Textron Inc. Ex Rel. Homelite Division v. Barber-Colman Co.

903 F. Supp. 1570, 26 Envtl. L. Rep. (Envtl. Law Inst.) 20633, 42 ERC (BNA) 1057, 1995 U.S. Dist. LEXIS 15801, 1995 WL 627973

Opinion

MEMORANDUM OF DECISION AND ORDER

ROBERT D. POTTER, Senior District Judge.

THIS MATTER is before the Court on Motion by Defendant Hoechst Celanese Corporation (“Celanese”) for Summary Judgment, filed December 16, 1994 (document #81). For the reasons stated herein, that motion will be denied.

An investigation by the United States Environmental Protection Agency (“EPA”) and the North Carolina Division of Solid Waste Management (“DSW’) confirmed the presence and threatened release of hazardous substances at the so-called “Harwell Road site.” That investigation also indicated that Textron Inc. (“Textron”) was a principal generator of those hazardous wastes. As a result, the EPA issued an administrative order, pursuant to Section 106 of CERCLA, 42 U.S.C. § 9606, that required Textron to remove the hazardous substances from the site. Textron voluntarily complied with the EPA’s order. Later, Textron filed this contribution action to recover some of those costs from the Defendants in this action.

I. BACKGROUND

In the late 1940’s, R.A. McKee began doing business in Gastonia, North Carolina. His business was waste disposal. In connection with that business, he purchased the 10.5 acre property that abuts Harwell and Beatty Roads in 1967. He used that property to dispose of waste picked up from his customers, and that property is now referred to as the Harwell Road site. 1

McKee did business with a number of industrial companies in Gaston County. He both constructed and maintained septic systems and removed sludge from their tanks. At some companies his practice was to dispose of the sludge removed from their tanks on their own property. With respect to other companies, however, he removed waste from their holding and treatment tanks by pumping sludge from those tanks into his tanker truck. He then transported that sludge and wastewater to the Harwell Road site and disposed of the waste there. The Harwell Road site was permitted for the disposal of sludge by the Gaston County Health Department.

McKee began using the Harwell Road property to dispose of septic sludge as soon as he purchased it in 1967. His practice was to dig trenches in the ground approximately 30" wide by 6' deep and anywhere from 12' to 20' long. He disposed of sludge by emptying his 2000 gallon tank truck into those trenches, allowing the sludge and wastewater to seep into the ground. Later, McKee would cover the sludge with dirt. McKee conducted his operations this way until about 1978 when he sold his business.

Ultimately, an investigation by the EPA and the North Carolina DSW indicated the presence and threatened release of hazardous substances at the Harwell Road site. That investigation also indicated that Tex-tron was a principal generator of those hazardous wastes. Therefore, in 1989 Textron received an administrative order from the EPA, later revised in May of 1990, pursuant to § 106(a) of CERCLA, 42 U.S.C. § 9606(a). The revised order stated that the migration of hazardous substances from the site constituted an actual or threatened release as defined in Section 101(22) of CERCLA. The order required Textron to remove the haz ardous substances from the Harwell Road site as well as conducting other operations designed to render the site safe.

Textron elected to voluntarily comply with the EPA’s order. It thereafter performed a variety of tasks including the installation of monitoring wells, hydraulic conductivity testing, metal detector surveys, as well as the sampling and analysis of soils, surface water, groundwater and septic pits. Pursuant to the EPA’s order, Textron also excavated, removed, sampled, and disposed of buried drums and their contents off-site, and performed other work required by the EPA. By virtue of the EPA’s investigation and Textron’s activities many hazardous substances have been detected at the Harwell Road site including zinc, copper, phenol, the BTEX compounds (benzene, toluene, ethyl-benzene and xylene), and a variety of other heavy metals, and volatile and semivolatile compounds.

Later, Textron filed this action to recover some of the costs incurred to comply with the EPA’s order from the Defendants in this action. According to Textron, the Defendants also generated some of the hazardous substances found at the Harwell Road site, and therefore, they are liable for some share of the costs Textron incurred to remove those wastes as well as any future costs incurred by Textron to comply with EPA mandates.

All of the Defendants have moved for summary judgment. In large part, the motions for summary judgment before the Court arise from the passage of time. McKee hauled waste generated by his customers to the Harwell Road site between 1967 and 1978 and he was 85 years old at the time of his deposition on May 19, 1994. Although he had kept the books for his business, those records were destroyed in a fire that occurred around 1979 or 1980. So he is the only source of information concerning his waste-hauling business. Similarly, the passage of time has limited the documentary and other evidence about the Defendants’ operations available through discovery. Textron’s efforts to discover the chemicals used in connection with the Defendants’ processes of production has met with little results. Thus, the documentary evidence upon which Tex-tron relies is fairly slim. And ultimately, it is this sparsity of evidence that drives all of the motions for summary judgment now before the Court. In one way or another, all of the Defendants assert that Textron has failed to produce evidence from which a reasonable person could conclude that their processes of production generated hazardous substances and/or that those hazardous substances were hauled by McKee to the Harwell Road site.

II. ANALYSIS.

In this case, Textron seeks contribution from the Defendants for their proportionate share of response costs etc. incurred by Textron in connection with its remediation of the Harwell Road site. As previously noted by this Court, according to the plain language of § 113 of CERCLA, 42 U.S.C. § 9613(f)(1), Textron’s contribution action in this ease arises under § 107 of CERCLA, 42 U.S.C. § 9607. That section imposes strict liability on certain classes of persons described therein, and does not require any showing that the specific hazardous wastes generated by a given defendant caused the release of hazardous substances that results in remedial action. U.S. v. Monsanto Co., 858 F.2d 160, 167-68 (4th Cir.1988). Liability under CERCLA is joint and several unless a defendant can show that damages are divisible. See e.g. Monsanto, 858 F.2d at 171 & n. 3; U.S. v. Alcan Aluminum Corp., 964 F.2d 252, 267-71 (3rd. Cir.1992) (discussing cases).

Taken together § 9613(f)(1) and § 9613(g)(3) confirm that Textron has a right to contribution from the Defendants if it can show that the Defendants are liable for response costs etc. under § 9607. In pertinent part that section provides:

§ 9607

(a) Covered persons; scope

* * * * * *

(3) any person who by contract, agreement, or otherwise arranged for disposal ... of hazardous substances owned or possessed by such person, by any other party or entity, at any facility owned or operated by another party or entity and containing such hazardous substances, shall be liable for

*ír •i' *i* ‡ ‡

(b) any other necessary costs of response incurred by any other person consistent with the national contingency plan;

42 U.S.C. § 9607(a). In this case, Textron must show that the Defendants arranged with McKee for the disposal of wastes, and those wastes are at the Harwell Road site.

The Defendants have moved for summary judgment pursuant to Fed.R.Civ.Proe. 56, alleging that Textron has failed to produce evidence from which a reasonable person could conclude that any Defendant generated hazardous substances that were disposed of at the Harwell Road site. “Under Rule 56(c), summary judgment is proper ‘if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.’ ” Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 2552, 91 L.Ed.2d 265 (1986) citing Fed.R.Civ.Proc. 56(c). However, Rule 56 does not require the moving party to produce evidence negating an opponent’s claim. Celotex Corp. v. Catrett, 477 U.S. at 323, 106 S.Ct. at 2553. That is, “under Celotex, ‘the moving party on a summary judgment motion need not produce evidence, but simply can argue that there is an absence of evidence by which the nonmovant can prove his case.’ ” Cray Communications v. Novatel Cmptr. Systems, Inc., 33 F.3d 390, 393-94 (4th Cir.1994) (citing 10A Charles A. Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 2720 at 10). This is because “[o]ne of the principal purposes of the summary judgment rule is to isolate and dispose of factually unsupported claims or defenses ...” Celotex, 477 U.S. at 323, 106 S.Ct. at 2553.

“Once a defendant makes a properly supported motion for summary judgment, the burden shifts to the plaintiff to set forth specific facts showing that there is a genuine issue for trial.” Sylvia Development Corp. v. Calvert County, Md., 48 F.3d 810, 817 (4th Cir.1995) citing Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247, 106 S.Ct. 2505, 2510, 91 L.Ed.2d 202 (1986). “[Tjhere is no issue for trial unless there is sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party. If the evidence is merely colorable, or is not significantly probative, summary judgment may be granted.” Anderson, at 250, 106 S.Ct. at 2511. Put another way, there must be a genuine issue for trial. “The mere existence of a scintilla of evidence in support of the plaintiffs position will be insufficient; there must be evidence on which the jury could reasonably find for the plaintiff.” Anderson, at 251, 106 S.Ct. at 2512. “The judge’s inquiry, therefore, unavoidably asks whether reasonable jurors could find by a preponderance of the evidence that the plaintiff is entitled to a verdict ...” Id. Where, as here, a party moves for summary judgment based on lack of proof as to material facts, “the judge must ask himself ... whether a fair-minded jury could return a verdict for the plaintiff on the evidence presented.” Anderson, at 252, 106 S.Ct. at 2512.

As with any summary judgment motion, “the court must draw any permissible inference from the underlying facts established in the record in the light most favorable to the non-moving party.” Austin v. Clark Equipment Co., 48 F.3d 833, 835 (1995). But in order for an inference to be permissible it must be reasonable, and “[wjhether an inference is reasonable cannot be decided in a vacuum; it must be considered in ‘light of the competing inferences’ to the contrary.” Sylvia, 48 F.3d at 818, citing Matsushita Elec. Indus. Co. v. Zenith Radio, 475 U.S. 574, 587, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986). As the Fourth Circuit has stated:

[I]t is the province of the jury to resolve conflicting inferences from circumstantial evidence. Permissible inferences must still be within the range of reasonable probability, however, and it is the duty of the court to withdraw the case from the jury when the necessary inference is so tenuous that it rests merely upon speculation and conjecture.

Sylvia, 48 F.3d at 818 citing Ford Motor Co. v. McDavid, 259 F.2d 261, 266 (4th Cir.1958) (brackets in original). In short, “[w]hen the moving party has carried its burden under Rule 56(c), its opponent must do more than simply show that there is some metaphysical doubt as to the material facts.” Matsushita Elec. Indus. Co. v. Zenith Radio, 475 U.S. 574, 587, 106 S.Ct. 1848, 1356, 89 L.Ed.2d 538 (1986).

In this case, Textron has relied heavily on the testimony of its expert and the Defendants have also albeit to a lesser extent. The admissibility of expert testimony in federal court is governed by Fed.R.Evid. 702. That rule provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

Fed.R.Evid. 702. Recently, in Daubert v. Merrell Dow Pharmaceuticals, Inc., — U.S. -, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the Supreme Court explained that portion of the rule relating to expert “knowledge” by stating, “the word ‘knowledge’ connotes more than subjective belief or unsupported speculation. The term applies to any body of known facts or any body of ideas inferred from such facts or accepted as truths on good grounds.” Id. at -, 113 S.Ct. at 2795. Under the Federal Rules, the trial judge must ensure that any and all expert testimony is not only relevant but also reliable. Daubert, — U.S. at -, 113 S.Ct. at 2786. Further, not every opinion offered by an expert is an expert opinion. Rule 702 “does not afford the expert unlimited license to testify ... without first relating that testimony to some ‘specialized knowledge’ on the expert’s part_” U.S. v. Johnson, 54 F.3d 1150, 1157 (4th Cir.1995). Put another way, an expert’s opinion “must be an ‘expert’ opinion (that is, an opinion informed by the witness’ expertise) rather than simply an opinion broached by a purported expert.” U.S. v. Benson, 941 F.2d 598, 604 (7th Cir.1991).

The Fourth Circuit has recognized that on a motion for summary judgment, the court’s screening role under Fed.R.Civ.Proc. 56, dovetails with the court’s “gatekeeper” function as to expert testimony, see Daubert, — U.S. at -, 113 S.Ct. at 2798-99, and “[t]he credibility of competing experts is a question for the jury only if the party with the burden of proof has offered enough evidence to sustain a verdict in its favor.” Alevromagiros v. Hechinger Co., 993 F.2d 417, 421 (4th Cir.1993). And just as “[a]n expert’s opinion should be excluded when it is based on assumptions which are speculative and are not supported by the record,” Tyger Const. Co. Inc. v. Pensacola Const. Co., 29 F.3d 137, 142 (4th Cir.1994), summary judgment should be granted where expert opinion suggests merely a possibility, as opposed to a probability, “precisely to guard against raw speculation by the fact-finder.” Sakaria v. Trans World Airlines, 8 F.3d 164, 172-73 (4th Cir.1993). The Fourth Circuit has indicated that an expert’s subjective opinion is insufficient to withstand a motion for summary judgment because “we are unprepared to agree that ‘it is so if an expert says it is so.’ ” Alevromagiros, 993 F.2d at 421 (citations omitted); see also Daubert, — U.S. at -, 113 S.Ct. at 2795 (expert knowledge must be more than subjective belief or unsupported speculation). Thus, the inferences a court is asked to draw by expert testimony must be reasonable in light of competing inferences. See Comprehensive Technologies v. Software Artisans, 3 F.3d 730, 737 (4th Cir.1993); see also Anderson, 477 U.S. at 251, 106 S.Ct. at 2512 (standard for summary judgment and directed verdict are the same).

Celanese has moved for summary judgment alleging that Textron has failed to produce evidence from which a reasonable fact-finder could find that it generated hazardous substances that were disposed of at the Har-well Road site. The evidence shows that during the 1960’s and 1970’s, Sou-Tex, a predecessor in interest to Celanese, manufactured textile chemicals and auxiliaries, surfactants, dyestuffs and other products. McKee has testified that on two occasions he was called out to the Sou-Tex plant for the purpose of disposing of some liquid waste. McKee states that on those two occasions, he picked up an extremely foul smelling liquid from two 55-gallon drums and an in-ground tank with a capacity of approximately 400-500 gallons that was located outside the plant. McKee estimated that he removed about a total of 1,000 gallons. According to McKee, the material was kind of grayish and slimy. McKee was told that the tanks’ contents were not toxic or hazardous. He also remembers being told that the liquid waste he picked up used to be discharged into the river.

It is uncontested by the parties that during the period in question the Celanese plant produced a sulfur dye material called Sulfur Black Brown AN Melt, and that “stinkwater” is a by-product of that process of production. The evidence produced by both parties confirms that stinkwater is aptly named and it is unanimously described as the most foul odor to have entered the nasal passages of every witness who testified. McKee testified that he stopped picking up Celanese’s waste because it was so obnoxious; in fact, his wife wouldn’t let him into his house until he had changed his clothes in the barn. McKee’s employees said their wives also barred them from the house pending their satisfaction of similar conditions. According to Celanese, this stinkwater was initially discharged into the Catawba River. However, a combination of increased regulation and complaints by citizens forced Celanese to find another method of disposal. By 1971 Celanese had modified its industrial production process so that the stinkwater was incinerated. Documents from 1969-70 indicate that until that system was in place, Celanese investigated various alternative disposal methods. One of those documents is a memo dated 1970 concerning the disposal of “industrial waste liquors” and McKee is listed as one of three private companies in the area that handled industrial waste.

Based on McKee’s testimony and this evidence, Textron has alleged that McKee hauled stinkwater from Celanese’s plant and disposed of it at the Harwell Road site. In this regard, Textron has offered the opinion of its expert, Mr. Douglas Chatham. Based on his review of the deposition testimony and documents describing how Sulfur Black Brown AN Melt was produced, Chatham concludes that McKee picked up stinkwater from the Celanese plant. Based upon those documents and that conclusion, Chatham then opines that the stinkwater hauled by McKee to the Harwell Road site contained hazardous substances; more specifically, he opines that the stinkwater that McKee hauled to the site contained hydrogen sulfide and methyl mercaptan, as well as the heavy metals zinc and nickel.

A. Evidence that McKee Hauled Cela-nese’s Stinkwater.

For the purposes of its motion for summary judgment, Celanese has assumed that McKee picked up some liquid waste from its plant. Nonetheless, Celanese has attacked Textron’s case and the opinion of Chatham on a number of grounds. First, Celanese argues that no reasonable person could conclude that McKee hauled stinkwater to the site. Here, Celanese emphasizes differences between the testimony of McKee and the testimony of Celanese employees about various details concerning the appearance of stinkwater, its storage and disposal. For example, while McKee described the substance he picked up as grayish, Celanese employees have testified that stinkwater was light brown or dark. Similarly, McKee testified that he pumped out two 55-gallon drums and a tank, which he thought held about 400-500 gallons, that was mostly in-ground; all of the tanks were outside. But some Celanese employees have testified that stinkwater was collected inside the plant and in portable containers before it was taken down and dumped in the river. Celanese also relies upon the testimony of Mr. Thompson, an employee of Celanese, who claims that stink-water was never disposed of off-site. Given these significant differences in testimony, Celanese argues, Textron’s assertion that McKee hauled stinkwater from its plant boils down to nothing more than an argument that what he hauled smelled bad, therefore it must have been stinkwater.

The Court believes that reasonable persons could find that McKee hauled stinkwa-ter from Celanese’s plant to the Harwell Road site. For one thing Mr. Thompson admitted that no odor from any other liquid waste produced by Celanese’s plant could be mistaken for stinkwater. He also testified that any other aqueous wastewater produced by the plant was “normal, and could be disposed of in the city system or in the waste treatment plant.” Although Mr. Thompson has said that no stinkwater was disposed of off-site, Celanese has also stated that there may have been a few times when stinkwater was disposed of in the city’s system. Likewise, Celanese says the tank used to hold the stinkwater was inside the plant, but there is evidence that this tank was brought outside to dump the stinkwater. The evidence shows that usually the stinkwater was dumped into the river behind Celanese’s plant, but that leaves open the possibility that stinkwater hauled outside may have been treated differently, and McKee’s testimony tends to support this conclusion. Under these circumstances, Textron’s inference that McKee hauled stinkwater is not patently unreasonable.

Celanese repeatedly emphasizes testimony that normal stinkwater was yellow-brown in color, whereas McKee said the substance he hauled was grayish. But this difference cannot be dispositive. One of Celanese’s employee’s testified that stinkwater was dark in color, and depending upon their amount and container, liquids can appear to have different degrees of color. And, at any rate, Cela-nese has not said why the color difference, if any, would change the chemical composition of the stinkwater. The same is true of Cela-nese’s argument that Chatham cannot offer any opinion about the “abnormal” stinkwater that Celanese says McKee must have picked up. Apart from differences in the color of the material, which are not clearly pronounced in any case, Celanese seems to think that because these quantities of stinkwater were disposed of differently from other batches of stinkwater, they must have had different chemical compositions. Celanese has not shown that any of the differences it relies upon necessarily indicate a change in the chemical composition of stinkwater — if that is what it was — and therefore Celanese is not entitled to summary judgment for this reason.

Similarly, Celanese’s argument that Chatham’s opinion must be rejected because he has no scientific basis for finding that McKee hauled stinkwater is misplaced. It is true that Chatham’s opinion rests upon his assumption that McKee must have hauled stinkwater because it smelled so bad and that Chatham’s opinion in this regard is largely that of a lay witness. But it is also true that his expertise plays some part in that opinion in so far as it rests upon his knowledge of the by-product created by Celanese’s production process, i.e., the methyl mercaptan he opines was in Celanese’s stinkwater, and the attributes of methyl mercaptan acknowledged by experts in his field (ie., it stinks). Also, it is certain that experts are allowed to premise their opinions on facts that find a basis in the record — which is the case here. Whether Chatham’s assumption that McKee hauled stinkwater is a valid basis for his opinion goes to its weight; but it does not mean that Chatham is speculating.

B. Evidence that Celanese’s Stinkwa-ter Contained Hazardous Substances.

Celanese also argues that no reasonable person could conclude that any stinkwater McKee may have hauled to the Harwell Road site contained hazardous substances. According to Celanese, Chatham’s claim that the stinkwater contained hydrogen sulfide and methyl mercaptan as well as the heavy metals zinc and nickel is sheer speculation. 2 As explained below, the Court finds that Textron has produced evidence from which a reasonable person could find that Celanese’s stinkwater contained hazardous substances.

1. Heavy Metals.

At the outset, the Court agrees with Celanese that no reasonable person could find that its stinkwater contained heavy metals. Initially Textron’s expert opined that the stinkwater hauled by McKee contained heavy metals based upon a 1970 test report that he believed was a test of the stinkwater. Of course, this conclusion was not based upon any expertise, but was merely his inference from the record. But the record simply provides no reasonable support for this inference and Textron’s expert admitted this. At Chatham’s deposition, Celanese pointed out that the test results concerned waters from a tank that had a flow rate of 3500 gallons over two hours while the testimony showed that, at best, Celanese’s plant produced 500 gallons of stinkwater per week. Confronted with this evidence, Chatham admitted that this difference in volume indicated that whatever wastewaters produced the 1970 test result, they were probably not stinkwater, and therefore, he could not use the test results to infer that stinkwater contained nickel and zinc.

The Court agrees with Textron’s expert that no reasonable person could find that Celanese’s stinkwater contained zinc or nickel based on the 1970 report. In its brief, Textron has pieced together circumstantial evidence which it claims shows that the 1970 test result was a test of the stinkwater. But the evidence presented by Textron is far removed from the dispositive point — what wastewaters produced the results recorded in the 1970 test — and it also neglects the fact that Celanese had several wastestreams including the “rinsewater” that was generated by its production processes. Finally, Tex-tron’s theory simply does not address the glaring disparity that exists concerning the amount of stinkwater generated by Cela-nese’s production of the sulphur melt. This evidence makes Textron’s explanation for the test result unreasonable in light of competing inferences. See Sylvia, 48 F.3d at 818. Although it is true that no witness knows exactly how much stinkwater was produced, they all agree that the amount was much smaller than the volume tested in 1970. As Celanese points out, even if one combined all of the estimated weekly flow rates for stinkwater cited by Textron, the total would still not reach the flow rate recorded in the 1970 test for a single two-hour period. Under these circumstances, there is no probative evidence that any stinkwater produced by Celanese and hauled by McKee contained heavy metals. Textron claims that the stinkwater could contain nickel from corrosion in the reactors at Celanese’s plant, but Textron’s scintilla of evidence on this point makes its argument sheer conjecture. The same is true of Textron’s claim that stinkwater contained heavy metals because some lignin sul-fonates — an ingredient for the sulphur melt — contains nickel and zinc. Textron has no idea whether the lignin sulfonate used to create the two batches of stinkwater that might have been hauled by McKee contained nickel or zinc.

2. Hydrogen Sulfíde.

For the purposes of this motion, Celanese has agreed that its stinkwater contained hydrogen sulfide. Nonetheless it denies that Textron has produced credible evidence that any stinkwater hauled by McKee contained that substance. The dispute here focuses on whether hydrogen sulfide would be present in the stinkwater whenever McKee came to pick it up. In its motion, Celanese has pointed out that when hydrogen sulfide is exposed to oxygen it reduces to elemental sulphur, which is not a hazardous substance. According to Celanese, Textron has not produced evidence from which a reasonable person could conclude that hydrogen sulfide was still in the stinkwater when McKee hauled it.

The Court cannot agree. It is true, as Celanese points out, that Textron’s expert admitted in his deposition that hydrogen sulfide would become elemental sulphur, a nonhazardous substance, when exposed to oxygen. He admitted he did not know how long it would take for any hydrogen sulfide in Celanese’s stinkwater to disappear. Further, he admitted that no one knows how long Celanese’s stinkwater was in the holding tanks before it was hauled off by McKee, or how much oxygen it was exposed to prior to its removal. However, there is evidence that Celanese generated up to 500 gallons of stinkwater per week, which is about the quantity of waste that McKee said he hauled on two separate occasions. Also, it is undisputed that because stinkwater reeks it was stored in a sealed container prior to dumping in order to avoid offending Celanese’s neighbors and employees. This evidence supports Textron’s argument that the hydrogen sulfide in Celanese’s stinkwater was not exposed to oxygen to allow biodegradation pri- or to its disposal at the site. Under these circumstances, Chatham’s opinion that any stinkwater hauled by McKee contained hydrogen sulfide is sufficient to withstand Cela-nese’s motion for summary judgment. 3

3. Methyl Mercaptan.

The final hazardous substance that Textron alleges was in the stinkwater that McKee hauled to the Harwell Road site from Celanese’s plant is methyl mercaptan. Here again Celanese alleges that Textron has not produced evidence from which a reasonable juror could agree with Textron on this point. More specifically, Celanese points to the testimony of Textron’s expert, Chatham. Celanese has highlighted the portion of Chat-ham’s deposition wherein he stated that methyl mercaptan “could” have been formed during the production of Sulfur Black Brown AN Melt. According to Celanese, Chatham’s opinion on this point is really just speculation.

The Court thinks Celanese has mischarae-terized Chatham’s testimony. As the Court sees it, when Chatham’s testimony is read as a whole he opined that it was highly likely that methyl mercaptan was generated during the production of the Sulfur Black Brown AN Melt. A review of Chatham’s deposition testimony indicates that he based this opinion on his knowledge of chemistry and the chemical structure of ammonium ligno-sulfonate, an ingredient of Celanese’s Sulfur Black Brown AN Melt, as depicted in the material safety data sheet for that chemical. In a footnote, Celanese argues that Chatham’s reliance on that material safety data sheet is misguided because it does not depict the whole molecular structure of the ammonium ligno-sulfonate; but Chatham testified that such sheets include only the active (and therefore important) part of the molecular structure. Chatham’s reliance on the material safety data sheet goes to the weight of his opinion. It is also true, as Celanese points out in a footnote, that its expert believes that Chatham’s theory concerning how methyl mercaptan would be produced “would be contrary to most of the premises of reaction chemistry,” but the Court cannot choose between these opinions at this point in the proceedings.

C. Evidence that Celanese’s Waste is at the Harwell Road Site.

Celanese also argues that it is entitled to summary judgment because there is no evidence that any hazardous substances contained in its stinkwater are located at the Harwell Road site. Here Celanese argues, and the Court agrees, that Textron has only produced evidence that is arguably sufficient to show that Celanese’s stinkwater contained hydrogen sulfide or methyl mercaptan. Celanese asserts that it cannot be held liable for contamination at the Harwell Road site, because neither of these substances has been detected at the site as of this date. And Textron’s experts have admitted that none of the testing or remediation activities conducted so far has been directed at these substances, so it seems clear that any hazardous substances Celanese generated have not yet caused any response costs. However, as noted previously, there is evidence from which a reasonable person could infer that McKee hauled stinkwater from Celanese’s plant to the site (ie., McKee’s testimony), and there is evidence from which a reasonable person could find that Celanese’s stinkwater contained hydrogen sulfide and methyl mercaptan (ie., Chatham’s testimony).

In its motion, Celanese argues that Tex-tron has failed to satisfy its prima facie case under CERCLA because it has failed to produce evidence that chemical compounds like those allegedly generated at Celanese’s plant and taken to the Harwell Road site are currently at the site. According to Celanese, Textron must at least show that the hazardous substance it allegedly contributed to the site or some similar chemical compound was present at the site at the time of the release in order to satisfy the requirements for so-called “generator liability.” In part, Celanese argues that this requirement follows from the language of § 107 of CERCLA which requires that a person arrange for disposal of hazardous substances with a person who owns a facility “containing such hazardous substances.” (Emphasis added). 42 U.S.C. § 9607(a)(3). Celanese also relies on language from the Fourth Circuit’s opinion in U.S. v. Monsanto Co., 858 F.2d 160, 169 (4th Cir.1988), that “[a]bsent proof that a generator defendant’s specific waste remained at a facility at the time of the release, a showing of chemical similarity between hazardous substances is sufficient.” (Emphasis added). According to Celanese, Monsanto requires Textron to produce evidence that chemicals similar to those generated by Celanese are at the site at the time of the release in order to establish a prima facie ease of CERCLA liability. Here, Celanese argues that any hazardous substances it may have contributed to the site biodegraded decades ago. Therefore, Celanese asserts, absent evidence that those substances were at the site when the EPA issued its order, Textron has failed to show that the Harwell Road site contained hazardous substances generated by Celanese at the time of the release. Thus, Celanese concludes, Textron has failed to make out a case for generator liability.

For a number of reasons, Celanese’s argument must be rejected. First, the ultimate issue before the Court is whether Textron has produced evidence that Celanese arranged for the disposal of its waste containing hazardous substances with McKee, and the Harwell Road site is a facility owned by McKee “containing such substances.” 42 U.S.C. § 9607(a)(3). So the question is whether Textron’s evidence, if believed, is sufficient to establish that the Harwell Road site contains Celanese’s hazardous substances. The Court finds that Textron’s evidence that McKee dumped stinkwater at the Harwell Road site, if believed, satisfies its prima facie case under CERCLA. If that testimony is believed, then it establishes that Celanese’s stinkwater was released at the Harwell Road site, and therefore, that the site contains Celanese’s waste. Despite Celanese’s assertion, there is no requirement that Textron use analytical evidence to show that the site contains its specific waste where other evidence does so. Thus, in the Court’s view, this case fits within the first part of the sentence relied upon by Celanese from Monsanto because it is a case where there is evidence that Celanese’s waste remained at the site: it was dumped there.

Given this evidence that Celanese’s waste was dumped at the site, Celanese’s argument is just a variation on the “causation” requirement that was rejected by the court in Monsanto. In that case, the government had shown that drums removed from the site bore the markings of the defendants. The court rejected the defendants’ argument that CERCLA required the government to show that the specific waste generated by the defendants caused the harm to the environment by being present at the time of the release. It noted that “CERCLA plaintiffs need not perform exhaustive chemical analyses of hazardous substances found at the disposal site. They must, however, present evidence that a generator defendant’s waste was shipped to a site and that hazardous substances similar to those contained in the defendants’ waste remained present at the time of the release.” Monsanto, 858 F.2d at 169 n. 15. In Monsanto the Fourth Circuit cited United States v. South Carolina Recycling & Disposal, Inc., 653 F.Supp. 984 (D.S.C.1984), a case affirmed in the Monsanto opinion. There, the district court observed that under CERCLA circumstantial evidence that a generator’s waste was disposed of at a site, “such as by identification of a generator’s drum at the site during cleanup or by way of documentary or circumstantial proof that the wastes were hauled to the site absent proof that they were subsequently taken away”, should be sufficient to satisfy the requirement that a site contain a generator’s hazardous substances. South Carolina Recycling & Disposal, Inc., 653 F.Supp. at 993 n. 6; See also U.S. v. Alcan Aluminum Corp., 964 F.2d 252, 256 n. 3 (3rd Cir.1992). Further support for this reading of Monsanto comes from the Fourth Circuit’s opinion in Nurad, Inc. v. William E. Hooper & Sons Co., 966 F.2d 837 (4th Cir.1992). In that case, the Fourth Circuit rejected the defendants’ argument that CERCLA required persons seeking cost recovery to pinpoint the precise moment in time when a release of hazardous substances occurred. The court emphasized that the plaintiff had presented evidence that the leaks of hazardous substances at issue had occurred gradually over a long period of time, and the defendants had not rebutted that evidence.

Both Nurad and Monsanto emphasize that Congress intended CERCLA to impose strict liability on persons who arranged for the disposal of hazardous substances on those generators and has allocated to defendants the burden of proving that any hazardous substances they generated have not caused response costs. The primary purpose underlying this allocation of burdens is to foster the prompt remediation of dangerous contamination. This Court’s reading of Monsanto fosters this goal, and is consistent with the statutory language requiring Textron to show that the Harwell Road site is a facility containing hazardous substances like those generated by Celanese. Therefore, the Court concludes that a party can satisfy its burden of showing that a facility contains hazardous substances generated by a given defendant by circumstantial evidence that such materials were dumped at the site and were not removed prior to the release that caused response costs. Textron has satisfied this burden.

Although Celanese asserts that its argument is consistent with Monsanto, the Court sees no significant difference between Cela-nese’s arguments and those rejected by the Monsanto court. In light of McKee’s testimony, Celanese cannot state that there is no evidence that the site contains its hazardous substances. Celanese admits as much when it argues that there is no analytic data that the site contains those hazardous substances. Celanese’s assertion that CERCLA requires Textron to show that its specific waste or a similar substance was at the site at the time of the release is just another way of asserting that CERCLA requires plaintiffs to show that a specific defendant’s hazardous substance caused harm to the environment and response costs. That requirement was rejected in Monsanto. Monsanto, 858 F.2d at 160, n. 15. Indeed, Celanese implicitly concedes as much in its argument. First, it emphasizes that its waste was “released” 20 years ago and therefore was not present at the time of the “release” that caused the EPA’s action and Textron’s response cost. Similarly Celanese states, in a footnote, that its waste has been “removed” through biode-gradation. In essence, Celanese is arguing that any hazardous substances in its waste were released at the site a long time ago, have long since biodegraded, and therefore, those hazardous substances have not caused the contamination that required remediation. Celanese’s arguments on these points are persuasive, but those arguments go to damages, not liability precisely because CERC-LA imposes strict liability on any party whose waste is contained at the site. Monsanto, 858 F.2d at 167-68; see also Bell Petroleum Services, Inc., 3 F.3d 889, 893 at n. 4 (5th Cir.1993) (citing cases). Textron has produced significantly probative evidence that the Harwell Road site is a facility containing hazardous substances generated by Celanese — if Textron is believed. Requiring Textron to prove that those chemicals are still there, and have not biodegraded because of their chemical properties, is to require the very “finger-printing” of waste rejected in Monsanto.

D. Joint and Several Liability.

This last point brings the Court to the final issue raised by Celanese’s motion: whether joint and several liability is appropriate in this case. In its brief Celanese has argued, and this Court agrees, that at best Textron has produced evidence that any stinkwater taken from Celanese and dumped at the site would contain methyl mercaptan and hydrogen sulfide. Next, Celanese emphasizes that none of the response activity to date has been directed towards the identification or removal of those hazardous substances; and Textron’s own experts forthrightly conceded that point. Based on this evidence, Celanese has argued that it has no liability for the contamination of the Harwell Road site. For the reasons given above, that argument is rejected.

However, Celanese has also asserted that any harm caused by the hazardous substances it generated is “divisible” such that it has no liability for response costs incurred thus far. In response, Textron argues that because it has produced evidence that the Harwell Road site contains hazardous substances generated by Celanese, Celanese is liable for certain response costs, such as the costs of testing at the site, even if other costs of remediation cannot be attributed to hazardous substances generated by Celanese. Also, Textron notes that it has sought declaratory relief concerning liability for future response costs incurred to decontaminate the Harwell Road site, and it may be that Cela-nese’s hazardous substances will require future remediation. In this regard Textron notes that CERCLA requires courts to enter declaratory relief in actions for cost recovery. 42 U.S.C. § 9613(g)(2) (“In any such [cost recovery] action ... the court shall enter declaratory judgment on liability for response costs or damages that will be binding on any subsequent action or actions to recover further response costs or damages.”) (emphasis added); see also Kelley v. E.I. DuPont De Nemours and Co., 17 F.3d 836, 844-45 (6th Cir.1994).

Celanese’s argument concerning divisibility has implications for its potential liability in this case. In Monsanto, the Fourth Circuit noted that “[w]hile CERCLA does not mandate the imposition of joint and several liability, it permits it in cases of indivisible harm.” Monsanto, 858 F.2d at 171. Whether joint and several liability is appropriate “focuses principally on the divisibility among responsible parties of the harm to the environment.” Id. at n. 22. Joint and several liability is not required where (1) there are distinct harms, or (2) there is a reasonable basis for determining the contribution of each cause to a single harm. Id. at 172, citing Restatement (Second) of Torts § U33A (1965). The defendant bears the burden of proving that the harm at issue is divisible. Id. at n. 24 citing Restatement (Second) of Torts § b33B(2) (1965). Since the Fourth Circuit’s decision, other courts have adopted its approach to determining whether joint and several liability is appropriate. See U.S. v. Alcan Aluminum Corp., 964 F.2d 252, 267-271 (3rd Cir.1992); U.S. v. Alcan Aluminum Corp., 990 F.2d 711, 721-23 (2nd Cir.1993); Matter of Bell Petroleum, 3 F.3d at 895-902.

The Fourth Circuit has not addressed the proper timing of the divisibility inquiry. Here, the question is whether this inquiry should be made at the liability or damages portion of CERCLA actions, which are typically bifurcated by reason of their complexity. 4 However, those Circuits that have considered this issue have concluded that the timing of this “divisibility” inquiry is best left to the discretion of the trial court. See Matter of Bell Petroleum, 3 F.3d at 901-02; Alcan, 990 F.2d at 723; Alcan, 964 F.2d at 270 n. 29. The Court believes that the Fourth Circuit would adopt this approach which is consistent with its observation that CERCLA does not require the imposition of joint and several liability. See Monsanto, 858 F.2d at 171.

The relationship between CERCLA’s liability principles is not easy to understand. On the one hand, it is well-settled that CERCLA imposes strict liability and does not require the government (or a private plaintiff) to show that the hazardous substances generated by a specific defendant caused response costs. Monsanto, 858 F.2d at 167 & n. 11. On the other hand, Congress did not want CERCLA to require joint and several liability in every case. Id. at 171. Theoretically, it is possible to distinguish strict (faultless) liability from joint and several liability. However, in cases like this one, allowing a defendant to show that a harm which generated response costs is divisible may allow a defendant to show that none of its hazardous substances caused response costs. Such a result cuts against the notion that CERCLA imposes strict liability and requires no showing that hazardous substances generated by a specific defendant caused response costs. See Alcan, 990 F.2d at 721-23 (acknowledging that the joint and several liability determination allows causation to be “brought in the back door”); Alcan, 964 F.2d at 269-71 (discussing relationship between causation and divisibility) and Id. at n. 29 (discussing relationship between divisibility and contribution inquiry); In re Matter of Bell Petroleum, 3 F.3d at 896-97 (discussing these matters). It seems the point is to shift the burden of proving causation to defendants. See Alcan, 964 F.2d at 269 5 ; Matter of Bell Petroleum, 3 F.3d at 896. And a defendant’s ability to evade joint and several liability functions as a defense that exists under the federal common law of CERCLA with the express approval of Congress. Also, under the Restatement approach that has been adopted to determine when joint and several liability is appropriate, there may be circumstances that justify holding a defendant liable even where he can show there are distinct harms or a reasonable basis for apportionment, see Matter of Bell Petroleum, 3 F.3d at 896, 901, but see Id. at n. 13. In any event, there is no question that the Fourth Circuit has recognized that CERCLA does not require joint and several liability and that is what counts.

The Court believes that the determination as to whether joint and several liability is appropriate in this case should be made in the liability portion of these proceedings. See Alcan, 964 F.2d at 270 n. 29. In this regard, the Court notes that Celanese is the only defendant to whom Textron attributed the generation of hydrogen sulfide and methyl mercaptan. Also, Celanese has offered some evidence that any such contamination it contributed to the site has long since biodegraded, and there is some evidence from Celanese’s expert that hydrogen sulfide currently present at the site would be a byproduct of the ongoing breakdown of hazardous substances contained therein. Further, it appears highly likely that Textron is liable for the vast majority of hazardous substances located at the site and that Celanese’s contribution, if any, is infinitesimal. But further discovery directed towards the apportionment of liability is likely to be time-consuming and expensive, and a determination that several liability is appropriate may well obviate the necessity for a contribution phase in these proceedings. See Matter of Bell Petroleum, 3 F.3d at 898 n. 10. Therefore, the Court will extend the discovery and briefing issues as to the liability portion of this proceeding in a separate Order entered simultaneously with this opinion so the parties can address this matter further.

NOW, THEREFORE, IT IS ORDERED that Defendant Hoechst Celanese Corporation’s Motion for Summary Judgment, filed December 16, 1994 (document # 81), be, and hereby is, DENIED.

1 . Like the parties, the Court refers to the Harwell Road site as such; but in his deposition testimony McKee refers to the properly as the Beatty Road site.

2 . Initially, Textron's expert also asserted that any stinkwater hauled by McKee would contain 2,4 diaminotoluene; but Textron has made no mention of this in its reply brief, which is prudent because on cross-examination Chatham admitted that he was merely speculating on this point.

3 . Later in its brief, Textron argues that a smell like that of rotten eggs detected by workers at the Harwell Road site is sufficient to show that there is hydrogen sulfide there. Assuming this is true, the Court cannot assume from the presence of hydrogen sulfide at the site, that Celanese produced that hydrogen sulfide. Textron cites the opinion of Norris, Celanese's expert, to show that Celanese’s hydrogen sulfide could be at the site. But the substance of his testimony is quite the opposite. Norris testified that he would expect hydrogen sulfide to be present at any site where sulphur compounds were present, even if the original waste was composed of non-volatile sul-phur compounds. If Norris is believed, the mere fact that there is hydrogen sulfide at the site says nothing about whether any hydrogen sulfide contributed by Celanese has caused Textron to incur response costs.

4 . In Monsanto, the court found that the harm was indivisible and the generator defendants conceded this point, Id. at 166, 172-73. Also, the district court had granted summary judgment for the defendant and found, in connection with that motion, that the harm was not divisible, so it appears that this decision was made in connection with the court's liability determination. See also, Alcan, 964 F.2d at 271 n. 30.

5 . In that case, the court held that CERCLA allowed defendants to show that a harm was divisible and stated:

Onr conclusions on this point are completely consistent with our previous discussion on causation, as there we were concerned with the Government’s burden of demonstrating liability in the first instance. Here we are dealing with liability otherwise established.

Id. at 269.