United States v. Vertac Chemical Corp.

MEMORANDUM OPINION AND ORDER GEORGE HOWARD, JR., District Judge. The Vertac Site in Jacksonville, Arkansas, has been the focus of litigation for 25 years. The history of the Vertac Site and Off-Site areas has been discussed in numerous prior decisions. 1 This Court previously found both Hercules Incorporated (“Hercules”) and Uniroyal Chemical Limited (“Uniroyal”) 2 jointly and severally liable under the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (“CERCLA”), 42 U.S.C. §§ 9601 et seq. for all response costs incurred and to be incurred by the United States at the Site and related areas, and entered judgment in the amount of $102,878,641.35. United States v. Vertac, 33 F.Supp.2d 769 (E.D.Ark.1998). In what was hoped to be the last decision in this matter, the Court allocated the costs between Hercules and Uniroyal. United States v. Vertac, 79 F.Supp.2d 1034 (E.D.Ark.1999). GENESIS OF THIS DECISION On October 12, 1993, the Court granted the United States’ motion for summary judgment and found that Hercules was jointly and severally liable under Section 107(a)(2) and (3) of CERCLA, 42 U.S.C. § 9607(a)(2) and (3) for the response costs incurred by the United States with regard to the Vertac Site. Hercules moved for reconsideration, at which time it argued that disputed issues of fact existed regarding divisibility. The Court, on November 1, 1993, denied the motion for reconsideration, finding, inter alia, that Hercules had failed to present evidence in support of its divisibility of harm argument. On December 19, 1999, Hercules appealed the Court’s summary judgment ruling, arguing in part that the harm at the Site is *945 divisible. On April 11, 2001, the Eighth Circuit Court of Appeals reversed the Court’s summary judgment against Hercules on the issue of liability and remanded the case for the Court to “address evidence supporting divisibility in light of the proper legal standards.” United States v. Hercules, Inc., 247 F.3d 706, 719 (8th Cir.2001). 3 The Eighth Circuit also vacated the Court’s judgments on response costs, United States v. Vertac Chem. Corp., 33 F.Supp.2d 769 (E.D.Ark.1998) (“Vertac IX”) and allocation, United States v. Vertac Chem. Corp., 79 F.Supp.2d 1034 (E.D.Ark.1999) (“Vertac X”) pending the Court’s reconsideration of Hercules’ claim of divisibility. 247 F.3d at 721. The Court scheduled a limited evidentia-ry hearing on the divisibility of harm defense raised by Hercules. The hearing was held between October 9 and 19, 2001 and December 11 and 12, 2001. The parties subsequently filed post-hearing briefs. The record, needless to say, is voluminous, comprising not only the numerous filed documents, but exhibits and transcripts admitted at the evidentiary hearing, exhibits and transcripts of the various trials in this action, evidentiary and deposition testimony of other federal actions involving Hercules, and the 52,000-page Administrative Record (AR) 'for the Site which has been filed in this action as 8 compact discs (see docket entry 2454). The Court has worked diligently to carefully review the record in considering this case. FACTUAL BACKGROUND The Vertac Chemical Plant Site (“Site”) consists of approximately 93 acres in Jacksonville, Arkansas. The Site was originah ly developed by the federal government in the 1930s as a munitions factory. Around 1948, Reasor-Hill Corporation (“Reasor-Hill”), a now defunct company, purchased the Site and first formulated finished insecticide products, primarily DDT, aldrin, dieldrin, and toxaphene. In the mid-1950s, Reasor-Hill modified the plant and began manufacturing phenoxy herbicides, primarily 2,4, dichlorophenoxyacetic , acid (“2,4-D”), 2,4,5 trichlorophenoxyacetic acid (“2,4,5-T”) and 2,4, 5 trichloropropionic acid (“2,4,5-TP” or “Silvex”). United *946 States v. Vertac Chem. Corp., 489 F.Supp. 870, 873 (E.D.Ark.1980) (“Vertac I ”). The chlorinated compound 2,4,5-triehlorophe-nol (2,4,5-TCP or TCP) was an intermediate in the 2,4,5-T manufacturing process. The manufacture of 2,4,5-T creates a highly toxic byproduct, 2,3,7,8-tetrachlorodi-benzo-p-dioxin (“dioxin” or “TCDD”) that is now viewed as hazardous to humans. Vertac XI at 712. All of the herbicide related compounds are listed hazardous wastes under the Resource Conservation and Recovery Act (RCRA), 42 U.S.C. § 6921, 40 C.F. R. § § 261.31, 261.33, and the compounds 2,4,5-T, 2,4,5-TCP, 2,4,5-TP are designated as “Acute Hazardous Wastes” under RCRA. 40 C.F.R. § 261.31. Additionally, all of the phenoxy herbicides and TCP are designated as hazardous substances under CERCLA. See 42 U.S.C. 9601(14); 40 C.F. R § 302.4. Reasor-Hill did not have a process to treat the waste water that was produced. United States v. Vertac Chem. Corp., 966 F.Supp. 1491, 1494 (E.D.Ark.1997) (‘Vertac VIII”). While Reasor-Hill operated the Site, an unknown quantity of untreated chemical wastes from the production process flowed through cooling ponds on the west side of the plant into a nearby stream. Reasor-Hill disposed of large quantities of pesticide (insecticide) and phenoxy herbicide wastes by leaving the wastes in drums on the Site. Vertac I at 874. Hercules bought the Site from Reasor-Hill in 1961 and continued to manufacture herbicides, including 2,4-D and 2,4,5-T, at the plant until 1971. United States v. Hercules, 247 F.3d 706, 712 (8th Cir.2001)(“Vertac XV). When it commenced operations at the Plant, Hercules found and then buried on-site thousands of drums of pesticide and herbicide wastes left by Reasor-Hill. Vertac VIII at 1494-95. The southeast corner of the Site where Hercules buried the drums was known as the Reasor-Hill drum burial area. Hazardous substances, including insecticides, phenoxy herbicides and TCDD dioxin, leached and flowed from the Reasor-Hill drum burial area into Rocky Branch Creek. Until late 1964, Hercules continued Rea-sor-Hill’s practice of discharging untreated wastewater directly into Rocky Branch Creek. Vertac I at 874. The waste water which resulted from the manufacturing of 2.4.5-T or 2,4,5-TP contained dioxin. Id. In 1964, Hercules constructed a waste water pretreatment system, which consisted of a neutralization trench designed to reduce the acidity of the water, an equalization basin designed to stabilize the rate of flow into the City of Jacksonville’s sewage system, and a pump and pipe to deliver the treated waste water to the Jacksonville sewage system. Id. The waste water pretreatment system did not remove the dioxin. The basin frequently overflowed directly into Rocky Branch Creek during heavy rainfalls, and leaked. Hercules manufactured formulations of 2.4.5-T, Silvex and 2,4,-D acids for commercial customers. It used some of the same equipment to manufacture both 2,4-D and 2,4,5-T products, but the equipment could not manufacture both simultaneously. In 1964, Hercules was awarded a contract to produce Agent Orange for the U.S. Department of Defense (“DOD”). 4 It produced and supplied Agent Orange to DOD through 1968. Agent Orange was a 50/50 mixture of the butyl esters of 2,4,5-T *947 and 2,4,-D acids; a much more concentrated form of the acids than that sold by Hercules to its commercial customers. Dioxin was formed as a by-product during the production of Agent Orange. Vertac VIII at 1494. Due to strict specifications in the Agent Orange contracts, Hercules was forced to discard as waste more significant quantities of material that did not meet specifications than was discarded during production for commercial customers. At the same time, Hercules continued manufacturing 2,4,5-T, silvex and 2,4-D for its commercial customers. As a result of the increased production, the volume of wastes generated at the Plant also increased. In 1965, Hercules discovered that in the manufacture of 2,4,5-T, dioxin was generated in the process, specifically in the de-chlorinator, also known as the “sputnik.” 5 In that same year, it began utilizing a “toluene extraction” system during which dioxin and other impurities that were extracted underwent a distillation process producing a residue (stillbottoms). The toluene stillbottoms were contaminated with TCDD and 2,3,7,8-tetrachloro-diben-zo-furan (TCDF). Hercules buried drums of toluene stillbottoms and 2,4-D wastes at the Site and disposed of them in the landfill areas. Vertac VIII, at 1494. See also Vertac I at 875. The drums were allowed to collect in the process areas, sometimes for several months, before being transported to the drum burial pit. The drums were thin gauge and were in poor condition. Leakage occurred at the point of storage and in transporting the drums to the drum burial area. Hercules’ practice was to scoop up any contaminated soil around the leaking drum and place it into the drum. Between June of 1964 and May of 1968, Hercules produced almost 3 million gallons of undiluted Agent Orange for DOD. After the Agent Orange contracts terminated in 1968, Hercules continued manufacturing 2,4,5-T, silvex and 2,4-D for commercial customers until June 1970. Vertac VIII at 1495-96. In late 1970, Hercules ceased production at the Site and cleaned it or “mothballed” it for the winter as Hercules began to seek a buyer or lessee of the Plant. 6 On October 1, 1971, Hercules leased the Site to Transvaal, Inc. (“Transvaal”) which later became Vertac Chemical Corporation (“Vertac”). Among the actions Transvaal took prior to starting operations was the redrumming of about 50 to 100 55-gallon drums of 2,4,5-T wastes, some of which were leaking. 7 Transvaal continued to manufacture 2,4,-D, 2,4,5-T and 2,4,5-TP and continued to bury wastes on site. In 1975, Transvaal began shipping its 2,4-D waste to off-site landfills and began to store its 2,4,5-T stillbottoms above ground with the hope of recycling the waste. 8 Vertac XI at 712. All waste burial took place while Hercules owned the plant. In August of 1976, Transvaal purchased the Site from Hercules and reorganized as Vertac. Vertac continued to make 2,4-D, 2,4,5-T and 2,4,5-TP. Vertac I at 874. Vertac continued its operations until 1986, when it abandoned the Site altogether. *948 On February 28, 1979, the Environmental Protection Agency (“EPA”) suspended many of the uses of 2,4,5-T. Vertac I at 875. On March 15, 1979, Vertac voluntarily ceased manufacturing 2,4,5-T and 2,4,5-TP and overpacked its 2,4,5-T stillbottom drums. M 9 In 1980, Vertac recommenced the manufacture of 2,4-D. It began redrumming, or overpacking, leaking drums that it had accumulated on site. 10 Spilled material along with dirt, dust, and debris were scooped up and placed into the overpack-ing drums. EPA found that the 2,4-D wastes had been contaminated with TCDD. On February 26, 1980, EPA issued a proposed rule under the Toxic Substances Control Act (“TCSA”) to prevent Vertac from disposing of TCDD wastes from the Site. The rule, known as the ‘Vertac Rule,” became effective March 11, 1980 11 and final May 19,1980. The Vertac Rule prohibited the off-site shipment of 2,4,-D wastes which contained TCDD. 12 However, if Vertac could show that a batch of 2,4-D was manufactured and produced wastes free of dioxins, it could ship those wastes and all subsequent wastes for off-site disposal unless it went back to manufacturing 2,4,5-T. Vertac IX at 780. According to testimony at the hearing on the proposed rule, Vertac had approximately 3200 drums of wastes resulting from the production of 2,4,-D. Analysis by Vertac of three samples from 700 drums of wastes resulting from the initial production of 2,4-D revealed TCDD levels of approximately twenty parts per billion (“ppb”). 13 Vertac then sent samples of its 2,4-D wastes to Wright State University and to Monsanto for testing. Monsanto was unable to detect TCDD with its analytical equipment, but Wright State detected TCDD in composite samples taken from 700 drums of “D” wastes of .7 ppb 2,3,7,9-TCDD. 14 Vertac IX at 780. It subsequently found .5 ppb in another sampling of 1000 drums and a third sampling done on an undisclosed number of drums found .3ppb TCDD. 15 Finding that dioxin was present in the 2,4-D wastes, Vertac stopped sampling and allowed the “D” wastes to accumulate. Subsequent testing by the State of Arkansas as well as the results of trial burns revealed the presence of TCDDs and TCDFs in the drums. Vertac IX at 781. 16 Both TCDD and TCDF were generated exclusively in the manufacture of 2,4,5-T and 2,4,5-TP, but not in the manufacture of 2,4-D. The 2,4-D waste drums, because of the acidic nature of their contents and their condition resulting from their storage and handling, failed at a rapid rate. Beside being stored on soil, the drums were also *949 stacked several deep, three high, and un-banded, by the thousands. Vertac ceased operations in 1986 and abandoned the Site in 1987. Vertac VIII at 1495. 17 When Vertac abandoned the Site in January of 1987, there were nearly 29,000 drums containing waste materials, including 2,4,5-T, 2,4,-D and dioxin. “Many of these drums had corroded and leaked, contaminating more soil, groundwater, and buildings at the site.” Vertac XI at 712. Over 15,000 drums stored outside were exposed to the elements. “Drums were stacked three high on deteriorating pallets.” Vertac IX at 772. The drums were failing at a rate of between five to 300 per week. 18 “EPA determined that the wastes on the site posed a threat to public health and welfare and the environment. Residents in the area could be exposed to hazardous substances and should a large release occur, in the event of a tornado, fire, or continued poor maintenance, the toxic waste could contaminate the environment. EPA initiated a removal action in 1987 to stabilize the drums and tanks and to provide site security.” Vertac IX at 772. Some of the drums were labeled “T” waste, some were labeled “D” waste, some were marked “T” and “D” and some were not marked at all. Many of the markings were indistinguishable or unreadable. Vertac, 671 F.Supp. at 606. “The drummed material was considered acutely hazardous waste and was characterized as ‘F-listed’ wasted under applicable ... RCRA regulations, 42 C.F.R. Part 261, App. VII.” Vertac IX at 772. EPA chose to dispose of the wastes by incineration. When the incineration of the drums commenced, the contents of the drums were found to have different wastes including 2,4,-D, 2,4,5-T, 2,4,5-TP, dioxins, dibenzofurans, and all sorts of various materials ranging from dirt to broken pipes. EPA completed the on-site incineration of the 2,4-D drummed waste in October, 1994. The remaining 3260 drums of 2,4,5-T wastes were shipped off-site for incineration. Off-site shipment was completed in March, 1996. Vertac IX at 774. SCOPE OF REVIEW Hercules argues that in its latest order the court of appeals vacated all previous findings of fact of this Court. That is, the Court is now free to consider new evidence and new facts. Hercules basically contends that the only evidence the Court may now consider is that which was introduced at the hearing on divisibility of harm that the Court conducted in 2001. The Court ordered a hearing to consider additional evidence on the issue of divisibility of harm. The Court finds that it can consider the existing record. The Eighth Circuit merely remanded the case for the Court to 'determine, based on the record, whether Hercules had established a basis for establishing divisibility of harm. The Court of Appeals did not remand for a new trial. The United States has already established, and the Court has previously found, that there were releases and threats of releases of hazardous substances during the time Hercules owned or operated the Site, which caused the United States to incur response costs. This is sufficient to establish Hercules’ liability under CERC-LA for response costs. However, it does not resolve the issue of liability completely. *950 Hercules can “escape joint and several liability” if it can demonstrate that the harm incurred at the Site is divisible. United States v. Alcan Aluminum Corp., 315 F.3d 179, 185 (2d Cir.2003). The universal starting point for divisibility of harm analyses in CERCLA cases is the Restatement (Second) of Torts, which provides for the apportionment of damages among two or more parties when at least one is able to show either (1) “distinct harms” or (2) a “reasonable basis for determining the contribution of each cause to a single harm.” Restatement (Second) of Torts § 433A (1965). Vertac XI, at 717. Section 433A of the Restatement (Second) of Torts entitled “Apportionment of Harm to Causes,” provides as follows: (1) Damages for harm are to be apportioned among two or more causes where (a) there are distinct harms, or (b) there is a reasonable basis for determining the contribution of each cause to a single harm. In other words, the defendant must prove that “there is a way to determine what portion of the harm (i.e. the hazardous substances present at the facility and the response costs incurred in dealing with them) is fairly attributable to the defendant as opposed to other responsible parties.” United States v. Manzo, 279 F.Supp.2d 558, 562 (D.N.J.2003) (quoting United States v. Rohm & Haas Co., 2 F.3d 1265, 1280 (3d Cir.1993)). “‘Distinct harms’ are ‘those that may properly be regarded as separate injuries.’ ” Vertac XI at 717. The burden is on Hercules to establish that there is a reasonable basis for divisibility of harm. That burden is substantial. United States v. Alcan Aluminum Corp., 315 F.3d 179, 185 (2d Cir.2003). The evidence in support of divisibility “must be concrete and specific.” Vertac XI at 718. “[T]he question whether there is a reasonable basis for apportionment depends on whether there is sufficient evidence from which the court can determine the amount of harm caused by each defendant.” In the Matter of Bell Petroleum Svcs., Inc., 3 F.3d 889, 903 (5th Cir.1993)(Bell Petroleum). “[Wjhere causation is unclear, divisibility is not an opportunity for courts to ‘split the difference’ in an attempt to achieve equity.” Id. A court should not settle on a compromise amount that it believes approximates the relative responsibility of the parties. If in doubt, the court should impose joint and several liability. United States v. Township of Brighton, 153 F.3d 307, 319 (6th Cir.1998). ANALYSIS Hercules advances several arguments in support of its divisibility defense. It advances a type of “geographical apportionment” theory, that is, that the harm at the Site is divisible on the basis of operable units. 19 Each operable unit should be considered, according to Hercules, a separate “site.” As to these different areas, Hercules argues that it is liable only for the harm that it caused. Different approaches are used, depending on the area. Thus, with regard to the incineration of the *951 drummed wastes, Operable Unit 1 (plant equipment) (“OU-1”) and the Jacksonville and Rogers Road landfills, Hercules contends that it hás no liability. As to the remaining sites, Hercules admits that it has some responsibility for contamination of on site soils and groundwater and the Off-Site Areas. It argues, however, that the Court should apply a volumetries approach for dividing the harm. EPA and Uniroyal assert that Hercules cannot establish that the harm is divisible because of the extensive cross-contamination and commingling of dioxin and other hazardous substances throughout all areas of the Site. They note that Hercules’ operations and extensive waste disposals at the Plant and on other areas of the Site from 1961 to 1971 contributed massive amounts of dioxin and other hazardous substances to the Site. Furthermore, Hercules also owned the Plant during the first five years of Transvaal/Vertac’s operations there, and it is therefore legally responsible under CERCLA for Vertac’s waste disposals during the period from 1971-1976. The EPA divided the Site into separate operable units for purposes of facilitating the remediation. The attempt to apportion the harm based on operable units was rejected by the court in United States v. Manzo, 279 F.Supp.2d 558, 574 (D.N.J.2003). “An operable unit is ‘a discrete action that comprises an incremental step toward comprehensively addressing site problems.’ 40 C.F.R. § 300.5. It is not a site, although it ‘may address geographical portions of a site.’ ” Furthermore, in Manzo as here, the operable units did not solely address geographical portions of the Site 20 The Court is not persuaded that EPA’s selection of remedial actions alone is- a basis for dividing the harm. See Washington v. United States, 922 F.Supp. 421, 428 (W.D.Wash.1996)(EPA’s selection of remedial actions provides no basis for apportioning harm) The Court will proceed to address the arguments raised by Hercules, and determine whether there is a basis to divide the harm in each of the areas. Drummed Wastes As discussed, above, when EPA arrived at the Site after Vertac abandoned in it found over 28,000 drums of waste stored on the Site. The drummed wastes were incinerated as part of a series of continuing removal actions that the State of Arkansas and the EPA began in 1987. There is no dispute that the 2,4-D waste drums were accumulated on the Site beginning in late 1979, at least three years after Hercules had sold the plant site to Vertac. Had the drums merely contained identifiable 2,4-D wastes, the Court might find that Hercules’ divisibility of harm ar: gument had merit. But, because they contained other hazardous substances which were produced while Hercules owned the plant, Hercules must demonstrate a basis for finding the harm divisible. Hercules makes several arguments to support its contention that it is not liable for the incineration of the drummed wastes. It has consistently contended that the approximately 25,000 drums of 2,4-D waste and approximately 3,000 drums of 2,4,5-T still bottoms left behind when Ver-tac abandoned the Site on January 31, 1987, were each “distinct” harms, to which *952 Hercules did not cause EPA’s response costs. Dioxin was found in the drums. The United States posits two sources of contamination of the 2,4-D drums with TCDD, TCDF,2,4,5-T and 2,4,5-TP One is the cross-contamination from the equipment, especially equipment used in both the production of 2,4-D and 2,4,5-T. The second source is the contaminated soil, trash, debris and other materials placed into the drums during the multiple over-packings. The soil could have been contaminated with dioxin produced during Hercules’ operation of the plant. No one can say that the dioxin in the “D” drums did not come from Hercules. Despite Hercules’ contention that the drums didn’t leak, there is ample evidence to show that the drums of 2,4,5-T still bottoms did leak on the ground when Hercules owned the plant. 21 The on-site soils, especially, in the process area, contained hazardous chemicals 22 The soil was contaminated with wastes generated by Hercules and Vertac. The drummed wastes stored on site leaked at one time or another and had to be over-packed. During overpacking, the chemical material that had leaked out on to the soil would be scooped up and put into the overpacked drum. Thus, the D-drums most likely contained soil contaminated by Hercules’ operations which had been scooped up when the corroding drums were overpacked. Hercules argues that it could not have caused the waste to be contaminated through the production process or through the placement of contaminated dirt into the drums at a later date. With regard to the production process, Hercules argues that it thoroughly cleaned out the production vessels in 1970, while looking for a buyer or lessee. The same equipment was used to produce 2,4,-D and 2,4,5-T. Testimony at the hearing revealed that the cleanup of the equipment was not as thorough as portrayed by Hercules, and that the plant equipment was cross-contaminated with 2,4-D, 2,4,5-T and dioxin. 23 Hercules also relies on a “flushing theory,” i.e., that the thousands of batches of material Vertac ran through the production vessels would have thoroughly flushed the dioxin from the vessels to the point when it would be virtually non-existent. 24 Thus, there should not have been, according to Hercules, any residual contamination in the 2,4-D drums which were generated and stored by Vertac between the fall of 1979 and 1986. Hercules’ flushing theory remains just that — a theory. It has not been subjected to scientific testing and validation. Rather the evidence basically disproves the theory. After nearly eight years of only 2,4-D production, dioxin and 2,4,5-T were present in the plant equipment. Hercules contends that the presence of dioxin in the 2,4-D waste drums did not *953 influence the decision to incinerate. Hercules points to testimony by Dr. Phyllis Moore, Director of the ADPC & E, who made the initial decision to incinerate the drums, that the presence of dirt in the drums did not influence the decision to incinerate. 25 Hercules further argues that the placement of contaminated dirt into the drums would constitute a separate “disposal” of hazardous substances under CERCLA. Additionally, the redrumming efforts took place after August, 1976, when Hercules sold the plant to Vertac and therefore Hercules had nothing to do with the redrumming efforts. None of these arguments withstand scrutiny. A number of witnesses, particularly from the State, testified about the rationale for incineration. There was no doubt that the drummed wastes presented a hazard to the area. EPA and the State of Arkansas were concerned that because of the manner in which the drums were stored, the risk of fire and explosion, as well as the possibility of tornadoes, could spread the dioxin into the environment. Additionally, witnesses testified that because time was of the essence, testing a large number of the drums to determine whether dioxin was present would have been costly and time-consuming. 26 It was not, as Hercules argues, the dirt in the drums that drove the incineration; rather it was the dioxin in the dirt that drove the incineration of the drums. Hercules also attacks the validity of the tests. The Court is not in a position to discredit the testing procedures or results. Witnesses from all parties maintained different positions with regard to the testing results. The Court cannot find as a matter of law that certain test results were not reliable. Hercules advances other innovative arguments to support its contention that it did not cause some of the drums to be contaminated with dioxin. One of these is that the placement of the dirt into the drums is in itself a separate disposal. “Disposal” under CERCLA has the same meaning as “disposal” under the Solid Waste Disposal Act. 42 U.S.C. § 9601(29) The Solid Waste Disposal Act defines “ “disposal” as the discharge, deposit, injection, dumping, spilling, leaking, or placing of any solid waste or hazardous waste into or on any land or water so that such solid waste or 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). Hercules “disposed of’ hazardous substances because it placed those hazardous substances in a manner that allowed them to enter the environment. Attempts to further subdivide actions in an attempt to escape liability borders on the absurd. Hercules also argues that it is relieved of liability under the doctrine of “superseding cause.” Restatement (Second) of Torts, § 440. Comment b. It points to a number of actions taken which would relieve it of responsibility. The Court cannot find that “superseding cause” is a basis for divisibility. The Restatement must be followed “only to the extent that it is compatible with the provisions of CERCLA.” 247 F.3d at 717. Recognizing a “superseding cause” theory in this situation “would not be compatible” with the strict liability provisions of CERCLA. Even assuming the theory is applicable in CERCLA cases, it is not applicable here. *954 The mere fact that other causes intervene between the original act of negligence and the injury for which recovery is sought is not sufficient to relieve the original actor of liability, if the injury is the natural and probable consequence of the original negligent act or omission and is such as might reasonably have been foreseen as probable. Butler v. Arkansas Power & Light Co., 186 Ark. 611, 54 S.W.2d 984 [(1932)]; Arkansas Power & Light Co. v. Marsh [195 Ark. 1135, 115 S.W.2d 825 (1938)], supra; Hayes v. Missouri Pac. R.R. Co., 208 Ark. 370, 186 S.W.2d 780 (1945). The original act or omission is not eliminated as a proximate cause by an intervening cause unless the latter is of itself sufficient to stand as the cause of the injury. Butler v. Arkansas Power & Light Co., supra; Arkansas Power & Light Co. v. Marsh, supra. The intervening cause must be such that the injury would not have been suffered except for the act, conduct or effect of the intervening agent totally independent of the acts or omission constituting the primary negligence. Pollard v. Union Pacific R. Co., 75 Ark.App. 75, 79, 54 S.W.3d 559, 562-563 (Ark.App.2001) Here, the Court cannot find that the harms associated with the drummed wastes, i.e., the contamination, would not have occurred except for the totally independent acts of EPA,,the State of Arkansas, or Vertac. “Moreover, the intervening act or omission of a third person is not a superseding cause when the original actor’s negligent conduct is a substantial factor in bringing about an injury, if the actor, at the time of his negligent conduct, realized that a third person might so act or if the intervening act is a normal response to a situation created by the actor’s conduct and the manner in which it is done is not extraordinarily negligent.” Ouachita Wilderness Institute, Inc. v. Mergen, 329 Ark. 405, 415, 947 S.W.2d 780, 785 (Ark.1997) There is no evidence that any action of any other person or entity was “extraordinarily negligent” so as to relieve Hercules of liability. Hercules has not demonstrated that the drums did not contain any hazardous substances that Hercules produced. Hercules cannot show that any of those hazardous substances in the drums that were attributable to Hercules can be distinguished from the hazardous substances in the drums that were attributable to Vertac. There is evidence of cross contamination of equipment. Vertac and Hercules used the same equipment to make 2,4,-D and 2,4,5-T. Subsequent years of only 2,4,-D production did not rid the equipment of dioxin. Dioxin was present in the soils; contaminated soil was placed in the D-drums when they were overpacked. Thé soil, dust, shredded pallets, shredded trash and the process areas were permeated with TCDD, TCDF, 2,4-D, 2,4,5-T, 2,4,5-TP, chlorinated benzenes and other hazardous wastes. The drum overpacking process therefore added TCDD, TCDF and other hazardous substances into the 2,4-D waste drums. The inner drums continued to corrode, combining the contents of the inner drums with, those used in the overpacking. Hercules cannot establish that it was not responsible for the dioxin in the soil which was ultimately placed in the drums. Even Hercules’ expert witnesses, Robert Fisher and Douglas Keilman, both of whom worked for Hercules could not say that the source of the dioxin in the “D” drums did not come from Hercules. 27 *955 The Court finds that Hercules has not provided any concrete and specific evidence that the hazardous substances in the drums were “distinct” such that Hercules should not be held responsible for the costs incurred with the incineration of the drummed wastes. OU-1 (Above Ground Media) Hercules argues that the Court must look to the remedies chosen and the situations which created the need for those remedies to determine whether the harm is divisible with respect to OU-1. Hercules divies up OU-1, declaring that it is responsible for some harm but not for other. 28 That is, the several different media should be characterized as “distinct” environmental harms. Hercules contends that it cannot be held responsible for hazardous materials in the tanks, vessels and equipment. It claims that it removed all of its hazardous wastes when it completely disassembled and cleaned out all the plant equipment in 1971. Testimony at the hearing revealed that the cleanup of the equipment was not as thorough as portrayed by Hercules, and that all hazardous wastes were not removed. 29 . Furthermore, that Hercules cleaned the plant in 1971 is irrelevant, because Hercules is liable as an owner while it leased the plant to Transvaal. As an owner, it is responsible for the contamination attributable to Transvaal’s operations. See Nurad, Inc. v. William E. Hooper & Sons Co., 966 F.2d 837, 846 (4th Cir.1992) cert. denied sub nom., Mumaw v. Nurad, Inc., 506 U.S. 940, 113 S.Ct. 377, 121 L.Ed.2d 288 (1992)(§ 9607(a)(2) imposes liability on owner of facility at time disposal of hazardous waste); United States v. R.W. Meyer, Inc., 889 F.2d 1497, 1507 (6th Cir.1989), cert. denied, 494 U.S. 1057, 110 S.Ct. 1527, 108 L.Ed.2d 767 (1990). In addition to its argument concerning the 1971 plant cleanup, Hercules relies on its “flushing theory” to establish that the hazardous substances found in the equipment were not its own. As discussed above, the theory is not valid. That 2,4,5-T and dioxin were found in the equipment after 8 years of just 2,4-D production disproves Hercules’ theory. Similarly, Hercules cannot establish that it was not the source of the contamination on the shredded trash and pallets. The pallets, which were used to store drummed wastes at the site, became contaminated with the 2,4,5-T, 2,4-D and dioxin leaked from the drums. 30 The pallets also became contaminated because they were in contact with the contaminated soil. Hercules also contaminated the buildings. Dioxin was found in leachate from the buildings. 31 Wipe samples from the buddings in OU-1 revealed reportable quantities of dioxin. Dioxin was found in storage vessels. 32 The 1974 Sputnik eruption contaminated nearby buildings. Hercules contends that the threat posed by the plant buildings was the threat of collapse, as well as asbestos contained in some building materials. Hercules argues that it did not cause these threats. Ver-tac, in abandoning the plant and letting it fall into disrepair, was responsible for the possibility of building collapse. The United States was responsible for asbestos being in the buildings when it built them to house the Arkansas Ordnance Plant. *956 Hercules is wrong. Asbestos was not the cause for the remediation. The buildings were demolished because the risk of collapse might cause release and human exposure to herbicide process contamination, including unacceptable, levels of TCDD. In sum, Hercules has not presented evidence to demonstrate that it was not the source of the 2,4,5-T and dioxin in the above-ground media, and therefore it is responsible for the costs incurred with regard to OU-1. OU-2 (Soils, Foundations and Underground Utilities) Hercules- argues that the harm at OU-2 can be characterized as “successive.” Hercules contends that the principles of Bell Petroleum apply with regard to OU-2. According to Hercules, dioxin, a single harm, drove the remedy. Synergistic effects, relative toxicity or differential migratory potential are not factors. The contribution of each harm can be reasonably estimated based on the production volume of 2,4,5-T and 2,4,5-TP. Hercules would apportion harm based on the production records of the 2,4,5-T and 2,4,5-TP that Hercules and Vertac produced during the years each of them owned the plant. According to the OU-2 Record of Decision (“ROD”), extensive dioxin contamination was found in soils in the blowout area, the east drum storage area, around and to the east of the broiler house, in the area of the existing and former chlorination plants, and around the maintenance area. Other contaminants of concern such as chlorophehols, chlorobenzenes and chloro-phenoxyacids were also found throughout the Site soils, in relative concentrations paralleling dioxin concentrations. 33 “Where, as in this case, hazardous substances are commingled, a defendant cannot rely on merely volumetric evidence. Evidence must be produced ‘disclosing the individual and interactive qualities of the substances deposited’ at the Site.” United States v. Vertac, 966 F.Supp. 1491, 1504 (E.D.Ark.1997) (quoting United States v. Monsanto Co., 858 F.2d 160, 172 (4th Cir.1988), ce rt. denied, 490 U.S. 1106, 109 S.Ct. 3156, 104 L.Ed.2d 1019 (1989)). Additionally, proof of “relative toxicity, migratory potential, degree of migration and synergistic capacity” of the hazardous substances at the Site is relevant to the divisibility of harm inquiry. United States v. Alcan Aluminum Corp., 315 F.3d 179, 186 (2d Cir.2003) (citation omitted) “[T]he volume of waste of a particular generator is not an accurate predictor of the risk associated with the waste because the toxicity or migratory potential of a particular hazardous substance generally varies independently of the volume ...” United States v. Chem-Dyne Corp. 572 F.Supp. 802, 811 (S.D.Ohio 1983). EPA contends that the Hercules has failed co present accurate production records. Hercules counters that exact numbers are not necessary; estimates suffice if they form a reasonable basis for apportioning the harm. “If the expert testimony and other evidence establishes a factual basis for making a reasonable estimate that will fairly apportion liability, joint and several liability should not be imposed in the absence of exceptional circumstances.” Bell Petroleum, 3 F.3d at 903. EPA asserts that Hercules’ reliance on sales records does not establish a reasonable basis for dividing the harm. Estimates of production were made because some of the production records were lost. Unlike the pure volumetric approach in Bell, the sales records fail to adequately account for all the production. They do not account for product made and not sold, *957 off-spec material disposed of at the Site, or for operational and waste disposal differences. The volumetric approach presupposes that leaks and spills took place at roughly the same amount and same frequency and severity during the production of 2,4,5-T and 2,4,5-TP. Douglas Keilman, Hercules’ project manager involved in the remediation and an expert witness for Hercules, conceded that in order to use volumetric evidence, he had to assume that the release of hazardous substances was constant over time. 34 If that assumption is correct, then, according to Hercules, the amount of dioxin contributed to the soils would be roughly proportionate to the production volume. Based on its calculations, Hercules contends that it is responsible for 70.74% of EPA’s response costs for OU-2. The evidence, however, does not support the assumption on which Hercules relies. There is no way to quantify the releases of wastes at the Site with production volumes, and there was not a constant ratio or direct correlation between the two. Leaks and spills occurred regularly during the years Hercules owned the Site. Production methods differed over the years, resulting in differences in the amount of waste generated. Until 1966, Hercules operated a blower or dryer at the acid plant that spewed dioxin-bearing flaked acid material around the Site. Thus, according to EPA, more TCDD would have been in the waste than in the product after Hercules installed the toluene wash process. The DOD changed the specifications for the formulation of Agent Orange during the time it was produced, precluding the assumption that the ratio of waste to production volume remained constant. The concentration of active ingredients in the DOD product was different than that in the commercial product. Additionally, waste disposal methods differed, with some methods producing more contamination than others. 35 The sputnik eruptions reveal the difficulty in relying on volumetries as a basis for divisibility of harm, and point to the conclusion that production is not directly correlated to the release of waste. The evidence reveals that there were as many as twenty sputnik eruptions at the Site while Hercules owned the plant, and another ten when it leased the plant to Transvaal. The eruptions spread dioxin over the Site and Off-Site areas. The worst eruption occurred in 1974 when Transvaal leased the Site from Hercules. 36 Here there were differences in the operations, as well as the generation and disposal of wastes, making this situation unlike that in Bell Petroleum, where there was only one media and one manner of disposal. There is no evidence in this case that the generation of waste was proportional to the production of product. The manufacturing process at the Site yielded hazardous substances other than dioxin. Hercules has provided no evidence *958 regarding the effect of these other substances on the migration or transport of dioxin. It has presented no evidence regarding the relationship between the volume of dioxin or other hazardous substances generated, the release of dioxin or other hazardous substances, and the harm at the Site. Hercules did not present evidence of the relative toxicity, migratory potential and degree of migration, and synergistic effects or capacities of the various hazardous substances deposited at the Site. It contends they are irrelevant as the decision to remediate was based solely on dioxin. Thus, these factors did not come into play in the selection and implementation of remedies. Hercules’ argument essentially places the burden on EPA to prove it considered the “synergistic effects” and that its failure to include mention of it in the ROD means that there is no need for Hercules to present evidence. While EPA may not have referred to synergistic effects of various contaminants, it did consider the various hazardous substances present at the Site, not relying solely on dioxin. As the court cautioned in Monsanto, “[v]olumetric contributions provide a reasonable basis of apportioning liability only if it can be reasonably assume, or it has been demonstrated, that independent factors had no substantial effect on the harm to the environment.” United States v. Monsanto Co., 858 F.2d 160, 173 n. 27 (4th Cir.1988)(emphasis added). Hercules has not demonstrated that independent factors had no substantial effect on the harm. For example, according to the testimony at the hearing, the half-life of dioxin is affected by different conditions. 37 Hercules did not present any evidence as to the rate the dioxin broke down at the Site and Off-Site areas based on photolysis, dissipation and other factors. It did not present any evidence of possible synergistic effects of the combinations of dioxin with other toxic materials, such as 2,4-D and 2,4,5-T. In sum, the Court finds that Hercules has not established a basis to divide the harm at OU-2. Off-Site Areas The Off-Site areas include soils, the floodplains and stream sediments of Rocky Branch Creek and Bayou Meto, sanitary sewer lines, and two municipal sewer plants and their structures. Hercules posits that the Off-Site environmental harm can be characterized as “successive.” As with OU-2, Hercules states that because the remedy was based entirely on the presence of dioxin, the synergistic effects, relative toxicity, and differential migratory potential were not factors. Hercules argues that the production volumes of Hercules and Vertac for 2,4,5-T and 2,4,5-TP are a basis to find divisibility of harm. Hercules further contends that because Hercules and Vertac never owned or operated these off-site areas, their respective liabilities can only be as an “arranger.” That Hercules did not own the off-site areas does not mean that it can only be held liable as an “arranger.” CERCLA defines “facility” broadly to include: (A) any building, structure, installation, equipment, pipe or pipeline (including any pipe into a sewer or publicly owned treatment works), well, pit, pond, lagoon, impoundment, ditch, landfill, storage container, motor vehicle, rolling stock, or aircraft, or (B) any site or area where a hazardous substance has been deposited, stored, disposed of, or placed, or otherwise come to be located; but does not include any consumer product in consumer use or any vessel. *959 42 U.S.C. § 9601(9). The Plant Site and Off-Site areas are not distinct facilities, but are one facility for purposes of liability under CERCLA.. The definition of facility “applies not only to traditional waste sites ... but also to any ‘area’ in and around which hazardous substances have ‘come to be located ... Axel Johnson, Inc. v. Carroll Carolina Oil Co., Inc. 191 F.3d 409, 417 (4th Cir.1999) (citations and internal quotations omitted). The Tenth Circuit recently reviewed how federal courts have interpreted “facility” under CERCLA, finding that the circuits that have applied the defined term “facility” have done so with a broad brush. See Uniroyal Chem. Co., 160 F.3d at 245 (applying broad definition of facility in an action under § 107 and stating “[i]n examining the contours of § 9601(9), it is apparent that facility is defined in the broadest possible terms, encompassing far more than traditional waste sites. It expressly includes buildings, pipelines, motor vehicles, rolling stock, wells, and aircraft. 42 U.S.C. § 9601(9)(A).”) (emphasis supplied); United States v. Rohm and Haas Co., 2 F.3d 1265, 1279 (3d Cir.1993) (examining liability under § 107 and noting “we think it evident from the broad statutory definition of ‘facility’ that Congress did not intend EPA to be straight-jacketed in this manner in situations involving a release transcending property boundaries”); 3550 Stevens Creek Assocs. v. Barclays Bank of Cal., 915 F.2d 1355, 1360 n. 10 (9th Cir.1990) (determining liability under § 107 and noting broad construction of the term “facility,” “such that in order to show that an area is a ‘facility,’ the plaintiff need only show that a hazardous substance under CERCLA is placed there or has otherwise come to be located there”) (internal citations omitted); New York v. Shore Realty Corp., 759 F.2d 1032, 1043 n. 15 (2d Cir.1985) (“CERCLA defines the term ‘facility’ broadly to include any property at which hazardous substances have come to be located”); see also Quaker State Minit-Lube, Inc. v. Fireman’s Fund Ins. Co., 52 F.3d 1522, 1525 (10th Cir.1995) (In a § 9607 action, “[t]he EPA designated the Ekotek Site [consisting of three surface impound-ments, piles and pits of waste material, underground tanks, and an underground drain field] a CERCLA facility pursuant to 42 U.S.C. § 9601(9) because of contamination by hazardous substances.”) (emphasis supplied). In a § 107 action, the Fourth Circuit noted that simply because “a property could be divided [into multiple facilities] does not, however, mean that it must be so divided for CERCLA purposes.” Axel Johnson, Inc. v. Carroll Carolina Oil Co., 191 F.3d 409, 418 (4th Cir.1999). There, the hazardous substances were not located only in the storage tanks and their associated spill areas; rather, they were located throughout the property. The court, after noting the “untenability of the contention that anything that could be designated as a separate facility must be so designated,”- applied the broad definition of facility, and noted that “[n]o court has held ... that any area that could qualify as a facility under the definition must be considered a separate facility.” Id. at 417-18 (emphasis supplied); see also Tyson, 299 F.Supp.2d at 709 (treating plant or “an area [that] is managed as a whole ... [as]-a single facility for CERCLA purposes”). We recognize that in a § 107 action, the hazardous substances may have contaminated large parts of a plant, while initially being emitted from only one tank. See Axel Johnson, 191 F.3d at 417. However, the underlying purpose of *960 § 103, that is, the alerting of government officials of a potential hazardous substance release that may require federal and local government response assistance, is best served through treating the Dorman Farm as a single facility. Solely owned by Seaboard, it is managed and operated as one facility, with one particular site purpose (producing swine products). See also Akzo Coatings, Inc. v. Aigner Corp., 960 F.Supp. 1354, 1359 (N.D.Ind.1996) (rejecting the argument that each contamination source is a separate facility because such argument “could have disastrous consequences, for ultimately every separate instance of contamination, down to each separate barrel of hazardous waste, could feasibly be construed to constitute a separate CERCLA facility”), aff'd in part, vacated in part by, 197 F.3d 302 (7th Cir.1999); Quaker State Minit-Lube, 52 F.3d at 1525 (10th Cir.1995) (entire site, which consisted of surface impound-ments, piles and pits of waste material, underground tanks, and an underground drain field considered a facility for § 9607); Cytec Indus. v. B.F. Goodrich Co., 232 F.Supp.2d 821, 836 (S.D.Ohio 2002) (“This court concludes that usually, although perhaps not always, the definition of facility will be the entire site or area,' including single or contiguous properties, where hazardous wastes have been deposited as part of the same operation or management.”); Clear Lake Props. v. Rockwell Int’l Corp., 959 F.Supp. 763, 767-68 (S.D.Tex.1997) (rejecting an attempt to create unnatural boundaries between a building and the site on which it is located). Sierra Club v. Seaboard Farms Inc., 387 F.3d 1167, 1174-1175 (10th Cir.2004) (finding that farm complex as a whole, as opposed to every barn, lagoon and land application area on complex, constituted single “facility” under CERCLA) In this case, the contamination from the Site migrated to the contiguous off-site areas. Hercules contributed to the contamination of the off-site areas when it operated the Plant Site from 1961 to 1971 and when it leased the Plant Site to Transvaal from 1971 to 1976, during which time Transvaal’s operations further contaminated the Plant Site and Off-Site Areas. Therefore, Hercules is liable as an owner under section 107(a)(2) for contamination in the Off-Site Areas. The Court has already discussed the flaws with applying a volumetries approach to the divisibility of harm with regard to OU-2. The same analysis applies to the Off-Site areas. The Court cannot find that the harm at the Off-Site areas is divisible based on production volumes. OU-S (Groundwater) Hercules states that it did not cause the groundwater contamination which resulted from Vertac’s ownership and operation of the Site after August, 1976. Hercules contends that the response costs associated with the remediation of groundwater contamination are divisible as a successive harm. Generalized organic compounds drove the remedy and, according to Hercules, the costs can be apportioned based on volumetries because there are no synergistic or interactive effects of the chemicals of concern, i.e., organics, upon each other. That is, the amount of organic chemicals generated in the production process, which would have been the source of contamination of the groundwater, would have been proportional to total production volume of all products, not just 2,4,5-T and 2,4,5-TP. Furthermore, Hercules contends that the manufacturing processes did not change materially over the time that Hercules and Vertac operated the plant. Here, Hercules posits that the Court should use total production volume to determine the amount of harm caused by Hercules. *961 The flaws associated with Hercules’ vol-umetries approach discussed above apply equally to its argument with respect to groundwater contamination. Furthermore, the evidence supports a finding that Hercules disproportionately contaminated the groundwater. There is no dispute that thousands of drums of 2,4-D and 2,4,5-T were buried on site during the time Hercules owned the Site. The burying of such wastes in unlined pits contributed to the groundwater contamination at the Site. Furthermore, when Hercules operated the Site, waste was poured into a well, resulting in contamination of the groundwater. 38 Hercules is liable for the contamination of the groundwater, not only as a result of its practices, but those of Transvaal. “[G]round water contamination at the site is complex, resulting from past water management and disposal practices. Sources of contamination include on-site landfills, spills and discharges into the central ditch, Reasor Hill well, and other parts of the central process area.” 39 The dioxin spewed from the sputnik eruption in 1974 contaminated the groundwater. 40 Sputnik waste which was probably contaminated with dioxin was dumped in the marshy area adjacent to the Boiler House. 41 Furthermore, the record establishes that Hercules’ operations contaminated other areas encompassed by OU-3, such as the process area and soils, including areas on the east side of the central ditch. Hercules tries to distinguish the drum burial areas from OU-3. Such an attempt is without merit. It admitted that it was liable for ground water contamination. Hercules buried drums containing hazardous wastes on site. The OU-3 remedial includes the burial areas and Hercules is responsible for the contamination resulting from the burial of the drums 42 The contaminants of concern in the ground water were released and disposed by Reasor-Hill, Hercules and Transvaal/Vertac. Hercules has not produced any specific evidence upon which to divide the harm. Landfills As Hercules never owned or operated the two landfills, its liability arises as an arranger for the disposal of hazardous substances at the landfills. Hercules argues that there is no evidence that it ever disposed of CERCLA hazardous waste at the landfills. Rogers Road Landfill The Rogers Road Landfill is a ten-acre tract that the City of Jacksonville acquired on September 16, 1953. Hercules contends that it did not send any of its waste, garbage or trash to the Rogers Road Landfill because it was closed during the years that Hercules operated the plant. Hercules states that the Rogers Road Landfill was opened from 1953 until 1959, closed between 1959 and 1973, and then reopened for a brief period from 1973-1974. The exact dates that the landfill was open are unclear; the Eighth Circuit in O’Dell v. Hercules, Inc., 904 F.2d 1194, 1197 (8th Cir.1990) found that “[t]he Rogers Road landfill had been closed by the city of Jacksonville in 1959. However, unofficial dumping occurred at Rogers Road.” 43 *962 Fifteen to 50 drums of dioxin contaminated waste were disposed of at the landfill. They were not labeled and were very corroded. 44 Hazardous substances identical to those generated by Hercules at the Site were found at the landfill, including 2,3,7,8-TCDD, 2,4-D, 2,4,5-T, and 2,4,5-TP. 45 In addition, Dieldrin, an herbicide produced by Reasor-Hill, was found. EPA in the ROD concluded that no detailed records were found to prove how the contaminants of concern were placed at the landfill. 46 To establish Hercules’ liability for harm at the Rogers Road landfill, EPA relies on the testimony of Alsie Glover, who lived next door to the Rogers Road landfill and in between it and the Graham Road landfill. She testified in the O’Dell-Bridges trial that the Rogers Road Landfill was opened from 1961 to 1967 and accessible during the years Hercules operated its Jacksonville Plant. 47 She saw barrels at the Rogers Road dump with the name “Hercules Manufacturing Company.” 48 Photographs of the drums at the landfill reveals no such labels or markings. The ROD’ states that there were no dis-cernable markings on the drums. 49 Vincent Dodson, who worked for Hercules from the summer of 1967 through 1969, testified in the O’Dell-Bridges trial that the 55-gallon barrels he took to the Graham Road landfill did not have Hercules’ name or logo. 50 Duane Reel, who was city engineer for the City of Jacksonville during the time of the remediation, testified that the drums he observed at the landfill did not have any labels on them. 51 Martin Swanson, a remedial project manager for the Jacksonville Site,' testified that there were no marking on the drums at the Rogers Road landfill and that there were no materials at the landfill that had Hercules markings on them. 52 Billy Roberts began working for Reasor-Hill in 1954, and worked almost continuously at the Site for 48 years. 53 He stated that Reasor-Hill disposed of its filter paper at the landfill. He recalled going to Rogers Road once and to the Jacksonville landfill twice for Reasor-Hill. He could not state definitely that he ever went to Rogers Road for Hercules. 54 Also supporting Hercules’ position that it did not dispose of wastes at Rogers Road is the presence of dieldrin. Dieldrin was produced by Reasor-Hill, who operated the Site during the time the Rogers Road landfill was opened. EPA states that Hercules may have disposed of some of the Dieldrin wastes that had been left by Reasor-Hill when Hercules purchased the plant. This, of course, is speculation. It is equally, if not more plausible, based on the testimony of Billy Roberts, that Reasor-Hill' disposed of the chemical wastes at the landfill. The Court also notes that the residents around the landfills used empty barrels for *963 their personal use, such as for burning trash or for barbecue grills. They also used the landfill to dump their wastes. It is quite possible that the barrels could have been placed in the landfill by the residents. 55 The Court finds that Hercules has met its burden of establishing that it is not responsible for the harm caused at the Rogers Road landfill. Thus, Hercules will not be responsible for the costs incurred by the government with respect to the Rogers Road landfill. Jacksonville Landfill (also knoim as the Graham Road Landfill or City Dump) The Jacksonville Municipal Landfill Superfund Site is an eight-acre plot purchased by the City of Jacksonville in June, 1960. The landfill has been referred to as the Jacksonville Landfill, the Graham Road Landfill, the Graham Road Dump and the Jacksonville City Dump. About forty of the eighty acres were used by the City for landfilling until the landfill was closed in 1973. 56 Between 15 and 50 drums were disposed of in an area referred to as the “drum disposal area,” located behind the pipeline on the south end of the landfill property. “Most of the drums have corroded completely away, leaving piles of white, fibrous, adsorbent-type materials.” 57 The contaminants of concern that were detected were 2,3,7,8-TCDD, 2,4-D, 2,4,5-T, 2,4,5-TP and 2,4-DCP. 58 Hercules admits that it took some waste to the Graham Road Landfill. However, Hercules states that the trash that was taken did not contain hazardous substances. Thus, Hercules contends that it was not responsible for the presence of the hazardous substances, particularly the filter paper in which the chemicals of concern were found, at the Jacksonville Landfill. The Court need not dwell long on Hercules’ argument. The record is replete with testimony of Hercules’ employees who took chemical wastes to the “city dump.” Vincent Dodson, who worked for Hercules in the late 1960s, testified that chemical wastes, including those that came from leaks and spills and the Sputnik explosions, were scooped up, placed in barrels and dumped at the Jacksonville landfill 59 Doyce Shurley, who worked for Hercules in the early 1960s, remembered going to the Jacksonville Landfill two or three times for Hercules to dispose of trash. He believed that he once disposed of a barrel of toluene stillbottoms. 60 Billy Honey also testified that when he worked for Hercules he hauled drums which contained chemical waste to the Graham Road landfill. 61 Similarly, LeRoy Jordan confirmed that he disposed of both liquid and solid waste barrels at the Graham Road site for Hercules. 62 That the drums only contained filter paper, as Hercules contends, does not absolve Hercules of arranging for the disposal of hazardous substances. Filter paper was used in the manufacture of 2,4-D, 2,4,5-T, and 2,4,5-TP and Hercules used it in its product formulation. Filter paper was caked with various wastes including TCDD. Hercules also relies on “geographic considerations” as a basis for limiting its harm. In United States v. Township of *964 Brighton, 153 F.3d 307, 320 (6th Cir.1998),the court stated that if the Township “could show that its ‘operating’ activities were completely limited to a discrete and measurable section of the property, and that the releases onto or from that section represented a discrete and measurable harm, this would provide a reasonable basis for apportionment.” The non-contiguous areas, according to Hercules, are the north and south sides of the Jacksonville landfill, areas which were divided by a natural gas pipeline and opened for use at different times. It contends that it only deposited waste on the north side of the landfill, and it is therefore not responsible for the response costs associated with the waste on the south side of the landfill. EPA presented a photograph of the Jacksonville Landfill dated July 1969 showing that there was waste disposal south of the pipeline right-of-way as of 1969. 63 Vincent Dodson testified that he disposed of wastes south of the pipeline. Hercules cannot show that the trash, filter paper, ánd other materials disposed of at the Jacksonville landfill were not contaminated with hazardous substances, including dioxin, 2,4,5-T and 2,4-D. It also cannot demonstrate that it is not the source of the contamination at the Jacksonville City landfill. Thus, the Court finds that Hercules has not established a basis to divide the harm at the Jacksonville City landfill. Constitutionality of CERCLA Hercules contends that the retroactive application of CERCLA to impose liability of $82 million upon Hercules with respect to the drummed wastes and OU-1 violates the Fifth Amendment under Eastern Enterprises v. Apfel, 524 U.S. 498, 118 S.Ct. 2131, 141 L.Ed.2d 451 (1998) because that liability is severe, unexpected, and grossly disproportionate to Hercules’ conduct. Hercules’ arguments rest on the assumption that it is not responsible for the disposal of Vertac’s drums and OU-1, and that it has shown divisibility. 64 The Eighth Circuit has previously held CERCLA to be constitutional. See United States v. Northeastern Pharm. and Chem. Co., 810 F.2d 726 (8th Cir.1986). It reaffirmed this holding and refused to accord Eastern Enterprises precedential value. United States v. Dico, Inc., 266 F.3d 864, 879-880 (8th Cir.2001) Recently, the Second Circuit addressed the argument that the retroactive application of CERCLA to an aluminum manufacturer’s waste disposal activities constituted an unconstitutional taking under the Fifth Amendment’s Takings Clause and a violation of the Fourteenth Amendment’s Due Process Clause. The court thoroughly discussed the issue, noting that the other courts that have considered the constitutionality of CERCLA in light of Eastern Enterprises have continued to hold that CERCLA is constitutional. United States v. Alcan Aluminum Corp., 315 F.3d 179, 189-190 (2d Cir.2003)(listing cases). *965 “[Cjourts considering the issue prior to the Eastern Enterprises decision consistently agreed that the retroactive liability scheme of CERCLA is constitutional. Like this Court, several other courts have been asked to reconsider this precedent in light of Eastern Enterprises. Those courts have uniformly held that CERCLA continues to pass constitutional muster.” Id. at 189. Hercules has not provided a basis for the Court to disregard the well-reasoned decisions of those courts that have considered the constitutionality of CERCLA post-Eastern Enterprises. The Court finds that the application of CERCLA to hold Hercules liable in this instance is not unconstitutional. Uniroyal’s Divisibility Defense Uniroyal asks that the Court reconsider Uniroyal’s divisibility defense. Uniroyal contends that if the Court accepts Hercules’ divisibility argument but rejects Uniroyal’s divisibility argument, liability for the divided portion of the harm will fall to Uniroyal alone. Uniroyal did not raise the divisibility defense on appeal and the Court was not directed by the Eighth Circuit to consider it. The Court will not go outside the scope of the remand. Even if it did, the Court would find no basis to reconsider Uniroyal’s divisibility of harm defense. As discussed above, the Court has rejected Hercules’ divisibility argument with respect to those areas for which Uniroyal was held jointly and severally liable. The Court could not find, therefore, that Uniroyal has established a basis to divide the harm. It remains jointly and severally liable for all areas except the landfills. CONCLUSION The Court has considered Hercules’ divisibility of harm defense according to the standards set forth in Vertac XII. The Court finds that Hercules has failed to establish a basis to divide the harm with the exception of the Rogers Road landfill. The Court finds that Hercules is not responsible for the costs incurred in connection with the Rogers Road landfill. The Court has already found that the costs incurred by EPA were reasonable and that it was entitled to recover those costs. The Court entered judgment against the parties in the amount of $102,878,641.35 plus any additional response costs incurred and to be incurred in this case. Vertac IX, at 787. The Court need not revisit its findings with regard to the reasonableness of the costs. Those findings remain valid. The judgment, however, must be reduced by the amount of costs incurred for the Rogers Road landfill. The Court further finds that the previous ruling regarding allocation remains valid. Thus, the previous finding that Uniroyal is responsible for 2.6 percent of the costs for which it is jointly and severally liable is affirmed. See United States v. Vertac Chem. Corp., 79 F.Supp.2d 1034, 1041 (E.D.Ark.1999). The government is directed to submit a proposed judgment within fifteen days of this Memorandum Opinion and Order which sets forth the modified costs consistent with this Order. Footnotes [1] United States v. Vertac Chem. Corp., 489 F.Supp. 870 (E.D.Ark.1980); United States v. Vertac Chem. Corp., 588 F.Supp. 1294 (E.D.Ark.1984); United States v. Vertac Chem. Corp., 671 F.Supp. 595 (E.D.Ark.1987), vacated, 855 F.2d 856 (8th Cir.1988); United States v. Vertac Chem. Corp., 756 F.Supp. 1215 (E.D.Ark.1991), aff'd, 961 F.2d 796 (8th Cir.1992); United States v. Vertac Chem. Corp., 966 F.Supp. 1491 (E.D.Ark.1997); United States v. Vertac Chem. Corp., 33 F.Supp.2d 769 (E.D.Ark.1998); United States v. Vertac Chem. Corp., 79 F.Supp.2d 1034 (E.D.Ark.1999). See also O’Dell v. Hercules, Inc., 687 F.Supp. 450 (E.D.Ark.1988)(O’Dell-Bridges), aff'd 904 F.2d 1194 (8th Cir.1990). [2] As of January 30, 2001, the name of Uniroyal Chemical Limited was changed to Cromp-ton Co./Cie. For ease and consistency, the Court will continue to use the name Uniroyal. [3] The Eighth Circuit's decision is somewhat confusing. The court stated that it was vacating United States v. Vertac Chem. Corp., 841 F.Supp. 884 (E.D.Ark.1993) ("Vertac V"). Vertac V however dealt only with the liability of the United States. In that case, Vertac, Hercules and the State of Arkansas argued that the United States should be held liable under CERCLA for its role in the production of Agent Orange. The United States filed a motion for partial summary judgment asking that the Court find that it is not liable. The Court found the United States not to be liable, denying the motions of Vertac, the State of Arkansas, and Hercules and granting the motion of the United States. The Eighth Circuit Court of Appeals affirmed the Court’s conclusion that the United States was not liable as an ''operator” or "arranger” in United States v. Vertac Chem. Corp., 46 F.3d 803 (8th Cir.1995) ("Vertac VI”). Vertac V did not grant summary judgment on the United States’ motion for summary judgment against Hercules on the issue of liability and did not siddress Hercules' argument concerning divisibility of harm. The finding of joint and several liability of Hercules was issued in an unpublished decision entered on October 12, 1993. (document no. 1826) (granting United States’ motion for partial summary judgment against Hercules on the issue of liability under CERCLA). The subsequent order denying Hercules' motion for reconsideration on the issue of divisibility, also unpublished, was entered on November 1, 1993 (document no. 1875)(denying Hercules' motion for reconsideration of October 12, 1993 Order). Should the Eighth Circuit again address this case, it might want to clarify the record. It is unclear whether the Court's decision in 841 F.Supp. 884 and affirmed in 46 F.3d 803 concerning the United States’ liability should have been vacated. [4] Agent Orange was an herbicide used as a defoliant in Vietnam. United States v. Vertac Chem. Corp., 46 F.3d 803, 805 (8th Cir.1995). [5] Tr.2001 at 877. The dechlorinator is the vessel in which dioxin in formed during the process of making trichlorophenol. Hercules later increased the number of "sputniks” from one to three. [6] The extent of Hercules' cleanup of the plant is in dispute and is discussed below. [7] December 12, 1988 Deposition of Jerry Keister, p. 6-7 [8] The drums of the toluene stillbottoms were still on-site when the State of Arkansas ordered them to be redrummed. 2001 Tr. at 734 [9] The recontainerization was pursuant to a June 15, 1979 Order by the Arkansas Department of Pollution Control and Ecology ("ADPC & E”). Vertac I at 875-876. [10] 2001 Tr. at 354 [11] 45 Fed.Reg. 15592 [12] Id. [13] 45 Fed.Reg. 32678 [14] Hercules contends that the Monsanto results of non-detect were more reliable than the Wright State University results, although Hercules' witness Robert Fisher acknowledged that Wright State had a high resolution mass spectrometer which Monsanto did not have. [15] 2001 Tr. at 1647 [16] See also e.g., 2001 Tr. at 1665-1678 discussing results obtained in 1993 by Richard Ehrhart, EPA Remedial Project Manager for the incineration project of about 400 drums of 2,4-D wastes. [17] See also United States v. Vertac Chem. Corp., 671 F.Supp. 595 (E.D.Ark.1987) for a description of Vertac’s actions in 1986 and 1987. The Eighth Circuit Court of Appeals vacated the decision because Inter-Ag, the entity who purchased Vertac's assets, had not been joined as a party to the action. 855 F.2d 856 (8th Cir.1988) (Table). [18] 2001 Tr. at 1427, Ex. 8073. [19] Operable unit means a discrete action that comprises an incremental step toward comprehensively addressing site problems. This discrete portion of a remedial response manages migration, or eliminates or mitigates a release, threat of a release, or pathway of exposure. The cleanup of a site can be divided into a number of operable units, depending on the complexity of the problems associated with the site. Operable units may address geographical portions of a site, specific site problems, or initial phases of an action, or may consist of any set of actions performed over time or any actions that are concurrent but located in different parts of a site. 40 C.F.R. § 300.5 [20] Randal Maud, Hercules’ expert in environmental remediation and Project Manager at the Site, testified that the Site was divided into “logical areas of study.” At first Operable Unit 1 was all engineering structures above the land surface and Operable Unit 2 was everything at land surface and below. Ground, water was spun off into a separate operable unit for administrative purposes. 2001 Tr. at 2068-69. [21] Robert Fisher, a chemist with Hercules and Vertac, and one of the persons considered most knowledgeable about Hercules' operations, testified that the deterioration of the T drums was one of the major sources of dioxin contamination at the Site. 2001 Tr. at 907. [22] See e.g. Remedial Investigation and Feasibility Study for OU-2. [23] See discussion of clean-up of OU-1, infra. [24] According to the "flushing theory,” if one makes a batch of 2,4,5-T and leaves 10% of the 2,4,5-T in the process vessel, each subsequent batch of 2,4-D made would reduce the amount of 2,4,5-T in the product such that in the period of time from 1971 to 1979 there would be less than one molecule remaining in any initial residual material. 2001 Tr. at 366-367. [25] Dr. Moore testified that the issue of "dirt” did not influence the decision. 2001 Tr. at 279. However, she stated that the major concern with the drums was the presence of dioxin. 2001 Tr. at 282. [26] See e.g. 2001 Tr. at 542, 1787. [27] 2001 Tr. at 907-908. See also testimony of Steven Quigley, an expert in environmental engineering for Uniroyal, who opined that there is no way to identify the source of the 2,3,7,8-TCDD in the D drums. 2001 Tr. at 1940. [28] It accepts responsibility for the costs related to the spent carbon. [29] See testimony of Reedie Ray, a former Hercules plant worker who was involved in the plant cleaning process. 2001 Tr. at 1369. [30] See exhibit 1149. [31] 2001 Tr. at 1771. [32] See Exhibits 1149 and 1150. [33] See ROD for OU# 2, Exhibit 6815, p. 30-35. [34] 2001 Tr. at 1323-24. [35] For example, Hercules used a "tee-pee” burner to dispose of its filter paper and cardboard drums that were contaminated with hazardous substances. The tee-pee burner was essentially a device for open burning. It was a large metal structure with a screen on the top, about 30 to 40 feet tall. 2001 Tr. at 143, 740. The filter paper was coated with 2,4,5-T, 2,4-D, as well as other chlorinated hydrocarbons. 2001Tr. at. 345-46. The cardboard drums contained TCB, a chlorinated hydrocarbon which may produce dioxins when burned. The release of hazardous substances into the air resulting from the burning may have caused an unknown amount of dioxin contamination at the Site. Vertac did not bum any wastes on the Site. [36] 2001 Tr. at 885. The most serious process upset was the 1974 sputnik eruption in terms of generation of dioxin. [37] 2001 Tr. at 1195. [38] 2001 Xr. at 1968-69. [39] Exhibit 6854, ROD for OU# 3, at p. 31. [40] See Exhibit 7513 at p. 124. [41] 2001 Tr. at 1245. [42] See Exhibit 6854, ROD for Operable Unit #3. [43] See also Exhibit 8099, Technical Assistance: Rogers Road Landfill. "From 1953 until 1974, approximately one-half of the site was used intermittently as a municipal waste disposal facility and an open dump.” But see *962 2001 Tr. at 349, deposition testimony of Stephen Veale, remedial project manager for EPA, who stated that Rogers Road was opened from 1953 to 1959, closed from 1959 to 1973 and then open again in 1973 to 1974. [44] Exhibit 6895, ROD for Rogers Road Landfill, pg. 1-4. [45] Id. at 1-32. [46] AR001821. [47] 2001 Tr. at 165. [48] 2001 Tr. at 171. [49] Exhibit 6895, p. 1-4. [50] Testimony of Vincent Dodson, O’Dell-Bridges trial, Feb. 16, 1988, p. 277. [51] 2001 Tr. at 151. [52] 2001 Tr. at 1524. [53] 2001 Tr. at 2050. [54] 2001 Tr. at 2054-55. [55] 2001 Tr. at 66-67, 191 — 194. [56] AR002384. [57] AR002389. [58] AR002396-97. [59] 2001 Tr. at 55, 70. [60] 2001 Tr. at 212. [61] 2001 Tr. at 144, 146-147. [62] 2001 Tr. at 200-202. [63] Ex. 8097, 2001 Tr. át 1610-12. [64] Hercules arrives at a somewhat outrageous conclusion that the Eighth Circuit, in remanding the case, had "clearly signaled its belief that the evidence supports only Hercules' contentions regarding its lack of liability for the drums and OU-1. As a matter of deference to the factfinder, however, the Eighth Circuit remanded so that this Court could take the step of entering judgment in favor of Hercules.” (Reply by Hercules in support of findings of fact and conclusions of law, docket entry 2488, p. 34 fn. 20). It is clear that the reversal and remand hardly amounts to a finding by the Eighth Circuit in favor of Hercules.