missouri law that juries may infer causation and the existence of product defects based on circumstantial evidence under a res ipsa-type theory such that some product liability claims may be submitted to juries without expert testimony that identifies specific product defects. Fain v. GTE Sylvania, Inc., 652 S.W.2d 163, 165 (Mo. Ct.App. 1983) (holding that, based on the elimination of other possible causes at the location of fire origin, a jury could infer that a television was the cause of a fire and could also infer that a defect existed in the television that caused the fire, because "common experience tells us that some accidents do not ordinarily occur in the absence of a defect"); Henderson v. W.C. Haas Realty Mgmt., Inc., 561 S.W.2d 382, 386 (Mo.Ct.App. 1977) (holding that a jury could infer from circumstantial evidence that wiring in a ceiling had caused a fire even though the wiring was destroyed and not available for inspection because, "[o]f all the things which ran through the floors and walls, the only thing which was capable of starting a fire was the electrical wiring"); Lindsay v. McDonnell Douglas Aircraft Corp., 460 F.2d 631, 635, 637-40 (8th Cir. 1972) (finding federal law regarding strict liability product defect claims under the Death on the High Seas Act, 46 U.S.C. § 761 et seq., to be "congruent" with Missouri law and holding that a jury could infer a product defect in a McDonnell Douglas F4B navy fighter jet based on circumstantial evidence that the jet had burst into flames during one of its maiden flights, even though the jet crashed into the ocean and experts could not examine the jet to identify a specific defect).

To prove a product liability claim by inference from circumstantial evidence without proof of a specific defect, a plaintiff must offer evidence that (1) tends to eliminate other possible causes of the injury or property damage, (2) demonstrates that the product was in the same basic condition at the time of the occurrence as when it left the hands of the defendants, and (3) the injury or damage is of a type that normally would not have occurred in the absence of a defect in the product. Fain, 652 S.W.2d at 165; Lindsay, 460 F.2d at 637-38. That having been said, "[a]n inference need not be justified beyond all doubt and is not precluded by a mere possibility that the contrary may be true." Fain, 652 S.W.2d at 165. Rather, a case may be given to a jury if circumstantial evidence is strong enough to support reasonable inferences necessary to the plaintiff's case free from speculation. The circumstantial evidence need not be so strong as to compel the jury to rule in a plaintiff's favor, and where multiple inferences may be possible, it is for the jury rather than the court to resolve the factual disputes.3

The three evidentiary points set forth above are useful tools to ensure that juries are not allowed to speculate in the process of inferring a product defect. We previously have recognized these limits on the use of the theory of inferred or implied product defects. See Martin, 464 F.3d at 830-31 (holding that where plaintiff had disposable lighters from multiple manufacturers, circumstantial evidence in that case was insufficient to permit a jury to determine which manufacturer's lighter had caused a fire; also holding that circumstantial evidence available in that case was insufficient to allow jury to determine whether fire was caused by product defect, product misuse, or damage to a product subsequent to the manufacturers' relinquishment of control); Weisgram, 169 F.3d at 520-22 (holding expert testimony inadmissible as unreliable and directing district court to enter a defense judgment notwithstanding the verdict where circumstantial evidence was too weak to permit the jury to adopt the plaintiff's theory of causation or to infer the presence of a product defect).

Limitations of this type amount to an assessment of the strength of the circumstantial evidence, and they exist in all cases involving circumstantial evidence, not only in product liability cases. See, e.g., Scheele v. Am. Bakeries Co., 427 S.W.2d 361, 365 (Mo. 1968) (discussing inferences from circumstantial evidence in the context of a claim of simple negligence). In Scheele, the Missouri Supreme Court explained limitations on a jury's use of circumstantial evidence:

Of course, a finding essential to recovery may be proved by circumstantial evidence; but, our appellate courts have said repeatedly that, in civil cases, the shown circumstances must be such that the facts necessary to support the finding may be inferred and reasonably must follow, that the existence of such facts may not depend upon guesswork, conjecture and speculation, and that the evidence should have a tendency to exclude every reasonable conclusion other than the one desired. Although an inference need not be justified beyond all doubt and is not precluded by a mere possibility that the contrary may be true . . ., the law does not countenance the drawing of forced and violent inferences which do not arise from a reasonable interpretation of the facts actually shown.

Id. (quoting Osterhaus v. Gladstone Hotel Corp., 344 S.W.2d 91, 94 (Mo. 1961)).

Having determined the extent of the admissibility of Mr. Schoffstall's testimony, the question that remains is whether, viewed in a light most favorable to the non-moving party, Mr. Schoffstall's admissible testimony taken together with the other evidence is sufficient to permit the jury to infer that the PowerChair contained a product defect that existed when Defendants' possessed the chair and that caused the fire. Under Missouri law, we believe the jury may make such an inference.

Accepting as fact that the area of the oval burn pattern in the living room was the location of the point of origin (as determined separately by Mr. Schoffstall and Mr. Owens), and also taking as fact Mr. Schoffstall's elimination of other possible sources of ignition at that location, we are left with one powered appliance at the point of origin, the PowerChair. "[C]ommon experience tells us that some accidents do not ordinarily occur in the absence of a defect," Fain, 652 S.W.2d at 165, and the spontaneous ignition of a battery-powered, rechargeable wheelchair would be such an accident. To the extent Defendants argue Mr. Schoffstall failed to eliminate certain other possible sources of ignition, we believe the Defendants have created issues of fact for resolution at trial. At trial, Defendants may succeed at showing that the other appliances in the living room were powered and positioned at locations such that they were possible sources of ignition relative to the point of origin. Taking the current record in a light most favorable to Mr. Hickerson, however, it does not appear that any of these possibly powered appliances were in the area of origin, and we find that a jury could reasonably infer a PowerChair defect as the cause without resort to "guesswork, conjecture and speculation." Scheele, 427 S.W.2d at 365. Where the evidence shows a point of origin and one source of ignition at that point, a jury need not resort to speculation to infer the cause of the fire and the presence of a product defect.

Finally, we note that Defendants rely on a number of cases in which expert testimony was excluded or a court found an inference of a product defect too speculative. These cases are distinguishable on their facts based on the strength of the inferences that reasonably could be drawn from the circumstantial evidence in each case. In Weisgram, 169 F.3d at 519, for example, we applied North Dakota law and held that a jury could not infer a product defect in a fire case. There, expert testimony was excluded and too much speculation was required to make the leap from the circumstantial evidence to the conclusion that a product defect existed and caused the fire. There was strong evidence to support a theory of causation different from the plaintiff's theory, and the plaintiff's theory of causation was speculative and did not enjoy the support of a reliable expert's identification of a point of origin. Rather, it depended on a detailed, problematic and speculative explanation of how a fire started at one location and moved throughout the house along a specific route. We similarly find the other authority cited by Defendants to be distinguishable on its facts and not to stand for the general proposition that circumstantial evidence is inherently infirm or incapable of proving the existence of a product defect. See, e.g., Winkler v. Robinett, 913 S.W.2d 817, 821 (Mo.Ct.App. 1995) ("[T]o make a submissible case on the issue of causation, such circumstantial evidence must fairly warrant the conclusion that the fire did not originate from some other cause."); Hills v. Ozark Border Elec. Coop., 710 S.W.2d 338, 341 (Mo.Ct.App. 1986) (reversing a jury verdict against an electric company in a case involving a fire allegedly of electrical origin, and stating, "Plaintiffs made no attempt to present evidence that might have eliminated other conditions that could have caused the fire").

Because we hold Mr. Schoffstall's testimony admissible to the extent discussed herein, and because Mr. Hickerson has presented sufficient evidence to support a plaintiff's verdict under the theory of Fain, we reverse the grant of summary judgment and remand for further proceedings not inconsistent with this opinion.