GEORGE C. PRATT, Circuit Judge:[fn*] [fn*] Of the U.S. Court of Appeals for the Second Circuit, sitting by designation.

In this multidistrict litigation Vietnam veterans and their relatives seek to recover damages from nine chemical companies, Dow Chemical Co. (Dow), Hercules, Inc. (Hercules), The Monsanto Co. (Monsanto), Diamond Shamrock Corporation (Diamond Shamrock), Hoffman-Taff (Missouri) (Hoffman-Taff), Thompson Chemical Corporation (Thompson), T.H. Agriculture Nutrition Co. (T.H.), Riverdale Chemical Co. (Riverdale), and Uniroyal, Inc. (Uniroyal), for injuries allegedly suffered as a result of their exposure to a herbicide called Agent Orange used by the military in Vietnam. Plaintiffs' claim is that 2,3,7,8 Tetrachloro-dibenzo-p-dioxin (dioxin) is extremely toxic, that it was produced as a by-product in the manufacture of trichlorophenol (TCP) which was a precursor chemical for 2,4,5-trichlorophenoxy acetic acid (2,4,5-T), which in turn was combined with 2,4-dichlorophenoxyacetic acid (2,4-D) to make Agent Orange. Any dioxin produced in the manufacture of TCP carried forward into 2,4,5-T and therefore into the Agent Orange. Thus the claims of the plaintiffs focus upon dioxin as a contaminant in the Agent Orange supplied to the government pursuant to contracts with the separate chemical companies.

In pretrial order no. 26, I recognized the possibility of a government contract defense to the plaintiffs' claims but denied defendants' motions to dismiss on that ground, finding that the motions presented issues of fact which precluded summary judgment. 506 F. Supp. 762, 796 (E.D.N.Y. 1980). The contours of the defense were developed in more detail in pretrial order no. 33:

[A] defendant in this case will be entitled to judgment dismissing all claims against it based on that defendant's having supplied "Agent Orange" to the government pursuant to a contract, if the defendant proves:

1. That the government established the specifications for "Agent Orange";

2. That the "Agent Orange" manufactured by the defendant met the government's specifications in all material respects; and

3. That the government knew as much as or more than the defendant about the hazards to people that accompanied use of "Agent Orange".

Because the issues presented by the government contract defense seemed to be separate and distinct from the general theories of liability then being advanced by plaintiffs, I ordered a separate trial of the defense, 506 F. Supp. at 796, and appointed a special master to supervise discovery for that trial scheduled to begin on June 27, 1983, 94 F.R.D. 173 (E.D.N.Y. 1982).

In April 1983, after almost eleven months of intensive discovery, I permitted any defendant who so elected to move for summary judgment with respect to the government contract defense. The basis of such summary judgment would, of course, be that there were no triable issues of fact with respect to the defense, and that the moving defendant was therefore entitled to dismissal of all claims against it as a matter of law. All defendants except Monsanto and Diamond Shamrock moved for summary judgment. In reaching a determination on these motions, I have reviewed all of the papers submitted, both in support and in opposition, including counsel's extensive memoranda and attached exhibits. In addition, I listened closely to the oral arguments presented by all parties.

The central issue raised by the government contract defense centers on its third element: whether the government knew as much as or more than the contracting defendant about the hazards to people that accompanied the use of Agent Orange. As this action has matured, the plaintiffs have concentrated their claims on dioxin as a contaminant present in 2,4,5-T; consequently, the knowledge in question is knowledge about dioxin. To focus upon this element, it is necessary to compare what knowledge the government had about dioxin in 2,4,5-T and about its contamination of Agent Orange, with what knowledge each of the moving defendants had about these matters.

Since discovery is not yet complete, the following discussion of facts and evidence does not constitute a finding of facts for any future purpose except for those specifically mentioned below in connection with the specifications for and performance of the government contracts.

RELATIVE LEVELS OF KNOWLEDGE

The Government

Even when all doubts are resolved in favor of the plaintiffs, as required by SEC v. Research Automation Corp., 585 F.2d 31, 33 (2d Cir. 1978), the record demonstrates that the government and the military had a considerable amount of knowledge about 2,4,5-T, about dioxin, and about the health hazards associated with both. The following general chronology, while not all-inclusive, gives some indication of both the extent of government knowledge in this area and of the fact that it was continually increasing.

During World War II, the military discovered the herbicidal properties of 2,4,5-T and conducted extensive testing of various possible herbicides. This research was conducted under the supervision of the Crops Division of the Army Chemical Corps. at Camp Detrick, Maryland.

Several years later, in 1949, Dr. Donald Birmingham of the Public Health Service visited Nitro, West Virginia, where there had been an explosion at Monsanto's 2,4,5-T plant. The report of Dr. Birmingham's colleague, Dr. Louis Schwartz, indicated a connection between chloracne and the chemicals produced in the plant.

There is uncontradicted evidence in the record that a number of people knew in the 1950s that dioxin was toxic although they may not have connected it with 2,4,5-T. Several factors contributed to this awareness.

First, in the early 1950s, C.H. Boehringer Sohn Company of Germany had serious cases of chloracne among workers engaged in the production of TCP, a precursor chemical used inter alia to manufacture 2,4,5-T. By 1955, the Boehringer company was forced to halt production at two plants. Dr. K.H. Schulz, a skin specialist, investigated the problem and in 1957 together with Professor J. Kimmig, reported his findings in an article entitled Chlorinated Aromatic Cyclic Ethers As the Cause of Chloracne. 44 Die Naturwissenshaften 337 (1957). In this article, the authors stated that they were able to isolate dioxin, which they believed to be the contaminant in TCP that was causing the health problems. While it is not established that anyone in the government read the Kimmig Schulz article at the time it was published, the article was available as part of the scientific literature and it appeared in a note to the report written by Friedrich Hoffmann concerning his trip to Europe in 1959.

The "Hoffmann Trip Report" was a second factor contributing to government knowledge during this period. Dr. Hoffman, who was searching on behalf of the military for potential chemical warfare agents, reported that he had received "startling information" regarding the toxicity of the compound dioxin. In his report, he described the deaths of several workers in a plant that produced wood preservatives containing trace amounts of dioxin. In addition, he reported that the compound could cause severe, indeed fatal, liver damage. At least 10 copies of the Hoffmann report were sent to the Army Chemical Corps Chemical Warfare Laboratories at Edgewood Arsenal, the governmental body responsible for investigating toxicity and analyzing chemical agents. Thus, the Hoffmann report on dioxin, coupled with the Kimmig Schulz article connecting dioxin to TCP, raises a strong possibility that personnel at Edgewood, even before 1960, were aware of the connection between dioxin and TCP as well as the use of TCP to make 2,4,5-T.

Deposition testimony of Edgewood research personnel confirms that people at Edgewood knew about the toxicity of dioxin. Dr. Bernard Jandorf, chief of the Army Chemical Research Laboratory, testified that people at Edgewood had been familiar with this fact since the late 1950s. Dr. Richard Horton, a toxicologist, testified that he knew dioxin was toxic in 1959, as did Dr. Thomas Simmons, who worked in the Agents Research Branch. Walter Sultan, a pharmacologist in the Toxicity Screening Branch, testified that he had read the Hoffmann report.

Further evidence of governmental knowledge is found in the article written by Dr. Birmingham of the Public Health Service in 1959, stating that in the manufacture of 2,4,5-T, intermediate hydrocarbons of the chlorine group had caused chloracne in more than 200 chemical workers at a manufacturing plant. Birmingham, New Causes of Occupational Dermatoses, 20 Industrial Health 489, 490 (1959). Dr. Marcus Key of the Public Health Service testified that he had learned of the association between hydrocarbons and chloracne and other diseases at the Harvard School of Public Health in 1953.

In the early 1960s, Dr. Bernard McNamara, Chief of the Toxicology Division at Edgewood, performed a study at Edgewood Arsenal of the toxicity of Agent Purple, another defoliant containing 2,4,5-T that was used by the military. This testing was conducted at the request of General F.J. Delmore, Commanding General, U.S. Army Chemical Corps, Research Development Committee. While the testing indicated that there was some toxicity, the results were not conclusive. At a meeting held at Edgewood Arsenal in 1963 to discuss and evaluate the toxicity of 2,4,5-T, the overall thrust of those reporting was that both 2,4,5-T and 2,4-D were safe for humans.

Other events occurring in 1963 give additional indications of governmental knowledge. The Institute for Defense Analysis wrote a report for the Advanced Research Project Agency, an agency within the Department of Defense. This report stated that herbicides were safe when used commercially, but that there could be increased hazards in military use because greater concentrations might be applied by less experienced personnel under the pressures inherent in battlefield use. The report noted the connection between chloracne and skin and respiratory irritations and their association with herbicides.

Dr. Key of the Public Health Service testified at his deposition that in 1963 he placed a sample of 2,4,5-T herbicide on his forearm to see if it would induce chloracne. He did this three times a week for three weeks and developed chloracne on his forearm. He also testified that he had read Kimmig Schulz and learned of dioxin from that article. When questioned concerning a June 1964 article by Dr. Jacob Bleiberg, Industrially Acquired Porphyria, 89 Archives of Dermatology 793 (1964), which discussed chloracne and porphyria in workers engaged in 2,4,5-T production, Key stated that he had reviewed the article at the time it was written and that it was only a more complete version of what they already knew.

The level of government knowledge appears to have increased much more rapidly during the mid-to-late 1960s. Defendants point to numerous instances of governmental knowledge which are not disputed by plaintiffs. Dr. Herbert Stokinger, the chief toxicologist of the Division of Occupational Health, testified that he knew dioxin was an impurity in 2,4,5-T sometime around 1965. Colonel Robert A. Shade, who was chief of the Chemical Operation Branch of Military Assistance Command-Vietnam and later on the staff of the chemical branch of the Assistant Chief of Staff Force Development, testified that he learned of the connection between dioxin and 2,4,5-T sometime between mid-1966 and summer 1968.

In July 1966, the director of the National Academy of Sciences wrote to the chief of the Bureau of Medicine and Surgery for the Navy advising him of the connection between 2,4,5-T and porphyria and chloracne. In August 1966, the National Academy of Sciences, in response to a request for information, wrote to the Army Surgeon General telling him that 2,4,5-T was toxic and that chloracne was associated with it.

Recent deposition testimony indicates that people closely associated with the White House were aware of hazards involved in the use of defoliants. Dr. Gordon MacDonald, a member of President Johnson's Science Advisory Committee (PSAC), testified that the issues of herbicides and dioxin in herbicides were informally discussed by a subgroup of PSAC sometime between April and June 1965. Dioxin as an impurity in 2,4,5-T was also discussed. He said there was discussion of the potential toxicity of dioxin, and while it was considered that the evidence was fragmentary and inconclusive, the subject of dioxin contamination deserved continuing attention. Dr. MacDonald testified that human health effects were discussed, and that he attended a meeting where the effectiveness of herbicides and the presence of dioxin in 2,4,5-T were discussed. According to MacDonald, Secretary of Defense Robert MacNamara attended this meeting.

Dr. Donald Hornig, Special Assistant to President Johnson for Science and Technology and Chairman of PSAC, testified at his deposition that by 1966 PSAC was discussing impurities in 2,4,5-T. He stated that this discussion occurred sometime between 1964 and 1966. He said that when he learned of the impurity, he felt that "one ought to be concerned" about what the magnitudes of the toxicological effects and of the exposures might be. He testified that he understood dioxin was a health hazard to human beings. However, he also testified that he did not relay the information to President Johnson.

An additional element of knowledge is found in a 1967 Rand report commissioned by the Advanced Research Project Agency of the Department of Defense. This report described "actual experience" of health hazards associated with the use of defoliants in Vietnam.

Finally, there is the study commissioned by the National Cancer Institute, Evaluation of Carcinogenic, Teratogenic, and Mutagenic Activities of Selected Pesticides and Industrial Chemicals (Bionetics Report). This study evaluated the carcinogenic, teratogenic, and mutagenic effects of various chemicals. The study was commissioned in 1963, and the report is dated August 1968. The study did result in a finding of some teratogenic effects connected with the use of 2,4,5-T. While it is not clear that defendants are correct in their assertion that portions of the study were available to the government earlier than August 1968, it is clear that by 1968 and 1969, the results of the study were available to the government.

This picture of knowledge shown to be in government hands is based almost entirely on uncontradicted and uncontested evidence. It reveals that the government and the military possessed rather extensive knowledge tending to show that its use of Agent Orange in Vietnam created significant, though undetermined, risks of harm to our military personnel. Against this picture we must examine what was known by the different defendants, keeping in mind that, by and large, much of the government's knowledge was classified and not shared with the defendants.

Dow Chemical Company

Dow supplied Agent Orange to the military pursuant to seven contracts. Deliveries were made from September 1965 to December 1968. The government established the specifications for the delivered product, and Dow performed to those specifications.

Dow began manufacturing 2,4,5-T in 1948. Dow admits that it knew prior to this that chloracne was an industrial health hazard present in the production of certain chlorinated hydrocarbons. It developed the "rabbit ear" test, which was non-specific, but was able to determine if a chloracnegen was present. Dow used this test until 1964.

Plaintiffs argue persuasively that Dow must have known about the 1949 explosion at the Monsanto plant in Nitro, West Virginia, in 1949, and the resulting cases of chloracne. Dow does not deny such knowledge.

In the 1950s, C.H. Boehringer Sohn Company had the chloracne problems in its plant discussed supra, and it wrote to Dow for help. In 1955, Dow replied by sending a data sheet describing the hazards due to toxicity and the precautions Dow was taking to prevent them. In 1957, C.H. Boehringer reciprocated by informing Dow of the methods it had developed to prepare TCP so as to avoid "chloracne exciters".

The Kimmig Schulz article was produced by Dow from its files during discovery. There is no indication in the record of when Dow obtained the article, other than the fact that it was referenced in a memo written by a Dow official in 1964.

Plaintiffs also argue that Dow knew about the outbreak of chloracne among workers at the Diamond Alkali (now Diamond Shamrock) plant in 1956. Dow does not deny this.

In February 1964 at Dow's plant in Midland, Michigan, more than 40 workers developed chloracne. These workers had been engaged in the manufacture of TCP. After shutting down the plant, Dow investigated the incident and found that there was a high concentration of chloracnegen in the waste stream from the plant. It determined that this chloracnegen was dioxin. Dow reported the incident to the Michigan Department of Health.

Around this time, Dow developed a method of using gas chromotography to detect dioxin in TCP and in 2,4,5-T at concentration levels as low as 1 part per million (ppm). Dow researchers were satisfied that there was no chloracnegenic response if the dioxin level was at or below 1 ppm. Dow thereupon instituted procedures to ensure that no TCP or 2,4,5-T left its plant with a dioxin level above 1 ppm. This is supported to a certain extent by Dow's exhibits 42 through 45 which indicate that Dow's dioxin levels in its 2,4,5-T were less than 1 ppm, although these exhibits do not appear to cover all of the 2,4,5-T and Agent Orange that Dow produced. However, nowhere in their papers do plaintiffs challenge Dow's contention that its 2,4,5-T was contaminated by 1 ppm of dioxin or less. Since Agent Orange was a 50-50 mixture of 2,4,5-T with 2,4-D, and since plaintiffs have abandoned their earlier claims against 2,4-D, this in effect establishes for purposes of this motion a contamination level for Dow's Agent Orange at .5 ppm or less.

In March 1965, Dow called a meeting attended by Hercules, Diamond Alkali, and Hooker Chemical Corporation to discuss health hazards involved in the production of TCP and 2,4,5-T. No one from the government was invited to the meeting, but Dow had not yet contracted to produce Agent Orange. It is not disputed that Dow knew that the dioxin problem arose during the manufacturing process and that any dioxin produced at that stage could carry forward into the delivered product. At the meeting, Dow explained that precautions were necessary to prevent health hazards, and stated that it had examined herbicides sold by some other companies and found some to contain "surprising high levels" of dioxin. In a memo to the file after the March 1965 meeting, Dr. Edward Chandler of Diamond Alkali indicated that Dow thought that repeated exposure to 1 ppm could be dangerous.

In June 1965, V.K. Rowe of the Dow Biochemical Research Laboratory wrote to Ross Mulholland of Dow Chemical of Canada. He described the chloracne problems Dow had experienced and stated that Dow did not want any of its customers to develop acne. The letter also indicates a fear of government intervention into and control of the entire herbicide industry, and further that Dow wanted to get the problem under control without governmental regulation. Mulholland was cautioned not to transmit this information to anyone else.

Dow claims, however, that on four occasions in 1967 it did transmit to the government information concerning health hazards related to the manufacture of Agent Orange and 2,4,5-T.

1. On February 29, 1967 A.P. Beutel, vice president of Dow, wrote a letter to Brigadier General J.A. Hebbeler concerning the government's plan for producing Agent Orange in a government plant. Beutel mentioned "certain health problems" inherent in the manufacturing process.

2. On April 20, 1967 Beutel wrote to H.G. Fredricks, Deputy Director of Procurement and Production, concerning the same subject — government production of Agent Orange. In the letter, Beutel mentioned a "serious potential health hazard" to workers, and stated that even with detection methods care is necessary in the handling of the product.

3. In August 1967 Beutel and two other Dow representatives told two officials from the office of the Secretary of Defense that caution should be exercised in producing 2,4,5-T.

4. On September 26, 1967 Beutel wrote to Andrew Anderson of Edgewood indicating that Dow would not bid on the government project to manufacture Agent Orange because of, among other reasons, the chloracne problem.

By 1970 there appears to have been widespread concern in various parts of the government concerning the hazards of the defoliation program. In March 1970, Dow briefed representatives of the military on the presence of dioxin as an impurity in TCP and 2,4,5-T.

In June 1970, after temporary suspension of Agent Orange use, Dow wrote to Secretary of Defense Melvin Laird recommending "strongly" that the government set appropriate specifications and controls to ensure that no 2,4,5-T be used if it contained more than 1 ppm dioxin. Dow specifically urged that standard for any 2,4,5-T used as a component of Agent Orange if it was to be used as a defoliant in Vietnam.

On a motion for summary judgment, the moving party must show that "no genuine issue as to any material fact" exists. Fed.R.Civ.P. 56(c); see, e.g., Landmark Land Co. v. Sprague, 701 F.2d 1065 (2d Cir. 1983). Here, several questions of fact preclude summary judgment in favor of Dow.

If there is a real difference in the level of knowledge between Dow and the government it focuses upon Dow's discovery in 1964 that dioxin was the chloracnegen in TCP, its development of a test to determine dioxin levels, and its development of techniques (partially through purchase from C.H. Boehringer Sohn Company) to reduce dioxin levels during the manufacturing process. One question of fact is whether this knowledge, if disclosed to the government, might have made a difference in the government's decisionmaking process.

Related questions of fact are the actual dioxin levels in Dow's product and the actual hazards involved in the use of the products at different levels of dioxin. Much of this may depend upon whether or not Dow's self-imposed maximum contamination of 1 ppm is or was safe, and this, too, presents a question of fact. Arguably, if Dow was selling a clean "safe" product to the government, then it told the government everything it needed to know. Of course, if the product was clean and safe, Dow would win on the causation issue as well, because a clean and safe herbicide, by definition, would not cause the injuries plaintiffs claim to have suffered from Agent Orange.

T.H. Agriculture Nutrition Co.

T.H. supplied the government with Agent Orange pursuant to contracts dated June 28, 1967, March 1, 1968, and May 20, 1968. As with Dow, the government established the specifications for the delivered product, and T.H. performed to those specifications. T.H. had manufactured 2,4,5-T for commercial sale as a herbicide prior to 1967. It did not make TCP but rather purchased it from other chemical companies.

According to its brief, T.H.'s first knowledge of any problems associated with 2,4,5-T occurred in 1964. In December of that year, Dr. David Groth of the Public Health Service wrote to T.H. requesting samples of 2,4,5-T. The letter stated that 2,4,5-T was associated with chloracne and that it was suspected that dioxin was the culprit. Groth stated that he was attempting to develop a method to isolate the contaminant.

In December 1964, Lindley S. DeAtley, a Vice President of Research and Development for T.H., wrote to Dr. R.C. Dosser, Laboratory Director at Dow, telling him of Groth's letter. Dosser replied in a telephone call that there might be some methods of production which led to a toxic compound. By January 1965, DeAtley had concluded that T.H. should conduct some tests of its own. At this time T.H. was not yet a government supplier of Agent Orange.

On February 19, 1965, DeAtley and D.W. Fuhlhage, T.H.'s Supervisor for Process Development, visited Dow's Midland plant. The minutes of the meeting indicate that DeAtley and Fuhlhage learned that cases of chloracne had recently been more severe, and that Dow had its workers changing clothing and showering at mid-shift.

T.H. admits that it had cases of chloracne among its workers, but claims that it did not know the cause until December 1964 at the earliest. T.H.'s argument is that by June 1967 when it first became a government contractor, the government certainly knew as much as it did. In June 1967, when T.H. negotiated its first contract, it did tell the government that there was a chloracne problem in the manufacturing process and that this was factored into the price.

As in the case of Dow, questions of fact remain with respect to T.H.'s assertion of the government contract defense. The level of dioxin in T.H.'s Agent Orange is not clear. According to plaintiffs, T.H. has not produced its records on dioxin levels, except for the years 1970 and 1971. Samples tested at Gulfport, Mississippi, indicate a level that went as high as 4.1 ppm. This raises a question of fact as to the potential for harm of T.H.'s product.

Plaintiffs emphasize the fact that T.H. knew that gas chromotography could determine dioxin level, but never told the military about it even after it began to supply Agent Orange to the government. Nor did it tell the government that its chloracne manufacturing problem was probably caused by dioxin, or that the contaminant very likely carried over into the delivered Agent Orange. Whether this information might have made a difference in the government's decisionmaking process is an issue of fact.

Uniroyal, Inc.

Uniroyal sold Agent Orange to the government pursuant to three contracts between October 6, 1966 and March 1, 1968. Uniroyal's delivered product, like that of the other moving defendants, complied with specifications established by the government. It appears to be undisputed that the product sold was not manufactured by Uniroyal, but rather, was supplied to Uniroyal by a Canadian subsidiary, now named Uniroyal, Ltd. (Ltd.). Uniroyal denies having had any knowledge of the toxicity of 2,4,5-T or that there were any health hazards connected with Agent Orange. Uniroyal has not presented evidence of the actual level of dioxin contamination of its product.

The only evidence of Uniroyal's knowledge of chloracne problems is found in a memo from Ltd. to Uniroyal dated June 11, 1962, which states that five employees at the Clover Bar plant in Canada had symptoms of chloracne. A reply affidavit of Arthur Gorman, technical liaison to the Ltd. plant at the time, states that the Clover Bar facility did not produce either TCP or 2,4,5-T in 1962. The inference is that this memo has nothing to do with this case.

There are at least two issues of fact with respect to Uniroyal which preclude summary judgment in its favor. The first concerns the memo of T.H. Evans of Ltd. dated August 10, 1965. This memo was found in the files of Walter Harris of Uniroyal who stated that he did not receive it until 1968. Evans stated in his affidavit that he did not send it to Uniroyal, and Arthur Gorman of Ltd. stated that copies of the Evans memo were not sent to Uniroyal until 1968. The actual date that Uniroyal received it, however, presents a triable issue of fact.

The memo is important because it contains Evans' description of his visit to Dow's Midland plant in 1965 and the discussions he had with Dow scientists concerning chloracne problems and dioxin. If a jury were to find that Uniroyal received the memo in 1965, then Uniroyal knew much more about the possible hazards of Agent Orange use than it now claims it knew when it entered into its first contract with the government in 1966.

The second issue of fact is whether Uniroyal is chargeable with the knowledge of its subsidiary, Ltd., because of their corporate relationship. While I do not fault Uniroyal for its irritation at plaintiffs' eleventh hour argument here, I do believe that plaintiffs and Uniroyal's cross-claiming co-defendants should be permitted to introduce evidence, if they can, that the two corporations were really the same entity, assuming without deciding the validity of Uniroyal's argument that there is a presumption of separateness, see Luckett v. Bethlehem Steel Corp., 618 F.2d 1373, 1379 (10th Cir. 1980); Bendix Home Systems, Inc. v. Hurston Enterprises, 566 F.2d 1039, 1042 (5th Cir. 1978); Williams v. McAllister Brothers, Inc., 534 F.2d 19, 22 (2d Cir. 1976). If the parties intend to pursue that line, I expect that they will do so in good faith and not just waste the court's time or attempt to confuse the issues. Counsel are reminded that 28 U.S.C. § 1927 (Supp. V) provides that an attorney "who so multiplies the proceedings in any case unreasonably and vexatiously may be required by the court to satisfy personally the excess costs, expenses, and attorneys' fees reasonably incurred because of such conduct." See Kline v. Wolf, 702 F.2d 400, 405 (2d Cir. 1983).

Riverdale Chemical Company

Riverdale states that it manufactured Agent Orange pursuant to a contract, that the government set the specifications, that it performed to specifications, and that it knew of no health hazard connected with Agent Orange.

Riverdale's motion for summary judgment is unopposed, except by its co-defendants who argue merely that it is too early to grant summary judgment for Riverdale. However, as Riverdale points out in its reply memorandum, opposition to a summary judgment motion "must set forth specific facts showing that there is a genuine issue for trial." Fed.R.Civ.P. 56(e).

The party opposing summary judgment must offer "some competent evidence that could be presented at trial showing that there is a genuine issue as to a material fact." Wright Miller, Federal Practice and Procedure: Civil § 2727. No such evidence has been presented, and there are no other circumstances as to why summary judgment cannot be fairly granted now to Riverdale.

Hoffman-Taff (Missouri)

Hoffman-Taff is in the same situation as Riverdale, and for the same reasons is entitled to summary judgment in its favor.

Thompson Chemical Corporation

Originally, Thompson declined to bid for Agent Orange contracts. Acting under appropriate statutory authority, The Defense Production Act of 1950, Sept. 8, 1950, ch. 932, 64 Stat. 798, amended, June 30, 1966, 80 Stat. 235 (current version at 50 U.S.C. App. §§ 2061 et seq.) however, the government required Thompson to supply Agent Orange pursuant to two contracts dated April 19, 1967 and May 24, 1968. It supplied 333,685 gallons between September 1967 and January 1969. As reflected in test results, the dioxin content of its product ranged from .1 to .3 ppm.

There is no evidence that Thompson knew of any toxicity problems associated with the use of any of its products up to the time that the government required it to produce Agent Orange. The company had no testing facilities on its premises. William T. Thompson, president of the corporation, testified that he had never heard of dioxin until recently.

Soon after it began to manufacture Agent Orange, Thompson experienced an incident that caused a few of its employees to develop what was believed to be chloracne. The principal evidence of Thompson's knowledge of this health hazard is found in two internal memos of Dow Chemical Company dated February 1967. The first states that M.S. Buckley, now deceased, Technical Director of Thompson, requested information from Dow to assist him in dealing with a "severe chloracne problem" among some of Thompson's employees. The second memo, written by V.K. Rowe, indicates that he did not give Buckley a very detailed description of what caused the problem because, in his words, "it was quickly apparent that Mr. Buckley had little understanding of the toxicological aspects of his problem. Had he asked for methods, etc., I would have agreed to send them to him."

There is no claim that Thompson disclosed this production accident to the government; however, the government already knew of similar and more serious problems that had occurred at Monsanto and Diamond Alkali, and, possibly, the government knew of Dow's 1964 chloracne problem.

At most, this incident establishes Thompson's knowledge of possible health hazards related to the manufacture of Agent Orange, something the government already knew. It establishes no knowledge in Thompson of hazards to users. Plaintiffs would have me infer such knowledge, perhaps even infer that the problem was discussed in the telephone conversation between Buckley and Rowe. The evidence presented, however, does not support the inference but instead invites mere speculation, particularly when we consider that the deceased Mr. Buckley would be unavailable to testify.

On the other side of the knowledge comparison, it is clear that by 1967, when Thompson first contracted to manufacture Agent Orange, the government had a significant amount of knowledge about dioxin, its contamination of 2,4,5-T, and its association with chloracne and some other health problems. Without question, the government's level of knowledge greatly exceeded that of Thompson in 1967.

Plaintiffs also argue that Thompson had knowledge of and the discretion to use alternative methods of manufacturing Agent Orange. Thompson points out, accurately, that there is no evidence to support that assertion. Even assuming it to be true, however, there is no evidence that Thompson knew of any risk to users that would call for the use of an alternative manufacturing method, and Thompson itself denies knowledge of any such risk. A party opposing summary judgment cannot rely on mere innuendo to defeat the motion. See Vantage Point, Inc. v. Parker Brothers, Inc., 529 F. Supp. 1204, 1213-14 (E.D.N.Y. 1981), aff'd, 697 F.2d 301 (2d Cir. 1982).

Under these circumstances, I conclude that Thompson has established that there is no issue of material fact remaining for trial on the government contract defense, and that Thompson is therefore entitled to summary judgment.

Hercules, Inc.

Hercules supplied to the government compounds containing 2,4,5-T pursuant to contracts dating from May 8, 1964 through May 20, 1968. There is no evidence in the record that Hercules knew anything concerning chloracne or other health problems related to the production of 2,4,5-T during the 1940s or 1950s. Dr. John P. Frawley, Hercules' general manager of Health, Environment and Safety, who has been with Hercules since 1956, testified at his deposition that he did not learn of Monsanto's 1949 chloracne problems until February 1965. With respect to Diamond Alkali's explosion in 1960, Frawley testified that he knew of the explosion but not of any toxicity associated with it.

Hercules began producing phenoxy herbicides in 1961. On July 3, 1963 Frawley wrote to V.K. Rowe of Dow concerning a request by Dr. John Leary of the United States Department of Agriculture that the chemical companies do some testing of phenoxy herbicides. Plaintiff contends that this letter is evidence of Hercules' knowledge of the problems associated with 2,4,5-T. However, as Frawley points out in his answering affidavit, the few problems of alleged health hazards mentioned in the letter relate to 2,4-D, not to 2,4,5-T. The only reference to 2,4,5-T concerned a lawsuit in which the plaintiffs were unsuccessful. Further, with the exception of two inconclusive incidents involving two children who ate sprayed fruit and some workers who worked in a sprayed area, plaintiffs present no concrete evidence of Hercules' knowledge of any problems prior to March 1965.

Frawley states that the first knowledge he had of industrial health problems associated with the production of 2,4,5-T occurred in February 1965, when he was told by Dow of its chloracne problem. In March 1965, Frawley attended the Dow meeting, where he received Dow's analyses of Hercules' product. They showed a very low level of dioxin. Later in 1965, Hercules improved its process of production so as to eliminate even the low dioxin level, and Hercules began to test its own product for dioxin contamination.

Dr. Frawley states in his affidavit that to his knowledge Hercules never had a case of chloracne among its workers from 1961 until 1970 when it ceased production. Further, he states that Hercules learned of possible teratogenicity only in 1969 when the government released the Bionetics Report. Plaintiffs have presented no evidence to contradict this.

From January 1966 through May 1970 Hercules' product contained no measurable dioxin except on one test in September 1966 when it measured .1 ppm. The results of U.S.D.A. tests in 1970 also showed no measurable dioxin. Hercules states in its reply brief that the only Hercules 2,4,5-T containing more than a trace amount of dioxin was tested in 1965 and was not sold to the government. Plaintiffs do not seriously contend that Hercules' product was contaminated by any significant amount of dioxin.

The lack of dioxin in Hercules' product leads to the conclusion that no issue of fact exists with respect to the relative knowledge of Hercules and the government. As will be discussed infra, since its product was dioxin-free, Hercules had no knowledge of harm from dioxin contamination caused by its product and thus did not know more than the government about hazards associated with the use of its product. Consequently, its motion for summary judgment is granted.

DISCUSSION

As indicated at the outset of this opinion, in pretrial order no. 33, I stated that the government contract defense would entitle a defendant to judgment dismissing all claims if the defendant proved:

1. That the government established the specifications for "Agent Orange";

2. That the "Agent Orange" manufactured by the defendant met the government's specifications in all material respects; and

3. That the government knew as much as or more than the defendant about the hazards to people that accompanied use of "Agent Orange".

Each defendant has established the first two elements of the defense — that the government established the specifications for Agent Orange, and that the Agent Orange manufactured by the defendant met those specifications in all material respects. Accordingly, pursuant to Federal Rules of Civil Procedure 56(d), I have determined that there is no substantial controversy with respect to any moving defendant over those two facts, and they are deemed established for purposes of the trial of this action. These are the only two facts I am determining for future purposes. All other discussion of facts and evidence in this decision relates only to these summary judgment motions.

Plaintiffs argue with considerable force that because defendants did not share with the government all their knowledge about Agent Orange, the resulting ignorance on the government's part affected not only what was put into or omitted from the specifications but also the standards applied to determine whether a particular defendant's product conformed to the specifications. I view this more as a means of talking about the central problem of knowledge than as a proper description of the elements of the government contract defense. For purposes of my discussion of that defense and its analysis, the various problems arising out of differing levels of knowledge between the government and the defendants are encompassed in the third element.

With respect to that third element — whether the government knew as much as or more than the defendant about the hazards to people that accompanied use of Agent Orange — four of the defendants, Riverdale, Hoffman-Taff, Thompson, and Hercules, have established that there is no triable issue and each of them is entitled to judgment dismissing all complaints and all cross-claims against it on the ground that the government contract defense shields it from liability.

As to all four of these defendants there is no triable issue over knowledge. As to each, on the record before the court the government clearly knew as much as or more than the defendant about any hazards to people that accompanied use of that defendant's product. Riverdale and Hoffman-Taff both deny any knowledge of hazards, and no other party has controverted their denials. Thompson was essentially an unwilling producer of Agent Orange who knew virtually nothing about its hazards. Clearly, in 1967 when it first began to supply the herbicide, the government's knowledge of the hazards of Agent Orange exceeded that of Thompson. Equally clearly, the government's knowledge at that time was increasing more rapidly than was Thompson's.

Hercules' dismissal rests on the same ground, but follows for different reasons. Unlike Riverdale, Hoffman-Taff and Thompson, Hercules was aware of possible dioxin contamination of 2,4,5-T. But since the Agent Orange produced by Hercules was free of the contamination, there were no dioxin-related hazards accompanying the use of its product about which Hercules could have had knowledge.

Summary judgment is denied with respect to Dow, T.H., and Uniroyal on the ground that on the papers before the court triable issues of fact are presented on the question of relative knowledge. Following our earlier schedule, this would have meant that the trial of the government contract defense would proceed for these defendants together with Monsanto and Diamond Shamrock on June 27, 1983.

However, for a combination of reasons I have concluded that a separate trial of the open issues on the government contract defense with respect to the remaining defendants is no longer appropriate. A separate trial might prejudice the plaintiffs by over-emphasizing the importance of this narrowly drawn defense as it has evolved in its present context; it might prejudice the defendants by requiring presentation of the defense based on hypothetical questions of causation that ultimately would have to be developed on a full record; and it would prejudice both sides due to the additional time, effort, and expense that would be required.

As this case has developed we all have raised our level of knowledge concerning this somewhat unique litigation. In 1980 when I entered an order anticipating phased trials, the idea of a separate trial of the fact issues raised by the government contract defense seemed appealing. It had the advantages of focusing on what appeared to be a discrete segment of the evidence with a possible early termination of the lawsuit and a saving to all parties of considerable unnecessary expense. The chief disadvantage of the separate trial appeared to be a relatively insignificant one of having to assume hypothetically certain facts about liability. On balance, the goals of obtaining a just, speedy, and inexpensive determination seemed to be well served by working toward a separate trial of the government contract defense. The premise underlying that conclusion was that the elements of the defense would be uniquely suited to consideration and adjudication separately and apart from the issues of liability, causation, and damages. In addition, it seemed to me at the time that as a practical matter discovery on those discrete issues would be rather narrow compared to the discovery that some of the other fact issues presented by this action might require.

What has actually happened, however, is that, as we all have learned more about the development and use of Agent Orange in Vietnam, the issues in the action have become clearer. Plaintiffs undoubtedly will strongly emphasize a negligent failure to warn as a basis for liability. But when the "knowledge" factor of the government contract defense is placed alongside a liability theory of a negligent failure to warn, the issues, unfortunately, no longer remain discrete or separate. On the contrary, they tend to merge to such an extent that the legal test for liability under a failure-to-warn negligence theory would fully encompass all the knowledge issues of the government contract defense except for the final test of whether the knowledge "would have" as opposed to "might have" affected the military's handling of Agent Orange purchases and use.

Separate application of the government contract defense has been possible as to the four defendants in whose favor I will be granting summary judgment. As to the remaining defendants, however, the central point of the dispute seems to have shifted. Whether or not the presence of a dioxin contaminant in Agent Orange gives rise to liability is complicated by a variety of circumstances, including the relative ignorance of virtually everyone about dioxin when our involvement in Vietnam commenced, the increasing level of everyone's knowledge about dioxin at varying rates until Agent Orange was no longer used in Vietnam, the changing level of technology which enabled the scientist to detect and measure smaller and smaller concentrations of dioxin over the relevant time period, and the dynamic, constantly changing attitudes of the military and political authorities about the use in Vietnam of herbicides.

The problem is illuminated by comparing the situation of Hercules with that of Dow. Hercules attended the 1965 meeting called by Dow to consider the problem of dioxin contamination in 2,4,5-T. Beginning in January 1966 the 2,4,5-T produced by Hercules was, with one exception, free of any detectable dioxin. This meant that if dioxin was present, it was there in concentrations of less than one tenth of one part per million. In one month out of the ensuing 39 months of production, Hercules' 2,4,5-T did show dioxin contamination, but at the minimum measurable level: .1 ppm. That one month when a trace was found becomes de minimis both as to the quantity affected and the degree of contamination when compared with the dioxin contained in the other companies' products. Since Hercules had no knowledge of its product creating hazards to people, its knowledge could not have exceeded that of the government, and Hercules has established the third element of the government contract defense, thereby entitling it to summary judgment in its favor.

Dow took an approach to its production that differed from that of Hercules. Instead of producing a dioxin-free product, it adopted a self-imposed contamination standard of 1 ppm for its 2,4,5-T. At the March 1965 meeting, Dow urged the others in attendance to adopt that single standard and to use it as the industry standard for 2,4,5-T. From the evidence in the present record it appears that Dow believed that standard to be within a reasonable margin of safety so that hazards to people would be eliminated.

The test results in the record show that neither Diamond Shamrock nor Monsanto followed Dow's recommendation. Dow itself lived up to its self-imposed standard, however, with the result that the Agent Orange it produced contained .5 ppm. It may eventually appear that the 1 ppm standard was safe. If so, then Dow could succeed in its government contract defense, just as Hercules has now done, by showing the safety of its product. In that event it could also succeed against any liability claim based on an allegedly defective product. The fact remains, however, that we do not know whether Dow's self-imposed 1 ppm standard was a safe level for dioxin contamination of the 2,4,5-T. Indeed, we do not know whether contamination even at the level of 140 ppm, the highest level for any product at any time reported in the papers now before the court, would produce Agent Orange that was hazardous to people. Thus, Dow's adoption of the 1 ppm standard raises an issue of fact that precludes summary judgment.

It may also eventually appear that if Dow had revealed its concerns about dioxin to the government in 1965 the military would have adopted the Dow 1 ppm standard regardless of what our present knowledge tells us about the safety of that standard. In that event, Dow would succeed against a claim of negligent failure to warn, but not necessarily on its government contract defense because the defense would fail merely if the information "might have affected" the military's decisionmaking. Rather, Dow would succeed on the causation issue, because its failure to disclose its concerns about dioxin would not have affected the military's judgment in using the product.

If we were to proceed at this time with a separate trial of the government contract defense as to those defendants who remain in the case, we would on that trial have to determine whether dioxin contamination of Agent Orange was harmful, and if so, in what concentration did it become harmful, and, if the defendant's product was unsafe and if the defendant had told the military in 1965 of Dow's fears about the effect of dioxin contamination, whether the military might have stopped using Agent Orange, or changed its specifications to provide a maximum level of contamination, or changed its method of using Agent Orange in the field, or imposed safety precautions in connection with its use in the field.

A trial to answer these questions would necessarily involve most of the evidence needed for trial of the issues relating to liability and general causation. Under these circumstances I conclude that justice would be served by combining what remains of the government contract defense issues with a trial on liability and general causation which will be scheduled after completion of the remaining discovery necessary for those issues.

Case Management

I will continue the action under the supervision of the special master, Sol Schreiber, who has thus far provided extraordinary, dedicated and effective assistance in bringing the case this far. His authority will continue as before, 94 F.R.D. at 174-76, with the objective being a trial at the earliest reasonable date covering the issues of liability, general causation, and the government contract defense. I will request the special master to recommend an appropriate timetable for the remaining discovery, any further motions, preparation of a pretrial order, and trial. I will also request from him as quickly as possible a formal recommendation with respect to the nature and form of class certification. And, despite the determined resistance to class certification by some of the dismissed defendants, I will nevertheless in the interest of justice defer entry of summary judgment on the present motions until after the class has been certified.

Confidentiality of Motion Papers

I will also ask the special master to reconsider with counsel whether the need for secrecy remains, or whether some of the restrictions on public disclosure of the affidavits, documents, and depositions cannot be lifted. From what I see on television and read in the newspapers, it appears that some counsel and some parties in this case do not take seriously either their ethical obligations or the orders of this court with respect to non-disclosure of information. I also understand that due to careless designations on the filing of some papers, some information which was intended to be under seal ended up in the hands of a newspaper reporter. Other documents as to which there have been no slip-ups in filing or designation for sealing, have also been quoted directly in newspaper articles, thereby confirming my own view in the circumstances of this case of the futility of attempting to keep information confidential.

Although keeping discovery materials subject to the protective order previously entered in this case, pretrial order no. 43, 96 F.R.D. 582 (E.D.N.Y. 1983), has already served several worthwhile purposes, I am not certain that those purposes are furthered by continuing the sealing with respect to the motion papers on these summary judgment motions. I request the special master to review the matter again with counsel and make an appropriate recommendation to me.

CONCLUSION

In summary, the motions of defendants Hercules, Thompson, Riverdale, and Hoffman-Taff for summary judgment on the ground of the government contract defense are granted, and all claims and cross-claims against those defendants based on their having supplied Agent Orange to the government pursuant to contract are dismissed. The motions of Dow, T.H., and Uniroyal are denied due to the existence of genuine issues of material fact precluding summary judgment. The trial of the government contract defense with respect to the five remaining defendants, Dow, T.H., Monsanto, Diamond Shamrock, and Uniroyal, will be combined with the trial on liability and general causation.

SO ORDERED.

ON MOTION FOR REARGUMENT

Defendants Dow Chemical Company, T H Agriculture Nutrition Co., Inc., and Uniroyal, Inc., move for reargument of the court's order dated May 20, 1983, which denied their motions for summary judgment on the government contract defense. After careful consideration of all of the materials submitted, the motions are denied.

SO ORDERED.