Nathan Bush v. Coleman Powermate, Inc. and Tecumseh Products Company

TEXAS COURT OF APPEALS, THIRD DISTRICT, AT AUSTIN NO. 03-04-00196-CV Nathan Bush, Appellant v. Coleman Powermate, Inc. and Tecumseh Products Company, Appellees FROM THE DISTRICT COURT OF SAN SABA COUNTY, 33RD JUDICIAL DISTRICT NO. 7871, HONORABLE GUILFORD L. JONES, JUDGE PRESIDING M E M O R A N D U M O P I N I O N Nathan Bush was injured by a fire that erupted while he was refueling a generator manufactured by Coleman Powermate, Inc., and powered by a Tecumseh Products Company motor. He asserted negligence and products liability claims against both Coleman and Tecumseh. The district court granted appellees' motions to exclude four of Bush's expert witnesses and granted summary judgment in favor of appellees on all claims. Bush appeals the exclusion of his expert witnesses and the summary judgment against his marketing defect claims under negligence and strict products liability theories. We affirm the judgment. BACKGROUND This background section is derived from the depositions and other documents filed in this case, as well as live testimony taken at the Robinson hearing. See E.I. du Pont de Nemours v. Robinson, 923 S.W.2d 549, 557 (Tex. 1995). At the time of the fire, Bush was constructing his home with tools powered by a Coleman generator equipped with a Tecumseh gasoline engine. While he was using a skill saw, the generator ran out of gas. The generator was just outside the back door. Bush checked the oil and removed the gas cap, then walked around the house to his truck to get a gas can. He also picked up a gallon jug of water, took a drink, and poured water over himself to cool down. Bush testified that he did not know how much time elapsed after the generator ran out of gas before he attempted to refuel it. He and Tecumseh's attorney had this exchange: Q. How much time elapsed from when the generator ran out of gas and you actually started pouring fuel in it? In other words, was that a 30-second period of time, walk to the truck and back-- A. Oh, no. Q. --or was it five minutes because you were messing around doing something else? A. Yeah. It was--I don't have an exact time. Long enough for me to check the oil and walk to my truck and get a drink of water. I mean, I didn't--I was in no hurry. It was pretty hot. Bush returned to the generator, rechecked the oil, replaced the oil cap, and began refueling. He testified that he did not spill any gasoline. As he bent over to monitor the level of gas, Bush heard a "woof" noise and dropped the gas can, splashing gasoline on himself. His face and neck caught fire. With some difficulty, he extinguished the flames on his body and sought help. The generator Bush used was purchased at a pawn shop by Bush's stepfather for Bush's brother. It did not come to Bush in a manufacturer's box, and he was unsure whether it was accompanied by an owner's manual. Bush previously had used this generator without mechanical problems or fire, and had used other generators while in the military. In June 2000, Bush sued Coleman and Tecumseh, alleging design and marketing defect claims under strict products liability and negligence theories. Appellees filed motions to exclude testimony by Bush's experts Lori Hasselbring, Joe Fowler, Waymon Johnston, Judd Clayton, and Paul Beaver. Hasselbring and Fowler were to testify about the auto-ignition of gasoline. Johnston was to testify about the need for a warning on the generator to wait two minutes after shutting it down before refueling. Clayton was to testify about the likely source of ignition. Beaver, the fire marshal, was to testify about his investigation of the fire. Appellees alleged that these witnesses were not qualified to give opinions as to design defects, dangers associated with refueling a hot generator, or causation. Appellees also claimed that Hasselbring and Fowler's opinions were based on flawed reasoning and methodology and that the underlying facts did not provide an adequate basis for their opinions. Appellees filed traditional and no-evidence motions for summary judgment as to all claims. These motions challenged the evidence Bush had adduced on design and marketing defects and his ability to prove causation. Bush defended the negligence and marketing defect claims; he expressly did not contest judgment against his design and manufacturing defect claims. At the Robinson hearing, held after the agreed discovery deadline passed, appellees' witnesses criticized Bush's expert witnesses' reliance on estimates of the closed-container auto-ignition temperature of gasoline in concluding that the generator had been hot enough to ignite gasoline vapors when Bush refueled it. Appellees' expert Dwight Pfennig (1) testified that the facts of this case required application of the auto-ignition temperature for open-air conditions, which is about 360 degrees Fahrenheit (360ºF) higher than the closed-container temperature. The district court stated at the hearing that it would exclude testimony from four of Bush's witnesses. (2) Subsequently, the district court sent the parties an e-mail "order/ruling of the court" that was "to be reduced to a formal order." The court found that the "fit" of the science to the facts of the case was fatally flawed because Bush's experts' failed to consider the open-air auto-ignition temperature of gasoline. The court also noted that Bush's experts had failed to get gasoline to auto-ignite during their tests. The court further found no evidence of causation in the record as a matter of law. It concluded that failing to attach a warning to the generator recommending a two-minute waiting period between turning the generator off and refueling, as Bush advocated, could not have caused the fire because (1) Bush "clearly waited at least two minutes" before refueling the generator, and (2) regardless of the wait, "open-air auto-ignition of this type of engine/generator has been proven to be all but impossible even without a cooling off period." The court granted summary judgment to appellees on all claims without specifying whether it relied on traditional or no-evidence grounds for summary judgment. The court subsequently signed separate orders granting summary judgment to Coleman and Tecumseh on (1) the contested design and manufacturing defect claims, and (2) all remaining claims. The court did not elaborate on the basis for its rulings. Bush filed a motion for new trial and to reconsider the exclusion of his experts. Bush also asked the court to consider reports submitted by two expert witnesses on December 31, 2003, after the court's rulings. This motion was overruled by operation of law, and this appeal ensued. DISCUSSION Bush appeals the district court's exclusion of his expert witnesses and granting of the summary judgment as to his marketing defect claims. Exclusion of experts We review the district court's exclusion of expert testimony for an abuse of discretion. See Robinson, 923 S.W.2d at 558. We determine whether the trial court acted arbitrarily, unreasonably, or without reference to any guiding rules or principles. See Downer v. Aquamarine Operators, Inc., 701 S.W.2d 238, 241-42 (Tex. 1985). The test is not whether, "in the opinion of the reviewing court, the facts present an appropriate case for the trial court's action." Id. at 241. We cannot conclude that a trial court abused its discretion merely because we would have ruled differently. See id. at 242. The trial court enjoys wide latitude in determining the admissibility of expert testimony. Wolfson v. Bic Corp., 95 S.W.3d 527, 532 (Tex. App.--Houston [1st Dist.] 2002, pet. denied). Texas Rule of Evidence 702 permits a witness qualified as an expert by knowledge, skill, experience, training, or education to testify on scientific, technical, or other specialized subjects if the testimony would assist the trier of fact in understanding the evidence or determining a fact issue. Tex. R. Evid. 702. The party offering the expert's testimony bears the burden to prove that the witness is qualified under Rule 702. Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 718-19 (Tex. 1998) (trial courts must "ensure that those who purport to be experts truly have expertise concerning the actual subject about which they are offering an opinion"). In addition to showing that the witness is an expert, the proponent must show that the expert's testimony is relevant to the issues in the case and is based upon a reliable foundation. Robinson, 923 S.W.2d at 556; see Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 589 (1993). To be relevant, the proposed expert testimony must be "sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute." Robinson, 923 S.W.2d at 556. Evidence that bears no relationship to the issues of the case is irrelevant and does not satisfy Rule 702's requirement that the testimony be of assistance to the jury. Id. To be reliable, the expert's method must be grounded in the methods and procedures of science and the opinion must be more than subjective belief or unsupported speculation. See id. at 557. If an expert relies on unreliable foundational data, any opinion drawn from that data is unreliable. Merrell Dow Pharms., Inc. v. Havner, 953 S.W.2d 706, 714 (Tex. 1997); Union Carbide Corp. v. Mayfield, 66 S.W.3d 354, 362 (Tex. App.—Corpus Christi 2001, pet. denied). Even when based on sound data, an expert's testimony is unreliable if the expert's methodology is flawed. Havner, 953 S.W.2d at 714. The trial court must decide only whether the analysis used to reach the conclusions is reliable, not whether the expert's conclusions are correct. Perez v. Blue Cross Blue Shield of Tex., Inc., 127 S.W.3d 826, 830 (Tex. App.—Austin 2003, pet. denied). The Robinson opinion provides a non-exclusive list of factors a trial court may consider in making the threshold admissibility determination under Rule 702; these include: (1) the extent to which the expert's theory has been or can be tested, (2) the extent to which the technique relies upon the subjective interpretation of the expert, (3) whether the theory has been subjected to peer review and/or other publication, (4) the technique's potential rate of error, (5) whether the underlying theory or technique has been generally accepted as valid by the relevant scientific community, and (6) the non-judicial uses which have been made of the theory or technique. Robinson, 923 S.W.2d at 557; see Volkswagen of Am., Inc. v. Ramirez, No. 02-0557, 2004 WL 3019227, at *4 (Tex. 2004). We will begin by considering together the exclusion of testimony from Hasselbring and Fowler because they filed a joint report and because their report underlies the other experts' testimony. We will then separately consider the exclusion of testimony from Johnston and Clayton. Hasselbring and Fowler Hasselbring and Fowler are both licensed professional engineers. Hasselbring holds a bachelor's degree in paper science and engineering and master's and doctoral degrees in chemical engineering. She has many years of experience in the petrochemical industry with an emphasis on plastics production, processing, and commercial applications. She has performed failure analysis and accident documentation and investigation, particularly in the chemical and petrochemical industries. Fowler holds a bachelor's, master's, and doctoral degrees in mechanical engineering. Fowler has consulted with more than two hundred companies regarding the mechanical design of process plant equipment, development of pipeline design criteria and operating procedures, fire safety of mechanical and consumer product equipment, fatigue design, finite element analysis, strain gauge testing, drilling and production equipment design and operation, failure analysis, and general mechanical engineering testing and design. In their joint report, Hasselbring and Fowler stated that, in order for gasoline to auto-ignite during refueling of a generator, gasoline vapors must contact some part of the generator that is heated at or above the closed-container auto-ignition temperature of gasoline, which they identified as being between 635ºF and 853ºF. Variables they identified that affect auto-ignition include vapor concentration, ambient temperature, wind speed, winter versus summer blends of gasoline, regular unleaded gasoline versus premium unleaded gasoline, gasoline additives, amount of gasoline poured, elevation of the can, whether gasoline has been spilled, and the exact sequence of events leading to a particular incident. Hasselbring and Fowler concluded that it is difficult to determine specific combinations of these variables that may lead to the auto-ignition of gasoline during refueling. Hasselbring and Fowler conducted a series of tests to determine if Bush's generator had reached the closed-container auto-ignition temperature of gasoline. They placed an exemplar Coleman generator in the corner of a firebox with the firebox doors left open. They measured surface temperature near the muffler exhaust, which they found was the hottest contact surface on the generator. Hasselbring and Fowler ran two different series of tests: "load tests" measuring the temperature of the generator when used to power a device, and "fire tests" in which they spilled gasoline on the generator. During the load tests, the generator rarely reached the minimum temperature threshold Hasselbring and Folwer identified for the closed-container auto-ignition of gasoline. The muffler temperature stayed near 600ºF with no load (594ºF in one test, 603ºF in the second), and fluctuated between 378ºF to 644ºF with loads. Within two minutes after the generator was shut off, the muffler temperature dropped to 441ºF. During the fire tests, Hasselbring and Fowler were not able to get gasoline to auto-ignite. Using regular gasoline with no load on the generator, the muffler reached a constant temperature of 558ºF and 585ºF in two tests. When they spilled 20 milliliters of gasoline onto the muffler, no ignition occurred; Bush's experts conceded that they were not surprised by this development because the muffler temperature was below the 635ºF temperature minimum they had identified for closed-container auto-ignition of gasoline. Nor did ignition occur when they ran the generator on premium unleaded gasoline, which raised the generator's maximum temperature. The muffler temperature reached 693ºF and 612ºF with no load, and 734ºF when a 150-watt load was added to the generator. (3) Within two minutes after the generator was shut off, the muffler temperature dropped to 495ºF to 500ºF. Once again, gasoline spilled on the motor did not ignite during any of the tests, even when the temperature was within the auto-ignition range for closed containers, although white smoke twice emanated from the rotor and from underneath the bottom of the generator. Hasselbring and Fowler offered conclusions that (1) there were insufficient warnings in operating manuals and on the generator and engine regarding possible auto-ignition of gasoline during refueling and contact with hot surfaces, and (2) both Coleman and Tecumseh were negligent in not testing the generator and engine to determine the possibility of auto-ignition of gasoline during refueling. We conclude that the court acted reasonably in excluding these witnesses because they were not shown to be experts in the relevant field and because their opinions do not satisfy Robinson. Although having expertise in other areas, Hasselbring had not performed any work on a generator before this case. She testified that she had no expertise or specialized knowledge with respect to internal-combustion engines, generators, product warnings or the origins and causes of fire. Fowler admitted that he had no expertise or specialized knowledge with respect to product warnings. Nor did Hasselbring and Fowler satisfy the first Robinson requirement that their theories on causation have been or can be proven through other scientifically valid testing. To the contrary, Hasselbring and Fowler failed to start a fire by dripping gasoline on a hot generator when testing their theory. (4) The conclusions drawn by Hasselbring and Fowler also do not satisfy the second Robinson factor because they are based on subjective interpretation, as the experts produced no objective evidence--and only unsuccessful tests--to support their theory of causation. Nor did they show that their theory is supported by other tests or studies, has been subjected to peer review or publication, is generally accepted as valid by the relevant scientific community, or has been used in non-judicial settings. See Volkswagen, 2004 WL 3019227, at *5. Under these circumstances, the potential rate of error with respect to the conclusions and opinions of Hasselbring and Fowler is high, and the gap between the theory and the facts of this case is too large. They were unable to obtain ignition even when pouring gasoline on the hot generator immediately after it stopped running (in contrast to the delay of uncertain duration in this case). The court could have concluded that their use of the auto-ignition temperature of gasoline for closed-containers was improper given that the generator was outside the back door in this case; use of the open-air temperature would require the generator to reach a temperature far above any achieved in Bush's expert's tests. There is a gap between the facts of this case and Hasselbring and Fowler's theory and tests because the latter are predicated on conditions that Bush himself denied existed. The district court did not abuse its discretion by excluding the testimony of Hasselbring and Fowler as unreliable. See Volkswagen, 2004 WL 3019227, at *5. Johnston Johnston, a licensed professional engineer and certified safety professional, holds a bachelor's degree in mechanical engineering and a master's and doctoral degree in industrial engineering. He taught human factors, product safety, and industrial safety engineering for 24 years at Texas A&M University. He has over ten years of industrial experience and has been involved in safety engineering and accident prevention activities in the military and the petroleum and aerospace industries. Johnston concluded that Coleman and Tecumseh knew or should have known that: (1) individuals occasionally would spill gasoline while refueling hot generator engines, (2) a fire could occur as a result, (3) serious injuries could occur from such a fire, (4) the warnings in their owner's manuals were inadequate with respect to these dangers, and (5) the generator and engine were defective and unreasonably dangerous with respect to the dangers associated with refueling a hot engine. He testified, however, that "I'm not giving any opinions about causation." In forming his opinion, Johnston relied on his review of materials furnished to him (including the report by Hasselbring and Fowler), his training, and his years of experience; he did not conduct any tests himself to determine what specific dangers were possible. Johnston testified that the generator should have borne a warning to wait two minutes after the generator stopped before refueling. He said that a two-minute warning is not perfect, but would have been adequate. Johnston explained that he used a two-minute period based on Hasselbring's data on how fast the engine cooled. He added that he had assumed that the two-minute period must have been based on test results because later models of engines manufactured by Tecumseh and others carried warnings that users should allow the generator to cool two minutes before refueling. Johnston admitted that he had only the data in Hasselbring's report and deposition and the two-minute warning on comparable engines. The district court did not abuse its discretion by finding Johnston's testimony unreliable. Johnston depended on Hasselbring and Fowler's report, which we have determined was properly excluded. He performed no testing or analysis to determine the scientific significance of the two-minute period. Johnston failed to satisfy the Robinson factors concerning objective, external support for his theory. See Robinson, 923 S.W.2d at 557. He also relied on underlying facts squarely rejected by Bush's uncontroverted testimony that he did not spill gasoline and that he had waited before refueling the generator. Because there is no evidence regarding the scientific significance of the two-minute period for the warning, the court did not abuse its discretion by excluding Johnston's opinion regarding its efficacy. See Volkswagen, 2004 WL 3019227, at *5. Clayton Clayton, who has a bachelor of science degree in electrical engineering, claims expertise regarding electrical product failures, electrocutions, fires, and explosions. Clayton was hired by Bush to evaluate the sources of ignition and determine which was the most probable. Clayton said that he has never been involved in a case or testing in which a fire was caused by a generator that was not running at the time of ignition. He did not perform any tests based on the facts in this case and did not review other testing in this case. Clayton said the ignition source was most probably the heat from the generator's exhaust system. In his report, he excluded all potential sources of ignition of the gasoline vapors except for the generator. He concluded that a defective condition of the generator, which allowed for ignition of gasoline, was a direct producing cause of this incident and Bush's injuries. Clayton's testimony is predicated on underlying facts contrary to Bush's testimony and the evidence of the temperature the generator reached in testing. Clayton stated that he was hired only to address the cause of the fire in terms of what available source of ignition existed. He conceded that a generator in the off position would have to be excluded as a potential ignition source if its surface temperature was lower than the auto-ignition temperature. With the exclusion of Hasselbring and Fowler's evidence, there is no evidence that the generator reached any recognized, applicable auto-ignition temperature; further, auto-ignition never occurred in any of their experiments. The court did not abuse its discretion by deciding that Clayton's testimony regarding a hypothetical generator that did reach the auto-ignition temperature would not help the jury. We conclude that the district court did not abuse its discretion by determining that Bush failed to show the relevance and reliability of these experts' testimony. Summary judgment On appeal, only Bush's claims for strict products liability marketing defect, negligence, and gross negligence remain. Bush argues (1) that Coleman and Tecumseh failed to present evidence to rebut the presumption that Bush would have heeded a warning had one been provided, (2) that the district court committed error by making a factual conclusion that Bush waited in excess of two minutes before refueling the generator, and (3) that the district court ignored evidence from Tecumseh and Coleman personnel as to the necessity of the "cool-down" period before refueling. Rebuttable presumption Bush contends that the district court erred by ignoring the rebuttable presumption regarding product warnings discussed in Magro v. Ragsdale Brothers, Inc., 721 S.W.2d 832 (Tex. 1987). Once a plaintiff proves that the lack of adequate warnings or instructions rendered a product unreasonably dangerous, a rebuttable presumption arises that the user would have read and heeded such warnings and instructions. Id. at 834. Bush, however, offered no evidence that a cool-down period is necessary before refueling or that the lack of a warning rendered the generator and engine unreasonably dangerous. Without such evidence, the rebuttable presumption of Magro did not arise. Factual determination Bush also alleges that the district court improperly made a factual determination regarding how much time passed before Bush refueled the generator. In its e-mail "order" granting summary judgment, the district court stated that Bush clearly waited at least two minutes before refueling and that, regardless of the wait, auto-ignition of gasoline on this occasion was proven to be all but impossible, even without a cooling-off period. The finding in the e-mail "order" is not binding. It is not the formal order granting summary judgment to which we must look for the court's reasons for ruling. See Sharpe v. Roman Catholic Diocese of Dallas, 97 S.W.3d 791, 796 (Tex. App.--Dallas 2003, pet. denied); Strather v. Dolgencorp of Tex., Inc., 96 S.W.3d 420, 426 (Tex. App.—Texarkana 2002, no pet.); see also Cherokee Water Co. v. Gregg County Appraisal Dist., 801 S.W.2d 872, 878 (Tex. 1990). Further, we cannot consider any finding of fact made in a summary-judgment context. See IKB Indus. (Nigeria) Ltd. v. Pro-Line Corp., 938 S.W.2d 440, 441 (Tex. 1997). Merits of the summary judgment As the district court's order did not specify whether it granted summary judgment on traditional or no-evidence grounds, we must affirm summary judgment if any of the summary judgment grounds are meritorious. Provident Life & Accident Ins. Co. v. Knott, 128 S.W.3d 211, 216 (Tex. 2003). We review the district court's grant of summary judgment de novo. Joe v. Two Thirty Nine Joint Venture, 145 S.W.3d 150, 156 (Tex. 2004). The standard for reviewing a motion for traditional summary judgment is well established: (1) the movant for summary judgment has the burden of showing that there is no genuine issue of material fact and that it is entitled to judgment as a matter of law, (2) in deciding whether there is a disputed material fact issue precluding summary judgment, we will take evidence favorable to the non-movant as true, and (3) we will indulge every reasonable inference in favor of the non-movant and resolve any doubts in its favor. See Tex. R. Civ. P. 166a(c); Nixon v. Mr. Prop. Mgmt. Co., 690 S.W.2d 546, 548-49 (Tex. 1985); Trilogy Software, Inc. v. Callidus Software, Inc., 143 S.W.3d 452, 458 (Tex. App.—Austin 2004, pet. filed). A defendant is entitled to summary judgment under Rule 166a(c) if it conclusively negates at least one of the essential elements of the plaintiff's cause of action or conclusively proves each element of its affirmative defense, thereby showing that it is entitled to judgment as a matter of law because no genuine issue of material fact remains. Rhone-Poulenc, Inc. v. Steel, 997 S.W.2d 217, 222-23 (Tex. 1999). Under the no-evidence summary-judgment rule, a party may move for summary judgment if, after adequate time for discovery, there is no evidence of one or more essential elements of a claim or defense on which the nonmovant would have the burden of proof at trial. Tex R. Civ. P. 166a(i); Jaimes v. Fiesta Mart, Inc., 21 S.W.3d 301, 303-04 (Tex. App.—Houston [1st Dist.] 1999, pet. denied). Once the movant specifies the elements on which there is no evidence, the burden shifts to the nonmovant to produce summary judgment evidence raising a genuine issue of material fact on those elements. Tex. R. Civ. P. 166a(i); Trilogy, 143 S.W.3d at 459. A general issue of material fact exists if more than a scintilla of evidence establishing the existence of the challenged element is produced. Ford Motor Co. v. Ridgway, 135 S.W.3d 598, 600 (Tex. 2004). More than a scintilla of evidence exists if the evidence "rises to a level that would enable reasonable and fair-minded people to differ in their conclusions." Id. at 601. But when "the evidence offered to prove a vital fact is so weak as to do no more than create a mere surmise or suspicion of its existence, the evidence is no more than a scintilla and, in legal effect, is no evidence." Id. The central flaw undermining Bush's claims is the absence of evidence that either a defect in the generator or the absence of a wait warning caused the fire and injuries. The only evidence is that he did not spill gasoline, and that the tested generator did not reach temperatures that caused or should have caused auto-ignition of vapors in open-air conditions. There is no evidence that the absence of warning recommending a waiting period before refueling caused any injury or that the presence of such a warning would have prevented Bush's injuries. Products liability: marketing defect Bush challenges the summary judgment against his marketing-defect theory of strict products liability. Bush contended that the generator and motor were defective for failing to warn of the need to allow the engine to cool for two minutes after shutoff before refueling. A marketing-defect cause of action consists of five elements: (1) a risk of harm that is inherent in the product or that may arise from the intended or reasonably anticipated use of the product must exist, (2) the product supplier must actually know or reasonably foresee the risk of harm at the time the product is marketed, (3) the product must possess a marketing defect, (4) the absence of the warnings and/or instructions must render the product unreasonably dangerous to the ultimate user or consumer of the product, and (5) the failure to warn and/or instruct must constitute a causative nexus in the product user's injury. Jaimes, 21 S.W.3d at 305-06. Bush did not present evidence of a causal link between any alleged failure to warn and his injuries. Some of his experts affirmatively denied they were providing evidence of causation. Johnston testified that determining whether the lack of a warning caused the accident "would be pure speculation." Johnston flatly stated, "I'm not giving any opinions about causation." Hasselbring and Fowler also testified in their depositions that they were not there to speak to causation. More important, there is no evidence that the generator and engine were unreasonably dangerous due to the absence of a warning. Bush's experts did not provoke auto-ignition even when they poured gasoline directly on the generator immediately after shut-off. Clayton, who testified regarding cause and fire origin, noted that if the generator did not reach the auto-ignition temperature of gasoline, then the generator would have to be ruled out as an ignition source. Without evidence that the generator met or exceeded an acceptable auto-ignition temperature, Clayton provided no evidence of causation. Because Bush did not produce more than a scintilla of evidence that the absence of a warning caused his injuries, the district court properly granted summary judgment as to the marketing defect claim. This failure obviates any harm from the court's alleged failure to consider evidence from Coleman and Tecumseh employees about their awareness of the need for a cool-down period at the time of marketing. Evidence that many employees were unaware of any need for a cool-down period would not help Bush because it does not show the awareness needed to prove a marketing defect. Although Coleman employee Kenneth Frank acknowledged that having a two-minute warning would make the product better than having no warning, we need not resolve whether this is relevant evidence because there is no evidence of causation. Bush also points to the fact that, although the 1989 Coleman generator used in this case did not have a warning regarding refueling, the same model generator manufactured in 1995 included a warning label that reads "Gasoline is flammable. Allow engine to cool at least two minutes before refueling." (5) But use of this warning on later models marketed by the appellees is not evidence that the generator in this case was unreasonably dangerous without the warning, and, in the absence of other evidence of causation, is no evidence that the lack of a warning on this generator caused Bush's injuries. Negligence and gross negligence Bush did not support his allegation that Coleman and Tecumseh were negligent by failing to provide adequate warnings and instructions regarding the hazards associated with the foreseeable uses of the generator. To prevail in a negligence cause of action, a plaintiff must show that (1) defendant owed plaintiff a duty, (2) defendant breached that duty, and (3) the breach proximately caused plaintiff's injuries. Greater Houston Trans. Co. v. Phillips, 801 S.W.2d 523, 525 (Tex. 1990). As discussed above, Bush's experts were unable to show that the generator created a hazard or danger during refueling. Without evidence that the gasoline vapors on this generator would auto-ignite during refueling, Bush cannot prove that a warning was needed or that the absence of a warning caused either the fire or his injuries. Summary judgment was proper against Bush's negligence claims. (6) The failure of Bush's negligence claims forecloses his gross negligence claims. See Trevino v. Lightning Laydown, Inc., 782 S.W.2d 946, 949 (Tex. App.--Austin 1990, writ denied) ("conduct cannot be grossly negligent without being negligent"); see also Tex. Civ. Prac. & Rem. Code § 41.001(11) (West Supp. 2004) (defining gross negligence). CONCLUSION We hold that the district court did not abuse its discretion in excluding the expert testimony of Bush's expert witnesses Lori Hasselbring, Joe Fowler, Waymon Johnston and Judd Clayton. In addition, we conclude that the district court did not err in granting summary judgment in favor of Coleman and Tecumseh on all claims that are the subject of this appeal. Accordingly, we affirm the judgment of the district court. __________________________________________ Bob Pemberton, Justice Before Justices B. A. Smith, Puryear and Pemberton Affirmed Filed: May 26, 2005 1. Dwight Pfennig has three degrees: a bachelor's in petroleum engineering and master's and doctor's in chemical engineering. His work focuses on potential fires and explosions from chemicals and hot equipment. 2. The district court said it did not exclude Beaver's testimony because Beaver did not offer an opinion regarding causation. 3. The temperature reached 756ºF in another test, but the engine ran out of gasoline before the testers could spill gasoline on it. Appellees' expert Pfennig said the generator reached 891ºF in one of his tests, but did not approach the open-air auto-ignition temperature of gasoline of 1155ºF, and still did not cause auto-ignition of gasoline vapors. 4. Bush attempted to file an additional expert report after summary judgment when the discovery deadline, the evidence-filing deadline, and the Robinson hearing all had passed. Late-filed evidence is not part of the summary-judgment record unless the district court granted leave to file it. See Tex. R. App. P. 166a(c); see also Benchmark Bank v. Crowder, 919 S.W.2d 657, 663 (Tex. 1996); Basin Credit Consultants, Inc. v. Obregon, 2 S.W.3d 372, 374 (Tex. App.—San Antonio 1999, pet. denied). Finding no indication that the district court granted such leave, we cannot consider this evidence on appeal. 5. Evidence of subsequent remedial measures currently is not admissible to prove negligence, culpable conduct, a defect in the product, a defect in the product's design, or a need for a warning or instruction. Tex. R. Evid. 407(a). When this case was filed, however, subsequent remedial measures were admissible in strict products liability cases. 6. Bush appears to have abandoned his claim for negligent design. Even if he did not, summary judgment against the claim was proper because no evidence of a reasonable alternative design was admitted into the record. See American Tobacco Co., Inc. v. Grinnell, 951 S.W.2d 420, 437 (Tex. 1997).