Full opinion text
MEMORANDUM DECISION BELOT, District Judge. Introduction This is an action for contribution both under Section 113(f) of the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA), 42 U.S.C. § 9613(f), and under state law, to recover response costs incurred by Morrison in connection with contamination from grain fumigant containing carbon tetrachloride at the Seoular elevator site in Salina, Kansas. Morrison seeks recovery of response costs incurred to date of trial, and a declaratory judgment concerning McShares’ liability for future response costs. The case originally included a cost recovery claim by Morrison under Section 107(a) of CERCLA, 42 U.S.C. § 9607(a), but the court entered summary judgment dismissing that claim on December 18,1996. (Doc.72). As the ease progressed toward trial, the court made a number of rulings. The ruling which bears directly upon some of the issues discussed and resolved herein is the court’s order precluding Morrison’s use of expert testimony at trial (Doc. 15). The court’s ruling became the subject of additional orders: Memorandum and Order of November 2, 1995 (Doe. 51); Memorandum and Order of December 18, 1996 (Doe. 72); and Order of August 19, 1997 (Doc. 86). The court did not retreat from its ruling excluding Morrison’s experts but did allow Morrison to offer lay opinion testimony pursuant to Fed. R.Evid. 701 (Doc. 72 at 16-18). The court’s exclusionary rulings notwithstanding, the court permitted Morrison to make both testimonial and written proffers of excluded testimony and to mark exhibits covered by the proffers so that the Court of Appeals will-have the benefit of both the evidence admitted and excluded. Tr. at 438-89 (Shepard); Doc. 91, Ex. 83; Tr. 273-96, 314 (Bean); Exs. 32-47; Tr. at 333-54 (Elder); Exs. 48-61C. The case came on for trial to the court on August 25-27, 1997. At the conclusion of Morrison’s case, the court took under advisement McShares’ motion pursuant to Fed. R.Civ.P. 52(c). Tr. at 560. The parties have submitted their suggested -findings of fact and conclusions of law (Does. 92, 93, 97, 98, and 101). Pursuant to Fed.R.Civ.P. 52(a), the court now makes its findings of fact and conclusions of law. Expert Evidence Revisited In its trial brief, Morrison reprises and expands upon its positions regarding the court’s exclusion of its use of expert evidence (Doc. 89 at 1-15). As Morrison candidly recognizes, the orders precluding its use of expert evidence resulted from its previous counsel’s repeated failures to comply with the rules of procedure, this court’s orders, and his own promises to make required expert disclosures. Nevertheless, Morrison continues to argue that it should be allowed to present expert evidence. To support an argument that it did identify experts as required, Morrison first notes that its previous counsel did file a designation of expert witnesses in the so-called 1992 case, No. 92-1641 (Doc. 20). According to the designation, three experts would testify, in substance, that the cause of the carbon tetrachloride contamination was consistent with the 1963 spill. Morrison’s designation was filed approximately two and a half months after the-deadline and after McShares filed a motion to preclude Morrison’s use of experts (Doc. 15). Now-retired Magistrate Judge Wooley granted McShares’ motion to preclude (Doc. 23). Ten days after filing its tardy expert designation, Morrison moved to dismiss the 1992 case (Doe. 21). Judge Theis granted Morrison’s motion to dismiss without prejudice on three conditions: (1) all completed discovery would be applicable in any second lawsuit; (2) Morrison was to pay McShares for attorney’s fees for McShares’ motion to prohibit Morrison’s use of expert witnesses; and (3) any refiling must be accomplished within 90 days (Doc. 36). Judge Theis’ order said nothing about whether the pleadings from the 1992 case would be applicable to a later-filed case. In the absence of an express condition, proceedings and pleadings in a prior case do not follow the parties into a refiled case. Morrison’s designation of experts in case no. 92-1641 did not therefore satisfy Morrison’s obligation to file a designation of experts in this case. Morrison next points to its former counsel’s statement at a May 22, 1995 conference to the effect that he had disclosed Morrison’s experts in the 1992 ease and that discovery carried over to this case. While it is true that there was no condition in the 1992 case dismissal that prohibited Morrison from using expert witnesses if the case was refiled, there was no expert discovery to be carried over to this case. Morrison does not contend that any of the experts named in the disallowed designation were deposed in the 1992 case. The designation itself, of course, does not constitute discovery. Morrison then argues that it has complied with the disclosure requirements of Rule 26(a)(2)(A) because two of the witnesses identified in the 1992 designation, Shepard and Christy, had been investigating the contamination at the Scoular site at Morrison’s request prior to the filing of the 1992 case. Morrison shared some of their work product with McShares. Morrison thus concludes that Shepard and Christy were properly identified as expert witnesses in this case as required by Rule 26(a)(2)(A). This argument could be valid only if the orders in the 1992 case are disregarded and the procedural requirements for expert witness disclosure in this case are ignored. Finally, Morrison contends that Shepard and Christy should be allowed to testify because a written report from them was not required under Rule 26(a)(2)(B) since neither were “retained or specially employed to provide expert testimony” within the meaning of the rule. Morrison likens Shepard and Christy to treating physicians and cites cases which hold that treating physicians can testify and give opinions on causation even though they have not provided written reports because they are not specially retained or employed. The court sees little, if any, merit in this analogy. Treating physicians seldom, if ever, are hired to testify against another treating physician. Shepard and Christy were not Morrison’s employees. Rather, Morrison retained their company, Kejr Science Group, to help investigate the contamination problem and, as MeShares points out (Doc. 90 at 3-4), to help Morrison prove up a case against MeShares, if necessary. While part of Shepard’s proffered testimony involved causation, it went farther. Just as a standard of care opinion goes beyond factual observations and opinions involving diagnosis, treatment and prognosis, the proof required of Morrison here goes beyond the expert opinions summarized in the designation filed in the 1992 case. Even if the document identified by Morrison’s former counsel at the May 22,1995 hearing, the June 1994 Cl Report prepared by Kejr, somehow could be analogized to a medical record, it does not purport to cover all the elements of Morrison’s proof (see Ex. 42, Section I, Executive Summary, pp. M002128-29). This is evident by comparing the Cl Report executive summary with Shepard’s proffered evidence (Ex. 83) and his testimony. Thus, the court cannot accept Morrison’s argument, premised on the treating physician analogy, that Shepard and Christy were not required to provide Rule 26(a)(2)(B) reports and therefore should be permitted to offer expert evidence. The parties and counsel are aware that the decision to disallow expert evidence has been troubling to the court. Because the court was acquainted with Roger Morrison in college and respects him, and because Mr. Morrison was without fault in the matter, the court personally wishes its decision could have been different. The decision, however, was not made on a personal basis. Rather, it was made on a legal basis, as all such decisions must be if the integrity of the judicial system is to be preserved. The court is not an adherent to “form over substance” application of rules. Nevertheless, Morrison’s prior counsel was a member of a respected law firm and was presumptively competent and responsible, yet his conduct was persistent, unexplained and inexcusable. MeShares was blameless and entitled to take the position it did. For the reasons stated herein and many times in previous orders, the court believes its decision was correct. The Parties 1. Plaintiff, Morrison Enterprises (Morrison), is a general partnership that owns land in Salina, Kansas, upon which its predecessor, Morrison Grain, Inc., constructed grain storage facilities and related improvements in the 1950s. Morrison operated these facilities until 1980 when they were leased to Scoular Grain, which has operated the facilities since that time. The Morrison facilities usually will be referred to herein as the Scoular site. 2. Defendant, MeShares, Inc., (MeShares) is the successor in interest to Research Products Company which supplied grain fumigants containing carbon tetrachloride to Morrison and Scoular for use at the Scoular site from the 1950’s until approximately 1985. 3. The Scoular site' is composed of a grain storage elevator consisting of three sets of multiple concrete silos, five metal grain storage tanks, three grain storage Butler buildings (known as “flat” storage) and other ancillary outbuildings. Pl.’s Ex. 4 and Def.’s Ex. WW, attached hereto. The total storage capacity during the relevant times was 10,096,000 bushels. The area surrounding the site is essentially rural, with a significant amount of farm ground and just a few residences. Tr. at 56-59 (Ellis), 617 (Morrison). Use of Fumigants at the Scoular Elevator 4. McShares sold two types of liquid grain fumigants to Morrison — Max High Life and Max Kill 10, both of which were registered pesticides. Max High Life was approximately 80% carbon tetrachloride. Max Kill 10 was approximately 70% carbon tetrachloride. Tr. at 605 (Allen); Trial Exs. 15 and 16. In applying liquid grain fumigants, Morrison followed the instructions on the labels, the procedures recommended by McShares’ personnel and literature provided by McShares. Tr. at 77, 84, 97-99; Trial Exs. 77 and 78 (Ellis). Morrison purchased liquid grain fumigants only from McShares. Tr. at 69, 71 (Ellis), 187-88 (Gebhart). 5. Max High Life was used to fumigate the grain in the concrete silos. It was stored in a tank on top of the elevator. The customary fumigation process was to fill each silo about half way with grain and spray the fumigant on the grain surface. Then the silos were filled to the top and the spraying process was repeated. Max High Life, which was heavier than air, penetrated all the way to the bottom of the silo. It was absorbed by the grain kernels and any infestation inside the kernel, as well as outside infestation, was killed. Tr. at 75-78 (Ellis). 6. The grain in the silos was “turned” periodically by moving grain between silos when conditions dictated, e.g., in hot summer months to reduce the temperature of the grain. On these occasions, the fumigation process was repeated. When turning was not required, or when there was no evidence of infestation, the grain was not refumigated. Tr. at 78-79,168 (Ellis). 7. The grain in the steel tanks and the Butler buildings (flat storage) was fumigated with Max Kill 10. Max Kill 10 was not stored on site, because the quantity of grain stored in the steel tanks and Butler buildings was much greater than that stored in the silos. Instead, McShares brought a tanker truck to the facility and Max Kill 10 was pumped directly from the tanker. The formulation of Max Kill 10 differed from Max High Life, which caused Max Kill 10 to move more slowly through the grain mass. Morrison’s employees applied the fumigant by spraying it over the surface of the grain. After applying Max Kill 10, the tanks and buildings then were sealed, and the fumigant, which percolated to the bottom, was allowed to work. Tr. at 80-84 (Ellis). The floors of the Butler buildings were concrete of unknown thickness. Tr. at 59, 63 (Ellis). Occasionally, spot treatment of flat storage was required. It was sometimes difficult to determine the exact area to be spot fumigated and dispersal of the fumigant could be uneven, with high concentrations in some areas and low concentrations in others. Tr. at 143-44 (Ellis). For this reason, Morrison discontinued use of Max Kill 10 to treat the flat storage when a pelletized product known as phostoxin became available. Tr. at 85-86 (Ellis); 179-80 (Gebhart). 8. Morrison did not use liquid grain fumigant for rodent control while Eldred Ellis was elevator manager from 1959 to 1979. However, Ellis admitted that he knew of others in the industry who used carbon tetrachloride to control rats and that he never asked Morrison’s employees whether they were doing so. Tr. at 103, 106, 141 (Ellis). Morrison used either a powder or a liquid product sold by McShares specifically for rodent control. Tr. at 102-03 (Ellis), 610 (Allen). 9. McShares has records of the amounts of liquid grain fumigant sold to Morrison and Seoular for certain periods from 1975 to 1984. They show that for that period, McShares sold Morrison and Seoular a total of 80,766 gallons of liquid grain fumigant, or an average of 8,076 gallons per year. Trial Exs. MM and UUUU. James Allen, McShares’ president, testified that based upon the general application formulas on the label — 1.5 to 2 gallons per thousand bushels of grain — the average annual usage of grain fumigant indicated by McShares records, Trial Ex. UUUU, would treat about 7 million bushels of grain. Tr. at 607. Allen speculated, without benefit of records from 1954-1975, that Morrison used approximately 240,000 gallons of fumigant from the period 1954 through 1985 and that Morrison’s use of fumigants during the period from 1959 to the middle 1960s would have been greater than the average figure set forth above. Tr. at 568, 589 and 595-600. 10. Seoular ceased using any liquid grain fumigant which contained carbon tetrachloride by January 1, 1985. After that date, no carbon tetrachloride was used at the Seoular elevator site for any purpose. Tr. at 172-73 (Gebhart). The Spill 11. The parties have stipulated that in about November 1963, a spill of Max High Life occurred at the Seoular site when a McShares employee, Walter Bernhardt, was preparing to unload fumigant from the front compartment of a trailer owned and operated by McShares. Pretrial Order (Doc. 80 at 2) and Trial Ex. 15. 12. McShares had purchased the trailer from Morrison sometime shortly after 1959. Tr. at 568, 608 (Allen). After McShares purchased the trailer, it cleaned out any rust from inside the trailer, visually inspected the trailer, repaired hose connections, installed new couplings, and made certain the trailer was in operating condition. Some of the work was done by Morrison’s shop. Tr. at 571-75, 609-10 (Allen). 13. The trailer contained three compart.ments which could hold a total capacity of approximately 2,400 gallons of grain fumigant. The capacity of the front compartment was 250-300 gallons. Tr. at 10-11 (Bernhardt). 14. The Max High Life used in the concrete elevator was stored in a 5,000 gallon tank located on top of the elevator approximately 130 feet above ground level. Tr. at 14 (Bernhardt), 72, 93 (Ellis). To fill the storage tank, a hose from McShares’ trailer was hooked to a quick coupler at the bottom end of a steel pipe which ran up the side of one of the silos. The fumigant was pumped up the pipe to the tank on top. Tr. at 17 (Bernhardt), 73-75 (Ellis); Trial Exs. 8 & 9. The pipe connected near the top of the storage tank. The fumigant was drawn off for use in the concrete silos through a pipe connected to the bottom of the storage tank. Tr. at 75 (Ellis). The storage tank was vented to the atmosphere. Tr. at 75 (Ellis), 187 (Gebhart). 15. The pump used to move the Max High Life from the trailer to the storage tank was located on the back of McShares’ tractor unit. On the day of the spill, Bernhardt recalled that he hooked the hose between the pump and the elevator pipe, Tr. at 18, 19, 39 (Bernhardt), but had not yet completed hooking up the hose between the pump and the front compartment of the trailer which held the grain fumigant. Tr. at 21, 22, 42, 45-46 (Bernhardt). Bernhardt normally did not turn on the pump until the hoses were all hooked up, and would not open the valves until the pump was turned on so that “you would have no pressure coming back at you” from fumigant which might be in the pipe. Tr. at 21, 22 (Bernhardt). At the time of the spill, Bernhardt knew that Eldred Ellis, the elevator manager, was standing near the valve on the pipe at the elevator, but he did not know whether the valve had been turned on. Tr. at 47, 51 (Bernhardt). Ellis testified that he had not turned on the valve on the pipe at the time of the spill. Tr. at 93,135 (Ellis). 16. The spill occurred while Bernhardt was down on his knees trying to hook up the hose from the pump to the front compartment of the trailer. Tr. at 49 (Bernhardt). The hose had a quick coupler which was pushed into the connection on the bottom of the front compartment of the trailer and then clamped in place. Tr. at 22 (Bernhardt). When Bernhardt pushed the hose up to get it connected, the bottom of the front compartment ripped loose at the connection or at a seam near the connection. Tr. at 22-23, 36-37, 49 (Bernhardt). The trailer started to leak right at the connection on the front compartment of the trailer, and there was no way to shut off the flow. Tr. at 23, 36 (Bernhardt). The fumigant came out of the break under pressure and hit Bernhardt in the face and eyes. Tr. at 45 (Bernhardt). Bernhardt testified that the pressure was from the fumigant in the trailer and he did not think that any fumigant came back through the pump. Tr. at 43, 46. The fumigant in the front compartment spilled onto the ground. Tr. at 23 (Bernhardt). 17. While- Bernhardt had no idea of the precise volume of fumigant which was spilled, he thought it was in the neighborhood of 200 to 250 gallons. Tr. at 23 (Bernhardt). 18. The location of the spill is identified by an “X” on Trial Exhibit 2, Tr. at ,16 (Bernhardt), and was in the general area of bin 113 and the dump identified on Def. Ex. WVY, attached. Tr. at 67 (Ellis); Trial Exs. 9 and 10. At that point, a gravel driveway sloped to the northeast toward an embankment between the driveway and the railroad tracks, and the railroad tracks were considerably lower than the driveway surface. Trial Exs. 7, 9, 10, 11, 12; Tr. at 67, 68 (Ellis). 19. Immediately after the spill, Ellis took a water hose and sprayed Bernhardt for some time to wash off the fumigant. Tr. at 24 (Bernhardt), 93 (Ellis). 20. The spilled fumigant and water flowed down the driveway and the embankment toward the railroad tracks. Tr. at 24 (Bernhardt), 94 (Ellis). The fumigant and water ran into a drain or discharge area which had been dug out between the railroad ties of the elevator’s siding tracks, and pooled there because it could not get beneath the mainline railroad tracks. Tr. at 94-95 (Ellis). The area where the pooling occurred is identified by the letter “B” on Trial Ex. 2. Tr. at 94 (Ellis). The spill occurred in mid-morning, and the fumigant and water remained in a pool in this area for a number of hours. Tr. at 95-97,139 (Ellis). Other Potential Sources of Carbon Tetrachloride Contamination 21. The sets of concrete silos were constructed on so-called “floating foundations” of approximately 86 inches of concrete. The exterior walls of the outer rows of silos came in contact with the ground. Depending on the elevation of the ground surrounding the elevators, portions of the silos could be 10-20 feet below ground level and, therefore, were not subject to visual inspection from the outside. The floors of the silos were concrete of unknown thickness. Tr. at 149-50, 156, 158, 167-68 (Ellis). 22. The interiors of silos on the outside rows of the elevator could be accessed by small doors on the outside wall of each silo. The outside silos had flat concrete floors. There were two types of inner silos between the rows of outside silos on A and B houses: rows of silos which were of the same size as the outside silos and smaller silos in the spaces or “interstice” between the larger silos. The interior of these inner silos could not be accessed by doors and their floors tapered to hoppers. Tr. at 615-17 (Morrison); Def. Ex. WW. 23. Below the concrete floors of the inner row of silos were five interconnecting tunnels which housed conveyor belts used to move grain. The ceilings of the tunnels, in effect, were the floors of some of the silos. However, only a portion of the floors was visible from the tunnels. The grain in the outside silos was extracted by means of chutes leading downward at a 45$ angle from the bottom of the silos to the conveyor belts in the tunnels. The grain in the inner silos was extracted through the hoppered bottoms of the silos. Tr. at 156, 158, 167-68 (Ellis). 24. The water table was so high in the vicinity of the elevators that there was a severe seepage problem in some areas of the tunnels. The problem was so serious that it could not be solved with a sump pump, so Ellis installed three “dewatering wells.” The purpose of the “dewatering wells” was to pump water from the groundwater table outside the tunnel walls and from below the bottom of the elevator tunnels. The “dewa-tering wells” functioned as intended. Tr. at 120-21,161,162,166-67 (Ellis). 25. During Ellis’ tenure as elevator manager (1959-1979), there were no spills of grain fumigant at the site other than the one by Bernhardt. Tr. at 99 (Ellis). As elevator manager, Ellis made regular inspections of the elevator two times each day by walking through the elevator, around the buildings, and through the underground tunnels. Liquid grain fumigant has a pungent odor, but Ellis did not recall ever detecting the presence of grain fumigant on his daily inspection tours, and did not see any indication of the presence of liquid grain fumigant in the tunnels below the concrete silos. Tr. at 87, 88, 147. 26. Before, during and after the 20-year period Ellis managed the Morrison Facility, the concrete elevator experienced “fallouts,” which happened when a “portion of the wall separates and allows the grain to exit that silo.” The cause of the fallouts was improper construction: “the steel was not tied behind the jack rod.” Tr. at 61, 116-18 (Ellis). Ellis never observed any evidence of leakage on the silo walls. Tr. at 146,164-65. 27. It was not Ellis’ practice to go into the silos when they were empty and he did so only on “rare occasions.” Although there was a suspension device that would have allowed a person to be lowered from the top to inspect the interior of each silo, it was never used for this purpose. Tr. at 117-18 (Ellis). 28. Ellis informed Roger Morrison that it was entirely possible that there were cracks in the floors of the flat storage buildings through which fumigant could have percolated over the years. Tr. at 118 (Ellis). 29. Ellis was present only about 20% of the time when the concrete silos were fumigated. Tr. at 142. He believed that the employees attempted to apply the proper rate of fumigant by using a meter located approximately 30 feet from the fumigant storage tank to keep track of the amount being pumped onto the grain. At least two persons were required to fumigate-one at the meter and another at the discharge point to apply fumigant to. the silo, which could be located a long distance away. The employee watching the meter was supposed to keep a record of the amount of fumigant used. Tr. at 79-80, 90 (Ellis). 30. The meter on the Max High Life storage tank malfunctioned and required service on more than one occasion. The person who ran the meter most of the time while Ellis was at the facility had an alcohol problem for many years and was finally fired after showing up drunk one day. Tr. at 109-10,121 (Ellis). 31. Even though Ellis claimed to have kept inventory records of fumigant at the Scoular site, he never attempted to perform any estimates of the amount used during the 20-year period he was there, nor has anyone ever requested him to do so. He testified that the records were at the site when he left, but he did not know what happened to them and has never asked anybody for them. Tr. at 125-26. 32. Leo Gebhart was hired by Morrison in 1979 to replace Ellis. In May 1980, Morrison leased the elevator to Scoular, and Scoular employed Gebhart as general superintendent of the elevator. Tr. at 170, 171 (Gebhart). Gebhart remained in that position until the spring of 1995. During the time Gebhart was at the Scoular site, he never observed any spills or leaks of grain fumigant, was never advised of any spills of grain fumigant, never detected the smell of liquid grain fumigant in walking about the. elevator premises, and never detected the presence of any liquid grain fumigant in the tunnels or around the outside of the storage buildings. Gebhart stated that if any amount of grain fumigant had leaked outside a silo, one would definitely know it. Tr. at 173,175-76,178. 33. Gebhart never made any effort to inspect the integrity of any of the structures at the facility to determine whether any fumigant was passing directly into the ground or into the groundwater. This was simply not an issue in his mind at the time. Tr. at 178. 34. The only way by which the employees determined that more Max High Life was needed was when the pump used to remove it from the storage tank “would pump dry.” Tr. at 183 (Gebhart). The main reason Scou-lar kept inventory records was because the “state requires that you keep a record of where this fumigant goes.” But Gebhart doesn’t think the state ever “came to look at those records while I was there.” Tr. at 184. 35. Gebhart doesn’t know what happened to the records Scoular kept of the amount of fumigant applied to the grain, has not asked anybody where they are, doesn’t know if they are still in the storeroom at the plant, and has never tried to retrieve them himself. Tr. at 185. 36. McShares offered an inspection service to elevators for approximately 5-6 years during the early 1960s, and James Allen did the inspections of the Scoular site. The purpose of the inspection was to determine the extent of insect and rodent activity, not leakage of grain fumigant. Tr. at 568, 612, 614 (Allen). A normal inspection took most of a day or a full day, Allen inspected the grain elevator, the storage bins, the outside area around the buildings and storage facilities, the upright silos, and basically the entire elevator site. The inspections took place approximately nine times a year. Tr. at 575-77, 611 (Allen). It was Allen’s judgment that the facility was not well kept or well maintained. However, during his inspections, Allen never observed any spill or leakage of grain fumigant, and did not smell any odors of grain fumigants (except for the single rat hole incident described in footnote 4, supra). Tr. at 577, 613. 37. McShares’ experience in storing liquid grain fumigant in its own tanks indicated that 3-5% of the stored fumigant could evaporate and be released to the atmosphere each month. Tr. at 582 (Allen). 38. Small amounts of carbon tetrachloride were used by the Exline Corporation in their old shop building which is located west of the Scoular site. Exline employees dipped rags in carbon tetrachloride and used the rags to wipe off oil film from parts during the manufacturing process. This use occurred from the early 1960’s until 1973-75 when Exline ceased using carbon tetrachloride. Tr. at 371, 378-79 (Exline). During that time, Ex-line never bought bulk quantities of carbon tetrachloride, and bought only one gallon cans. Tr. at 372 (Exline). There was no evidence that Exline improperly disposed of carbon tetrachloride. Discovery of Carbon Tetrachloride Contamination 39. In 1988, the Kansas Department of Health and Environment (KDHE) tested residential water wells adjacent to the Scoular site and discovered the presence of carbon tetrachloride above the minimum action levels. Tr. at 27 (Bernhardt); Tr. at 496-98 (Morrison); Trial Ex. LL. One of the wells where the KDHE found carbon tetrachloride was at Bernhardt’s residence, which is just across the railroad tracks to the northeast from the elevator. Tr. at 7, 26-27 (Bernhardt); Trial Ex. 2. 40. Carbon tetrachloride was also detected in the dewatering wells at the Scoular site in connection with Exline’s remediation activities. Tr. at 374-75, 496. Morrison’s dewa-tering wells were being used as a part of a treatment program previously instituted by Exline to remediate chromium contamination of the groundwater caused by Exline’s operations, Tr. at 362-66, 369-71. Exline installed an air stripper at its facility to remove the carbon tetrachloride from the water being pumped from the dewatering wells. Tr. at 375-78. 41. There was no evidence that carbon tetrachloride had been used or spilled anywhere else in the vicinity of the water wells which were tested by KDHE and found to contain carbon tetrachloride contamination above minimum action levels. 42. Immediately after discovery of the carbon tetrachloride in the residential water wells, Morrison provided alternative drinking water to all area residents through rural water districts, and paid the initial cost of rural water installation together with future water bills for ten years. Tr. at 27 (Bernhardt), 499-501 (Morrison); Trial Ex. 76 (1988 Summary of Expenses). KDHE Involvement 43. The KDHE issued an administrative order in 1989 requiring Morrison to conduct an investigation of the Scoular site, and Morrison, after initially appealing the administrative order, decided to conduct the investigation. Tr. at 505, 511 (Morrison); Trial Ex. 31. When it learned that KDHE was not going to look to McShares as a responsible party, Morrison made demand on McShares to participate in the costs of responding to the KDHE order. Tr. at 509, 527 (Morrison); Trial Exs. 21,21A. 44. Kejr Science Group, Inc. (Kejr) was hired by Morrison to investigate the source and extent of carbon tetrachloride contamination at the Scoular site. Tr. at 391-92 (Shepard); Trial Ex. 20. Kejr provides environmental consulting services, including soil and groundwater investigation and remediation. Tr. at 386. Max Shepard was one of the principals of Kejr who was involved in the Scoular investigation. Tr. at. 387-89. From 1989 through the time of trial, Kejr’s activities at the Scoular site included: conducting a soil gas survey of 200 samples which were analyzed in the field using a gas chromatograph, Tr. at 397-99; installing an air stripper at the adjacent Exline facility, Tr. at 401-02, 405-06; taking and analyzing soil samples in 1991 and again in 1993, Tr. at 407-09, 420; conducting groundwater probing and testing, Tr. at 420-22; obtaining quality assurance/quality control reports, Tr. at 426; obtaining a lab analysis of soil content, Tr. at 430-31; making quarterly reports to KDHE, Tr. at 431; drilling of nine monitor wells and two recovery wells, Tr. at 432-33; conducting an aquifer test; and performing regular sampling and analysis of the wells, Tr. at 434. 45. Morrison entered into a Consent Order for Comprehensive Investigation and Corrective Action Study with the KDHE, Case No. 91-E-214, effective December 23, 1992 (Consent Order). Pretrial Order (Doc. 80 at 3); Trial Ex. 31. McShares was not a party to, and did not sign, the Consent Order. Pretrial Order (Doc. 80 at 3). Morrison’s decision to enter into the Consent Order was influenced, to an extent, by his company’s experience with an EPA-managed clean-up in Nebraska. Tr. at 539-42 (Morrison). 46. The Consent Order contained “boilerplate” provisions which describe the work to be performed and related activities to be undertaken by Morrison. It required quality assurance plans, gives the KDHE the right to approve or disapprove of activities or reports submitted by or on behalf of Morrison and requires that all work be performed under the supervision of a professional engineer or geologist with expertise in hazardous waste investigation. Tr. at 206, 210-21 (Bean). 47. The findings contained in the Consent Order were the subject of negotiation between the KDHE and Morrison, Tr. at 301. They are not the result of a formal hearing where evidence was presented under oath and witnesses were subject to cross-examination, Tr. at 301. 48. It was the function of the KDHE to review and either approve or disapprove of the work performed by the engineer or consultant retained by Morrison under the Consent Order. The Order required the development of a Work Plan, which is to be followed by the Comprehensive Investigation Report (“Cl”) and a Corrective Action Study (“CAS”). When performing such reviews, the KDHE staff was required to utilize knowledge not possessed by a layman and necessary to form professional and scientific judgment with regard to what goes on at such sites and whether the requirements imposed by the KDHE have been satisfied. Tr. at 212, 213, 227, 286 (Bean). 49. Kejr prepared a Workplan on behalf of Morrison in response to the Consent Order. After review, comment and input from the KDHE, the Workplan was formally approved as contemplated by the Consent Order by the KDHE in its letter of May 13, 1993. Pretrial Order (Doe. 80 at 3). 50. Kejr also prepared Morrison’s Cl Report. After review, comment and input from the KDHE, the Cl Report was formally approved as contemplated by the Consent Order by the KDHE in its letter of October 31, 1994. Pretrial Order (Doe. 80 at 3). 51. Finally, Kejr prepared Morrison’s Corrective Action Study'(CAS). After review, comment and input from the KDHE, the CAS was submitted to the KDHE by letter of June 25,1996. By letter of June 17, 1997, as contemplated by the Consent Order, the KDHE conditionally approved the CAS, subject to clarification of seven identified points. Pretrial Order (Doc. 80 at 3). The CAS was finally approved by the KDHE in August 1997, and this ended Morrison’s responsibilities under the Consent Order covering the Seoular site. Tr. at 515 (Morrison), 437 (Shepard). 52. Morrison funded the investigation of the Seoular site as required by the Consent Order. Tr. at 513-15 (Morrison); Trial Ex. 76. As of June 30, 1997, Morrison had paid $432,936.55 in response costs directly attributable to the Seoular site. Trial Ex. 76. All of the costs incurred by Morrison are reasonable and necessary response costs of the Seoular elevator site project. Tr. at 550-51 (Court). 53. Morrison has performed all activities required of it under the Consent Order with the KDHE. Tr. at 515 (Morrison). The Pilot Program Between KDHE and EPA 54. In October 1994, KDHE and the U.S. Environmental Protection Agency (EPA) entered into a Workplan for Cooperative Agreement which is known as the state deferral pilot program (Phot Program). Tr. at 215-16 (Bean). There were later communications between EPA and KDHE about issues pertaining to the Pilot Program, including compliance with CERCLA and the National Consistency Plan. Trial Exs. 62 and 63. The Workplan for Cooperative Agreement was revised as of June 7, 1995, and was approved by the EPA by letter of July 25, 1995. Trial Exs. 64 and 65; Tr. at 310 — 11 (Bean). The final Work-plan for Cooperative Agreement was modified to address the concerns raised in the letters between KDHE and EPA about the National Consistency Plan and other issues. Tr. at 310-11 (Bean). 55. The Pilot Program was intended to make the federal procedure applicable to sites within the program, to comply with the National Consistency Plan, and to give participants in the program assurance that the federal government would not come back on them if participants followed the KDHE comprehensive investigation and corrective action procedures. The Pilot Program was to be the functional equivalent of going through the federal Superfund process with the EPA. Tr. at 217, 308, 315-16 (Bean). 56. One purpose of the Pilot Program was to negotiate a division of responsibility between KDHE and EPA for sites not yet listed as Superfund sites whereby the state would take the lead and an expanded oversight responsibility for sites within the program. Tidal Ex. 65 at 1. As part of the Pilot Program, EPA determined that KDHE’s site investigation and cleanup process is consistent with the requirements of CERCLA and the National Consistency Plan, and KDHE agreed to recommend a CERCLA-protective cleanup for all state-lead sites under the Pilot Program. Trial Ex. 65 at 3. 57. In the Workplan for Cooperative Agreement, EPA defined its oversight responsibilities for the sites in the Pilot Program as consisting of the receipt of quarterly and annual reports from KDHE, semi-annual meetings between EPA and KDHE to review the program status and resolve potential problems, and the conduct by EPA of annual review of the pilot projects and review of any remedy selected by the State. Trial Ex. 65 at 3-4; Tr. at 219 (Bean). KDHE agreed to make available to EPA upon request, all remedy selection decision documents and supporting analysis pertaining to all sites included in the Pilot Program. Trial Ex. 65 at 5. If EPA determined that the state-lead cleanup process or remedial action was inadequate to achieve a CERCLA-protective cleanup, EPA retained the right to remove any site from the Pilot Program and have it processed for placement in the Superfund program. Trial Ex. 65 at 5. 58. The Scoular site was accepted into the Pilot Program in 1995. Trial Exs. 66, 69, 70, 71 and 72 (Exs. 70-72 received for limited purposes); Tr. at 241 (Bean), 328-32 (Elder). 59. KDHE submitted quarterly and annual reports to EPA concerning sites in the Pilot Program, which included the Scoular site. The purpose of the reports was to ensure that KDHE was complying with the Workplan. Tr. at 225-29 (Bean); Trial Ex. 68 at 9 (received for a limited purpose). 60. EPA has conducted program reviews of the Pilot Program, including an annual review at KDHE’s offices on December 17, 1996. Tr. at 237 (Bean), 346 (Elder). EPA has never taken a site out of the Pilot Program for the purpose of exercising direct control over the site. Tr. at 238 (Bean). 61. KDHE has never received any complaints from EPA about how the Pilot Program is being implemented, and EPA has been complimentary of KDHE as to its handling of the pilot program. Tr. at 239, 251-52 (Bean) (testimony received for limited purpose). 62. EPA has never made any complaints to KDHE about KDHE’s handling of the Scoular elevator site. Tr. at 239 (Bean). KDHE’s Public Information Program 63. In the Workplan for Cooperative Agreement, EPA and KDHE agreed that community relations activities at all sites in the Pilot Program would be conducted in accordance with KDHE’s Public Information Program (PIP), and KDHE agreed to involve the affected community in the remedy selection process. The State agreed that its PIP is comparable to the procedures required by the National Consistency Plan and ensures early, effective and meaningful community involvement. Trial Ex. 65 at 4. 64. KDHE’s PIP policy was in effect in 1993, and was to apply to KDHE-lead sites complying with the National Consistency Plan and for all sites being worked with the state cooperative program. Trial Ex. 73. Under the PIP policy, the KDHE project manager shall develop a written Public Relations Strategy not more than one page in length. Trial Ex. 73, ¶ A. 65. A PIP for the Scoular site was in place before the site was accepted into the pilot program in 1995. Tr. at 233-36, 240 (Bean). The Public Relations Strategy for the Scoular site was sent to Morrison for review and comment in December 1993 and was approved by the KDHE. Tr. at 236 (Bean); Trial Exs. 74, 75 (received for a limited purpose). 66. During the investigation stage, the KDHE PIP policy states that an announcement or news release should be published in a local newspaper declaring the Consent Order and initiation of field work for the Cl/ CAS, and any such news release must be approved by the KDHE project manager. Trial Ex. 73, ¶ B(l). During the investigation stage, it is up to the KDHE to determine whether there was any need for any availability sessions with the public. Trial Ex. 73, ¶¶ B(2) & (3). When the Public Relations Strategy was sent to Morrison in December 1993 under the KDHE’s 1993 PIP policy, no press release was issued about the consent order and no public availability sessions were held, because the Consent Order had already been in place for a year and the investigation was well underway. Tr. at 313 (Bean). When EPA accepted sites for inclusion in the Pilot Program in 1994r-1995, including the Scoular site, EPA knew the status of the sites, whether there was a consent order in place, and whether a Cl investigation was underway. Tr. at 323 (Bean). 67. The KDHE PIP policy states that after KDHE has prepared a Draft Corrective Action Decision (CAD), the KDHE shall conduct certain community relations activities, including publication of a news release or announcement, time for submission of comments on the Draft CAD, the opportunity for an availability session for public comment, preparation of a written summary of comments received during the public comment period, and publication of a summary of the Final CAD. Trial Ex. 73, ¶ C(l-5). The Public Relations Strategy adopted for the Scou-lar site provides for each of these activities, Trial Ex. 75, but the draft CAD has not been prepared by the KDHE and the time for performing these activities has not yet arrived. Tr. at 313 (Bean); Trial Ex. 74. 68. Morrison and its consultants have held meetings with the three individuals who owned homes or businesses immediately across the railroad tracks from the elevator (Bernhardt, Weber and O’Neal), and with other persons who owned land to the northeast of the elevator (Hilton, Morgan, Rosener, Allen and Madison). Morrison shared with them the test results of the water sampling, maps delineating the plume, information concerning the legal status of the project, and what the plans were for the site. Morrison had a policy of open disclosure of any information regarding the site. Tr. at 547-49 (Morrison). Conclusions of Fact and Law In a contribution action under Section 113(f) of CERCLA, 42 U.S.C. § 9613(f), Morrison first must make a prima facie showing that MeShares is “liable or potentially liable” under Section 107(a), 42 U.S.C. § 9607(a). Thus, Morrison must prove that: (A) the site is a facility as defined in § 9601(9); (B) a release of a hazardous substance has occurred at the facility; (C) MeShares is a covered person under § 9607(a); (D) the release caused Morrison to incur costs; (E) Morrison’s response actions were consistent with the National Contingency Plan (NCP); and (F) Morrison’s response costs were reasonable and necessary. Bancamerica Commercial Corp. v. Trinity Industries, 900 F.Supp. 1427, 1451 (D.Kan.1995) (Bancamerica I), aff'd in part, rev’d in part, sub nom. Bancamerica Commercial Corp. v. Mosher Steel, 100 F.3d 792 (10th Cir.1996) (Bancamerica II); Greene v. Product Mfg. Corp., 842 F.Supp. 1321, 1323, 1325 (D.Kan.1993). Elements (A) and (B) The parties have stipulated that the Scou-lar site is a facility as defined in 42 U.S.C. § 9601(9). There is no dispute that carbon tetrachloride is a hazardous substance as defined in 42 U.S.C. § 9601(14) and 40 C.F.R. §§ 302.1 and 302.4 (1994). There is no question that a release of carbon tetrachloride occurred at the Scoular site as a result of the 1963 spill within the meaning of 42 U.S.C. § 9601(22). Thus, elements (A) and (B) have been proven. Element (C) The parties’ first dispute concerns whether Morrison has proved element (C), that MeShares is a “covered person” as defined by CERCLA. A covered person includes “the owner and operator of a ... facility.” 42 U.S.C. § 9607(a)(1). A facility is defined to include any “motor vehicle ... where a hazardous substance has been deposited, stored, disposed of, or placed, or otherwise come to be located.” 42 U.S.C. § 9601(9). The term “disposal” includes dumping, spiffing, leaking, or placing of any ... hazardous waste into or on any land or water so that such ... hazardous waste or any constituent thereof may enter the environment or be emitted into the air or discharged into any waters, including ground waters. 42 U.S.C. §§ 6903(3) & 9601(29). The term “ground waters” means “water in a saturated zone or stratum beneath the surface of land or water.” 42 U.S.C. § 9601(12). Morrison asserts that McShares is a “covered person” because the spill was from McShares’ trailer, which was being operated by McShares’ employee, and McShares owned the fumigant. (Doc 92 at 20). McShares denies liability on two bases (Doc. 98 at 2-4). First, it asserts that Ellis caused the 1963 spill by opening the valve before Bernhardt had connected the hose and started the pump. Next, McShares confusingly casts itself in the role of either a common carrier, or a supplier, but not both. Both parties rely on Amcast Industrial Corp. v. Detrex Corp., 2 F.3d 746 (7th Cir.1993), cert. denied, 510 U.S. 1044, 114 S.Ct. 691, 126 L.Ed.2d 658 (1994). The pertinent conclusions of Amcast sere summarized as follows: Each of the tanker trucks owned by De-trex in which it delivered TCE to Elkhart constituted prima facie a “facility” within the meaning of the Superfund law, § 9601(9)(A), contained a hazardous substance, namely TCE, and “disposed of’ it when the truck spilled it, because the statute defines disposal to include spiffing. §§ 6903(3), 9601(29). We conclude that Detrex was liable under the Superfund statute for the spillage from its own trucks (though it may be able to shift some of that liability back to Elkhart by means of its suit for contribution) but not the spillage from the trucks of the common carrier that it hired. Id. at 750-51. McShares can be substituted for Detrex, Max High Life for TCE and Morrison for Elkhart. In this case, the supplier itself, McShares, delivered the hazardous substance, Max High Life, on the only relevant “disposal,” the 1963 spill. The facts do not support McShares’ argument that Morrison’s manager, Eldred Ellis, caused the spill by opening the valve (Doe. 98 at 2-3). Rather, the court concludes that the spill occurred when Bernhardt, McShares’ employee, was attempting to connect a hose owned by McShares to a trailer owned by McShares. The spilled carbon tetrachloride was owned by McShares. This is not a products liability case and Morrison was not required to prove the existence of a defect in the hose or trailer. Similarly, Morrison was not required to prove exactly what “caused” the spill. Morrison satisfied its burden by showing that the spill occurred when Bernhardt was doing something to McShares’ trailer and that its employee, Ellis, did nothing to “cause” the spill. Accordingly, the court concludes that McShares is a “covered person” and that Morrison has proved element (C). Element (D)- The parties next dispute whether Morrison has proved element (D), that the release caused Morrison to incur response costs. 42 U.S.C. § 9607(a)(4)(B) makes a covered person liable for necessary response costs incurred by any other person consistent with the NCP. It is here that the court’s orders regarding expert evidence first come into play. Morrison’s position is that KDHE’s discovery of carbon tetrachloride in the water wells immediately adjacent to the site of the 1963 spill in turn caused Morrison to incur response costs (Doc. 92 at 19). McShares responds that Morrison cannot meet its burden without resort to expert testimony (Doc. 93 at 36 ¶ 104 & Doc. 98 at 4-6). The court discussed the “causation of response costs” issue in its December 18, 1996 Memorandum and Order (Doc. 72 at 9-16). Neither party sought reconsideration of the order, nor have they argued about the court’s analysis in their trial briefs (Docs. 89 & 90) or their post-trial submissions. However, McShares has cited cases not discussed by the court, principally Acushnet Co. v. Coa ters, Inc., 937 F.Supp. 988 (D.Mass.1996), for the proposition that Morrison must prove by expert evidence a causal connection between the waste discharged and the need for response costs (Doc. 98 at 5). It is not necessary to recite Acushnet’s facts. The court summarized the plaintiffs CERCLA recovery theory as follows: Under the form of legal test proposed by Plaintiffs, they do not need to prove that a toxic substance in NETT’s waste (PAHs, for example) caused them to incur response costs. Instead, Plaintiffs assert that the only elements they must prove are: (1) that the [defendant] generator disposed of waste material; (2) at a facility which contains hazardous substances of the type found in the defendant’s waste; (3) there is a release or a threatened release of that or any hazardous substance; (4) which triggers the incurrence of response costs. In other words, Plaintiffs assert that they do not need to prove any connection between any toxic substance in NETT’s waste and the response costs being incurred by a plaintiff. Under Plaintiffs’ legal theory, as long as any response costs are being incurred by a plaintiff, any party that disposed of any hazardous substance is hable to compensate that plaintiff. It does not matter what type or amount of hazardous substance was disposed of by the party. Any hazardous substance in any quantity will open the floodgates of liability, and will do so even if the hazardous substance disposed of by the party is not causing any harm, is not threatening to cause any harm, and is not any part of the reason a response is needed and costs of that response are incurred. Id. at 993. The court rejected this far-out theory after giving it more consideration than it probably deserved. Morrison does not rely on Acushnet. The causation element of a prima facie case does not require expert evidence, at least in this ease. This is illustrated by Dedham Water Co. v. Cumberland Farms Dairy, 770 F.Supp. 41 (D.Mass.1991), aff'd, 972 F.2d 453 (1st Cir.1992), a case cited in Acushnet and by McShares. The facts aré as follows: in 1974, Dedham Water Company discovered heavy metal contamination in its water well fields. • It commissioned an emergency study, which recommended, in 1980, the construction of a new water treatment plant. In 1981, the plaintiff discovered that similar contamination had been released on defendant’s property. The plaintiff.commissioned a hydrogeology study which concluded, in 1982, that defendant was the source of the contamination in plaintiffs well field. Plaintiff eventually constructed a new water treatment plant and then started a CERCLA action to recover from defendant the costs of building and operating the plant. When the case was tried, it was determined that the contamination was not “caused” by defendant. 689 F.Supp. 1223 (D.Mass.1988). On appeal, the contamination “causation” finding was not challenged. However, the case was remanded for a determination of whether the treatment plant was built in response to a threat of release of contaminants fi"om defendant’s facility into plaintiff’s well field. The court made it clear that it is not necessary to prove actual contamination in order to establish liability for response costs under 42 U.S.C. § 9607(a)(4)(B). 889 F.2d 1146, 1156 (1st Cir.1989). On remand, the district court found as follows: I find that, at that point [1982], it was reasonable for the plaintiffs to believe that the VOCs had migrated from the defendant’s facility to the well field and might continue to do so in the future. Even though it has now been judicially determined in the prior trial in this case that the offending VOCs were not released by the defendant, I find that there is a rather remote possibility (a “potential threat” to use Dr. Guswa’s redundant expression) that VOCs from the defendant’s facility might migrate to the plaintiffs’ well field at some time in the future if geological conditions should change. As far as I can tell from this record, however, the response for which the plaintiffs seek reimbursement was not related to the “potential threat.” The two-stage treatment plant recommended by Metcalf & Eddy in 1980 is substantially what was built. To be sure, the plant was not constructed until 1985, after the time when the plaintiffs reasonably believed that the defendant’s facility was the source of existing pollution and a threat of future pollution of the well field. Many of the necessary authorizing votes of the Board of Directors were taken after 1982, but they concerned regulatory delays, zoning problems and objections of abutters. The only action that appears to me to relate to the G. & M. report is the decision in 1982 to commence this action for damages. 770 F.Supp. at 42-43. Judgment was entered for defendant. The plaintiff appealed. The second appellate opinion is long and contains big words which required this court to consult a dictionary to make sure that they had no real import in the decision. The bottom line of the opinion affirming the district court is this: The actions of the plaintiff which lead to the response costs which the plaintiff sought to recover, i.e., the cost of building and operating the new plant, were not “caused” by anything done by the defendant because the plaintiff had decided to build the new plant before it received the 1982 hydro-geology report which (erroneously) blamed the contamination on the defendant. So, in considering element (D) — whether the 1963 release caused Morrison to incur response costs — the focus is on what “caused” or motivated Morrison to take the action which resulted in the costs. Expert evidence is not required because, under the rationale of Dedham, it is irrelevant whether the party from whom the response costs are sought actually caused the contamination. Thus, it follows that it is irrelevant for purposes of element (D) whether Morrison can prove, without expert testimony, that McShares actually “caused” the contamination. See Doc. 72 at 9-15. The court finds that McShares, the person responsible for the 1963 spill, “somehow, or in some way” (Dedham, 972 F.2d at 460) “caused” Morrison to incur response costs. The fact that KDHE, not Morrison, discovered the contamination is immaterial. The undisputed fact of the 1963 spill justified Morrison’s reasonable belief beginning at or near the outset of the KDHE 1988 discovery and 1989 administrative order that McShares was responsible for the contamination. This is evidenced by Morrison’s early demand that McShares participate in the costs of responding to the KDHE order. McShares rejected the demand and Morrison funded the project, thereby incurring costs. The court concludes that Morrison has met its burden to prove element (D). Element (E) Element (E) requires Morrison to prove that its response actions were consistent with the NCP. .42 U.S.C. § 9607(a)(4)(B). The parties’ dispute regarding this element is multi-faceted. (Docs. 92 at 21-27 and 35-36; 93 at 36-38; 97 at 12-29; 98 at 6-16 and 101 at 1-12). 1985 Versus 1990 NCP The first dispute concerns whether the 1985 NCP (40 CFR Part 300.71 (1988)) or the 1990 NCP (40 CFR Part 300.700 (1990)) is applicable. Morrison understandably argues for the 1990 version because it adopts a more lenient “substantial compliance” standard for a plaintiff seeking response costs. The court views the discrete issue of which NCP version to apply as one of law, not fact. The evidence established that Morrison’s response actions took place both before and after the 1990 change. When faced with the same situation, courts have taken sensible compromise approaches. For example, the court in Tri-County Business Campus v. Clow Corp., 792 F.Supp. 984 (E.D.Pa.1992), found that “if Tri-County’s clean-up was ‘underway’ on April 9, 1990, the 1990 NCP is the applicable plan,” at least to the extent that the 1990 plan does not impose new and more onerous requirements on private parties seeking recovery. Id. at 990 (citing Con-Tech Sales v. Cockerham, No. 87-5137, 1991 WL 209791 (E.D.Pa. Oct. 9, 1991)). In a later case from the same court, a different judge observed: Since CERCLA’s passage in 1980, the EPA has issued three versions of the NCP — the 1982 NCP, the 1985 NCP and the 1990 NCP. Courts evaluate response costs incurred by referring to the NCP in effect at the time the costs were incurred. Reading Co. v. City of Philadelphia, 823 F.Supp. 1218, 1239 (E.D.Pa.1993); Con-Tech Sales Defined Benefit Trust v. Cockerham, Civ. A. No. 87-5137, 1991 WL 209791, at *3 (E.D.Pa. Oct. 9, 1991). In comments to the 1990 NCP, the EPA stated that “although some people have expressed the opinion that the new NCP should not apply to private response actions initiated prior to the effective date of the revised NCP ... EPA does not believe that it is appropriate to grandfather cleanups that are already underway.” 55 Fed.Reg. 8795 (March 8,1990). The 1990 NCP became effective on April 9, 1990. Thus, the court will apply the requirements of the 1990 NCP to costs incurred after April 9, 1990, or any response actions already underway as of April 9, 1990. The requirements of the 1985 NCP will apply to any cleanups complete prior to April 9, 1990. Bethlehem Iron Works, Inc. v. Lewis Industries, Inc., No. 94-0752, 1996 WL 557592, at *54 (E.D.Pa. Oct. 1, 1996). McShares does not cite a case, and the court has not found one, which holds that only the 1985 or 1990 versions of the NCP must be applied when response costs were incurred both prior to and after 1990. The court finds Bethlehem Iron Works persuasive and will follow it. Removal Versus Remedial The second dispute concerns whether Morrison’s response actions were “removal” or “remedial.” The distinction is important because the requirements for NCP compliance are different, depending on how the response action is characterized. McShares asserts that expert testimony is required on this issue (Doc. 98 at 7-8 n. 2 and 12-13). Morrison argues that its actions to date have been “removal” in nature (Doc. 92 at 21-27). Citing Channel Master Satellite v. JFD Electronics Corp., 748 F.Supp. 373, 386 (E.D.N.C.1990), Morrison contends that the proper classification of a cleanup as “removal” or “remedial” is an issue of law not appropriate for expert testimony. Courts have followed Channel Master in this regard. See Yellow Freight System, Inc. v. ACF Industries, Inc., 909 F.Supp. 1290, 1299-1300 (E.D.Mo.1995) and Hatco Corp. v. W.R. Grace & Co., 849 F.Supp. 931, 962 (D.N.J.1994). These cases focus on the general rule that expert testimony cannot be used to define the legal parameters under which a jury exercises its fact-finding function. Specht v. Jensen, 853 F.2d 805, 809-10 (10th Cir.1988), cert. denied, 488 U.S. 1008, 109 S.Ct. 792, 102 L.Ed.2d 783 (1989). This general rule loses its force when applied to a case, such as this one, which is tried to the court. In a bench trial, the court is presumed capable of separating its fact-finding function from its obligation to interpret and apply the law. Thus, the question is not whether expert testimony is required to resolve the issue. Rather, the question is whether the non-expert evidence in the record is sufficient for the court to reach a legal conclusion whether Morrison’s actions were “removal” or “remedial.” As a preliminary matter, the court is aware of the Tenth Circuit’s ruling in Bancamerica Commercial v. Mosher Steel of Kansas, 100 F.3d 792, 797-98 (1996) that a determination of “removal” versus “remedial” is “irrelevant” when response actions are taken pursuant to 40 C.F.R. § 300.700(c)(3)(ii)(1995). Subsection (c)(3)(ii) provides that any response action carried out in compliance with the term