Lemond v. Lone Star Gas Co.

897 S.W.2d 378 (1994)

Phares E. LEMOND, Individually and as Surviving Spouse of Nova Ellen Lemond, and as Independent Executor of the Estate of Nova Ellen Lemond, Deceased; Phares E. Lemond, Jr.; William Clark Lemond; and All Survivors and Beneficiaries at Law of Nova Ellen Lemond, Deceased, Appellants,
v.
LONE STAR GAS COMPANY, an Unincorporated Division of Enserch Corporation, Appellee.

No. 2-92-128-CV.

Court of Appeals of Texas, Fort Worth.

February 28, 1994. Rehearing Overruled April 13, 1994.

David L. Evans, Jeffrey H. Kobs, Richard C. DeBerry, Hooper & Evans, Fort Worth, Dottie Murphy, Law Offices of Dottie Murphy, Wichita Falls, David Keltner, Haynes & Boone, Fort Worth, Larry Cain, Law Offices of Larry Cain, Sherman, for appellants.

Eric D. Archer, Enserch Corp., Dallas, Robert Q. Keith, Amy Keith, Keith & Weber, P.C., Johnson City, Glynn Purtle, Fillmore & Purtle, Wichita Falls, for appellee.

Before WEAVER, HICKS and FARRAR, JJ.

OPINION

WEAVER, Justice.

Phares Lemond and various family members (appellants or the Lemonds) brought suit seeking compensation for the death of Phares' wife, Nova, injuries to Phares, and other damages related to a natural gas explosion at the Lemonds' home in the city of Iowa Park on August 16, 1988. The appellee (Lone Star)^[1] supplied natural gas to the home. Appellants' action, grounded on product liability, alleged defects in design, manufacturing and marketing of the gas. They also brought a negligence action. No jury issue was submitted on the manufacturing defect. Following jury findings favorable to Lone Star on the issues of design defect, marketing defect and negligence, the trial court entered a take nothing judgment against appellants. We reverse as to the marketing defect claims and otherwise affirm the judgment of the trial court.

By four points of error the appellants assert the trial court erred: 1) in failing to submit a jury question on manufacturing defect; 2) by including in the jury question on marketing defect a surplus instruction regarding a duty to warn; 3) in failing to grant a new trial because the jury's answer on the issue of marketing defect was against the great weight and preponderance of the evidence; and 4) in refusing to submit an instruction that Lone Star owed a duty of high care.

The undisputed facts show the explosion was caused by natural gas collecting under the crawl space of the Lemond home and traveling up into the home where it was ignited by the pilot light of the hot water heater. The ensuing explosion caused the damages for which appellants sought to recover.

Appellants' liability claims were based on the theories that the gas leaked from Lone Star's poorly maintained gas distribution line into the crawl space under the house or otherwise migrated there through the soil; that the gas which collected under the house emitted no odor; that the warning odorant in the gas had been "scrubbed out" of the gas by soil, a phenomenon known as "odor fade", and as a result of which there was no odorant or insufficient odorant in the gas to warn of its presence; or that liquid hydrocarbons, and their heavier hydrocarbons such as propane, in Lone Star's transmission lines, leached out or chemically reacted with the odorant, known as "masking" or also as "odorant fade", thus eliminating the effectiveness of its odor and thereby reducing or diminishing its ability to warn of the presence of the natural gas. Lone Star counters that the gas was properly odorized, that the gas which caused the explosion leaked from pipes at the house which were under the Lemonds' control, and not from its gas lines, and that Lone Star's activities were not the proximate or producing cause of the appellants' damages.

The jury failed to find that Lone Star's negligence was the proximate cause of the explosion, and appellants present no arguments on this appeal regarding the negligence issue. Their design defect issue was based on Lone Star's use of a product known as Captan to odorize the gas, claiming that it should have used other more efficient odorizing agents which were available and which were less subject to being masked or leached out in the soil. The jury also failed to find that there was a design defect in the natural gas, at the time it left the possession of Lone Star, that was a producing cause of the explosion. Likewise, the design defect issue is not before us on this appeal.

In their first point of error, the Lemonds claim the trial court erred in refusing to submit their manufacturing defect question to the jury. They assert that they pled a cause of action for manufacturing defect, that the escaped gas which caused the explosion was improperly manufactured by Lone Star, that the gas was defective, and that the defect was the failure of the gas to contain enough odorant to warn consumers such as the Lemonds of its presence.

As the Texas Supreme Court has stated:

So long as matters are timely raised and properly requested as part of a trial court's charge, a judgment cannot be permitted to stand when a party is denied proper submission of a valid theory of recovery... raised by the pleading and evidence.

Exxon Corp. v. Perez, 842 S.W.2d 629, 631 (Tex.1992) (footnote omitted). If both manufacturing defect and design defect are alleged and proved, jury questions on each theory should be separately submitted. Pool v. Ford Motor Co., 715 S.W.2d 629, 637 (Tex. 1986) (opinion on reh'g). The trial court may decline to submit a relevant issue only if there is no evidence to support it. Garza v. Alviar, 395 S.W.2d 821, 824 (Tex.1965).

Lone Star contends the trial court correctly refused to submit the manufacturing defect issue to the jury because it was not supported by any pleading, the case was not tried on that theory, and there was no evidence to support the same. We first address the attack on the lack of pleading.

A pleading must give full and fair notice of the theories of recovery so a party may prepare a defense. Roark v. Allen, 633 S.W.2d 804, 810 (Tex.1982); Sun Power, Inc. v. Adams, 751 S.W.2d 689, 696 (Tex.App.— Fort Worth 1988, no writ); Haddock v. Arnspiger, 763 S.W.2d 13, 15 (Tex.App.—Dallas 1988), aff'd, 793 S.W.2d 948 (Tex.1990). Each of those cases stands for the proposition that the opposing party must have notice of the theory and facts for which it is being sued.

The Lemonds' live trial pleadings included the following allegations regarding their claims against Lone Star:

Plaintiffs ... purchased natural gas which was manufactured, marketed and sold by one or more of the Defendants.... [N]atural gas supplied by Defendants escaped from Defendants' system or the piping through which it was carried causing the occurrence made the basis of this suit.... Defendants' natural gas contained improper, defective or inadequate levels of ordorization....

Defendants sell, market, manufacture and design the product and place [it] into the stream of commerce.

. . . .

Defendants, in addition to the above, sold, marketed, and distributed an unreasonably dangerous product ... in a defective condition.

[D]ue to the ultrahazardous nature of this product and the unreasonably dangerous and defective condition in which it was sold, marketed and distributed to Plaintiffs, Nova ... was killed and Phares ... was severely injured.

At the charging conference, the Lemonds' attorney, in urging the submission of a manufacturing defect, stated:

[W]e believe that we have the right to submit a manufacturing defect in that there is evidence in this case of a manufacturing defect, to-wit; the testimony about dead odorizers, about liquid hydrocarbons in the line that mask odorant, and all those things that tend to make the product as it comes out and is sold to the customers flawed in the manufacturing process, in addition to design defect and marketing defect.

The trial court overruled the request without comment.

This point was next raised at the hearing on the Lemonds' motion for new trial, at which the Lemonds' attorney argued that there was evidence to support an issue on manufacturing defect and that the issue was raised by the pleadings. The trial judge was apparently of the opinion that the manufacturing defect had been pled, as reflected by his statement that "[t]he Court doesn't argue that you pled the matter." In response to comments of the attorney for the Lemonds concerning the distinction between a manufacturing defect and a design defect, the trial court stated: "Well, you're going after Captan two different ways. It looks to me like you're trying to get two bites at the same apple." The court overruled the motion for new trial without further comment on that issue.

To the extent that the trial court's reasoning in refusing to submit the manufacturing defect issue was because the same matter was included in the submission of the design defect, such reasoning is rejected. Where both manufacturing defect and design defect theories are alleged and supported by the evidence, issues on both theories should be separately submitted. Pool, 715 S.W.2d at 637. However, the ruling of the trial court, in this instance the refusal to submit the manufacturing defect issue, is to be upheld if such ruling comports to a theory allowing such action and which is supported by the record, notwithstanding that the reasoning of the trial court in making such ruling may have been based on another expressed theory. Luxenberg v. Marshall, 835 S.W.2d 136, 141-42 (Tex.App.—Dallas 1992, no writ). In this case, Lone Star claims the refusal to submit such issue was proper because there was no evidence to support the same.

We hold the allegations in the Lemonds' pleadings that the natural gas was manufactured and sold by the Defendants, the gas contained improper, defective or inadequate levels of ordorization, and the gas was distributed in a defective condition, were sufficient to place Lone Star on notice of a manufacturing defect claim. We agree with the trial court's conclusion that such cause of action was raised by the pleadings.

Having found that manufacturing defect was pled, and the Lemonds having requested a jury issue on the same, we must reject Lone Star's claim that the case was not tried on that issue if there was evidence to support the same.

We will now determine whether there was evidence bearing on the manufacturing defect which would require a jury submission on that issue. Lone Star directs us to much record evidence showing that the gas was properly manufactured. However, the question is whether there was any probative evidence showing that the gas was not properly manufactured. Such may be shown by circumstantial evidence. See Morgan v. Compugraphic Corp., 675 S.W.2d 729, 733 (Tex.1984); Darryl v. Ford Motor Co., 440 S.W.2d 630, 632 (Tex.1969). Expert testimony can also be used as proof of a manufacturing defect. Fitzgerald Marine Sales v. LeUnes, 659 S.W.2d 917, 918 (Tex.App.— Fort Worth 1983, writ dism'd). However, there must be more than a mere scintilla of evidence. If the evidence is so weak as to do no more than create a mere surmise or suspicion of its existence, it is no more than a scintilla and, in legal effect, is no evidence. There is some evidence, more than a scintilla, only if the evidence furnishes some reasonable basis for differing conclusions by reasonable minds as to the existence of the vital fact. Kindred v. Con/Chem, Inc., 650 S.W.2d 61, 63 (Tex.1983).

The Lemonds claim there was evidence to show manufacturing defect in two separate areas of the manufacturing process. One was failure of the gas to contain enough odorant to warn consumers of the presence of gas. They claim this was due to the malfunctioning of odorizer machines which injected the odorant into the gas stream. The other claim was the failure of Lone Star to remove impurities from the gas, namely liquid hydrocarbons, which they apparently claim leached out or chemically reacted with the odorant so as to eliminate the effectiveness of the odorant which was injected into the gas stream.

The transporting and delivery of the natural gas to the consumers in Iowa Park fell under the responsibility of two separate divisions or departments of Lone Star, being the transmission department and the distribution department. The gas was gathered from producers or various processing plants and transported to the "city gate" in Iowa Park by transmission pipelines operated by the transmission department. That department was responsible for odorizing the gas and for removing the liquid hydrocarbons from the gas prior to turning it over to the distribution department at the "city gate." Alton Gates was the superintendent of the transmission department for the Wichita Falls district, or northern region, which transported the gas to Iowa Park. The distribution department was responsible for delivering gas to customers in the city through its distribution lines and system. That department had no responsibility for odorizing the gas or removing the liquid hydrocarbons.

The transmission department delivered gas to the "city gate" in Iowa Park through two separate transmission lines. One was Line A which brought gas from the north, and the other was Line 73 which brought gas from the south. Line A was the source of the majority of the gas supplied to the "city gate." That line received its gas from two other lines, which were a part of what was referred to as the T System, one being Line T which brought in gas from the north. The other was Line TK which received gas from the New York storage facility located south of Henrietta. Gates testified that no gas from the New York storage facility was flowing into Lone Star's lines to Iowa Park in August of 1988. The other transmission line, Line 73, supplied gas to the "city gate" from the south, and had an odorizer near New Castle which odorized the gas in that line. Only a small amount of the gas supplied to the "city gate" came from Line 73.

Lone Star introduced the testimony of Gates and other evidence showing that the gas was properly manufactured. This included evidence that the liquid hydrocarbons were removed from the gas in the transmission system prior to the time it was odorized, there was no excess amount of liquid hydrocarbons in the gas flowing to the Lemonds' home during August of 1988, and the odorizers on the transmission lines were operating and were injecting odorant into the gas stream during the period of June through September, 1988. We will now review the evidence which is claimed by the Lemonds as showing a manufacturing defect. Although much of that evidence was offered and introduced for another specific purpose, such as bearing on negligence or on the design or marketing defects, we will nevertheless review it with respect to the manufacturing defect since we have held that cause of action was pled.

Failure to Properly Odorize the Gas

The Lemonds rely upon the following evidence as showing that the gas was not properly odorized due to the malfunctioning of odorizing machines.

Gas not smelled at house: One of the Lemonds' claims, that the gas was not properly odorized, was based on circumstantial evidence showing the failure of various persons to smell gas at the house. The Lemonds presented several witnesses who testified they did not smell gas at the Lemonds' house. Lahoma Adams, a friend of the Lemonds who visited in their home from time to time, testified she never smelled gas there. However, she was unable to say how long it had been since she was there before the explosion, but testified it could have been three weeks or longer. Roy Wright, a neighbor, testified he mowed the lawn at a house next to the Lemonds' house, probably once a week during the summer of 1988, that he was around the Lemonds' house mowing the lawn next to it or driving by it, probably on a daily basis before the explosion, and he never smelled natural gas around the house. He was last at the Lemonds' house, visiting Phares in the garage, about a month before the explosion, and testified that he did not smell gas at that time. He was not outside on the morning before the explosion. James Spruiell, who lived across town, testified he drove by the Lemonds' house in his pick-up with his windows down at 6:00 a.m. in the morning before the explosion and did not smell gas.

Phares' two brothers both testified they attended a family reunion at the house over the weekend prior to the explosion and smelled no gas, nor did any one at the reunion mention smelling gas. Likewise, Phares testified he got up about 4:00 a.m. on the morning of the explosion, went outside to look for the newspaper, returned to the house and went back to bed, and that he smelled no gas at that time.

The Lemonds also rely upon the following answer of Richard A. Dyer, Lone Star's expert witness, to a hypothetical question posed during his deposition:

[I]f you have a dwelling where there is a slow natural gas leak that's occurring in the crawl space that is lower instead of occurring in an attic where there is ventilation to go out above the people, and there are a considerable number of people present at the dwelling during this slow leak and in and around the dwelling and you have a ventilated crawl space, the lack of smelling the gas would indicate to me the lack of odorization.

That exchange was followed by another hypothetical situation posed to Dyer, being whether the failure of one or two occupants in one room of a house to smell gas means that the gas was not odorized, to which he replied:

[I]f you are talking about a very rapid, significant gas leak—and that gas does not migrate to the location where the occupants are, then the occupants not smelling the gas would not have considerable weight in determining that the gas was not properly odorized.

The Lemonds direct us to no evidence showing the nature of any gas leak at their house, the extent of any such leak, whether it was a slow or a rapid leak, or the duration of any such leak.

Malfunctioning Odorizers: The Lemonds also rely upon the evidence discussed below as showing a failure of the gas to contain enough odorant to warn consumers such as the Lemonds of the presence of gas, and thus as proof of a manufacturing defect.

Three dead or malfunctioning odorizers were found on Lone Star's system as shown by its Odorizer Inspection Reports. Two of these were found at the New York storage facility in December 1987. The reports show that one odorizer was dead and was repaired and left operating satisfactorily and that the other had a small leak in the odorizer line. As mentioned above, Gates testified that no gas was transmitted from the New York storage facility to Iowa Park in August of 1988. The third report showed an odorizer was found dead near Newcastle on November 22, 1988, some three months after the explosion at the Lemonds' house. The report shows that valves were replaced and the odorizer was left running satisfactorily.

The Lemonds claim that certain other events, which they refer to as "similar incidents", also constitute evidence bearing on the manufacturing defect, being the failure of the gas to contain enough odorant. One was an occurrence in Leonard, Texas, in 1981, when tests showed that the odorant concentration in the gas was below acceptable levels. Another such incident occurred in Nocona, Texas, where Lone Star was cited by the Railroad Commission for an alleged violation of odorization requirements because a representative of the commission reported that he could not smell the gas. Another "similar incident" relied upon by the Lemonds is the testimony of Bobby Clark, a Lone Star employee, that he had heard rumors within the company about improperly odorized gas reaching the point of use by customers. Clark testified that he had heard "rumors, talk, whatever", but stated that he did not have any specifics on exactly where those conversations took place, or by whom, or when.

The Lemonds direct us to no evidence in the record which indicates that any gas was transmitted and delivered to Lone Star's distribution at Iowa Park which failed to contain sufficient amounts of odorant at the time it was manufactured. We hold that the above mentioned evidence which the Lemonds submit as bearing on Lone Star's failure to properly odorize the gas is so weak as to do no more than create a mere surmise or suspicion of its existence, is no more than a scintilla and, in legal effect, is no evidence. Kindred, 650 S.W.2d at 63.

Failure To Remove The Liquid Hydrocarbons From The Gas

It is undisputed that sufficient quantities of liquid hydrocarbons in natural gas can mask the odorant and cause it to fade and lose its effectiveness as a warning to consumers. The Lemonds claim that the removal of liquid hydrocarbons from the gas is a part of the manufacturing process, and that the evidence showing that Lone Star failed to remove the liquids from the gas is sufficient to support the submission of their requested issue on the manufacturing defect.

In support of this claim, the Lemonds direct us to the testimony of five witnesses whose testimony primarily bears on the characteristics of liquid hydrocarbons to mask with the odorants in the gas. Alton Gates, testifying from the company's Operating Procedure Manual, confirmed that the existence of liquid hydrocarbons in sufficient quantities can render the malodorant injected at normal rates ineffective in serving its intended purpose. He testified that propane is one such hydrocarbon.

Robert Lindsey, who had worked for Lone Star for over twenty years, testified that the phenomena of odorant fade can occur through oxidation, such as when it reacts with a rusty pipe, and that it can occur due to adsorption and absorption until it reaches an equilibrium.

Douglas Chisholm testified that most natural gases contain liquid hydrocarbons. The heavier hydrocarbons such as propane can exist as liquids under pressure, and, due to failure to remove such hydrocarbons from the gas, the same can enter the pipeline stream, where they can leach out or chemically react with odorants such as Captan and eliminate the effectiveness of its odor.

Paul O'Connor is an employee of Haag Engineering. That firm is in the business of failure and damage analysis consulting, including investigation of fire explosion scenes. O'Connor testified that he and a Haag employee took soil samples inside the perimeter beam of the Lemonds' house on August 17, 1988, and the lab report revealed propane and methane in the soil samples.

David Flores, Regional Engineer for the Railroad Commission, investigated the Lemond explosion. He was familiar with the concept of odorant being rendered undetectable by smell if there are liquid hydrocarbons in the lines. He testified that one way a high concentration of propane under the Lemond house could get there would be because liquid hydrocarbons were in the gas lines leading up to the Lemond residence.

None of these witnesses testified that liquid hydrocarbons were present in the gas being delivered to the Lemonds' house at or prior to the time of the explosion. The testimony of O'Connor merely showed that propane was present in the soil samples. According to Flores, one way a high concentration of propane could have gotten under the house was by liquid hydrocarbons being in the gas. Even if we were to assume that propane from the soil samples came from Lone Star's gas distribution system, the Lemonds direct us to no testimony showing that such propane was of a high concentration, or in any way showing in what proportions it was so concentrated in the gas delivered by Lone Star. There is no testimony showing, nor can an inference be drawn from the testimony which shows, that the gas which was delivered to the Lemonds at or prior to the explosion contained liquid hydrocarbons in sufficient quantities to cause it to mask with the odorant and thereby cause it to lose its effectiveness to warn. At best there is only a mere scintilla of evidence, based on possibilities or conjectures, which we hold to be no evidence on this point. Having found that there is no evidence to support the Lemonds' requested issue on the manufacturing defect, point of error one is overruled.

In their second point of error, the Lemonds assert the trial court erred by adding an extraneous instruction to the jury question relative to the marketing defect. Ordinarily, natural gas is colorless and odorless and cannot be detected by smell without the addition of an odorant. Under certain circumstances, even if initially injected into the gas in sufficient quantities, odorants added to the gas may mask or fade to the extent that they no longer warn of the presence of the gas. The Lemonds' marketing defect complaint was based on the failure of Lone Star to warn that odorants in the gas may mask or fade and of the dangerous characteristics of gas when the concentration of odorants is reduced to levels insufficient to warn of the presence of the gas. That Lone Star failed to give such warning in this case is uncontested.

In Texas, the existence of a duty to warn of dangers or instruct as to the proper use of a product is a question of law. See Munoz v. Gulf Oil Co., 732 S.W.2d 62, 65 (Tex.App.—Houston [14th Dist.] 1987, writ ref'd n.r.e.). "A product may be proven to be defective if ... it is unreasonably dangerous because adequate warnings or instructions are not provided." Lucas v. Texas Industries, Inc., 696 S.W.2d 372, 377 (Tex.1984) (opinion on reh'g). Even a product which is safely designed and manufactured may be unreasonably dangerous as marketed because of a lack of adequate warnings or instructions. Id. at 377. In marketing defect cases, the defective or unreasonably dangerous condition of the marketed product arises not from the product itself, but because the seller fails to warn the user of an improper use of the product. Technical Chemical Co. v. Jacobs, 480 S.W.2d 602, 604-05 (Tex.1972).

The jury question submitted by the trial court, inquiring as to whether there was a defect in the marketing of the gas that was a producing cause of the injuries, followed the language recommended by section 71.06 of the Texas Pattern Jury Charges, except it also contained the following instruction of which the Lemonds complain under this point of error:

A seller's duty to warn arises only where the dangers to be warned of are reasonably foreseeable and are such that a consumer cannot reasonably be expected to be aware of them.

The Lemonds attack the submission of the instruction as constituting an erroneous comment on the evidence. In doing so, they refer us to a number of cases which have dealt with the propriety of jury instructions of this nature in strict liability cases. In Fleishman v. Guadiano, 651 S.W.2d 730, 731 (Tex.1983), a design defect case, the supreme court upheld the trial court's refusal to include an instruction on the plaintiff's negligence. In an earlier conscious design defect case, Turner v. General Motors Corp., 584 S.W.2d 844 (Tex.1979) (opinion on reh'g), the court stated that "henceforth in the trial of strict liability cases involving design defects the issue and accompanying instruction will not include either the element of the ordinary consumer or of the prudent manufacturer;..." and rejected the notion of instructing the jury to balance specifically enumerated factors. Id. at 847.

The court's position respecting additional instructions in design defect cases, as voiced in Fleishman and Turner, was further addressed in Acord v. General Motors Corp., 669 S.W.2d 111 (Tex.1984). In that closely contested design defect case, the court held that an instruction which singled out that General Motors was neither an insurer nor a guarantor of a perfect or accident-proof product was a comment on the case as a whole, and as such constituted harmful error. Id. at 115-16. In Lemos v. Montez, 680 S.W.2d 798 (Tex.1984), a traffic accident negligence case, the supreme court again expressed its position regarding surplus jury instructions, stating that the court "has treated addenda to the charge as impermissible comments that tilt or nudge the jury one way or the other." Id. at 801.

Wilson v. Kaufman & Broad Home Systems, 728 S.W.2d 874 (Tex.App.—Beaumont 1987, writ ref'd n.r.e.), a manufacturing defect case, is another example of the trial court being reversed due to surplus jury instructions being added as a departure to the pattern jury charges. The Beaumont court amply called attention to the supreme court's position on this matter by the following language:

Our highest court has made it abundantly clear that to deviate from the pattern jury charges in products liability cases is a perilous journey. We believe their admonitions, in this regard, are wise and should be heeded.

Id. at 875-76.

None of the cases cited above specifically deal with marketing defect. However, we believe the prohibition against such surplus jury instructions is equally applicable to products liability cases in general. See id. Based on the above mentioned cases, we find the surplus instruction which was included in the jury question concerning marketing defect was an erroneous comment on the evidence. We will address the harm analysis later in this opinion.

Lone Star does not respond directly to the "comment on the weight of the evidence" arguments made by the Lemonds pursuant to the above cited cases. It neither admits nor disputes the existence or validity of that theory of law regarding surplus instructions in product liability cases. It does, however, claim that the instruction was proper for two reasons.

Lone Star's first contention is that it rebutted the Lemonds' contention that the warning would have been some benefit to them. Apparently this argument relates to the rule that when a manufacturer fails to give adequate warnings concerning the use of a product, a rebuttable presumption arises that the user would have read and heeded such warnings. See Technical Chemical Co., 480 S.W.2d at 606. According to the Lemonds, the presumption in this case was that they would have either chosen not to use gas or that they would have taken safety measures by using gas detectors to warn them of the build-up of explosive levels of unodorized gas if they had been warned of the dangers.

Lone Star contends that by its having rebutted such presumption, the instruction given to the jury was proper, relying upon Magro v. Ragsdale Bros., Inc., 721 S.W.2d 832 (Tex.1986). We do not necessarily consider Magro as authority for the proposition tendered by Lone Star, but we will nevertheless review the testimony to which it refers us for this purpose.

In support of its contention that it rebutted this presumption, Lone Star directs us to testimony of David Biegler, chairman of Lone Star Gas Company division of Enserch Corporation. He acknowledged odorant was required to be put in the gas so it can be detected by consumers for their safety and that consumers can't know it's there unless they can smell it. He said Lone Star, through advertisements, tells the public what to do if they smell gas, namely to leave the house and call the gas company. He acknowledged it would be hypothetically feasible to tell the public through advertisements that there are certain conditions and mechanisms and instances in which they may not smell gas that escapes in their home. More specifically, the testimony relied upon by Lone Star to show the presumption was rebutted, is reflected by the following answers by Mr. Biegler:

Q. Can you think of any reason not to tell the public about something so important to their safety as the propensity or the possibility of your gas losing its ability to warn them by odor, by smell?

A. I think you're making a value judgment in the question that makes it awfully difficult to answer because your value judgment is that that propensity is so great and affects their safety such that they need to be. I can think of reasons why that isn't so. I can think of reasons why that value judgment results in a conclusion that it is not.

. . . .

Q. Has Lone Star Gas made a value judgment that the risk is not that great regarding odorant fade?

A. That is our position now, as I understand it.

Q. Part of your answer was—your earlier answer was that there are a lot of reasons that might affect a decision not to warn about that sort of thing. Tell us what some of those reasons are.

A. I think the public being warned about something that isn't a risk can create other problems for the public. I think hypothetically, which is what we're creating, hypothetical scenarios, I mean, it would be similar to warning the public that, let's say, there was one bug in an air controller pattern in Seattle, and they are in imminent danger of an airplane landing on their head.

I don't think that is in the area of beneficial public good. I think in this instance you run the counter risk to the public by an inadequate explanation of risk that what you can do is mask true risks, and if you overrode the system for detecting true risks whereby valid leaks that are smelled, et cetera, are not corrected in a timely manner, et cetera, I think there is a counterbalancing risk to the public.

We view Mr. Biegler's testimony as bearing more on the reasons or justifications for Lone Star not giving a warning. We don't view it as rebutting the presumption that the Lemonds would have heeded the warnings had they been given. We reject Lone Star's argument that the jury instruction was proper because it rebutted such presumption.

Lone Star next argues that the instruction was proper because it was a correct statement of the law. In support of this argument, Lone Star relies on Beans v. Entex, Inc., 744 S.W.2d 323 (Tex.App.—Houston [1st Dist.] 1988, writ denied). In Beans, the court stated that a "seller's duty to warn arises only where the dangers to be warned of are reasonably foreseeable and are such that a consumer cannot reasonably be expected to be aware of them." Id. at 325. Lone Star also cites Houston Lighting & Power v. Reynolds, 765 S.W.2d 784 (Tex. 1988), for the proposition that where the use of a product is not reasonably foreseeable or where the warning would prove no benefit that the instruction here under review is proper. Neither of those cases dealt with the propriety of surplus instructions in product liability cases. An erroneous instruction is not saved in these product liability cases merely because it correctly states the law. In Acord, 669 S.W.2d at 116, the supreme court specifically disapproved the addition of any other instructions in such cases, however correctly they may state the law, and we reject Lone Star's argument that the instruction in this case was proper because it did so.

Lone Star then argues that the instruction given in this case, even if improper, was harmless. It contends that the instruction actually benefitted the Lemonds, that it does nothing more than reiterate the duty that the Lemonds attempted to prove, that is "the dangers to be warned of must be reasonably foreseeable" and "are such that a consumer cannot reasonably be expected to be aware of them", and reinforced the proof and arguments of the Lemonds that consumers of natural gas are not familiar with the concept of odor fade and must be warned of such. It also argues that the instruction benefitted the Lemonds by focusing on the need for the warning and did not serve to divert the jury's attention from that issue. The Lemonds counter that the instruction focused the jury's attention on what danger they should have been expected to know rather than on what dangers Lone Star did know about and whether its failure to warn of such dangers constituted a marketing defect. The marketing defect question submitted to the jury made no reference to a seller's duty to warn other than as contained in the instruction here under review. Such jury question, in language approved in the Pattern Jury Charges, defined a marketing defect as:

[T]he failure to give adequate warnings of the products [sic] dangers that were known, or by the application of reasonably developed human skill or foresight should have been known, or failure to give adequate instructions to avoid such dangers, which failure rendered the product unreasonably dangerous as marketed.

We hold that the extraneous instruction given by the trial court in this case conflicted with and diverted the jury's attention from the above quoted language of the charge, was confusing to the jury, was calculated to and probably did tilt or nudge the jury one way or the other, and prejudiced the jury verdict, resulting in harmful error. Having so held, we need not address the Lemonds' further attack on the extraneous instruction as not being supported by the evidence. The second point of error is sustained.

Also, in light of our ruling on the second point of error, we do not reach the third point of error by which the Lemonds contend that the jury finding on the marketing defect question was so against the great weight and preponderance of the evidence as to render it manifestly erroneous.

Finally, in their fourth point of error, the Lemonds contend the trial court should have instructed the jury that Lone Star owed a "high" degree of care to its customers, rather than the "ordinary" care instruction that was given with the negligence question. This court has previously addressed this issue and has decided to the contrary. See Gray v. Enserch, Inc., 665 S.W.2d 601, 603-04 (Tex.App—Fort Worth 1984, writ ref'd n.r.e.). Although we have examined the authorities cited by appellants, we conclude that we are bound by the holding in Prudential Fire Ins. Co. v. United Gas Corp., 145 Tex. 257, 199 S.W.2d 767, 772 (1946). The fourth point of error is overruled.

The judgment of the trial court is reversed and remanded for new trial on the Lemonds' marketing defect claims. In all other respects, the judgment of the trial court is affirmed.

NOTES

[1] Other original defendants were dismissed from the lawsuit, leaving only Lone Star as a defendant when the judgment was entered.