OPINION

BRIGHAM, Justice.

In this case concerning alleged permanent injury to land, the primary issues we must determine are whether appellees timely filed their lawsuits against appellant Mitchell Energy Corporation (Mitchell) and whether the evidence establishes that Mitchell caused appellees' alleged injuries. We hold that most of appellees' claims are time-barred and that there is no evidence of causation. Consequently, we reverse the trial court's judgment and render judgment that appellees take nothing from Mitchell.

I. Background Facts and Procedural History.

Appellees are all landowners in rural Wise County. They obtain their residential water from wells located on their land. For many years, the water in the vicinity of appellees' land has had a bad taste and odor, often compared to the smell of rotten eggs. The smell of the water has long been a matter of general knowledge in the area, and many of appellees knew of the poor quality of the water when they bought their land.

Mitchell is an oil and gas exploration company. Mitchell began drilling and operating gas wells in Wise County in the 1950s and has drilled approximately 3,700 gas wells in Wise and adjoining counties. Currently, Mitchell has approximately 2,000 active wells in the area.

The Texas Railroad Commission (TRC) regulates oil and gas operations in Texas and has designated the area as the Boonsville Bend Conglomerate Gas Field (BBC). In the early 1950s, oil and gas operations were governed by TRC's statewide rules. In November 1957, TRC enacted field rules that governed operations in the BBC (the BBC rules).

Most BBC gas is produced from the Atoka formation, which is from 5,000 to 7,000 feet below the ground. However, gas also exists in areas as shallow as 500 feet. The Atoka formation is found in Paleozoic age rock. Overlying the Paleozoic rock is Cretaceous period rock. In Wise County, the Trinity aquifer exists within the Cretaceous formation. An aquifer is an underground rock stratum with sufficient permeability to permit movement of water through it. The Trinity aquifer consists of two separate aquifers, the Paluxy, which is found at depths of 90 to 120 feet, and the Twin Mountain, which is found at depths of 200 to 400 feet. Appellees' wells draw water from the deeper, Twin Mountain aquifer.

Appellees bought their land at different times between 1977 and 1993. In April 1987, the Bartletts sued Mitchell, alleging that it was responsible for polluting the water under their land. The other appellees filed their lawsuits in January 1995, and all of the cases were consolidated.

Appellees alleged that their water had been polluted by twelve of Mitchell's gas wells. Appellees contended that Mitchell placed insufficient surface casing on the wells, failed to properly repair casing leaks in the wells, failed to place cement over potentially productive zones in the wells, failed to properly plug one of the wells, and failed to properly cement surface casing in one of the wells. Appellees alleged causes of action for nuisance, negligence, trespass, violation of TRC and BBC rules, and fraud. They also sought punitive damages based on alleged gross negligence, fraud, and malice.

The jury found in appellees' favor on all five of their causes of action. The jury also found that appellees had brought suit within two years of discovering the cause of their damages. The jury awarded appellees compensatory damages for mental anguish, physical pain, "discomfort, annoyance, and inconvenience," out-of-pocket expenses, property damages, and diminished value of real estate.

The jury found that Mitchell's conduct constituted gross negligence and was committed with malice. The jury awarded appellees punitive damages of $200,000,000, allocated equally among each individual. The trial court rendered judgment for appellees on the jury's verdict, and this appeal followed.

Mitchell raises 31 points of error on appeal, contending that:

• the trial court improperly denied Mitchell's motion for judgment notwithstanding the verdict (MJNOV) because the claims for nuisance, negligence, TRC and BBC rules violations, and trespass of all appellees except the Bartletts are barred by limitations;

• the trial court improperly denied Mitchell's MJNOV because the fraud claims of all appellees except the Bartletts and the Barans are barred by limitations;

• the evidence is legally and factually insufficient to support most of the jury's findings;

• the trial court improperly denied Mitchell's motion for new trial;

• the punitive damages award is excessive and violates Mitchell's constitutional rights; and

• the trial judge was disqualified from presiding over the trial.

II. Statute of Limitations.

A. The trial court improperly premised the statute of limitations jury issue on when appellees discovered the cause in fact of their injury.

In its second point, Mitchell contends that the jury question on when appellees discovered their injury was improper. Jury Question 6 asked the jury:

Do you find that the plaintiffs listed below filed their lawsuits within two years of the date the plaintiffs discovered, or in the exercise of reasonable diligence, should have discovered the cause in fact of the pollution, if any, and the injury arising from the pollution, if any? [Emphasis added.]

Mitchell contends that this question is improper because, under the discovery rule, the test is not discovery of the cause of the injury; rather, the test is discovery of the injury itself. We agree.

Appellees' personal injury and property damage claims are governed by the two-year statute of limitations and were required to be brought within two years from the date of accrual. See TEX. CIV. PRAC. REM.CODE ANN. § 16.003(a) (Vernon Supp. 1997); Hues v. Warren Petro. Co., 814 S.W.2d 526, 529 (Tex.App. — Houston [14th Dist.] 1991, writ denied) (negligence, trespass, and nuisance claims based on alleged gas leaks and improper brine disposal subject to two-year statute of limitations); Matysek v. Medders, 443 S.W.2d 929, 929-30 (Tex.Civ.App. — Amarillo 1969, writ ref'd n.r.e.) (two-year statute of limitations applicable to cause of action for damages caused by pollution of sub-surface strata of fresh water); William R. Keffer, Drilling for Damages: Common Law Relief in Oilfield Pollution Cases, 47 SMU LAW REVIEW 523, 532 (1994) (claims for permanent injury to property are barred in most jurisdictions if brought more than two years after discovery of the injury).

Ordinarily, the statute of limitations begins to run when a particular cause of action accrues. See S.V. v. R.V., 933 S.W.2d 1, 4 (Tex. 1996). Determining what rule of accrual to apply is a question of law. See Moreno v. Sterling Drug, Inc., 787 S.W.2d 348, 351 (Tex. 1990); Willis v. Maverick, 760 S.W.2d 642, 644 (Tex. 1988). Generally, a cause of action accrues when a wrongful act causes an injury, regardless of when the plaintiff learns of the injury. See Murray v. San Jacinto Agency, Inc., 800 S.W.2d 826, 828 (Tex. 1990); Moreno, 787 S.W.2d at 351.

However, Texas courts have applied the discovery rule of accrual to causes of action for damage to property. See Bayouth v. Lion Oil Co., 671 S.W.2d 867, 868 (Tex. 1984); Hues, 814 S.W.2d at 529. The discovery rule is a judicially constructed test used to determine when a plaintiff's cause of action accrued. When applied, this rule operates to toll the running of the limitations period until the plaintiff discovers or through the exercise of reasonable care and diligence should have discovered the " fact of injury." Moreno, 787 S.W.2d at 351, 357 (emphasis in original); see also Bell v. Showa Denko K.K., 899 S.W.2d 749, 753-54 (Tex.App. — Amarillo 1995, writ denied) (discovery rule defers accrual until "an injury . . . is discovered, or until the party acquires knowledge of facts which, in the exercise of reasonable diligence would lead to discovery of the injury."); Hoover v. Gregory, 835 S.W.2d 668, 671 (Tex.App. — Dallas 1992, writ denied) (discovery occurs when plaintiff has knowledge of facts that would cause a reasonably prudent person to make inquiry that would lead to discovery of cause of action).

Regarding claims for permanent injury to property, Texas law is especially clear that the action accrues and the statute of limitations begins to run "upon discovery of the first actionable injury and not on the date when the extent of the damages to the land are fully ascertainable." Bayouth, 671 S.W.2d at 868; Hues, 814 S.W.2d at 529. An action for permanent damages to property must be brought within two years from the time of discovery of the injury. Bayouth, 671 S.W.2d at 868. Moreover, knowledge of facts that could cause a reasonably prudent person to make an inquiry that would lead to discovery of the cause of action is "in the law equivalent to knowledge of the cause of action for limitation purposes." See Bayou Bend Towers Council v. Manhattan Constr. Co., 866 S.W.2d 740, 747 (Tex.App. — Houston [14th Dist.] 1993, writ denied).

This rule is borne out by common sense. "Permanent injuries to land result from an activity of such a character and existing under such circumstances that it will be presumed to continue indefinitely; the injury must be constant and continuous, not occasional, intermittent or recurrent." Bayouth, 671 S.W.2d at 868. Because a permanent injury to land is continuous and ongoing, a rule that did not require limitations to begin running upon discovery of the injury would invite abuse. For instance, an injured party might be tempted to wait an indefinite period before ascertaining the full extent of damages and filing suit, in hopes of maximizing a potential damages award.

In this case, appellees obtained a jury finding that the injury to their property was "ongoing and continuous." All of the damages appellees claimed flowed from the alleged injury to their property. Thus, appellees' claims were for permanent injury to property, and their respective causes of action accrued for limitations purposes when they first discovered or should have discovered their injuries.

Because a plaintiff's cause of action for damage to property accrues when the plaintiff discovers or should have discovered the fact of injury, the trial court erred when it asked the jury in Question 6 to determine when the plaintiffs discovered the cause, if any, of their polluted water.

Appellees assert that Mitchell has waived its complaint about the improper submission of Question 6 because:

• "although Mitchell states in its brief that it objected to question 6, it did not. . . ." Appellees contend that Mitchell only objected to the trial court's jury instruction defining "discovery" and that the court adopted Mitchell's proposed definition;

• Mitchell invited the erroneous instruction by telling the trial court that it could instruct the jury to determine when appellees discovered the cause in fact of their injuries rather than the injuries themselves.

These arguments mischaracterize the record.

The record shows that Mitchell made the following objections, among others, to Question 6:

• "Question No. 6 . . . instructs the jury that the Plaintiff is not deemed to have discovered the injury until the injury is complete when, in fact, under the laws of Texas, all the Plaintiff need know is that some injury has occurred";

• "by instructing the jury that the Plaintiff does not discover until the Plaintiff knows the cause, in fact, of the harm, the Court's instruction implies that the Plaintiff must . . . know as part of the cause . . . know the identity of the person causing the injury when Texas law does not require any such knowledge";

• "[Texas law] requires only that the Plaintiff have knowledge of such facts as would incite a reasonably prudent person to make inquiry as to the cause in fact of the injury."

We hold that Mitchell did not waive its complaint regarding Question 6. To the contrary, Mitchell's objections tracked Texas law and properly preserved error on this issue. See TEX.R. CIV. P. 274. The trial court's defective submission of Question 6 was reversible error. See TEX.R. APP. P. 44.1(a); Reinhart v. Young, 906 S.W.2d 471, 473 (Tex. 1995). We sustain Mitchell's second point.

Ordinarily, the remedy for a trial court's submission of a defective material fact issue like Question 6 is to remand the case for a new trial with instructions to submit the issue properly to the jury. See Spencer v. Eagle Star Ins. Co., 876 S.W.2d 154, 157 (Tex. 1994). However, as we discuss in the next section, the evidence conclusively establishes that all of the appellees except Richelle Wallace and the Bartletts discovered their injuries more than two years before they sued Mitchell. As a matter of law, all of the remaining appellees' claims that are governed by the two-year statute of limitations are time-barred, and there is nothing for the jury to decide with respect to this issue. Accordingly, in the interest of judicial economy we review the evidence in light of the proper inquiry:

Did appellees file their lawsuits within two years of the date they discovered, or in the exercise of reasonable diligence should have discovered, the fact of their injury?

B. The claims of all appellees except Richelle Wallace and the Bartletts that are governed by the two-year statute of limitations are time-barred.

1. Most appellees discovered their injuries long before they filed their lawsuits.

In its fifth point, Mitchell contends that the trial court improperly overruled Mitchell's MJNOV because, as a matter of law, the evidence established that the claims for nuisance, negligence, TRC and BBC rules violations, and trespass of all appellees except the Bartletts are time-barred. Under the circumstances of this case, we hold as a matter of law that all of the appellees except the Bartletts and Richelle Wallace discovered the injuries underlying their causes of action more than two years before they filed their lawsuits in January 1995. Thus, with these exceptions, appellees' claims are barred by the two-year statute of limitations.

The record contains the following evidence about when appellees discovered their injuries:

a. The Gilberts

• In 1978 Shelva Gilbert, who was then Shelva Wallace, first noticed that the

Wallaces' water had a "real dreadful odor" that was like rotten eggs. The water was undrinkable.

• In 1979 or 1980, the Wallaces knew that their neighbor, Mr. Russ, also had water that smelled bad and was undrinkable.

• Before he married Shelva in July 1991, Quention Gilbert knew the water on their property was bad.

b. The Reynoldses

• The Reynoldses first contacted TRC regarding their water problems by 1978 — within a year or two of when they moved onto their property. At that time, they knew that their water smelled like rotten eggs and had "other" problems.

• The Reynoldses could see a Mitchell gas well from their property. By 1978 or 1979, William Reynolds suspected that Mitchell might be a cause of their bad water problems.

c. The Drurys

• When the Drurys purchased their property in 1985, the seller told Robert Drury not to drink the water because it had gas in it. But Robert thought he could drill another well and get good water because two of his neighbors had done that.

• Drilling a new well (in 1986 or 1987) didn't work — the new well developed a "real bad odor" after three to four months.

d. G.W. and Betty Manning

• Immediately after the Mannings' well was drilled in 1979, G.W. Manning knew the water smelled terrible, like rotten eggs. It also tasted bad.

• Mr. Lewis, the man who drilled the well, told G.W. in 1979 that G.W. had both gas and water in his well.

e. Ronnie and Gina Manning

• In 1980, when the Mannings moved onto their property, the water on the property smelled bad, and the Mannings could not use it for drinking or cooking.

f. The Wallaces

• In 1988, Tommy Wallace knew the water on his property smelled like rotten eggs. With this knowledge, he moved from Arlington, Texas back onto the property in August 1993.

• Richelle Wallace never knew the water smelled bad until around August 1993. She and Tommy married in August 1989 and lived in Arlington, Texas for four years. They never had water problems before they moved onto the property in approximately August 1993.

g. The Barans

• The Barans noticed the smell of their water when they moved onto their property in November 1991.

• In the spring or summer of 1992, Carrie Baran noticed corrosion on many types of metal things in and around the Barans' home.

It is beyond dispute that all of these appellees except Richelle Wallace knew that their water smelled like rotten eggs and was unfit for human consumption for between three and sixteen years before they filed their lawsuits against Mitchell in January 1995. As a matter of law, this knowledge would cause a reasonably prudent person to make an inquiry that would lead to discovery of the cause of action and is therefore equivalent to knowledge of the cause of action. See Bayou Bend Towers, 866 S.W.2d at 747; Hoover, 835 S.W.2d at 671. Thus, the statute of limitations began running when appellees discovered that they could not use their water because of its terrible smell and taste. Moreover, the statute of limitations began running more than two years before January 1995. Accordingly, the claims of all appellees except Richelle Wallace and the Bartletts that are governed by the two-year statute of limitations are time-barred.

2. The fraudulent concealment doctrine does not save appellees' claims.

Appellees contend that, regardless of when they knew of their bad water, Mitchell fraudulently concealed their causes of action from them. Appellees point to the jury's finding of fraudulent concealment on Mitchell's part. In its fourth point, Mitchell contends that appellees failed to establish fraudulent concealment because there is no evidence that appellees believed or relied on allegedly fraudulent statements made by either Mitchell or TRC. We agree.

In determining a "no evidence" point, we are to consider only the evidence and inferences that tend to support the finding and disregard all evidence and inferences to the contrary. See Catalina v. Blasdel, 881 S.W.2d 295, 297 (Tex. 1994); In re King's Estate, 150 Tex. 662, 244 S.W.2d 660, 661-62 (1951). A "no evidence" point of error may only be sustained when the record discloses one of the following: (1) a complete absence of evidence of a vital fact; (2) the court is barred by rules of law or evidence from giving weight to the only evidence offered to prove a vital fact; (3) the evidence offered to prove a vital fact is no more than a mere scintilla of evidence; or (4) the evidence establishes conclusively the opposite of a vital fact. See Juliette Fowler Homes, Inc. v. Welch Assocs., 793 S.W.2d 660, 666 n. 9 (Tex. 1990); Robert W. Calvert, "No Evidence" and "Insufficient Evidence" Points of Error, 38 TEX. L.REV. 361, 362-63 (1960). To withstand a "no evidence" challenge, circumstantial evidence still must consist of more than a scintilla. The jury may not infer an ultimate fact from evidence that merely gives rise to any number of inferences, none more probable than the other. See Blount v. Bordens, Inc., 910 S.W.2d 931, 933 (Tex. 1995).

Fraudulent concealment is based on the doctrine of equitable estoppel. Where applicable, fraudulent concealment estops a defendant from relying on the statute of limitations as an affirmative defense to the plaintiff's claim. Borderlon v. Peck, 661 S.W.2d 907, 908 (Tex. 1983). Unlike the discovery rule, which determines when the limitations period begins to run, the doctrine of fraudulent concealment tolls or suspends the running of limitations after it has begun because the defendant concealed from the plaintiff facts necessary for the plaintiff to know that he had a cause of action. See DiGrazia v. Old, 900 S.W.2d 499, 504 (Tex.App. — Texarkana 1995, no writ). But the estoppel effect of fraudulent concealment is not permanent. Knowledge of facts, conditions, or circumstances that would cause a reasonable person to make inquiry leading to the discovery of a concealed cause of action is in the law equivalent to knowledge of the cause of action for limitations purposes. See Bayou Bend Towers, 866 S.W.2d at 747; Stephens v. James, 673 S.W.2d 299, 303 (Tex.App. — Dallas 1984, writ ref'd n.r.e.).

The elements of fraudulent concealment are: (1) the existence of the underlying tort; (2) the defendant's knowledge of the tort; (3) the defendant's use of deception to conceal the tort; and (4) the plaintiff's reasonable reliance on the deception. See DiGrazia, 900 S.W.2d at 502, 503; Arabian Shield Dev. Co. v. Hunt, 808 S.W.2d 577, 585 (Tex.App. — Dallas 1991, writ denied). The party asserting fraudulent concealment — in this case, appellees — bears the burden of establishing each of these elements. See Weaver v. Witt, 561 S.W.2d 792, 793 (Tex. 1977); Arabian Shield, 808 S.W.2d at 584.

The gist of fraudulent concealment is the defendant's active suppression of the truth or failure to disclose when the defendant is under a duty to disclose. The plaintiff asserting fraudulent concealment must have reasonably relied on the defendant's lies or improper silence. See DiGrazia, 900 S.W.2d at 503. Once a plaintiff knows or should know of the deceit, reliance is no longer reasonable, and the tolling effect ends. See Arabian Shield, 808 S.W.2d at 585.

In this case, the record does not contain any evidence of appellees' reliance on either Mitchell or TRC, whether reasonable or otherwise. To the contrary, the evidence shows as follows:

a. The Gilberts

• Before 1988, Shelva Gilbert realized that she might have gas in her water well. She knew that John Wallace (her then husband) believed the water was probably unsafe and took water samples to various places for testing.

• Before 1988, Shelva signed a petition to sue Mitchell because of the water problems.

• John called Mitchell about the bad water. Someone from Mitchell tested the water and said the bad water "wasn't Mitchell's problem," but John did not believe Mitchell. Just before he died in January 1991, John told Shelva that Mitchell was "sweeping everything under the rug" and that he had a lawyer who was going to represent them. By that time, Shelva was also suspicious that Mitchell was the cause of their bad water.

• In 1991, Shelva knew that her tenant (Mr. Covington) had complained to TRC about water problems.

• Before they married in July 1991, Shelva told Quention Gilbert about the problems she and John Wallace had had with their water, including her suspicion that Mitchell was to blame for the problems.

• Quention never had any contact with Mitchell or the TRC concerning the Gilberts' water. There is no evidence that Shelva contacted TRC.

b. The Reynoldses

• The Reynoldses first contacted TRC regarding their water problems by 1978. At that time, they knew that their water smelled like rotten eggs and had "other" problems.

• The Reynoldses could see a Mitchell gas well from their property. By 1978 or 1979, William Reynolds suspected that Mitchell might be a cause of their bad water problems.

• In 1979 or 1980, Patricia Reynolds and some of her neighbors attended a meeting where they discussed their concerns about the water problems. One of the neighbors indicated that he thought the bad water was caused by a gas well in the area.

• In the mid-1980s, Patricia attended a second neighborhood meeting. At that meeting, a Mr. Nelon said he was going to sue Mitchell because of problems with his water. At that time, Patricia wondered if Mitchell was the cause of problems with the water in her well.

• Before 1992, Patricia was aware that other people in her area had had "battles" with Mitchell about their water. In November 1992, Carrie Baran told Patricia that Carrie was going to write to Phil Gramm and complain about water contamination in the area.

• In November 1992 Patricia made a formal complaint to TRC about their water condition. The Reynoldses expected TRC to resolve the problem.

• Neither of the Reynoldses ever spoke with anyone at Mitchell about their water or relied on anything Mitchell said about it.

c. The Drurys

• In 1986 or 1987, the Drurys' neighbors told Robert Drury that they suspected Mitchell was the cause of the bad water.

• By early 1987, the Drurys had contacted TRC about their water. [RR 8:2741-44] Robert relied on TRC to help him out.

• TRC took some samples of the Drurys' water. In March 1987, TRC informed Robert by letter that the rotten egg smell and taste could be caused by bacteria, but that there was "no evidence to suggest oilfield-related contamination in this case."

• After receiving the report, Robert did not consider his complaint resolved to his satisfaction. He had no further contact with TRC.

• Robert testified that TRC's letter "kind of pointed in another direction," i.e., away from Mitchell. However, Robert later contacted Mitchell about his bad water because he was suspicious that a nearby Mitchell oil and gas well might be causing it. Ray Stephens from Mitchell went to the Drurys' property and discussed the matter with Robert. Stephens said he didn't think Mitchell was causing the water problems, but Robert didn't believe him. Robert was

pretty sure that Mitchell was causing the problems.

• On January 7, 1992, Stephens sent Mitchell a letter indicating that Robert had serious bacterial contamination in his water. Stephens advised Robert to take a sample of his water to the Fort Worth health department for analysis. Robert did not follow this advice.

• After receiving the January 1992 letter from Stephens, Robert had no further contact with Mitchell. Dorothy Drury did not testify.

d. G.W. and Betty Manning

• Mr. Lewis, the man who drilled the Mannings' well, told G.W. Manning in 1979 that G.W. had both gas and water in his well. [RR 9:2792-93, 2827] Lewis informed G.W. that his property was surrounded by Mitchell wells and that some of them had been leaking for up to ten years.

• From 1979 on, G.W. believed Mitchell was the cause of the Mannings' bad water problems.

• Neither of the Mannings or their sons ever talked with anyone from Mitchell. There is no evidence that the Mannings had any contact with or relied on TRC. Betty did not testify.

e. Ronnie and Gina Manning

• In 1980, Ronnie believed Mitchell was responsible for the problems the Mannings were having with their water. Ronnie had always believed that Mitchell was at least partially responsible for the problems. Ronnie had known many people who "over the years" had tried to get TRC or Mitchell to do something about their water problems.

• Ronnie never had any contact with anyone from TRC or Mitchell about the water problems. Gina did not testify.

f. The Wallaces

• John Wallace (Shelva Wallace Gilbert's first husband) was Tommy Wallace's father. Before John died in 1991, he may have told Tommy that the gas wells in the area were turning the water bad. Tommy knew that some of the wells in the area were Mitchell wells.

• No one from Mitchell ever made any representations to Tommy about his water quality or the cause of his water problems. There is no evidence that Tommy or Richelle Wallace had any contact with TRC.

g. The Barans

• Around April 1992, Carrie Baran became concerned about the Barans' water problems. Her neighbors told her they were very sure that Mitchell was responsible for their bad water.

• After talking with Patricia Bartlett, Carrie suspected that gas in Mitchell's wells might be the cause of the Barans' water problems. Carrie contacted the EPA, TRC, the Texas Water Commission, Mitchell, and Senators Phil Gramm Kay Bailey Hutchison.

• In November 1992, Carrie wrote a letter to Senator Phil Gramm about the water problems. In her letter, Carrie told Gramm that "concerned citizens" blamed Mitchell for their water problems and that there had been numerous battles with Mitchell over the water problems.

• After Carrie contacted Mitchell, Brad Hansen from Mitchell told Carrie he believed her water problems were naturally occurring. Carrie did not believe Hansen; she was convinced that Mitchell was responsible.

• Carrie visited TRC to see what was in TRC's files regarding complaints about Mitchell. She found complaints filed by the Reynoldses, the Bartletts, and other individuals who were not parties to the underlying suit.

• Someone from TRC also visited Carrie's property. On June 7, 1993, TRC sent Carrie a letter telling her that TRC could not make a "conclusive determination" about her water problems. When she received this letter, Carrie did not feel like she had been "helped" by TRC.

• Carrie testified that, as a citizen, she relies upon what TRC has told her. Yet, even after reading the public records concerning the complaints made against Mitchell and TRC's and the Water Commission's conclusions that they

couldn't determine the cause of the water problems, Carrie still was not convinced that Mitchell was not causing the problem.

• John Baran did not testify.

A few appellees stated that they relied on TRC to resolve their water problems, but this is not evidence that they relied on what TRC told them. For instance, although Robert Drury and Carrie Baran relied on TRC to resolve their water problems, neither of them was satisfied with TRC's reported findings. Further, neither of them relied on their communications from TRC. After hearing from TRC, Robert Drury still suspected Mitchell and contacted Mitchell about his bad water. Despite Mitchell's representations that it was probably not causing the water problems, Drury remained "pretty sure" that Mitchell was causing the problems. Likewise, after reading reports from TRC and the Water Commission that concluded they could not determine the cause of her water problems, Carrie Baran was still not convinced that Mitchell was not causing the problems. She "knew there was more to it."

Appellees contend that Mitchell concealed facts necessary for them to know that they had a cause of action — either through Mitchell's direct communications with the appellees or through the information Mitchell provided TRC — and that they did not know until 1994 that gas from Mitchell's wells was in their water. The record shows, however, that many of appellees — Quention Gilbert, both Manning families, and the Wallaces — had no contact with Mitchell or TRC about their water problems. In addition, the Drurys and Barans contacted both Mitchell and TRC but did not believe either entity.

In essence, then, appellees' contention is that the estoppel effect of fraudulent concealment tolled the running of limitations until they were able to obtain definitive test results from some source showing that gas from Mitchell's wells caused their water problems. This contention is incorrect, and such a rule would discourage plaintiffs from diligently pursuing their claims. "The statutory period does not await a plaintiff's leisurely discovery of the full details of the alleged scheme." Humphrey v. J.B. Land Co., 478 F. Supp. 770, 772 (S.D.Tex. 1979). Rather, the dispositive inquiry is when the plaintiff knew enough that a reasonable person in the plaintiff's place would disbelieve the defendant's denial of improper conduct. At that point, reliance becomes unreasonable, and the tolling effect ends. Arabian Shield, 808 S.W.2d at 586.

Instead of demonstrating reasonable reliance, the evidence in this case conclusively establishes just the opposite: appellees either did not contact Mitchell or TRC, or they did not rely on what they were told. Because there is no evidence that appellees reasonably relied on TRC or Mitchell, and because the evidence instead conclusively establishes that appellees did not believe or rely on either entity, we need not determine whether appellees established the other three elements of fraudulent concealment. See Juliette Fowler Homes, 793 S.W.2d at 666 n. 9. The doctrine of fraudulent concealment will not save appellees' claims from limitations in this case. We sustain Mitchell's fourth point.

3. The continuing tort doctrine does not save appellees' claims.

Appellees also contend that their claims are saved from the limitations bar by the continuing tort doctrine because the jury found that the injury to their property was "ongoing and continuous." One exception to the statute of limitations in Texas is the continuing tort doctrine. In some types of cases, if a defendant's tortious conduct and the plaintiff's resulting injury are ongoing and continuing, the statute of limitations begins to run as of the date the tortious conduct ceased. See Newton v. Newton, 895 S.W.2d 503, 506 (Tex.App. — Fort Worth 1995, no writ) (holding that doctrine applies to claim for intentional infliction of emotional distress). However, the continuing tort doctrine does not apply in this case because all of appellees' claims were for damages arising from permanent injury to land.

Appellees sued Mitchell for nuisance, negligence, violation of TRC and BBC rules, trespass, and fraud. Texas courts have not applied the continuing tort doctrine to these causes of action if they are based on permanent injury to land. Instead, courts have held that the continuing tort doctrine does not apply. See, e.g., Tennessee Gas Transmission Co. v. Fromme, 153 Tex. 352, 269 S.W.2d 336, 337 (1954) (negligence and trespass claims based on permanent injury to land must be brought within two years of date of first actionable injury); Austin N.W. Ry. Co. v. Anderson, 79 Tex. 427, 15 S.W. 484, 485 (1891) (where a nuisance is permanent and continuous, the damages should all be estimated in one lawsuit); Hues, 814 S.W.2d at 529 (claims for negligence, gross negligence, trespass, nuisance, and misrepresentation based on permanent injury to land are time-barred if not brought within two years of first injury).

The character of an injury as either permanent or temporary is determined by its continuum. Permanent injuries to land result from an activity of such a character and existing under such circumstances that it will be presumed to continue indefinitely; the injury must be constant and continuous, not occasional, intermittent or recurrent. Temporary injuries, however, have been found where the injury is not continuous, but is sporadic and contingent upon some irregular force such as rain.

Bayouth, 671 S.W.2d at 868; accord Atlas Chem. Indus., Inc. v. Anderson, 524 S.W.2d 681, 684 (Tex. 1975)

In this case, the jury found that the injury to appellees' property was "continuous and ongoing," or permanent, under Texas law. As we have previously noted, "[a]n action for permanent damages to land accrues, for limitation purposes, upon discovery of the first actionable injury and not on the date when the extent of the damages to the land are fully ascertainable." Bayouth, 671 S.W.2d at 868 (citing Tennessee Gas Transmission Co., 269 S.W.2d at 337). A lawsuit based on permanent injury to land must be brought within two years of the time of discovery of the injury. Bayouth, 671 S.W.2d at 868. Appellees discovered their injuries between three and sixteen years before they filed their lawsuits; thus, their claims are time-barred.

The cases appellees cite are inapposite because they did not involve permanent injuries to land. See Newton, 895 S.W.2d at 506 (intentional infliction of emotional distress); Upjohn Co. v. Freeman, 885 S.W.2d 538, 542 (Tex.App. — Dallas 1994, writ denied) (products liability and negligence).

Nonetheless, appellees contend that their claims did not just involve permanent injuries to land because they recovered damages for personal injuries as well as for injuries to real and personal property. We are unpersuaded by this argument for several reasons.

First, the jury's finding that appellees' injuries were "ongoing and continuing" forms the basis of their argument that continuing tort doctrine applies to their claims. But the jury found that the injury to appellees' property was ongoing and continuing. This finding buttresses Mitchell's contention that all of appellees' claims involved permanent injuries to land; it does nothing to further appellees' position.

Second, appellees asserted throughout trial that their damages were caused by Mitchell's permanent pollution of the water underlying their property. For instance:

• During his opening statement, appellees' counsel argued, "[a]s we sit here today, as we come tomorrow, as we were here yesterday, the pollution continues. It's an everyday thing."

• During his closing argument, appellees' counsel stated, "[i]t's about the pollution of water . . . that can't be replaced, can't be replenished, can't be restored. Once polluted, it's lost." . . . "Mitchell has destroyed our water." . . . "These people live surrounded by [the pollution] every day."

• Carrie Baran testified that the pollution on her property and the property of the other appellees continues day in and day out, "[e]very day, 365 days a year."

Appellees based all of their claims for damages, ranging from mental anguish to diminished value of their property, on their contention that Mitchell had permanently injured their property by polluting their water supply. In short, appellees' "personal" injuries, if any, were caused by the permanent injury to their water supply. Thus, all of appellees' claims involved permanent injury to their land.

Third, when the continuing tort doctrine is applied, a plaintiff can only recover damages for the two years immediately preceding the date the lawsuit was filed. See Bayouth, 671 S.W.2d at 869. Appellees did not limit the relief requested to the damages they incurred during the two-year period before they filed their lawsuits. Rather, the court submitted appellees' requested damage questions that asked the jury to find all damages — including future damages — that appellees allegedly sustained because of Mitchell's conduct. Having elected to submit their case on the theory of permanent injury to property, and having received an award based on that theory, appellees are limited to that theory on appeal. See Davis v. Campbell, 572 S.W.2d 660, 662 (Tex. 1978) ("parties are restricted on appeal to the theory on which the case was tried"); B.L. Nelson Assoc. v. City of Argyle, 535 S.W.2d 906, 910 (Tex.Civ.App. — Fort Worth 1976, writ ref'd n.r.e.) (same).

Because the record conclusively establishes that appellees' claims were for permanent injury to land and because the jury found that the injuries to appellees' property were ongoing and continuing, appellees were required to file their lawsuits within two years from the time they first learned or should have learned of their injuries. The continuing tort doctrine will not save appellees' claims from limitations.

4. Our holding that most of appellees' claims are time-barred does not violate their constitutional rights.

Appellees also argue that applying the statute of limitations to their claims would violate federal and state constitutional due process and equal protection doctrines and the open courts provision of the Texas Constitution. Appellees do not elaborate on how their constitutional rights would be violated except to state that application of limitations "would effectively preclude any claims that challenged Mitchell's tortious conduct." This argument is without merit.

Applying the statute of limitations does not completely shield Mitchell's conduct from challenge; it only precludes claims that are not brought within two years of the time a plaintiff discovered or should have discovered that his or her water was tainted. Application of limitations does not even preclude all of the appellees' claims in this case. We have held that Richelle Wallace's and the Bartletts' claims are not barred by limitations.

Moreover, the open courts provision does not act to toll running of the statute of limitations until a plaintiff discovers his cause of action. Rather, it requires only that a plaintiff be given a reasonable time to discover the injury and bring suit before the cause of action is barred by limitations. See Morrison v. Chan, 699 S.W.2d 205, 208 (Tex. 1985); Bayou Bend Towers, 866 S.W.2d at 745; see also Computer Assocs. Int'l v. Altai, Inc., 918 S.W.2d 453, 458 (Tex. 1996) (open courts provision not violated unless application of limitations is unreasonable and arbitrary when balanced against the purpose and basis of the statute). A primary purpose of limitations is to prevent litigation of stale claims. See S.V., 933 S.W.2d at 6. Thus, preclusion of a legal remedy alone is not enough to justify a judicial exception to the statute. See id.

In this case, the appellees whose claims are barred knew of their injuries between three and sixteen years before they filed their lawsuits. We hold as a matter of law that application of the statute of limitations will not violate those appellees' constitutional rights. Point five is sustained. In light of our disposition of points four and five, we need not consider Mitchell's first and third points.

III. There is no evidence to support the jury's finding of fraud.

In its twelfth point, Mitchell contends that there is no evidence to support the jury's fraud finding.

Jury Question 5 asked the jury whether Mitchell committed fraud against appellees, and the jury responded in the affirmative. The question contained definitions of fraudulent concealment and fraudulent misrepresentation. Both of these types of fraud contain a common element: reliance by the injured party. See Trenholm v. Ratcliff, 646 S.W.2d 927, 930 (Tex. 1983); DiGrazia, 900 S.W.2d at 502.

In section II(B)(2), infra, we sustained Mitchell's fourth point because the record does not contain any evidence that appellees relied on Mitchell's alleged deception, whether directly or through TRC. That holding is also dispositive of this point as to all appellees except the Bartletts. Because there is no evidence that these appellees relied on Mitchell's alleged deception, they failed to prove at least one element of their fraud claim. Accordingly, we sustain Mitchell's twelfth point as to all appellees except the Bartletts.

For clarity, we summarize our holdings thus far:

• There is no evidence to support the jury's finding of fraud against Mitchell as to all appellees except the Bartletts; and

• The other claims of all appellees except Richelle Wallace and the Bartletts are barred by limitations.

IV. There is no evidence that Mitchell caused appellees' alleged injuries.

Because of the "rotten egg" smell to their water, appellees contend that Mitchell's failure to properly maintain its gas wells caused hydrogen sulfide to migrate from the gas wells into the Trinity aquifer, the source of appellees' water supply, and that the hydrogen sulfide caused appellees' injuries. (Hydrogen sulfide is often associated with a "rotten egg" odor.) In its thirteenth through sixteenth points, Mitchell contends there is no evidence that gas from Mitchell's wells was the source of the hydrogen sulfide in appellees' water and therefore no evidence that Mitchell caused any of appellees' injuries.

In determining a "no evidence" point, we are to consider only the evidence and inferences that tend to support the finding and disregard all evidence and inferences to the contrary. See Leitch v. Hornsby, 935 S.W.2d 114, 118 (Tex. 1996); In re King's Estate, 244 S.W.2d at 661-62. If there is more than a scintilla of such evidence to support the finding, the claim is sufficient as a matter of law, and any challenges go merely to the weight to be accorded the evidence. See Leitch, 935 S.W.2d at 118. "More than a scintilla of evidence exists when the evidence supporting the findings as a whole, 'rises to a level that would enable reasonable and fair-minded people to differ in their conclusions.'" See Merrell Dow Pharmaceuticals, Inc. v. Havner, 953 S.W.2d 706, 711 (Tex. 1997) (quoting Burroughs Wellcome Co. v. Crye, 907 S.W.2d 497, 499 (Tex. 1995)).

To recover from Mitchell, appellees were required to show that hydrogen sulfide from Mitchell's wells migrated into appellees' water and caused their injuries. See Morgan v. Compugraphic Corp., 675 S.W.2d 729, 731 (Tex. 1984) (plaintiff must plead and prove both that defendant's conduct caused an event and that event caused plaintiff's injuries). Causation cannot be established by mere guess or conjecture; it must be established by evidence of probative value. See McClure v. Allied Stores of Texas, Inc., 608 S.W.2d 901, 903 (Tex. 1980).

In du Pont v. Robinson, 923 S.W.2d 549, (Tex. 1995), the Texas Supreme Court held that, in addition to showing that an expert witness is qualified to testify under TEX.R. CIV. EVID. 702, the proponent of expert testimony bears the burden of establishing that the expert's testimony is both relevant to the issues in the case and based upon a reliable foundation. See Robinson, 923 S.W.2d at 556. Unreliable evidence is of no assistance to the trier of fact and is therefore inadmissible under Rule 702. See id. at 557 (citing Kelly v. State, 824 S.W.2d 568, 572 (Tex.Crim.App. 1992)). The issue in Robinson was admissibility of evidence, but the Texas Supreme Court recently ruled that Robinson also applies to a no evidence review of scientific evidence. Havner, 953 S.W.2d at 714. If the foundational data underlying expert opinion testimony are unreliable, "the expert's scientific testimony is unreliable and, legally, no evidence." Id. at 714.

Appellees relied on the testimony of Dr. Randy Bassett, a geochemist who specializes in "isotope geochemistry," and Joe Neal, a petroleum engineer, to establish that both their water and Mitchell's wells contained hydrogen sulfide and that the hydrogen sulfide in their water came from Mitchell's wells. We hold that neither Dr. Bassett's nor Neal's testimony is evidence that the hydrogen sulfide in appellees' water came from Mitchell's wells.

1. Joe Neal's testimony is not evidence of causation.

Neal's testimony centered around his conclusion that Mitchell had not properly maintained its gas wells and that, as a result, gas from some of Mitchell's wells probably leaked into the Trinity aquifer and appellees' water supply. But Neal did not testify that gas from Mitchell's wells was found in any of appellees' water. More importantly, Neal did not testify that either Mitchell's wells or appellees' water supply contained hydrogen sulfide. Thus, Neal's testimony is no evidence that hydrogen sulfide from Mitchell's gas wells migrated into appellees' water supply and caused their injuries.

2. Dr. Randy Bassett's testimony is not evidence of causation.

Dr. Bassett testified about the chemical composition of appellees' water and the gas in Mitchell's wells. Dr. Basset formed his opinions after reading unspecified "published literature" and reviewing the chemical test data on appellees' water wells and four of Mitchell's gas wells generated by appellees' experts, Mitchell's experts, and TRC. After reviewing all of these materials, Dr. Bassett concluded "with a reasonable degree of scientific certainty" that there was gas in appellees' wells and that the gas in Mitchell's four wells had the same chemical makeup as the gas in appellees' water wells.

Dr. Bassett also testified that he had concluded to a reasonable degree of scientific certainty that hydrogen sulfide was in appellees' water. In addition, he testified that, "from the analyses that I've seen, it's very clear that hydrogen sulfide is a natural component in the Mitchell gas." Dr. Bassett would not say, however, that Mitchell's wells were a "probable" source of the hydrogen sulfide in appellees' water. Instead, he equivocated, stating only that the hydrogen sulfide "certainly could have come" from Mitchell's gas wells.

In addition, the "analyses" on which Dr. Bassett based his conclusions consisted of a single test run on two of Mitchell's wells, which indicated that hydrogen sulfide was present in those wells on July 9, 1976. In August 1976, a second test was run on the two wells, the results of which indicated that hydrogen sulfide was not present. Although Dr. Bassett testified that he also took other "analyses" into consideration in concluding that Mitchell's wells contained hydrogen sulfide, he did not indicate what they were.

An expert's use of the "magic" words "reasonable degree of [scientific] certainty," alone, will not establish causation. See Havner, 953 S.W.2d at 712. Unless the expert provides supporting facts, his bare conclusion is not evidence. See id. at 712 ("expert's bald assurance of validity is not enough"). We hold that the 1976 test results, which showed only that hydrogen sulfide was present in two of Mitchell's gas wells on a single day and was absent a month later, are no more than a scintilla of evidence. Such evidence is legally insufficient to allow a reasonable juror to conclude that hydrogen sulfide from Mitchell's gas wells polluted appellees' water supply. See id. at 713.

Likewise, Dr. Bassett's testimony on "isotopic signatures" is not evidence that hydrogen sulfide was present in Mitchell's gas or appellees' water. Dr. Bassett testified that he had examined the isotopic signatures of the gas in appellees' water wells and Mitchell's wells. He concluded that the isotopic signature of the gas in Mitchell's wells was "essentially identical to" that of the gas in appellees' wells. Dr. Bassett supported his conclusions with "produced gas analysis tables" that he had prepared using all the chemical data on Mitchell's gas wells and appellees' water wells produced by both sides' experts. These tables do not indicate that hydrogen sulfide or sulfur in any form was present in Mitchell's gas or appellees' water, nor do the tables show that hydrogen sulfide or sulfur was absent. Thus, the tables and Dr. Bassett's testimony based on them are no evidence that Mitchell's gas wells contained hydrogen sulfide, much less that hydrogen sulfide migrated from Mitchell's wells to appellees' water. Moreover, a TRC study on which Dr. Bassett also relied heavily specifically indicates the absence of hydrogen sulfide from one of Mitchell's gas wells that was allegedly polluting appellees' water supply.

In addition, Dr. Bassett only reviewed isotopic data from four of Mitchell's gas wells. Although he admitted the possibility that the gas in all the wells in the area would have similar, if not identical, isotopic signatures, Dr. Bassett did not gather any evidence at all about the chemical make-up of the other gas wells in appellees' vicinity. The record shows that Mitchell was one of 22 operators with gas wells in the area, and it operated less than half the wells.

Dr. Bassett testified that the scientific method requires an expert to ask

How many different possibilities exist for a given circumstance? . . . You find all the information you can about each one of these hypotheses or possibilities, and evaluate them, or reduce them to one that is the highest probability . . . . [a] scientific probability.

Yet Dr. Bassett did not rule out the possibility that the hydrogen sulfide in appellees' water could have come from a source other than Mitchell's gas wells. As noted above, he did not rule out the possibility that the hydrogen sulfide could have come from gas wells other than Mitchell's. He even inferred that his failure to sample non-Mitchell wells made his investigation less than "scientific."

In addition, evidence was presented at trial that the hydrogen sulfide in appellees' water wells could have been caused by bacteria introduced on the surface of the wells (rather than by gas migrating into the aquifer). Dr. Bassett did not rule out the possibility that surface bacteria could have caused the hydrogen sulfide in appellees' wells. An expert who is trying to find the cause of something should carefully consider and rule out alternative causes. See Robinson, 923 S.W.2d at 559. Dr. Bassett's failure to rule out other causes of the presence of hydrogen sulfide in appellees' water renders his opinion "little more than speculation." Id. Mere guess or conjecture is not probative evidence. McClure, 608 S.W.2d at 903.

Because Dr. Bassett did not present more than a scintilla of evidence that hydrogen sulfide was present in Mitchell's gas wells, and because Dr. Bassett did not rule out the other possible causes of hydrogen sulfide in appellees' water, his testimony is no evidence that hydrogen sulfide from Mitchell's wells had polluted appellees' water. Consequently, there is no evidence that Mitchell's conduct caused appellees' injuries, and appellees are not entitled to recover any damages from Mitchell. We sustain Mitchell's thirteenth, fourteenth, fifteenth, and sixteenth points. We also sustain Mitchell's twenty-second and twenty-third points, in which it complains that the exemplary damages award is improper. In light of our holdings with regard to these points, we need not consider Mitchell's remaining points.

V. Conclusion.

For the foregoing reasons, we reverse the trial court's judgment and render judgment that appellees recover nothing from Mitchell. See TEX.R.APP. P. 43.2(c).