Full opinion text
GEE, Circuit Judge: Plaintiffs Anadarko Petroleum Corporation (“Anadarko”) and its wholly-owned subsidiary Pan Eastern Exploration Company (“PEEC”) own the natural gas rights in certain sections of land in the Panhandle Field of Texas. Certain of the defendants own the oil and casinghead gas rights in the same sections. The plaintiffs sued, claiming that some defendants wrongfully produced their gas, and that other defendants purchased, processed, and transported the converted gas. The defendants argued that they did not take the plaintiffs' gas; also, they asserted that even if they did the taking was with the consent of the plaintiffs or their affiliate corporation Panhandle Eastern Pipe Line Company (“PEPL” or “the pipeline”) whose acts were binding on the plaintiffs. Some defendants made counterclaims against the plaintiffs and claims against PEPL. The jury rejected all claims by all parties. Plaintiffs and certain defendants appeal from various orders and from judgment entered on the jury verdict. We affirm in large part and remand in part for further proceedings. A. The Facts Beneath the surface of the Texas Panhandle is a vast hydrocarbon reservoir that since 1917 has produced huge quantities of oil and gas. The parties vigorously dispute the precise geological structure of this reservoir, but roughly speaking the upper formations — the so-called Red Cave, Brown Dolomite, and Moore County Lime formations — tend to be more productive of gas while the lower formations — the Granite Wash being the only one relevant here— tend to be more productive of oil. Throughout much of this region the right to produce gas from beneath a given tract has been severed from the right to produce oil and casinghead gas, so that one party may own the gas rights and another party the oil and casinghead gas rights throughout all horizontal formations beneath the very same piece of land. The division of gas rights from oil rights in the Panhandle Field of Texas apparently occurred many years ago, but only recently — under new market incentives to produce valuable “undedicated” natural gas — the oil and casinghead gas owners and their lawyers discovered that the division of rights is a rich deposit of legal ambiguities waiting to be mined. With the advent of new drilling and legal strategies, the so-called “split leases” have now for several years produced a steady flow of gas, controversy, and litigation. This massive case may be one of the last surges in production, reaching our Court as Texas administrators and courts have started to limit the flow. 1. The Parties PEPL held gas leases and owned the gas wells on the disputed sections for many years. Anadarko was its wholly-owned subsidiary, formed to hold and develop various properties including some of the leases at issue in this case. In the early 1970s, PEPL assigned the ownership of the wells to PEEC, a newly-formed subsidiary of Anadarko. This assignment, undertaken with the permission of the Federal Power Commission, allowed PEEC to charge more for the gas under federal regulations than could its super-parent PEPL. In the early 1980s, PEPL spun back a new corporate parent called Panhandle Eastern Corporation, turning itself into an affiliate rather than the parent of the plaintiffs. The scorecard then looked like this: Panhandle Eastern Corporation — the tiew corporate parent — owned all the stock of PEPL and of plaintiff Anadarko; Anadarko in turn owned all the stock of plaintiff PEEC. At all times relevant to this dispute PEPL was an affiliate of the plaintiffs. Plaintiff Anadarko was a fully-functioning exploration and production subsidiary of Panhandle Eastern Corporation with hundreds of employees and various corporate activities. Plaintiff PEEC, on the other hand, was what a non-lawyer would call a “shell” corporation. It owned several former PEPL wells but had only one full-time employee. Anadarko took care of the day-to-day management of PEEC’s affairs, while PEPL continued to operate PEEC’s gas wells (formerly PEPL’s wells) under a contract denominating it an “independent contractor” of PEEC. PEPL’s responsibilities for PEEC’s wells included administrative, legal, field, and office work, including regulatory, purchasing, insurance, taxation, negotiation, and accounting services on behalf of PEEC. As is typical of such a corporate “family,” Panhandle Eastern Corporation (the parent), Anadarko and PEEC (the plaintiffs), and PEPL (the affiliate) had interlocking but not identical officers and directors. The profits and losses of all members of the Panhandle Eastern family were aggregated for purposes of financial statements and periodic disclosures to the Securities Exchange Commission (Panhandle Eastern Corporation was a publicly held corporation). At all times, however, all corporate formalities were strictly observed and there is not a hint in the record of any legal irregularities in the way the various affiliates operated. The plaintiffs sued a number of defendants for conversion and other wrongs. The parties have divided the defendants into functional groups. The “operator defendants” are the persons and entities that drilled and produced the oil wells or that owned a working interest in the wells; this group includes the two main operators Hufo Oils and Ted True, Incorporated and their owners. The “processing defendants” are Mitchell Energy & Development Corporation and its subsidiary Liquid Energy Corporation; these defendants installed a small gas plant on the property that processed the gas produced by the operator defendants before it was sold to the pipeline defendants. The “pipeline defendants” are Houston Natural Gas and its subsidiaries Houston Pipe Line Company and Intra-tex Gas Company; along with PEPL, the operator defendants’ first customer, these pipeline companies purchased gas from the operator and processing defendants. Finally, the “bank defendants” are Canadian Commercial Bank and First National Bank in Albuquerque; they loaned money to Ted True, Inc. to drill and complete many of the wells at issue in this case. 2. Producing Casinghead Gas for Fun and Profit (a) The Beginnings of the Dispute In 1981, Hufo Oils, a partnership of defendants Lynn Hunt and Carl Foulds, obtained an oil and casinghead gas lease covering certain sections of land in which the plaintiffs held the gas rights. Hufo Oils assigned working interests to a number of investors. The investors elected Hufo Production Corporation, also owned by Hunt and Foulds, to be the operator of the leases. Hufo’s leases were initially limited by horizon to the Red Cave and the Granite Wash formations, the most shallow and deepest formations in the Panhandle Field. After drilling the first well, Foulds thought that the formations between the Red Cave and Granite Wash — including the Brown Dolomite and Moore County Lime— looked promising for oil and casinghead gas production. Hufo later obtained new leases from the landowners, giving it the oil and casinghead gas rights throughout all these formations. Meanwhile, Hunt approached PEPL. gas purchase agent John Gurche to see if the pipeline would be interested in purchasing the casinghead gas that Hufo planned to produce. Apparently, Hunt sought out PEPL because it was already purchasing gas from Hufo at other locations in the Panhandle Field. The pipeline was eagerly looking for new supplies of gas because at that time there was a tremendous unmet demand for gas. At first, Gurche and pipeline engineers studied Hufo’s proposal and rejected it because the initial projected volumes from the Red Cave and Granite Wash were too small to justify the necessary expenses in hooking up the wells to the pipeline. In late 1981, Hunt once again approached Gurche with an offer to sell casinghead gas to the pipeline. This time, the pipeline’s engineers projected total gas production more than 20 times greater than that of the original proposal; a satisfied Gurche then sent Hufo a proposed long-term casinghead gas contract. Before the contract was executed, however, Gurche’s supervisor Richard Dixon ordered that the pipeline cease all casing-head gas purchases until further notice. Dixon was the senior vice president of gas supply at PEPL; he was also the Chief Executive Officer of plaintiff PEEC and a member of plaintiff Anadarko’s board of directors. Dixon wanted to review the growing controversy in the Panhandle Field over “white oil” operators (such as Hufo) before proceeding with gas purchases from them. (b) The “White Oil” Controversy No geological formation contains either natural gas or crude oil alone; there are always some liquid hydrocarbons in a formation of gaseous hydrocarbons, and vice versa. In consequence, all oil wells produce at least trace amounts of gas and gas wells produce at least trace amounts of oil. Frequently, legitimate oil wells produce a great deal of gas, and gas wells a measurable amount of oil. There is no inherent or metaphysical distinction between oil wells and gas wells; it just depends on what comes up. Therefore, the State of Texas sensibly classifies oil wells and gas wells based on the ratio between the amount of oil produced and the amount of gas produced: if the gas-to-oil ratio (“GOR”) is greater than 100,000 cubic feet of gas per barrel of oil, the well is a gas well; if the GOR is less than or equal to 100,000 cubic feet of gas per barrel of oil, the well is an oil well. See Tex.Nat.Res.Code § 86.002(5)-(6) (1978). For several years before Hufo began drilling the wells in this case, operators had been buying up oil and casinghead gas rights in the Panhandle Field, then drilling putative oil wells and producing some oil along with large quantities of gas. At that time, the market for natural gas was booming, and the value of new “undedicated” gas was high. The primary economic value of these oil wells was in the casinghead gas. These new “casinghead gas wells” were controversial for several reasons. First, there was a general feeling (among the gas producers, anyway) that it was wrong for “white oil” operators to produce casing-head gas from the Brown Dolomite and Moore County Lime formations because, although there had long been oil production from these formations in a few locations around the Panhandle Field, the Brown Dolomite and Moore County Lime had historically been viewed as “dry gas” formations. Second, and more to the point, there were two completion and production practices associated with the “white oilers” that were of dubious legality. The first gave them their name: many such operators asserted that they could count the volatile hydrocarbon liquids called “white oil” created at the surface from casinghead gas with “low temperature extraction” or LTX units as “oil” for purposes of computing their GOR and classifying their wells as oil wells. See generally Colorado Interstate Co. v. Hufo Oils, 802 F.2d 133 (5th Cir.1986). Since a good deal of the volume of the gas can be converted to liquids, if this practice was legal a well that would otherwise be a statutory gas well could be converted into a statutory oil well by installing an LTX unit at the surface. The second dubious practice — typically referred to as the “high perforations” issue — was that the oil operators would drill their wells down to the deeper oil-producing strata, but would perforate (shoot holes in the casing and the rock) at higher levels (usually in the Brown Dolomite and Moore County Lime) where gas wells had been producing for years with little show of oil. These “high perf” wells frequently would qualify as statutory oil wells even without the use of LTX units. The cautious oil operators asserted that they were producing both oil and gas through the high perforations; they did so in order to meet the definition of casinghead gas as “gas or vapor indigenous to an oil stratum and produced from the stratum with oil.” Tex.Nat.Res.Code § 86.002(10) (1978). But the gas owners were suspicious that the oil produced by these wells was coming predominantly — if not totally — from the low perforations in the oil zones while the gas was coming from the high perforations in “their” gas zones. This case more concerns the second of these problems than the first: The operator defendants used an LTX unit on a few of their oil wells, but neither used the LTX liquids to qualify the wells as statutory oil wells nor asserted a right to do so. The central issue in the case, as will soon appear, was whether the defendants wrongfully produced the plaintiffs’ gas through high perforations in the Brown Dolomite and the Moore County Lime. (c) The Hufo Contract So Richard Dixon, vice president of gas supply for PEPL and CEO of plaintiff PEEC, decided to slow down, stand back and take stock of the situation. The pipeline was buying large quantities of putative casinghead gas from oil operators in the Panhandle Field, and it was planning on buying more. But the “white oil” practices were already being challenged in a. widely-publicized administrative proceeding before the Railroad Commission, and there were rumblings on the “high perforations” issue as well. In consequence, all purchases of casinghead gas ceased in February 1982 while PEPL officials tried to assess the risks involved in purchasing gas from “white oil” operators such as Hufo. Dixon and other PEPL officials met in early February to discuss the problems surrounding the purchase of casinghead gas. After the meeting, Dixon and the others decided to proceed with the Hufo contract. This decision and the circumstance surrounding it are the key to the defendants’ affirmative defenses. We will discuss it in detail below in section A.3. The contract between PEPL and Hufo called for a steady increase in Hufo’s cas-inghead gas production, with a commensurate, three-phase obligation on the part of PEPL to improve and expand its gathering facilities and to take or pay for larger and larger quantities of gas. The first phase involved minor improvements to pipeline’s facilities and purchases of up to 6,500 mcf per day. The second phase, beginning July 1982, was to involve a $350,000 expansion of facilities and the take or pay purchase of 10,000 mcf per day. And the third phase, beginning August 1983 and requiring FERC approval, was projected to cost PEPL in the range of $5 million to $9 million and to require the purchase of up to 60,000 mcf per day depending on supply. PEPL officials, well aware by this time that Hufo’s wells might drain the plaintiffs’ gas reserves, inserted a specially-crafted definition of casinghead gas designed to protect from and indemnify against claims based on the illegality of Hufo’s production practices. The parties signed the 15-year contract in mid-February 1982, and PEPL began purchasing gas from Hufo’s wells a few days later. Hufo drilled and completed several wells on the disputed sections. These wells passed the initial Railroad Commission GOR tests and were classified as oil wells. Thereafter, Hufo installed and operated a central LTX unit that removed liquids from the gas coming from all the wells. Hufo then mixed the extracted liquids with the crude oil and reported the total liquid amounts as the oil production on monthly Commission reports. This practice arguably was required by contemporaneous Commission rules and opinion letters. For reasons that are not clear from the record, Hufo’s performance did not satisfy the working interest owners, and they replaced operator Hufo with Ted True, Incorporated in November 1982. Mr. Ted True, not surprisingly the owner of Ted True, Inc., soon acquired all Hufo’s interest in the operations. (d) True’s Progress True began an ambitious drilling program. He drilled 76 wells, 21 of which became a part of this dispute. Once again, the True :wells were initially tested and passed GOR requirements without an LTX unit. While True continued to gather and to report as oil the LTX liquids from the five original Hufo wells, he did not install any LTX units on the 21 new wells. Defendant Billy Mack Gideon was an employee of True in charge of drilling, completing, and perforating the wells.' True gave him the authority to choose where to drill and where to perforate. Gideon did not have any ownership interest in the wells. Meanwhile, as True geared up for greater and greater gas production in late 1982 and early 1983, PEPL was facing a sudden turn-around in the gas market. The low supply and high demand of early 1982 disappeared, as demand slacked while new gas inventories — developed during the gas shortage — began to come on-line. Dixon ordered his buyers to stop new purchases and, in early 1983, PEPL sent telegrams to all producers telling them to defer new drilling and workovers that would increase their production. In its annual report, Panhandle Eastern Corporation (the parent corporation of PEPL and the plaintiffs) stated that the oversupply problem had become so serious that it “created uncertainties concerning the continued viability of the pipeline company.” Gurche met with True in early 1983 to find out about True’s development plans; True reported that he planned to develop the properties at the rapid pace called for in the contract. PEPL was in a bind: under the original Hufo contract, to which True had succeeded, it was obligated to complete a multi-million dollar improvement of its gathering facilities by late summer 1983 and to purchase up to 60,000 mcf of gas. If it failed to make the necessary improvements, it still had to pay True for the gas he made available. In consequence, PEPL set out to persuade True either to limit his production or to find another buyer. PEPL’s legal department considered various strategies for discouraging True from increasing production. It seized on Hufo’s failure to meet a June 1982 contractual deadline for development plans until November 1982, sending a letter to Hufo stating that it would not meet its third-phase obligations in August 1983 because of Hufo’s failure. Although the pipeline knew that True had replaced Hufo and in fact had discussed these matters with True, the letter went to Hufo because True had never formally succeeded to Hufo’s contract with the pipeline; and legal formality was part of the pipeline’s new strategy. Pipeline officials eventually decided in May or June 1983 to try to terminate the Hufo/True contract. In the meantime True, needing funds to finance his drilling operations, proposed to borrow money from the bank defendants. After reviewing independent engineering reports on True’s potential for gas production and visiting the drilling sites, defendant Canadian Commercial Bank (“CCB”) loaned True $15 million in June of 1983, funding half the loan then and the remainder during the next six months. Defendant First National Bank in Albuquerque, whose loan officer had originally explored the True deal and offered it to CCB, purchased a participation interest of $3 million in the True loan in August 1983. At about this time PEPL informed True that it would not begin phase three of the gas purchase contract because of asserted breaches of the agreement by Hufo and True. True calculated that although he could win a breach-of-contract suit against the pipeline, he could not afford the delay and cost of legal remedies. His financial pressures shot up during this time as his production income dropped to zero: first, PEPL closed its pipeline for repairs for several weeks and asserted immunity from financial liability by reason of the force majeure contract provision; second, the pipeline suspended payments for a time under appropriate contract provisions because True was being sued by another gas producer who claimed that he was stealing its gas on adjacent leases. True decided to seek other buyers for his gas rather than sue PEPL, and eventually he agreed to terminate the PEPL contract at the end of 1983. True entered into agreements with defendants Liquid Energy Corporation (“LEC”) and Houston Pipe Line Company to purchase his casinghead gas. During the negotiations, Bruce Withers, president and chief operating officer of LEC and president of the Transmission and Processing Division of LEC’s parent Mitchell, telephoned his friend Dan Kelley, then the vice president of gas supply for PEPL, to inform him of the negotiations between LEC and True and to check True’s representation that PEPL did not want the gas and was willing to cancel the contract when True found another buyer. Kelley confirmed that PEPL would release the gas as True had stated. In response to Withers’ question, Kelley stated that he had no objections to LEC’s proposed contract with True. Withers testified that he “certainly relied” on Kelley’s statements in proceeding with the True contract. LEC built a gas processing facility on the leases at a cost of $12 million. It bought some compressors from PEPL that were used in the plant; PEPL knew that LEC was building a gas processing plant and that the compressors were to be used to separate liquids from True’s casinghead gas. The plant began operation in December 1983, separating liquids in the casing-head gas from the gas stream. These liquids were mixed with the produced crude oil in storage tanks and, pursuant to advice from the Railroad Commission staff, counted as oil on Commission reports. Houston Pipe Line purchased the gas residues from the tailgate of the LEC plant. Defendant Intratex, an affiliate of Houston’ Pipe Line, later took over performance of the True contract. (e) True’s Regress True’s casinghead gas business struggled along for a time but never got off the ground. By early 1984, he was unable to pay trade creditors as accounts became due; and he fell behind on his loan payments to the bank defendants. The bank defendants then tried to exercise their contractual right to receive directly LEC’s purchase payments to True, but it is unclear from the record whether they actually received any money in this manner. In May 1984, True — both Mr. Ted True and Ted True, Inc., that is — went into bankruptcy, where they remained at the time of trial. As his financial difficulties increased, True became unable to maintain the wells properly so as to keep them producing at their full capacity. Wells in the Panhandle Field require regular maintenance to prevent paraffin from clogging the perforations. Paraffin is a thick, waxy substance that accumulates in conjunction with the production of crude oil. Moreover, oil wells must be regularly maintained and repaired because of mechanical problems such as broken pumps, holes in tubing, and parted rods. Any of these problems can reduce or eliminate a well’s oil production, but they tend to have a lesser effect on the well’s production of gas. Consequently, an oil well’s GOR tends to increase as maintenance declines; in other words, a poorly-maintained well can lose its statutory status as an oil well because it produces more than 100 mcf of gas per barrel of oil. Initially, True treated the wells for paraffin buildup about every other week, but his financial troubles caused him to cut back on the paraffin treatments to less than one a month, and finally to eliminate treating the wells altogether. In October 1984, an independent testing company performed GOR tests on nine True oil wells, eight of which are among • the 26 in dispute. The testing company counted only the crude oil that was produced and not any LTX liquids in calculating the wells’ GORs. Despite the wells’ deteriorating condition, all but one of them qualified as oil wells under the test, and, in the opinion of a witness from the testing company, the well that failed would probably have passed had it been properly maintained. After the Railroad Commission’s decision in the Phillips proceeding, in the spring of 1985, eight wells connected to an LTX unit on the lease were tested; five wells failed the test and the remaining tests were inconclusive. Finally, after this suit commenced the district court directed that the wells be re-tested or shut in. True was long in bankruptcy and was no longer operator of the wells. The working interest owners formed their own operating company but had little success with the wells as they were in bad condition, clogged with paraffin, with parted rods and pump jacks down. The investors treated and tested some of the wells but none passed the GOR tests. All wells were then shut in by order of the district court pending the outcome of this suit. There was some gas production between the time the plaintiffs brought suit and the time the wells were shut in. 3. Confusion Above: The Plaintiffs’ Ambivalence The defendants presented compelling evidence that the plaintiffs’ managers were aware that because of Hufo and True’s drilling and production practices the operator defendants might be producing gas that belonged to the plaintiffs, but nonetheless decided to proceed with the PEPL casing-head gas purchase contract because of PEPL’s overriding need for more gas. The defendants’ circumstantial evidence was elaborate and far-ranging, and we will not attempt to summarize all of it here. Instead, we focus on the behavior of three protagonists: Richard Dixon, CEO and Chairman of the Board of plaintiff PEEC, a director of Anadarko, and senior vice president of PEPL in charge of gas supply; Robert Allison, CEO and director of plaintiff Anadarko, director of PEEC, PEPL, and Panhandle Eastern Corporation, and group vice president of Panhandle Eastern Corporation; and Richard O’Shields, Chairman of the Board of Panhandle Eastern Corporation (the overall parent), PEPL, Anadarko, and director of PEEC (later Chairman of the Board of PEEC). Richard Dixon was in a tough situation. He was in charge of gas supply at PEPL and PEPL needed gas — lots of gas in a hurry. Dixon testified that he was “anxious” to find new sources of gas to meet the demands of his customers. Much of PEPL’s new gas supply was casinghead gas produced by independent “white oil” operators such as Hufo. In early 1982, Dixon learned of the Phillips proceeding, in which certain traditional gas operators were asking the Railroad Commission to forbid the use of LTX units and to prohibit high perforations in the gas zones. Dixon realized that his casinghead gas sources might be jeopardized by a ruling adverse to the independents, and this concerned him. So while his gas purchase agent John Gurche was in the middle of negotiating a contract with Hufo, Dixon ordered a moratorium on new casinghead gas purchases while he became better informed on the issues. Dixon, Gurche, and several other PEPL agents, lawyers, and engineers met in early February with a lawyer from Austin who was considered an expert on the Phillips proceeding and the “white oil” issues. Gurche memorialized the February meeting in a classic “smoking gun” memorandum to the file: Note to the File 2/8/82 Re: Hufo Oils — Potential Casinghead Gas Contract; Brent Lease Moore Co. Tx. Due to many questions surrounding cas-inghead gas in Texas which have arisen as a result of the Phillips petition to the Texas Commission, Mr. Matheson arranged for Mr. David Nelson of Akin, Gump, Hauer and Feld, law firm of Austin, Texas to come to Kansas City and brief Mr. Dixon and our company attorneys on the split lease situation in Texas with respect to oil (casinghead gas) and dry gas. Mr. Nelson, who represents some of the independents, outlined the arguments on both sides of the issue in depth. Later, as a result of the meeting it was decided that if Panhandle [PEPL] failed to follow through on its offer for the Hufo gas it would end up being marketed to someone else, and, if as Phillips contends, there is dry gas being eo-min-gled with the casinghead, then our dry wells under contract on that acreage [the plaintiffs’ wells] would be drained and we would lose the dry gas entirely. It was decided, therefore, to proceed with the Hufo contract, but to modify the definition of casinghead gas to state that it has to be in conformity with Texas Commission rules regarding gas oil ratios. Dixon was the senior PEPL official at that meeting, and presided over the decision to proceed; he was also the CEO of plaintiff PEEC. Like Dixon, Robert Allison was in a tough situation. He became concerned in the early 1980s that the independent “white oil” operators were perforating the dry gas zones in the Brown Dolomite. As a director of PEPL, he was concerned that the pipeline was buying stolen gas; as CEO of Anadarko, and as effective manager of PEEC’s wells under management contracts between PEEC and Anadarko, he was concerned that his gas was being stolen. Sometime in 1982, Allison began complaining. First, he called Richard Matheson, a PEPL vice president in charge of gas supply. Allison told Matheson that the pipeline should not buy casinghead gas from the “white oil” operators because they were stealing it from the legitimate gas owners. Matheson responded that PEPL did not intend to change its purchasing practices. Dissatisfied, Allison called Matheson’s superior, Dixon. Allison told Dixon that the pipeline was buying gas from Hufo and that Hufo was probably stealing the gas from Anadarko and Dixon’s own company PEEC; he suggested that the pipeline stop buying casinghead gas from the independents. Allison testified about Dixon’s response as follows: Q. And Mr. Dixon responded to you that the pipeline was going to continue doing this [buying casinghead gas from independents], didn’t he? A. Yes. Q. And he told you that if the pipeline didn’t buy the gas someone else would; isn’t that right? A. That was certainly his comment, well, you know, if Panhandle Eastern Pipe Line doesn’t buy it, some other pipeline would. So our rights, Anadarko’s rights would be hurt in either case. Understandably, Allison was still not satisfied. He turned to the top man in the Panhandle family of corporations, Richard O’Shields. Allison told O’Shields what that he believed that the pipeline was buying gas that was being converted from Anadar-ko and PEEC. He told O’Shields that the pipeline should stop buying casinghead gas from the “white oil” operators. Q. And Mr. O’Shields responded to you that he did not intend to change anything, didn’t he? A. Well, I can’t say he responded exactly that way. He listened to what I had to say, said he wasn’t, as I recall dimly, said he, you know, wasn’t aware of the details and that is the last I heard of it. In other words, there was really very little response, almost no response from him. It just died. Q. And he did not express any particular concern to you, did he, about this? A. No. Q. And as time went on you did not ever go back to the people at the pipeline again to raise this with them because you thought the subject was closed; isn’t that right? A. Yes. Q. And you believed the decision had • already been made; isn’t that right? A. Yes. Q. And the pipeline company was going to continue to buy this gas; correct? A. Yes. My assumption was that was the case. Allison instructed his staff to keep tabs on the activities of Hufo and True, whom he characterized as “crooks”; several years after PEPL’s need for gas had subsided and turned to a need to get rid of gas, Anadarko and PEEC brought this suit. 4- Confusion Below: Lessons in Law and Geology (a) Defining “Casinghead Gas” The parties agree that “casinghead gas” is, in the words of one of the plaintiff’s title documents, “gas produced with oil from an oil sand, whether obtained from the casing-head of the well or from separators or otherwise,” or in the words of the Texas statute, “any gas or vapor indigenous to an oil stratum and produced from the stratum with oil.” Tex.Nat.Res.Code § 86.002(10) (1978). Both sides agree that “sand” and “stratum” have the same meaning in this context. Each side, however, has a particular extension of the definition of “oil sand” or “oil stratum” that, if accurate, means that it wins the case based on undisputed facts. Also, each side has a fall-back argument based on disputed facts and the un-embellished definitions just quoted. The plaintiffs’ addition is based on various statutes, regulations, and title documents: they assert that an “oil sand” is a formation capable of producing oil in commercial quantities. In other words, the plaintiffs assert that even gas produced with small but non-trivial quantities of oil from the same interval does not qualify as casinghead gas because the value of the oil is too small to justify production from the well. According to the plaintiffs, only when a well produces oil in sufficient quantity for the oil production alone to make the well profitable can the operator then legally produce casinghead gas with the oil. On this basis, the plaintiffs boldly assert they have “title” to the entire Brown Dolomite and Moore County Lime formation because those formations are incapable of producing oil in commercial quantities. The defendants, on the other hand, assert that an “oil sand” or “oil stratum” is synonymous with an “oil reservoir.” They point out that the entire Panhandle Field is regulated by the Texas Railroad Commission as one reservoir, that all areas of the field (other than the topmost formation called the Red Cave) exhibit the same internal pressure and therefore are in communication with each other, and that the entire Panhandle Field is thus an associated reservoir of oil and gas. According to the defendants, any production of gas from a properly classified oil well in the Panhandle Field is ipso facto the production of casing-head gas, regardless of whether the oil comes from one horizon and the gas from another. Finally, both sides argue about the actual geology of the Brown Dolomite and Moore County Lime formations, and about whether the defendants produced more than trace amounts of oil from the perforations they made in those predominantly gas-producing formations. The plaintiffs argue that the defendants in fact produced gas and only traces of oil from those formations; and, of course, the defendants assert that they produced both oil and gas in the proper ratios from those formations. Thus, there are three possible definitions of casinghead gas. Each side posits one definition, and then falls back to a common one. (1) The plaintiffs assert that an “oil sand” must produce oil in commercial quantities. All parties acknowledge that the casinghead gas from the defendants’ wells was far more valuable than the oil produced, and that the wells did not produce enough oil to make them worth drilling for oil alone. Therefore, under their preferred definition the plaintiffs win. (2) The defendants assert that the entire Panhandle Field is an associated reservoir of oil and gas, that it is an “oil sand” or “oil stratum” within the meaning of the relevant definitions. Under this definition any properly classified oil well produces only casinghead gas. Therefore, because their wells were properly classified — at least there was sufficient evidence for the jury to conclude that they were properly classified — the defendants win. (3) Both sides acknowledge (at least implicitly) that the proper definition of “oil sand” or “oil stratum” may be any horizontal interval in any formation that produces any oil. If that definition is accurate, the only remaining question is the strictly factual one of whether the defendants’ wells produced enough oil with the gas from the high perforations in the Brown Dolomite and the Moore County Lime. Each side asserts that, based on the geological facts they presented at trial, they win under this neutral definition. (b) Geological Evidence At trial, the parties presented sharply conflicting evidence on the geological facts. The plaintiffs’ evidence tended to show: The Brown Dolomite and Moore County Lime formations are gas formations. While there are occasional small pockets of oil in the gas formations, the oil and gas in the Panhandle Field conform to the law of gravity, with the heavier oil in the lower formations and the lighter gas in the upper formations. The oil-gas contact — the horizontal level where the heavier liquid hydrocarbons and lighter gaseous hydrocarbons meet — is roughly uniform throughout the region, and occurs near the bottom of the Moore County Lime or the top of the Granite Wash. The defendants must have known that they were stealing the plaintiffs’ gas because they tried to hide their illegal drilling and producing practices: while their well logs show that they penetrated their putative oil wells throughout the gas formations above the oil-gas contact, their public filings with the Railroad Commission omit reference to most of these high per-formations. The defendants’ evidence tended to show: There are small but producible quantities of crude oil throughout the Brown Dolomite and Moore County Lime formations trapped in cracks and fissures in the rock. There is no uniform gas-oil contact in the Panhandle Field; gas and oil exist in varying ratios throughout the vertical column. Wells drilled in nearby areas showed producible quantities of oil throughout the Brown Dolomite and Moore County Lime. The plaintiffs’ gas wells — completed primarily in the Brown Dolomite and Moore County Lime but also in the Granite Wash — have long produced small quantities of crude oil (which the plaintiffs have failed to report to the Railroad Commission and to the royalty owners). The defendants’ well logs show markings both where the wells were actually perforated and where the operator defendants considered penetrating; although no one is now certain, the public filings with the Railroad Commission may be a more accurate description of where the wells are perforated than the markings on the well logs. The plaintiffs’ main expert recommended that a $1500 test be performed to determine at precisely what level certain wells were perforated, but the plaintiffs refused to pay for the test. In any case, the public filings clearly show perforations in the zones that the plaintiffs assert to be “gas” zones, so the defendants were not trying to hide anything. Finally, based on the plaintiffs’ own yearly Railroad Commission submissions of data on well pressure versus production, the steady pressure decline in the plaintiffs’ gas wells proves that there was no drainage from their wells throughout the period in which they claim the defendants were stealing huge quantities of their gas. B. The Trial, the Verdict, and the Appeal In the words of one defendant’s brief: “In the fall of 1985 PEEC and Anadarko sued everyone who had ever been remotely connected with the production and sale of oil and casinghead gas on [the disputed sections] ... with the notable exception of [their] sister corporation PEPL.” The plaintiffs’ main claim on the facts was for conversion of their gas, but the legal theories in the complaint ranged from trespass on personal property, common law fraud, civil conspiracy, and negligence to breach of the defendants’ leases and gas sales contracts, with the plaintiffs tacitly presenting themselves as third-party beneficiaries. Virtually every party in the case was a citizen or resident of Texas, and almost every claim was based on Texas law. Federal jurisdiction, however, was pegged on a claimed violation — apparently de rigueur in commercial litigation these days — of the federal RICO statute, with the remainder of the massive case dangling ponderously below, suspended by the doctrine of pendent jurisdiction. The defendants quickly brought in PEPL on third-party complaints, asserting among other things that PEPL was an “alter ego” or agent of the plaintiffs. They raised a host of affirmative defenses and brought counter-claims analogous to the plaintiffs’ offensive theories, alleging that the plaintiffs’ gas wells had illegally produced their casinghead gas. The defendants also cross-complained against each other, but wisely settled their differences (at least temporarily) in order to present a unified defense at trial. The parties agreed to divide the case into separate liability and damages phases, and the liability trial began in September 1987. The district court directed verdicts for the bank defendants and Billy Mack Gideon upon completion of the plaintiffs’ case-in-chief. After extensive testimony from the parties’ principals and from a parade of experts on both sides of the geological questions, the court submitted to the jury 42 pages of instructions and 53 special interrogatories, many of which required separate answers for each of the 29 defendants. The jury sent everyone packing. It rejected each of the plaintiffs’ claims and each of the defendants’ cross-claims. For good (if slightly inconsistent) measure, the jury found in favor of the defendants on all affirmative defenses — including entrustment, license, release, waiver, ratification, and estoppel — except the defense of laches. The plaintiffs bring this appeal primarily on the basis of insufficient evidence to support the verdict and errors in the jury instructions. The defendants together defend the jury verdict, and some bring cross-appeals that we discuss as necessary. C. The Unasked and Unanswered Question The most astonishing thing about this complex case is what has not been resolved. The jury was never asked the key questions: Did the defendants take the plaintiffs’ gas? If so, did the plaintiffs consent to the taking, thus absolving the defendants of conversion? If not, are the plaintiffs nevertheless bound by the consent of their authorized agents? If not, are they nevertheless bound (through some form of “corporate disregard” theory) by the consent of their corporate affiliate PEPL? As a matter of fact, the court did ask the jury these questions; but it asked them all at oncel Like the logic professor’s request for a yes-or-no answer to the question “have you quit beating your wife?,” the key interrogatory asked the jury a complex question: did the defendants convert the plaintiffs’ gas? Conversion was defined as taking without consent, and consent was defined to include consent by the plaintiffs or “their agents or alter egos, if any.” Thus, the jury’s finding of no conversion is profoundly ambiguous: did the jury find that the defendants did not take the plaintiffs’ gas? or did it find that the defendants took the plaintiffs’ gas but with plaintiffs’ consent? or did it find (in true lawyerly fashion) that if the defendants took the plaintiff’s gas, they did so with the plaintiffs’ consent? or did it find that although the plaintiffs did not consent, their “agents or alter egos” consented? Assuming that the taking issue was a question of fact for the jury under the neutral definition of “oil sand,” we would have gladly affirmed a jury finding that the defendants extracted only casinghead gas or, more precisely, that the plaintiffs had failed to prove that the defendants had extracted anything other than casinghead gas. The parties gave the stalwart jury a highly technical and almost unmanageable task: determine both the geological structure underlying the disputed tracts and the nature of the hydrocarbons produced from the defendants’ high perforations on the basis of elaborate, well-reasoned, convincing, and absolutely conflicting evidence and expert testimony. As bearers of the burden of persuasion, by definition the plaintiffs assumed the risk of non-persuasion. In other words, the jury reasonably could have thrown up its hands and determined that, whatever the truth of the matter, the plaintiffs had failed to establish that their version was true; therefore, the defendants were entitled to win. Unhappily, no such determination was made. More precisely, we cannot infer that it was made. Our duty is to consider the jury answers as a whole and to make sense of each in the context of the trial and the charge. Davis v. West Community Hospital, 755 F.2d 455, 465 (5th Cir.1985). If the jury had found no taking of the plaintiffs’ gas, that would have ended the case. It would not have been necessary to answer any other interrogatories except those on counterclaims. On the other hand, if the defendants took the plaintiffs’ gas but with their consent, then the affirmative defenses found by the jury are much more pertinent, and (as we shall see) they may be dispositive. Moreover, the plaintiffs recognized this part of the basic ambiguity in the conversion interrogatory and objected to it, and the district court assured them that it would rule on the taking question — what the parties have called the “title” issue — after the jury returned its verdict. Finally, the defendants do not contend on appeal that we should interpret the jury’s answer to the conversion issue as a finding that they did not take the plaintiffs’ gas. We therefore assume that in answer to the key interrogatory the jury determined only that if the defendants took the plaintiffs’ gas, the plaintiffs (or some person or entity whose decision was binding on them) consented to it. It remains to be seen whether the jury’s finding of consent and its benediction of the various affirmative defenses can support the take-nothing verdict judgment. D. Consent to the Taking The remaining issues are wound together in an analytic knot. Did the plaintiffs consent to the taking of their gas? Did their affiliate PEPL consent? Could PEPL consent for the plaintiffs, was it an agent of the plaintiffs or their “alter ego” such that its activities were binding on the plaintiffs? Are there sufficient reasons for disregarding the corporate distinctions between the plaintiffs and their parent and sister companies? Even if the jury finding of consent by the plaintiffs or its “agents or alter egos” is not legally supportable, are the plaintiffs nevertheless “estopped” from claiming conversion because of their own behavior or the behavior of their “agents or alter egos”? In this section of the opinion, we will begin to untangle the knot by determining whether the jury’s implicit finding of consent is supported in fact and law, and whether it alone can sustain the verdict. We hold that the finding of consent is supportable, and can sustain much of the verdict, but that the plaintiffs’ consent ended at the latest upon commencement of legal proceedings against the defendants. In the following section, we will determine whether the affirmative defenses can sustain the remainder of the take-nothing verdict. 1. Some Initial Problems with Consent (a) Remaining Ambiguities in the Interrogatory We have already commented on the ambiguities inherent in the jury’s finding of no conversion and held that we would construe it to be a finding of “taking with consent” rather than a finding of “no taking.” But a further ambiguity remains: did the jury find that the plaintiffs consented, or that their agents consented, or that their alter egos consented? The instruction would have permitted the jury to answer “no” if it found any or all of these. The question now is: what happens if we find only one or two but not all three of these possible findings legally sufficient? The parties have not raised or briefed this issue — the plaintiffs because they never admit even for the sake of argument that any of these findings is legally sufficient, the defendants because they never admit even for the sake of argument any is not sufficient. “Thus we have one more small failure of the adversary system, which relies so heavily on the parties to do the requisite research and analysis.” Mobil Chemical Co. v. Blount Brothers Corp., 809 F.2d 1175, 1183 n. 8 (5th Cir.1987). In contrast to their precise objection to the ambiguity on the taking issue, the plaintiffs did not object to this inherent ambiguity in the conversion charge. Instead they simply argued the “inclusion of a reference to agency or alter ego is not part of a proper definition of conversion” and that neither the agency nor the “alter ego” theories should be submitted to the jury. In subsequent objections to similar language in the instructions on the affirmative defenses, the plaintiffs did point out the problem of discerning whether the jury would find that the plaintiffs consented or that PEPL consented, and did ask generally for a separate instruction on agency and alter ego, but there was no mention of the need to distinguish between agency and alter ego. A party must ordinarily object precisely to the wording of jury instructions and interrogatories, and it is doubtful under these strict standards whether the plaintiffs’ objections were adequate. On the other hand, when a case is submitted to the jury on a general verdict, the failure of evidence or a legal mistake under one theory of the case generally requires reversal for a new trial because the reviewing court cannot determine whether the jury based its verdict on a sound or an unsound theory. See, e.g., Nowell ex rel. Nowell v. Universal Electric Co., 792 F.2d 1310, 1312 (5th Cir.), cert. denied, 479 U.S. 987, 107 S.Ct. 578, 93 L.Ed.2d 581 (1986); Smith v. Southern Airways, Inc., 556 F.2d 1347 (5th Cir.1977). Our Court has recognized that even a single question in a special verdict involving alternative theories has the same troublesome characteristic: the reviewing court cannot tell which theory formed the basis of the jury’s categorical answer. See Dougherty v. Continental Oil Co., 579 F.2d 954, 960 n. 2 (5th Cir.1978), vacated, 591 F.2d 1206 (5th Cir.1979). In consequence, we have reversed and remanded for a new trial when an interrogatory contained a complex question and the jury’s categorical answer could have been to either or both of the sub-questions. See Prudential Insurance Co. v. Morrow, 339 F.2d 411, 412 (5th Cir.1964). It seems likely that in the case of a potentially ambiguous general verdict all the complaining party must do to protect his rights is to object to the charge and the submission vel non of the questionable theory or theories; probably he need not object to the ambiguity inherent in its submission, as the ambiguity arises from the nature of general verdicts and no party has a right to a particular kind of verdict, general or special. Thus, if we analogize the submission of the conversion interrogatory to the submission of a general verdict on several alternative theories, we might conclude that the plaintiffs’ objection to the submission of each theory is enough to preserve for appeal the argument that the case must be retried because one theory is unsound and it is impossible to tell which theory was the basis for the jury’s answer. If, however, we apply the normal strict rules requiring objection to the form of the charge and interrogatories, we might conclude that the plaintiffs’ objections were inadequate to preserve the error and that the verdict can be sustained based on any one of the possible alternative grounds. We are thus confronted by a close and difficult question. Happily, we can leave its resolution to another day. Even on appeal the plaintiffs fail to argue that the cause should be reversed and remanded for retrial if we find any one of the alternate theories of consent legally insupportable. They maintain the all-or-nothing posture that none of the possible findings of consent was legally supportable; they do not argue in the alternative that if one is unsound the case must be remanded because we cannot determine if the jury relied on the unsound theory. We need not consider issues or arguments not raised in the appellant’s brief. See In re Texas Mortgage Service Corporation, 761 F.2d 1068, 1073-74 (5th Cir.1985). We conclude that we may affirm the jury’s finding of consent if any one of the underlying theories is legally and factually sufficient. (b) Consent: Element of Conversion or Affirmative Defense? In its definition of conversion, the district court instructed the jury that “the complaining party must prove that each offending party took the gas owned by the complaining parties without their assent.” The plaintiffs argue that the court erred by requiring that they prove lack of consent, rather than requiring the defendants to prove consent as an affirmative defense. The plaintiffs contend that consent is a species of the genus “assumption of the risk” and that under Rule 8(c) and Texas authority they were not required to show iack of consent. There may be some confusion in Texas law about which party must prove or disprove the existence of consent. After reviewing the cases cited by the defendants, however, we conclude that although the matter is not without doubt, the district court properly required the plaintiffs to show lack of consent. Conversion is a tort against the possession of property. Taking another’s property is not conversion; wrongful taking is. Older Texas authority and some modern cases speak of the owner’s lack of consent as the sine qua non of wrongful taking: When the word “conversion” is used to signify a tort, it may be defined as any distinct act of dominion wrongfully asserted over another’s personal property in denial of the owner’s rights or inconsistent with them. The essence of conversion is not the actual taking of the owner’s property or carrying it away; it is wrongfully depriving the owner of its use and possession. The taking must be wrongful, for without the element of wrong no tort can be committed and conversion does not occur. To be wrongful, the conversion [sic ] must be wholly without the owner’s sanction or assent, either expressly or impliedly. Staats v. Miller, 240 S.W.2d 342, 344 (Tex.Civ.App.—Amarillo), rev’d on other grounds, 150 Tex. 581, 243 S.W.2d 686 (1951). Thus, the Texas courts have held that for venue purposes the tort of conversion does not occur where the taking occurred if the taking was with consent; while these are venue cases, they broadly and categorically define the tort of conversion in terms of the lack of consent. See, e.g., Conlee Seed Co. v. Brandvik, 526 S.W.2d 795, 798 (Tex.Civ.App.—Amarillo 1975, no writ) (“there can be no conversion by the taking of property where the owner has expressly or impliedly consented to the taking”); American Mortgage Corp. v. Wyman, 41 S.W.2d 270, 272 (Tex.Civ.App.—Austin 1931, no writ) (“It is indispensable to the maintenance of the cause of action for conversion for the complaining party to prove that the original taking of the property was without his consent.”). Because only wrongful taking is conversion, and because the lack of consent establishes the wrongfulness of the taking, lack of consent is an element of the tort of conversion which the plaintiff must plead and prove.- Affirmative defenses usually have the structure of “confession and avoidance,” that is to say, they admit the initial sufficiency and completeness of the claim while asserting other grounds for avoiding the normal consequences of that concession. But if the defendant to a conversion claim was required to prove that the plaintiff consented, rather than requiring the defendant to show that although a conversion occurred he has some legal excuse for it, that would be requiring him to prove that no conversion occurred at all. This analysis is consonant with traditional common law treatment of the problem. Therefore, we conclude that the district court did not err in requiring the plaintiffs to show lack of consent. 2. Who Consented? At trial, the defendants presented three factual and legal bases for the finding that the plaintiffs consented to the taking of their gas. First, the defendants argued that the plaintiffs and their affiliate PEPL were “alter egos” and that the acts of PEPL were therefore binding on the plaintiffs. Second, the defendants contended that PEPL was an agent of the plaintiffs and that its acts were therefore binding on the plaintiffs as principals. Third, the defendants asserted that through their corporate management the plaintiffs themselves consented to the taking. Technically, of course, the last contention is also an agency argument, but it is quite different from the second. Presumably because of PEPL’s extensive knowledge of and participation in much of the activity of which the plaintiffs complain, most of the arguments on appeal focus on the first two theories, particularly the issue of corporate disregard and to a lesser extent the issue of the pipeline as an agent of the plaintiffs. It seems to us, however, that the issue is most easily resolved by focusing on the evidence of the plaintiffs’ own consent to the taking of the gas through their corporate management. Because of the nature of the consent issue, the other theories present difficulties. 3. Consent by the Plaintiffs The evidence recited above in section A.3 speaks for itself. It supports the conclusion that Dixon, head of plaintiff PEEC, decided to sacrifice the plaintiffs’ gas rights in order to advance the interests of their much larger and more profitable sibling PEPL; that Allison, head of plain,tiff Anadarko, objected to this strategy and complained to O’Shields; and that O’Shields, head of all the Panhandle companies, sided with Dixon and the interests of PEPL, making the final decision to allow PEPL to buy gas that all suspected was in part stolen from Anadarko and PEEC. Dixon, Allison, and O’Shields were the CEOs of PEEC, its parent Anadarko, and Anadarko’s parent Panhandle Eastern Corporation, respectively; and O’Shields was Chairman of the Board of most Panhandle companies. These men had the power and the duty to maximize profits for the entire family of corporations. Just as they once transferred PEPL wells to PEEC to increase the price that the family could receive for the gas, they had the power to risk gas losses by the plaintiffs in order to help PEPL in its time of gas shortage. Whether, as a matter of their duties of care and loyalty to the plaintiffs, they should have so decided is a different question, which we discuss briefly below. The point here is that they had the power and authority to act for the plaintiffs, even to their detriment; and they did so. To find that the plaintiffs consented to the taking, the jury did not need to find that the plaintiffs’ management was happy to give away the gas in return for an increased gas supply to the pipeline; it did not need to find that management was totally satisfied with this arrangement; it did not need to find that CEO Allison did more than acquiesce in the decision by Dixon and O’Shields. The word “consent” is used interchangeably in Texas tort jurisprudence with the word “assent”; both words connote acquiescence in a decision or a state of affairs. The jury needed only to conclude that the plaintiffs’ management accepted the loss of gas in order to further other corporate goals. There was evidence to support that conclusion, and we affirm it. 4. The Plaintiffs’ Arguments Against the Finding of Consent (a) The Part-Time Agent and the Adverse Agent The plaintiffs make two arguments against the jury’s finding that they consented. The first is that their corporate management’s knowledge of and acquiescence in the defendants’ taking should not be imputed to them because, in the words of their brief: even when the agency relationship is established, the acts of an agent will not bind the principal absent proof the agent was acting within the course and scope of its authority and in the best interest of the principal. * 5ji * * % It is equally well established in Texas, knowledge of an officer or director of one corporation that is acquired while acting as an officer or director of another corporation is not imputable to the first corporation. Furthermore, where an officer or director is adversely interested in a transaction, his or her knowledge is not imputable to the corporation at all, even though the officer is, at the time, the managing agent of the corporation. (Emphasis in original; citations omitted.) This breaks down into two contentions. (1) The knowledge of Dixon and O’Shields and others about the de