dissenting.
I dissent. I would deny Marley’s motions for a new trial and judgment as a matter of law. The jury verdict has adequate support from properly admitted expert testimony. The plaintiffs’ theory of the case relied on the testimony of two properly qualified fire investigators and a properly qualified metallurgist. This testimony provided evidence that the fire originated with the heater and that defects in the thermostat contacts and the placement of the high limit control contributed to the fire. The experts arrived at their conclusions by personally inspecting the evidence from the fire scene using accepted investigative techniques. Although each expert may have testified on matters outside his particular area of expertise, these matters went to the weight and not the admissibility of the testimony. Moreover, the experts’ experience lent support to this testimony.
Fire cases differ from most accident cases because fires tend to destroy evidence of causation. As a result, theories about the cause of fires inevitably rest on circumstantial evidence. Arson and insurance cases, as well as product liability cases like this one, require expert evaluations to determine the cause of fires. The courts traditionally permit qualified fire investigators to express opinions on the cause of fires.
For example, in Talkington v. Atria Reclamelucifers Fabrieken BV (Cricket BV), 152 F.3d 254 (4th Cir.1998), a product liability case, the court affirmed a jury verdict for the plaintiffs which rested on the testimony of two fire investigators. The fire investigators discovered the body of a child in close proximity to a cigarette lighter and a sofa. See id. at 260. They also conducted a test in which they compared the sofa to a sofa ignited with a cigarette and an open flame. See id. at 265. At trial, the investigators testified that the fire occurred because the child ignited the sofa with the cigarette lighter. See id. at 264-65. The court upheld the admission of this testimony. The experts qualified as fire investigators and gave reasoned explanations for rejecting alternative explanations, including the defense theory that a cigarette started the fire. See id. at 264-66.
Similarly, in Marshall v. Humble Oil & Refining Co., 459 F.2d 355 (8th Cir.1972), this court affirmed a jury verdict finding the defendants liable for an explosion in a poorly ventilated gas station storeroom. The plaintiffs contended that the explosion occurred when a spark from an air compressor switch ignited accumulated gas vapors. See id. at 357-58. This theory rested on the discovery *523of soot in the storeroom after the fire, testimony that the storeroom door was found closed immediately after the explosion, and the fact of the explosion itself. See id. at 358-60. This court held that the plaintiffs presented sufficient evidence of causation even though the compressor switch itself showed no fire damage. See id. at 359-60.
In this case, the jury was required to determine the cause of the fire that killed Bonnie Weisgram in her home. Plaintiffs Weisgram and State Farm argued that a defect in the baseboard heater caused the fire. The court must determine whether the plaintiffs presented reliable evidence of them theory by experts who possessed the necessary qualifications.9 Under Rule 702, a witness can testify as an expert if his knowledge qualifies him to offer an opinion that will assist the trier of fact in arriving at the truth. See Fed.R.Evid. 702; Pelster v. Ray, 987 F.2d 514, 526 (8th Cir.1993). Rule 702 reflects an attempt to liberalize the rules governing the admission of expert testimony. See Jenson v. Eveleth Taconite Co., 130 F.3d 1287, 1298 (8th Cir.1997). The rule “ ‘is one of admissibility rather than exclusion.’ ” Id. (quoting Arcoren v. United States, 929 F.2d 1235, 1239 (8th Cir.1991)). We review district court decisions on the qualifications and reliability of experts for an abuse of discretion. See General Elec. Co. v. Joiner, 522 U.S. 136, 118 S.Ct. 512, 517, 139 L.Ed.2d 508 (1997). With this standard in mind, I review the district court’s decision to admit the testimony of the plaintiffs’ experts.
The plaintiffs first called Captain Dan Freeman of the Fargo Fire Department. Freeman was the first person to arrive at the fire scene. Freeman testified as a fact witness and to some extent as an expert. Freeman testified that the fire originated in the entryway. Upon arrival, he saw a fire burning on the south side of the entryway. He saw no other fire in the house. After searching for survivors and removing Bonnie Weis-gram’s body, Freeman began to investigate the cause of the fire. He first considered the sofa as a source of the fire, but discarded that possibility when he inspected the sofa and the surrounding area. The sofa exhibited signs of smouldering combustion but not flaming combustion. Freeman saw no charring on the floor beneath the sofa or damage to the ceiling above the sofa. Furthermore, a piano located within P/¿ feet of the sofa remained in good condition. As Freeman continued to investigate, he determined that the fire originated in the entryway and spread to the sofa. The sofa suffered the most damage on its backside, which was exposed to the entryway, and there was charring on the woodwork inches from the backside of the sofa. Freeman testified that a fire strong enough to char the woodwork could have ignited the sofa. Freeman also saw evidence of flaming combustion in the entryway. A two-foot long hole had been burned through the floor underneath the heater and the ceiling above the heater had suffered significant damage.
The plaintiffs next called Ralph Dolence, a professional fire investigator, to testify as an expert witness. Dolence listened to Freeman’s description of the fire scene and inspected the heater and photographs of the fire scene. Dolence also inspected and tested an exemplar heater from a nearby house. Dolence testified that the fire originated with the heater. In his opinion, a throw rug pushed up against the heater caused heat to accumulate in and around the heater.10 The excess heat caused the glue on the vinyl flooring beneath the heater to vaporize and ignite. The fire spread to the walls and ceiling in the entryway and eventually to the sofa in the living room. To support his *524conclusions, Dolence testified that burn patterns on the heater indicated that it had been partially covered during the fire and that it had been exposed to extreme heat underneath. Dolence also testified that asphalt-based glue used in vinyl flooring has an ignition temperature of approximately 450°. Dolence’s testing of the exemplar heater demonstrated that the heater could reach a temperature of 750°.
As their third expert, the plaintiffs called Sandy Lazarowicz, a metallurgist. Lazarow-icz had performed failure analysis on the thermostat contacts and had inspected the high limit control. Lazarowicz’s testimony provided evidence of defects in the thermostat contacts and the plácement of the high limit control that resulted in their failure to shut down the heater at the time of the fire. Lazarowicz testified that the thermostat contacts were welded together at the time of the fire. He testified that serrations on the contact surfaces and mismatching of the contacts caused electrical arcing and material transfer, which in turn caused the contacts to weld.11 Lazarowicz also testified that the high limit control did not detect the heat build-up before the fire because it was placed behind a deflector shield.
Freeman and Dolence possessed ample qualifications to testify as experts on the cause of the fire. Freeman was a certified fire investigator and had investigated over 100 fires in addition to completing 400 hours of training in fire and arson investigation. Dolence had been investigating fires for over fifteen years and had directed fire investigation units in two municipalities. While neither Freeman nor Dolence had worked with the type of heater found in the Weisgram home, they both had experience investigating fires caused by heaters and other electrical appliances. Thus, in my view they were qualified to offer opinions on whether the heater caused the Weisgram fire. Dolence possesses particularly relevant qualifications. He is a master electrician who has investigated a number of electrical fires and has written a book on the subject. He was well-qualified to give his opinion that a malfunction in the electrical components of the heater caused the fire. The district court did not abuse its discretion in permitting Freeman and Dolence to testify.
Lazarowicz also possessed sufficient qualifications to testify about defects in the thermostat contacts and high limit control. While Lazarowicz may not have been a specialist in heaters per se, he was a metallurgical engineer and had inspected defective electrical contacts for years as part of his employment. These qualifications made him competent to testify about the post-fire condition of the thermostat contacts and high limit control, which were in essence electrical contacts. The district court did not abuse its discretion in permitting him to testify.
Moreover, even if the plaintiffs’ experts lacked specific expertise in heaters, this matter went to the weight and not the admissibility of their testimony. See Arkwright Mut. Ins. Co. v. Gwinner Oil, Inc., 125 F.3d 1176, 1183 (8th Cir.1997) (“ ‘[T]he factual basis of an expert opinion goes to the credibility of the testimony, not the admissibility, and it is up to the opposing party to examine the factual basis for the opinion in cross-examination.’ ”) (quoting Hose v. Chicago Northwestern Transp. Co., 70 F.3d 968, 974 (8th Cir.1995)); Sylla-Sawdon v. Uniroyal Goodrich Tire Co., 47 F.3d 277, 283 (8th Cir.1995) (“ ‘Once the trial court has determined that a witness is competent to testify as an expert, challenges to the expert’s skill or knowledge go to the weight to be accorded the expert testimony rather than to its admissibility.’ ”) (quoting Fox v. Dannenberg, 906 F.2d 1253, 1256 (8th Cir.1990)); Williams v. Pro-Tec, Inc., 908 F.2d 345, 348 (8th Cir.1990) (holding that an expert’s self-acknowledged lack of medical expertise went to the weight rather than to the admissibility of his testimony).
The majority opinion views the testimony of the plaintiffs’ experts as unreliable. Certainly expert testimony must be reliable. *525See Maj. Op. at 517, 518. Reliability in the context of this case and under Rule 702 relates to the qualifications of the experts, the foundation for their opinions, and their helpfulness to the jury. An expert’s opinion should be excluded if it is fundamentally unsupported so that it can offer no assistance to the jury. See Arkwright, 125 F.3d at 1183. The plaintiffs’ experts based their opinions on personal inspection of the evidence and applying knowledge acquired from training and experience. Their opinions rested on factual support. The district court did not abuse its discretion in determining that their opinions were reliable and could be considered by the jury.
Lazarowiez personally inspected the thermostat contacts and the high limit control. He saw that the surface of the contacts was serrated and that the contacts did not properly connect. He concluded in light of his knowledge and experience that these features contributed to the fire by causing the contacts to weld at the time of the fire. He saw that the high limit control was located behind a deflector shield. He determined, again in light of relevant training and experience, that the high limit control had failed to shut down the heater because the deflector shield prevented it from detecting the heat build-up. When determining whether such evidence is sufficient to support the jury’s verdict, the evidence must be viewed in the light most favorable to the plaintiffs, giving them the benefit of all inferences. See Stockmen’s Livestock Market, Inc. v. Norwest Bank of Sioux City, 135 F.3d 1236, 1240-41 (8th Cir.1998). Lazarowicz’s observations of the contact features and the placement of the high limit control provided enough factual support for his opinions to justify submitting the issue of a product defect to the jury.
Freeman and Dolence personally inspected the evidence from the fire scene. Their opinions about the cause of the fire did not lack factual support. The condition of the sofa and the charring in the entryway supported Freeman’s and Dolence’s determination that the fire originated in the entryway. The charring in the entryway and the hole in the floor supported their determination that the fire originated near the heater. Freeman discovered a partially burned throw rug near the hole. Three frequent visitors to the Weisgram home testified that a throw rug in the entryway was frequently pushed up against the heater. These facts supported the determination that the throw rug was pushed up against the heater and caused heat to build up in the heater. Freeman and Dolence knew that the throw rug and the glue used on vinyl flooring had flashpoints within the range of temperatures produced by baseboard heaters. This knowledge supported their determination that the heat build-up ignited the throw rug and the glue. Dolence knew the heater could not overheat if the thermostat and the high limit control functioned properly. This justified his determination that those parts failed to shut down the heater at the time of the fire.
The plaintiffs’ theory of causation depended primarily on circumstantial evidence. However, as discussed earlier, theories of fire causation must rely heavily on circumstantial evidence. There was enough evidence here for a reasonable jury to conclude that the fire originated as Freeman and Do-lence testified. The fire investigators saw no other source of heat in the entryway other than the heater and no combustibles were found other than the throw rug and the glue.12 The district court did not err in submitting the issue of causation to the jury.
The majority errs in a second regard. See Maj. Op. at 517 n. 2. If the district court erred in admitting the testimony of the plaintiffs’ experts, the relief to be awarded is a new trial, not judgment as a matter of law. In Midcontinent Broadcasting Co. v. North Central Airlines, Inc., 471 F.2d 357 (8th Cir.1973), this court held that a new trial is the proper remedy for an error in the admission of expert testimony. See id. at 360. See also Schudel v. General Elec. Co., 120 F.3d 991, 995 (9th Cir.1997); Jackson v. Pleasant Grove Health Care Ctr., 980 F.2d 692, 695-96 (11th Cir.1993); Douglass v. Eaton Corp., 956 F.2d 1339, 1343-44 (6th Cir.1992). Neither Wright v. Willamette Industries, Inc., 91 F.3d 1105 (8th Cir.1996) nor Peitzmeier v. Hennessy Industries, Inc., 97 *526F.3d 293 (8th Cir.1996) has overruled Mid-continent Broadcasting. The new trial question was not before the court or discussed in either of those two cases.
In sum, the parties presented two reasonable theories about the cause of the fire. The experienced and able trial judge admitted the testimony of the plaintiffs’ experts. He also permitted expert witnesses to testify to Marley’s theory of the case. A North Dakota jury evaluated the evidence and determined that the plaintiffs should recover damages. As a court, we are only called upon to determine whether the district court abused its discretion in permitting the experts to testify. Had the jury rendered a verdict for Marley, we would not be in a position to say that the district court abused its discretion in admitting the testimony of the defense experts. This controversy represents a typical case to be decided by a jury. This court ought not overturn both the trial judge and the jury. Accordingly, I dissent.
. In answering that question in the negative, the majority cites Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) for support but also recognizes that the Daubert analysis for the admission of expert scientific testimony may not apply to non-scientific testimony. Maj. Op. at 515 n. 3. As the majority has noted, Carmichael v. Samyang Tire, Inc., 131 F.3d 1433 (11th Cir.1997), cert. granted sub nom., Kumho Tire Co. v. Carmichael,-U.S.-, 118 S.Ct. 2339, 141 L.Ed.2d 711 (1998), now before the Supreme Court, raises the issue whether the factors enunciated in Daubert for the admissibility of scientific evidence also apply to non-scientific evidence.
. Three frequent visitors to the Weisgram home testified that an entryway throw rug would frequently get pushed up against the heater by the movement of the entryway door.
. Defense experts Michael Phy, Richard Moore, and Vincent Acampora testified that the thermostat contacts showed evidence of electrical arcing and material transfer. Dolence testified that he observed serrations on tire surface of the contacts, something he had never seen before even though he had inspected thousands of contacts. Marley’s own witness, Richard Moore, likewise testified that he had seen over 1,000 contacts, none with serrations.
. Freeman had the hole beneath the heater checked for accelerants. None were found.