Jackson v. Anchor Packing Co.

WOLLMAN, Circuit Judge.

This is a “tireworker” asbestos case arising under the district court’s diversity jurisdiction. Ten former employees of the Mohawk Tire and Rubber Company plant in West Helena, Arkansas, (“Mohawk”) filed suit against numerous manufacturers of asbestos-containing products.¹ After considerable discovery, the district court² granted the defendants’ motions for summary judgment on the ground that the plaintiffs had failed to produce sufficient evidence that exposure to the defendants’ products had proximately caused the plaintiffs’ injuries, an essential element of their claims under Arkansas law. On appeal, the plaintiffs contend that the district court erred in adopting a proximate causation standard that is more stringent than Arkansas law requires. Alternatively, the plaintiffs contend that their evidence in fact met the causation standard that the district court employed. We affirm.

I.

Plaintiffs worked in the Mohawk tire plant for terms ranging from sixteen to twenty-three years during the plant’s operation from 1956 to 1979. They performed various jobs at the plant and will be divided into three categories for purposes of this opinion. The first group of plaintiffs performed general “fireworker” duties and did not directly handle any of the defendants’ asbestos products (the “bystander plaintiffs”). This group includes the following plaintiffs: Claudy Jackson (1963-79), Horace Griffin (1957-79), John Thrower (1957-79), LaVaughen Freeman (1959-79), Johnny Griffin (1958-79), John Holloway (1960-79), David Treadaway (1960-79), and David White (1960-79). The second category consists solely of plaintiff George Bokker, who worked as a boiler operator at the plant from 1958-74 and occasionally replaced gaskets and packing materials that allegedly contained asbestos.³ The third category consists only of plaintiff Billy Carr, who worked as a maintenance mechanic from 1956-62 and as a machinist from 1962-79. As part of his duties as a maintenance mechanic, Carr directly handled pipe insulation, gaskets, and packing material that allegedly contained asbestos.

Likewise, the defendants fit into categories according to the type of products they manufactured. The first category consists of those defendants that manufactured pipe-covering insulation and block insulation products: Owens-Corning Fiberglas Corp. (“Owens-Corning”), which produced asbestos-containing “Kaylo” pipe and block insulation from 1956 until 1972; Owens-Illinois, Inc. (“Owens-Illinois”), which produced “Kaylo” pipe and block insulation from 1948 until 1958; Fibreboard Corp. (“Fibreboard”), which produced PABCO pipe insulation; and Keene Corp. (“Keene”), which is involved only in the suit brought by plaintiff David Treadaway as the alleged successor to Ehret & Hill Manufacturing Co., which manufactured Ehret pipe insulation. The second category consists of the manufacturers of gaskets and packing material: Anchor Packing Co. (“Anchor”); A.W. Chesterton Co. (“Chesterton”); John Crane, Inc. (“Crane”); and Garlock, Inc. (“Garloek”). The third category consists solely of Rexnord Corporation, which the plaintiffs are suing as the successor to Stearns Division, which manufactured asbestos-containing brake linings. (For simplicity, we will refer to the alleged predecessor and successor as “Rexnord.”) The fourth category consists solely of Georgia Talc Co. (“Georgia Talc”), which the plaintiffs allege produced asbestos-containing talc or soapstone that was used to keep the rubber at the plant from sticking to itself and the machinery.⁴

The plaintiffs filed their initial complaint on January 5, 1990. The complaint alleges that the plaintiffs were exposed to the products of all the defendants during their work careers at the Mohawk plant. Additionally, plaintiff David Treadaway alleges that he was exposed to insulation manufactured by Owens-Corning, Owens-Illinois, and Keene while he was stationed aboard the U.S.S. Boston from 1957 to 1960. The plaintiffs’ complaint and the deposition testimony of their doctors state that all of the plaintiffs have developed asbestos-related medical conditions, including asbestosis, lung cancer, and thickening of the lungs’ outer membrane (the pleura).⁵ The plaintiffs allege that the defendants are liable for their injuries on the theories of negligence, strict products liability, and market share or alternative liability.

After both sides had conducted exhaustive discovery, all of the defendants moved individually for summary judgment on the ground that the plaintiffs had failed to discover sufficient evidence to support a finding that they had been exposed to a particular defendant’s products and that such exposure was the proximate cause of their conditions. The defendants argued that Arkansas negligence and strict liability law require a prima facie showing by a plaintiff that a specific defendant’s products constituted a substantial factor in producing the plaintiffs injuries.

In response, the plaintiffs contended that Arkansas law requires them to show only that they have contracted diseases caused by exposure to asbestos and that the defendants supplied asbestos-containing products to the Mohawk plant. The plaintiffs based their contention on the Arkansas Supreme Court’s discussion of alternative liability theories in Woodward v. Blythe, 249 Ark. 793, 462 S.W.2d 205, 209-10 (1971) (“Woodward II”), and on the “fiber drift” theory of causation.⁶ They argued that their product identification deposition testimony had placed all of the defendants’ products in the Mohawk plant during the plaintiffs’ work careers and that their expert witness, John M. Dement, PhD., had supplied sufficient evidence of the ability of asbestos fibers to travel on air currents throughout the large, open Mohawk plant.

In three separate orders, the district court considered the defendants’ motions for summary judgment. The court first addressed the proper standard of causation under Arkansas substantive law and the impact of the plaintiffs’ evidence supporting the “fiber drift” theory. Relying heavily on its own opinion in Rogers v. Armstrong World Indus., 744 F.Supp. 901 (E.D.Ark.1990), the district court concluded that Arkansas courts have not embraced the concepts of alternative or market share liability, but require in all cases a showing that the defendant’s conduct constituted a substantial contributing factor in causing’the plaintiffs injuries. The court held that what has become known as the “frequency, regularity, and proximity” test of causation provides the correct measure of the substantial factor causation requirement in asbestos-exposure cases.

The district court, however, did not discard entirely the plaintiffs’ evidence of fiber drift. Following the Third Circuit’s lead in Robertson v. Allied Signal, Inc., 914 F.2d 360 (3d Cir.1990), the district court held that the plaintiffs’ “fiber drift” evidence could be used to expand the area of possible exposure under the proximity prong of the causation test, but could not by itself satisfy the full proximate cause requirement.

After thoroughly reviewing the plaintiffs’ proffered evidence under this standard, the court granted summary judgment in favor of all defendants. The district court found that, as to some defendants, the plaintiffs had failed to show that they had been exposed to products of the defendant that had contained asbestos. With respect to the remainder of the defendants, the court found that the plaintiffs had failed to introduce sufficient evidence that the defendant’s products had been regularly and frequently used in proximity to the plaintiffs’ work areas.

On appeal, we must initially decide whether the district court adopted the proper test of proximate causation in an asbestos-exposure case under Arkansas law. Second, we must examine the evidence proffered by the plaintiffs to determine whether it would support a finding of proximate causation under the standard we adopt.

II.

The district court properly looked to the substantive tort law of Arkansas for the appropriate causation standard. See Erie B.R. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 822, 82 L.Ed. 1188 (1938); American Home Assurance Co. v. Major Tool & Mach, Inc., 767 F.2d 446, 447 (8th Cir.1985). Because the Arkansas Supreme Court has not addressed the proper standard of proximate causation in asbestos-exposure cases, the district court’s task was to predict how the Arkansas Supreme Court would resolve the issue if confronted with it. Commissioner v. Estate of Bosch, 387 U.S. 456, 465, 87 S.Ct. 1776, 1783, 18 L.Ed.2d 886 (1967). On appeal, we review de novo the district court’s prediction that the Arkansas Supreme Court would require a plaintiff to satisfy the “frequency, regularity, and proximity” test. Salve Regina College v. Russell, 499 U.S. 225, -, 111 S.Ct. 1217, 1225, 113 L.Ed.2d 190 (1991).

Courts across this country have struggled with the appropriate quantum of proof that asbestos plaintiffs should be required to produce in order to submit the issue of causation to a jury. The courts have recognized that, given the nature of asbestos exposure in large industrial settings and the long latency periods for asbestos-related diseases, plaintiffs (especially bystanders) face a formidable task in showing, after many intervening years, exposure to a particular defendant’s asbestos product and that exposure’s causation of the plaintiffs injuries. See, e.g., Blackston v. Shook & Fletcher Insulation Co., 764 F.2d 1480, 1485 (11th Cir.1985). On the other hand, the courts have also recognized the countervailing interest that states and defendants have in retaining the longstanding requirement of proximate cause in tort law. See id. at 1482-83 (summarizing cases that discuss the reasons for retaining the proximate cause requirement in asbestos-exposure eases).

In response to this problem, a majority of courts have adopted the “frequency, regularity, and proximity” standard set forth by the Fourth Circuit in Lohrmann v. Pittsburgh Corning Corp., 782 F.2d 1156, 1162-63 (4th Cir.1986), as the proper articulation of the proximate cause requirement. Under the Lohrmann test, a plaintiff must present “evidence of exposure to a specific product on a regular basis over some extended period of time in proximity to where the plaintiff actually worked” in order to survive a defendant’s motion for summary judgment. Id. While most courts have adopted this articulation of the causation requirement in asbestos cases, they have applied it in differing ways to reach opposite results. Compare Lohrmann, 782 F.2d at 1163; Eckenrod v. GAF Corp., 375 Pa.Super. 187, 544 A.2d 50, 52 (granting summary judgment in spite of plaintiffs evidence of direct handling of defendants’ products), appeal denied, 520 Pa. 605, 553 A.2d 968 (1988), with Slaughter v. Southern Talc Co., 949 F.2d 167, 171-72 (5th Cir.1991) (reversing summary judgment where bystander plaintiff asserted merely that defendant’s products had been used “all over the plant,” but could not specifically show their use near the plaintiffs work area).

The plaintiffs urge us not to adopt the Lohrmann test as the standard under Arkansas law. They contend that although Arkansas law typically requires a plaintiff to prove proximate causation, the Arkansas courts would allow a lower standard of proof in situations such as asbestos-exposure cases, where plaintiffs are often unable to prove exposure to a particular product and a disease resulting from that exposure. They claim that the Arkansas Supreme Court has adopted the theory of alternative liability in such concurrent-cause cases, allowing a plaintiff to show the defendants’ connection to the plaintiffs injuries and then shifting the burden to the defendants to absolve themselves of liability. See Woodward II, 462 S.W.2d at 209-10.

We note at the outset that Arkansas tort law has traditionally required plaintiffs to prove proximate causation whether they are proceeding under a negligence or strict liability theory. See, e.g., Ark.Code Ann. §§ 4-86-102(a) and 16-116-102(5) (Michie 1991 & 1987) (strict products liability); Ellsworth Bros. Truck Lines, Inc. v. Canady, 245 Ark. 1055, 437 S.W.2d 243, 244 (1969) (negligence). Arkansas courts have defined a proximate cause as a “cause which, in a natural and continuous sequence, produces damage and without which the damage would not have occurred.” Rogers v. Armstrong World Indus., 744 F.Supp. at 904 (citing Ark. Model Jury -Instr. — Civil, 3d ed. 501). Where several defendants’ negligent acts may have produced the plaintiffs injury, to be considered a proximate cause an individual defendant’s tortious conduct must constitute a substantial contributing factor in bringing about the plaintiffs damages. Rogers, 744 F.Supp. at 905.

We now turn to the question whether the Arkansas Supreme Court’s decision in Woodward II altered the traditional proximate causation standard in certain situations. The Woodward cases addressed an automobile crash in which the plaintiffs car had been struck by two automobiles. See Woodward v. Blythe, 246 Ark. 791, 439 S.W.2d 919, 920 (1969) (“Woodivard /”) (describing the accident in detail). The accident occurred when Kermit Blythe swerved into the oncoming lane to avoid hitting one defendant’s stalled, unlighted automobile that was partially occupying Blythe’s lane. Id. When Blythe crossed the center line, his car was struck by an oncoming automobile, causing Blythe’s ear to spin around and end up in its original lane facing backward. Id. Defendant Woodward’s ear, which had been closely following Blythe’s car, then smashed into the left front end of Blythe’s automobile, pushing it 50 feet down the road. Blythe was found dead in his car. Id.

After a jury verdict against him, Woodward appealed to the Arkansas Supreme Court, contending that Blythe’s estate had failed to show that Blythe was alive at the time of the second impact and that Woodward’s car was thus a proximate cause of Blythe’s death. Id. at 921. The court agreed that although Blythe’s estate had established Woodward’s negligent conduct, it had not introduced any evidence that Blythe was still alive at the moment of the second impact. Id. at 922. Thus, the court concluded that any finding of causation against Woodward would necessarily rest upon speculation and conjecture. Id. The court did, however, remand the case for a new trial so that the estate could attempt to supply the missing proof. Id.

After a retrial and a second verdict for the estate, Woodward once again appealed to the Arkansas Supreme Court, again alleging that the estate had failed to prove proximate causation. This time the Arkansas Supreme Court disagreed. The court noted that the estate at retrial had introduced expert testimony that the first collision with the oncoming cai' had probably not caused the severe brain injury and internal injuries that had killed Blythe, but rather that the impact of Woodward’s ear on the left front of Blythe’s car had produced Blythe’s fatal injuries. Woodward II, 462 S.W.2d at 208. The court held that, given the expert testimony, “the jury certainly had substantial evidence ... that appellant’s negligence was a proximate or contributing cause of the fatal injuries.” Id.

Nonetheless, Woodward argued that the estate should have been able to recover only by “specifically delineating the precise injuries proximately caused by the second impact.” Id. The court rejected this argument as requiring a plaintiff to overcome a nearly insurmountable barrier. The court stated:

If proof of the precise damage proximately caused by each tortfeasor were nonetheless required, as appellant seems to insist, then joint and several liability would be an unnecessary luxury for a plaintiff since by so proving proximate cause he would simultaneously be apportioning damages among the tortfeasors and thereby limiting his recovery in accordance therewith.

Id. at 209.

The court then proceeded to “note that some jurisdictions have extensively minified or even completely eliminated the requirement of proving proximate causation in factual situations similar to the case at bar.” Id. The court discussed several cases in which other states had adopted alternative liability theories, citing the case most familiar to lawyers and law students, Summers v. Tice, 33 Cal.2d 80, 199 P.2d 1 (1948). It is on this discussion that the plaintiffs focus, asserting that the Arkansas Supreme Court thereby adopted alternative liability as a substitute measure of proximate cause in concurrent negligence cases.

We believe that the plaintiffs misread Woodivard II. In both Woodward I and Woodward II, the Arkansas Supreme Court analyzed whether the estate’s evidence was sufficient to support a finding of proximate cause. The court specifically found in Woodivard II that the expert testimony on retrial had cured the defect found in Woodward I and had satisfied the traditional proximate causation requirement. Woodward II, 462 S.W.2d at 208 (quoted above). Thus, the court did not need to adopt a lower standard of alternative liability to uphold the verdict for Blythe’s estate. In rejecting Woodward’s attempt to require the estate to delineate the precise injuries proximately caused by Woodward’s negligence, the court held only that a plaintiff need not apportion damages among concurrent tortfeasors after the plaintiff has initially established that each tortfeasor’s conduct constituted a substantial factor in causing the plaintiffs injuries. See id. at 209. Moreover, the court’s subsequent discussion of alternative liability cases appears to be nothing more than dicta. Id. (“It is interesting to note that some jurisdictions have extensively minified....”). Thus, we conclude that Arkansas has not adopted alternative or market share liability, but has retained the traditional requirement of proximate cause in all tort cases.

Accordingly, asbestos plaintiffs in Arkansas must introduce sufficient evidence to allow a jury to find that more likely than not their exposure to a particular defendant’s product was a substantial factor in producing their injuries. We agree with the district court that the “frequency, regularity and proximity” test, as we interpret it, provides the correct articulation of this standard in asbestos-exposure eases. Consequently, to survive a motion for summary judgment under Arkansas law, an asbestos plaintiff must show that the defendant’s asbestos products were used with sufficient frequency and regularity in locations from which asbestos fibers could have travelled to the plaintiffs work areas that it is probable that the exposure to the defendant’s asbestos products caused the plaintiffs injuries.

Moreover, we also agree with the district court’s adoption of the Third Circuit’s reasoning in Robertson v. Allied Signal, Inc. that a plaintiffs evidence of “fiber drift” may be used to widen the area of probable exposure surrounding the plaintiffs work station. See Robertson, 914 F.2d at 383. We should not ignore competent evidence based on scientific studies and specific fact-based analysis that asbestos fibers may be disseminated on air currents for certain distances. Indeed, the proximity prong of the Lohrmann standard recognizes that plaintiffs need not handle the defendant’s asbestos products directly, but need only be within a reasonable proximity to the use of the defendant’s products. Id. at 375. “Fiber drift” evidence by itself, however, does not show that the plaintiffs were exposed to a defendant’s asbestos often enough to support a reasonable conclusion that the exposure more likely than not was a substantial factor in producing the plaintiffs’ diseases.

The plaintiffs’ evidence supporting their “fiber drift” theory consists solely of the affidavit of Dr. John Dement, who holds a PhD. in industrial hygiene. Dr. Dement’s affidavit provides little support for the plaintiffs’ case. The affidavit states in a conclusory fashion that asbestos fibers could have travelled throughout the entire Mohawk plant, creating a “potential for some asbestos exposure” to all Mohawk employees because asbestos fibers can be suspended and resuspended by activity to and around asbestos-containing products. Dr. Dement’s affidavit provides no scientific studies that demonstrate the movement of asbestos fibers throughout a manufacturing plant the size of the Mohawk facility. Additionally, it is evident that Dr. Dement did not visit the Mohawk plant to conduct tests to determine air flow patterns and the extent to which asbestos fibers could have been disseminated. The lack of scientific evidence gathered from studies done in facilities similar to the Mohawk plant or fact-based evidence gathered from the Mohawk plant itself render Dr. Dement’s testimony unhelpful to the court’s determination concerning how far asbestos fibers could have travelled in the Mohawk plant. Conelusory affidavits, even from expert witnesses, do not provide a basis upon which to deny motions for summary judgment. See Fed.R.Civ.P. 56(e) (affidavits responding to opponent’s summary judgment motion must “set forth specific facts showing that there is a genuine issue for trial”); Miller v. Solem, 728 F.2d 1020, 1024 (8th Cir.) (“conclusive assertions of ultimate fact are entitled to little weight”), cert. denied, 469 U.S. 841, 105 S.Ct. 145, 83 L.Ed.2d 84 (1984).

III.

Having concluded that the district court employed the appropriate standard for analyzing proximate causation, we now turn to the district court’s application of that standard to the plaintiffs’ evidence. “In reviewing a grant of summary judgment, we apply the same standard as that applied by the district court. We thus will affirm the lower court’s grant of summary judgment if there is no genuine issue as to any material fact and ... the moving party is entitled to judgment as a matter of law.” Meester v. IASD Health Servs. Corp., 963 F.2d 194, 196 (8th Cir.1992) (quoting Meyer v. Barnes, 867 F.2d 464, 466 (8th Cir.), cert. denied, 493 U.S. 825, 110 S.Ct. 86, 107 L.Ed.2d 51 (1989)); see Fed.R.Civ.P. 56. Rule 56 requires the entry of judgment “against a party who fails to make a showing sufficient to establish the existence of an element essential to that party’s case, and on which that party will bear the burden of proof at trial.” Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 2552, 91 L.Ed.2d 265 (1986). In determining whether the plaintiffs’ evidence has established a genuine issue of material fact, we must view the evidence in the light most favorable to the plaintiffs (as the nonmoving party) and give them the benefit of all reasonable inferences. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986); Alpine Elec. Co. v. Union Bank, 979 F.2d 133, 135 (8th Cir.1992).

A. Owens-Corning Fiberglas Corp.

The plaintiffs’ evidence establishes at least a probability that Owens-Corning’s Kaylo pipe and block insulation products were present in the Mohawk plant. The deposition testimony of Rivers Billingsley, a storeroom clerk at the Mohawk plant, states that he saw Owens-Corning insulation products in the plant. Billingsley was unable, however, to state how much Owens-Corning material had been present and during which time periods. Pauline Moore, a secretary in the Mohawk purchasing department, also testified that Mohawk had purchased products from Owens-Corning.

Several deposition witnesses further testified that they had installed Owens-Corning insulation on the steam pipes at the plant. Homer Steinbeck, a welder and steam pipe repairer, testified that he had used Owens-Corning pipe insulation when repairing sections of steam pipe. Jack Kimberling, an employee of an insulation contractor that did some work at the Mohawk plant, also testified that he had used Kaylo insulation at Mohawk. Robert Brackin, who worked at the Mohawk plant from 1956 to 1979, stated that he had installed some new insulation at the plant and had performed pipe insulation patchwork during his career. Brackin testified that he had used Kaylo insulation in his work, but was not able to state when he had used it or in what quantities. When asked to describe when he had done the patchwork on the steam pipe insulation, Brackin replied that the patchwork was “an ongoing process” that he was not sure was ever fully completed.

The only plaintiff who may have directly handled Owens-Corning products was Billy Carr. Carr testified that during his six years as a maintenance mechanic he did a considerable amount of pipe insulation repair work and some new installation work, although most of the large installation jobs had been awarded to outside contractors. He could not recall the brand names of any of the insulation products that he had used because he and his coworkers “weren’t interested in the manufacturer at the time, all we was [sic] interested in was grabbing a piece out of the box and putting it on [the pipe].” Carr also testified that during his years as a machinist he had worked on machine parts that were covered with insulation dust and carbon black when they were brought into his machine shop.

The only other possible exposure to an Owens-Corning asbestos product involves David Treadaway’s service aboard the U.S.S. Boston from 1956 to 1960. Treadaway himself was not able to state that he had been exposed to Owens-Corning insulation on board the Boston. He did offer the deposition testimony of James Bruce, who worked in the Norfolk Naval Yard from the 1950’s until approximately 1967. Bruce testified that he and his fellow workers had used Kaylo insulation, among others, in their work at the naval yard and that he had worked aboard the Boston, at some point during his career.

With respect to the bystander plaintiffs, we find that the evidence outlined above is insufficient to support a jury finding that Owens-Corning’s products were more likely than not a proximate cause of the plaintiffs’ injuries. The plaintiffs’ witnesses have not stated how much Kaylo insulation was used at the Mohawk plant, how often it was used, and where in the plant it was used. They have not given an estimate as to the amount of Kaylo used in relation to the other insulation products installed in the plant. They have, therefore, failed to show the frequency and regularity of the use of Kaylo insulation. Any evidence of fiber drift is consequently unhelpful.

We find the evidence against Owens-Corning to be close to the situation in Robertson v. Allied Signal, Inc. In Robertson, several product identification witnesses had testified that they had used asbestos pipe covering and cement products manufactured by Eagle-Picher in a tire plant. 914 F.2d at 370-71. None of the plaintiffs could state that they had used Eagle-Picher products or that they had worked near any location where Eagle-Picher products had been installed. Id. Additionally, the product identification witnesses could not, for the most part, testify as to the amount of Eagle-Picher products used or to the locations where such products had been used. Id. The Third Circuit found that this evidence, which merely placed Eagle-Picher products in the plant in unspecified quantities, was insufficient to support a jury finding of proximate cause. Id. at 378.

The plaintiffs assert that their evidence is more similar to the amount and type found in Slaughter v. Southern Talc Co., 949 F.2d 167 (5th Cir.1991). In Slaughter, product identification witnesses were able to state that Owens-Corning’s Kaylo insulation had been delivered to and used in a tire plant. Id. at 170. Although the witnesses could not pinpoint the locations in the plant where the Kaylo insulation had been installed, two witnesses did testify that Kaylo insulation had been installed “all over the plant.” Id. Moreover, all of the plaintiffs in Slaughter testified that they had worked near insulated pipes and that insulation dust had regularly covered them when the pipe insulation had been repaired. Id. The Fifth Circuit found that this circumstantial evidence satisfied the plaintiffs’ burden under the Lohrmann test. Id. at 171.

The plaintiffs in the present case have not . presented evidence equivalent to that presented in Slaughter. They have not produced deposition testimony that Owens-Corning’s Kaylo insulation was used “all over” the Mohawk plant or that they worked close enough to insulated pipes that they were regularly covered with insulation dust during repairs. Thus, their reliance on Slaughter is unavailing.

Plaintiff Billy Carr’s case against Owens-Corning presents a closer question. Although Carr testified that he had done a substantial amount of pipe insulation work, especially repair work, at the beginning of his career at Mohawk, he was unable to state which company’s insulation products he had used. Because no witness has identified which insulation products were in the plant during Carr’s years as a mechanic (and Owens-Corning did not begin selling Kaylo until 1958, two years after Carr began his work as a mechanic), we have no way of inferring whether Carr ever used Owens-Corning products or the amounts he used. Thus, we must find that Owens-Corning’s summary judgment motion was properly granted against him as well. To hold otherwise would amount to imposing alternative liability on Owens-Corning.

Plaintiff George Bokker has not presented any evidence that he ever used Owens-Corning products or that he worked near locations where Owens-Corning products had been installed. Thus, Owens-Corning’s motion for summary judgment against him was properly granted.

Finally, we address plaintiff David Treadaway’s alleged exposure to Owens-Corning products aboard the U.S.S. Boston. James Bruce testified that Kaylo insulation, among others, was regularly used by insulation installers in the Norfolk Naval Yard during the 1950’s and 1960’s and that he personally had done some insulation work aboard the U.S.S. Boston during his career at the Norfolk Naval Yard. Plaintiff Treadaway stated in his deposition that he had worked in the Boston’s boiler room during his service from 1957 to 1960. Neither Bruce nor Treadaway, however, stated that Kaylo insulation was installed anywhere on board the Boston during Treadaway’s service. Treadaway stated that he could not recall the names of any products that he personally had used or that others may have used in proximity to him. The evidence, at best, demonstrates only a possibility that Kaylo may have been used on board the Boston at some point during a nearly two-decade period of time. Given our conclusion that Arkansas has not adopted alternative or market share liability, the mere possibility of Kaylo’s use and Treadaway’s exposure does not satisfy Treadaway’s burden of showing probable exposure to Owens-Corning’s asbestos, much less its causation of his injuries.

B.Owens-Illinois, Inc.

In their response to Owens-Illinois’s ten separate motions for summary judgment, the plaintiffs adopted the arguments and evidence that they had proffered in response to Owens-Corning’s motion for summary judgment. Owens-Illinois manufactured and sold Kaylo insulation from 1943 until 1958. In the 1950’s, Owens-Illinois and Owens-Corning entered into agreements which provided that Kaylo insulation products would be marketed under Owens-Corning’s name. Ultimately, Owens-Illinois sold its production facilities for manufacturing Kaylo insulation products to Owens-Corning.

Given our conclusion in the preceding section that the plaintiffs’ evidence is insufficient to support a jury finding that their exposure to Kaylo insulation constituted a substantial factor in causing their injuries, we must conclude that the plaintiffs have likewise failed to meet their burden with respect to Owens-Illinois.

C. Fibreboard Corp.

The plaintiffs direct us to the deposition testimony of two witnesses in support of plaintiffs’ claims against Fibreboard. Rivers Billingsley stated that he may have seen pipe insulation boxes labeled PABCO (a Fibre-board product) during his years at the Mohawk plant, but he “couldn’t swear to that at all.” Robert Brackin testified that he had seen “PABCO” on boxes of pipe insulation at some point during his work at Mohawk from 1956 to 1979, but could not recall any unique characteristics of the PABCO insulation or the amounts of PABCO insulation present at Mohawk. Brackin also testified that he recalled using PABCO finishing cement, but could not state whether PABCO’s cement was “the one [he] liked the best.”

The plaintiffs’ evidence establishes only that Fibreboard’s products were probably present in the Mohawk plant in some unspecified quantity at some unspecified point in time during the plant’s operation. This evidence is insufficient to support a finding that the plaintiffs were more likely than not exposed to asbestos fibers from PABCO insulation and that such exposure produced their injuries.

D. Keene Corp.

The only evidence relating to Keene involves David Treadaway’s service aboard the U.S.S. Boston from 1957 until 1960. As with Owens-Corning and Owens-Illinois .pipe insulation, James Bruce testified that Ehret pipe insulation manufactured by Ehret & Hill Mfr. Co., the alleged predecessor to Keene, was regularly used in the Norfolk Naval Yard during the 1950’s-1960’s. As discussed earlier with respect to Owens-Corning, this evidence fails to present a submissible case against Keene on the causation issue.

E. Anchor Packing Co.

Anchor sold a variety of industrial sealing products, including packing and gasket material. Some of Anchor’s sealing products contained asbestos and some did not.

The plaintiffs’ evidence demonstrates a substantial probability that some of Anchor’s products were present in the Mohawk plant during its operation. Rivers Billingsley testified that he had Anchor products in his storeroom beginning in the 1960’s or early 1970’s. He stated that the Anchor products, including a teflon packing material, were brought in “to replace some of the older .style asbestos and lead packings.” Billingsley stated that he would not be able to testify under oath that any Anchor product present at Mohawk contained asbestos.

Pauline Moore testified that she remembered ordering gaskets and packing materials directly from Anchor. When questioned about the orders, she replied that she could not testify as to which specific Anchor products Mohawk had ordered. Moreover, when asked to give an estimate of the dollar amount ordered from Anchor, she responded, “Not that much.” At one point in her deposition, she testified that it seemed to her that the word “asbestos” had appeared on some purchase orders for Anchor products, but she later stated that she could not testify under oath that she had ordered any asbestos-containing materials from Anchor.

Robert Brackin testified that he was certain he had used Anchor products during his years at the Mohawk plant. He stated that he had probably used Anchor braided packing at Mohawk, but that he would not be able to go into detail about the types of Anchor products he had used, including whether those products had contained asbestos.

Similarly, plaintiff Billy Carr testified that he had regularly used gaskets and packing materials in his duties as a mechanic. He also stated that he remembered specifically using Anchor products, including Anchor’s Anklon packing and ring packing. Carr could not state, however, whether any of the Anchor products that he had used had contained asbestos.

With respect to the bystander plaintiffs, the evidence summarized above does not create a basis for a jury finding that the bystander plaintiffs were exposed to asbestos from Anchor products or that such exposure caused their injuries. The evidence does not show where and how often Anchor products were used in the plant and whether the locations of Anchor’s products were close enough to the bystander plaintiffs’ work stations to support a reasonable inference of exposure. Moreover, no witness was prepared to testify under oath even that any Anchor product present in the Mohawk facility actually contained asbestos. Even with respect to plaintiff Billy Carr, who appears to have directly handled Anchor products, there is no support in the record for finding the presence of asbestos in the Anchor products he used. In the absence of a showing that nearly all Anchor sealing products had contained asbestos or that Anklon packing in particular, mentioned by Carr, had contained asbestos, no inference arose that asbestos was present in the Anchor products used at the Mohawk facility. Without evidence to demonstrate the likely presence of asbestos, none of the plaintiffs can ever make the ultimate requisite showing that asbestos from Anchor products more likely than not caused his injuries.

F. A.W. Chesterton Co.

Like Anchor, Chesterton produced industrial gaskets and packing materials that were used to seal valves, pumps, and pipe flanges. Some of Chesterton’s sealing products contained asbestos and some did not.

The plaintiffs’ evidence against Chesterton is almost identical to that offered against Anchor. Rivers Billingsley testified that he had stocked Chesterton-gaskets and packing in the storeroom at Mohawk. He specifically recalled stocking “many a box” of a style number 64 Chesterton packing during the entire time that he had worked at Mohawk, but he was unable to state whether that product had contained asbestos. When asked generally about the presence of asbestos in sealing products, Billingsley stated that he had only a general belief from his experience that some of the gasket and packing material at the Mohawk plant contained asbestos. He said that he could not remember seeing the word “asbestos” on any product labels, but then qualified his answer by stating: “Some of the packing, like the Chesterton that we would get, it would have [“asbestos”] wrote (sic) on the box....”

Pauline Moore testified that she believed she had seen the word “asbestos” on order forms for Chesterton products. She later stated, however, that she could not testify under oath that she had ordered Chesterton products that had contained asbestos.

Three other workers in the Mohawk plant gave deposition testimony that they had used Chesterton gaskets and packing at the Mohawk plant. They could not, however, state whether any of the Chesterton products they had used had contained asbestos. Additionally, they could not say where in the Mohawk plant they had used Chesterton products or in what quantities.

As noted earlier, Billy Carr testified that he had often used gaskets and packing materials during his six years as a maintenance mechanic. He also specifically recalled that he had used Chesterton products. He could not assert, however, that any of the gasket and packing materials, including the Chesterton products, had contained asbestos fibers that he might have inhaled.

The final testimony concerning Chesterton products came from Tommy Polk, the purchasing agent for the Mohawk plant from 1956 until 1974. Polk stated that he had purchased at most $50 worth of materials from Chesterton in any one year.

With respect to George Bokker and the bystander plaintiffs, we must agree with the district court’s conclusion that this evidence is not sufficient to withstand a motion for summary judgment. Although Rivers Billingsley’s remark that “asbestos” appeared on some Chesterton packages provides support that some Chesterton asbestos-containing products were used at the Mohawk plant, there is no evidence as to where those products were used or how often they were used. Thus, a jury could only speculate that asbestos emitted from Chesterton’s gaskets and packing more likely than not had been a substantial contributing cause of the injuries to either the bystander plaintiffs or to George Bokker, who never mentioned Chesterton products as ones that he had directly handled.

We also agree with the district court’s assessment that Billy Carr’s claim against Chesterton presents a more difficult determination. Rivers Billingsley’s remark provides some support for a finding that Chesterton’s products used at Mohawk contained asbestos. Moreover, Carr testified that he had worked with gaskets and packing “many times” during his years as a mechanic. Thus, it appears at first glance that Carr has satisfied the “frequency, regularity and proximity” test. Carr’s case has a missing link, however. Although Carr said that he had used gaskets and packing many times, the record contains no evidence that he used Chesterton gaskets and packing many times. We have no way of knowing whether Carr selected Chesterton products for two jobs or two hundred jobs. This gap in Carr’s proof is spotlighted by Tommy Polk’s testimony that Mohawk never purchased more than $50 in Chesterton products in a year. Moreover, some portion of those few purchases presumably may have been for Chesterton products that did not contain asbestos because Chesterton manufactured materials with and without asbestos. Given Arkansas’s proximate cause requirement, we cannot say that Carr’s evidence raises a legitimate issue for the jury concerning causation.

G. Garlock, Inc.

The plaintiffs point to four deposition witnesses to support their claim against Gar-lock. Rivers Billingsley testified that he carried Garlock’s gaskets and packing in the Mohawk storeroom. He also stated that he had a general understanding that the gasket and packing material present in the storeroom, including the Garlock products, had contained asbestos.

Robert Brackin also stated that Garlock’s gaskets and packing had been used in the Mohawk plant. In general, he stated that he had assumed that “since [a gasket] was made for heat that it had [asbestos] in it, now that don’t [sic] mean it did.”

Plaintiffs Billy Carr and George Bokker testified that they had directly handled Gar-lock products. Neither man could say, however, that the Garlock products they had used had contained asbestos.

Once more, we must find that this evidence does not rise to a level sufficient to support a finding of causation against Gar-lock. There is no evidence that any Garlock product was used in a location from which asbestos fibers could have been emitted into the work areas of the bystander plaintiffs. Moreover, although Rivers Billingsley stated that he had a general belief that some Gar-lock products contained asbestos, there is no evidence that the Garlock products used by Carr and Bokker contained asbestos. Additionally, Carr and Bokker have not stated how many times they specifically used Gar-lock products. Thus, the district court’s grant of summary judgment in favor of Gar-lock must be upheld.

H.John Crane, Inc.

As with the other defendants, the plaintiffs have presented evidence that Crane products were present in the Mohawk plant. Additionally, they have introduced Crane’s interrogatory answers, in which Crane admits that it has manufactured sealing products since the 1930’s, some of which contained asbestos and some of which did not. The plaintiffs have failed, however, to demonstrate that Crane asbestos products were regularly used in locations in the Mohawk plant from which asbestos fibers could likely have reached the bystander plaintiffs. Also, they have failed to show that Plaintiffs Bokker and Carr specifically used Crane’s asbestos-containing products with any level of regularity. Finally, the testimony of Pauline Moore demonstrates that Mohawk purchased only a small amount of Crane products. She stated twice that “We did not buy that much from [Crane].”

I. Rexnord Corp.

The plaintiffs’ evidence demonstrates that Stearns brakes probably were used in the Mohawk plant. The district court held, nonetheless, that the plaintiffs had failed to show exposure significant enough to support a finding of causation. Having carefully reviewed the plaintiffs’ evidence, we agree with the district court’s conclusion. Moreover, we find that on the record before us the plaintiffs have not introduced sufficient evidence to support a finding that the Stearns brake disks actually contained asbestos. Accordingly, we do not address Rexnord’s alternative argument that it is not the successor to the Stearns Division that manufactured brake parts during the Mohawk plant’s existence.

J. Georgia Talc Co.

The plaintiffs’ evidence demonstrates that tale or soapstone was used in several areas of the Mohawk plant as a detactifying agent to keep the rubber from sticking to itself and to the machinery. Several of the plaintiffs testified that as part of their job duties they had personally “dusted” tires with soapstone as the tires had moved along the production line. They stated that this soapstone had covered their shirt sleeves and arms when they had applied it to the tires and had created significant amounts of dust if applied carelessly.

With respect to Georgia Talc specifically, the plaintiffs direct us to test results on samples taken from the mines of Georgia Talc and Southern Talc, a sister corporation, that show the presence of asbestiform minerals. Additionally, the plaintiffs have introduced evidence that Georgia Talc distributed its products through a chemical supply company in Ohio that in turn -sold substantial amounts of talc to Mohawk’s plant in Akron, Ohio. Affidavits signed by Nyle Oswalt and ■ Tommy Polk, two former' Mohawk employees, state that Mohawk’s West Helena plant regularly requisitioned raw materials, including talc, from the Akron facility. Finally, Oswalt flatly stated that in his duties at Mohawk he had become familiar with the suppliers of talc to the West Helena plant and that Georgia Talc had been one of those suppliers.

The plaintiffs’ evidence demonstrates a probability that Georgia Talc’s asbestos-containing products was present in the West Helena plant in some unspecified quantity. Although no witness specifically recalled that the West Helena plant had requisitioned Georgia Talc’s product from Akron, the frequent practice of intra-company transfer of materials and Oswalt’s statement support a reasonable inference that some talc produced by Georgia Talc probably made its way into Mohawk’s West Helena plant.

Thus, the plaintiffs have placed Georgia Talc’s product in the plant and have shown that talc or soapstone produced by unknown manufacturers, possibly including Georgia Talc, was regularly and frequently used in the vicinity of several of the plaintiffs. Nevertheless, there is a missing link in the plaintiffs’ evidence against Georgia Talc. The plaintiffs have been unable to show that talc produced by Georgia Talc was used near their work stations. This is a necessary part of proving proximate causation under the “frequency, regularity, and proximity” test. See, e.g., Robertson, 914 F.2d at 377 (plaintiffs who worked on level where defendant’s talc bags were identified survived summary judgment while plaintiffs on other levels who had worked with or near talc products of unspecified origin were dismissed); cf. Roehling v. National Gypsum Co. Gold Bond Bldg. Products, 786 F.2d 1225, 1227-28 (4th Cir.1986) (evidence regarding use of specific defendant’s products at location where plaintiff worked sufficient to reverse grant of summary judgment); Odum v. Celotex Corp., 764 F.2d 1486, 1488 (11th Cir.1985) (same).

Even in Slaughter, which represents one of the most lenient applications of the “frequency, regularity, and proximity” test, the plaintiffs evidence specifically named Kaylo insulation as having been installed “all over the plant.” 949 F.2d at 171. Only after the plaintiffs had demonstrated that Owens-Corning’s product in particular likely had been installed throughout the plant was the court willing to accept the inference that the plaintiffs had been exposed to Owens-Corning’s asbestos products. Id. at 172. The Fifth Circuit did not state that the plaintiffs had shown causation merely because they had worked near insulated pipes and Owens-Corning had shipped some unknown quantity of pipe insulation to the plaintiffs’ tire factory.

To allow the plaintiffs to obtain a jury verdict against Georgia Talc by showing only that Georgia Talc’s product was present at the Mohawk plant in unknown quantities and that the plaintiffs worked near some manufacturer’s talc would be tantamount to adopting a theory of alternative or group liability. As discussed earlier, Arkansas law does not recognize such theories. Accordingly, we must affirm the district court’s grant of summary judgment to Georgia Talc.

We recognize that the plaintiffs from the start have faced a formidable task in establishing whose asbestos products caused their injuries, especially where the plant in which the asbestos exposure is alleged to have occurred closed in 1979 and where the employer’s purchasing and production records have been subsequently destroyed. As a federal court in a diversity suit, however, we are not free to fashion rules of law from whole cloth. We are bound to apply the law of the state as we are able to discern it from the rulings of the state’s courts. In this ease, we have found no indication that the Arkansas courts would abandon their proximate cause requirement.

Accordingly, the judgment of the district court is affirmed with respect to all defendants.

1 . The employees' wives asserted claims for loss of consortium. For simplicity, however, we will refer to the employees as "the plaintiffs.”

2 . The Honorable G. Thomas Eisele, Senior United States District Judge for the Eastern District of Arkansas.

3 . Bokker died after this action was filed, but for simplicity we will refer to him as one of the plaintiffs.

4 . The thirteen other defendants named in the complaint have been dismissed from the case for various reasons, including the fact that most of them are in bankruptcy.

5 . The plaintiffs' medical conditions and their connection to asbestos exposure have not been either established or challenged because the defendants focused their motions for summary judgment solely on the issue of proximate causation. Our resolution of this case does not require us to address this issue.

6 . As used by the plaintiffs, the "fiber drift” theory entails more than the scientific recognition that asbestos fibers may break off from their products and become disseminated through the surrounding area by air currents. The plaintiffs use the term "fiber drift” as a label for a legal theory of causation which states that, given expert evidence of the movement of asbestos fibers through the air, a plaintiff in an asbestos case can survive a motion for summary judgment whenever he can show that the defendant's asbestos-containing products were used anywhere in the plaintiff's place of employment. See Robertson v. Allied Signal, Inc., 914 F.2d 360, 375-76 (3rd Cir.1990) (discussing more fully what is meant by the phrase "fiber drift”).