Date: 20000612
Docket: T-577-87
BETWEEN:
IMPERIAL OIL LIMITED
and its subdivision PARAMINS
Plaintiff
- and -
THE LUBRIZOL CORPORATION
and LUBRIZOL CANADA, LIMITED
Defendants
- and -
DR. FRED W. BILLMEYER JR.
Intervenor
REASONS FOR JUDGMENT
NADON J.
Introduction
[1] These are unusual proceedings. The Plaintiff, Imperial Oil Limited ("Imperial") seeks to set aside a judgment of this Court rendered on September 17, 1990 by Cullen J.1 Imperial brings these proceedings pursuant to former Rule 17332 of the Federal Court Rules, which provides:
| 1733. A party entitled to maintain an action for the reversal or variation of a judgment or order upon the ground of matter arising subsequent to the making thereof or subsequently discovered, or to impeach a judgment or order on the ground of fraud, may make an application in the action or other proceeding in which such judgment or order was delivered or made for the relief claimed. |
[2] The basis of Imperial"s proceedings is fraud on the part of the intervenor, Dr. Fred W. Billmeyer Jr., an expert witness who testified on behalf of the Defendants, and fraud on the part of the Defendants. Specifically, Imperial alleges that Dr. Billmeyer, during the course of the trial of the issues which led to the judgment of Cullen J., deliberately gave false evidence to deceive the Court or was reckless when giving his evidence. Imperial further alleges that the Defendants, the Lubrizol Corporation and Lubrizol Canada ("Lubrizol") gave Dr. Billmeyer a mandate to match test results which they had predetermined. The specific allegations against Dr. Billmeyer appear at paragraphs 21, 22 and 23 of Imperial"s Statement of Claim:
| 21. In May and June 1990, Dr. Billmeyer falsely testified to this Court |
| (a) that he had drawn his preliminary calibration curve based on Mn values for the "narrow fractions" based on VPO data; |
| (b) that he had adjusted his preliminary calibration curve to match proper VPO results for the anchor samples before testing any of the PIB samples that he had testified about; |
| (c) that the final calibration curve did in fact "match" the VPO value of the two anchor samples, thereby suppressing the fact that the Waters 200 GPC result on his critical Amoco H-1500 anchor sample was 7.5% higher than the VPO result (when belatedly received) on that sample; |
| (d) that the Mn values of the PIBs he offered to the Court were the product of the independent scientific inquiry he had described; |
| (e) that this Court could regard Parapol 2200 and Parapol 2225 and ECA 10788 as indistinguishable for the purpose of determining the "true" Mn of the starting PIB of the accused products; |
| (f) that he had run Exxon"s AM-S 320.06 test method exactly except for the solvent used; |
| (g) that Exxon"s AM-S 320.06 test method understated the true Mn of the PIB by 4 per cent when compared with results obtained on his purportedly properly calibrated Waters 200 GPC; |
| (h) that the Mn values of the Certificates of Quality Assurance tendered by Imperial Oil in respect of the starting PIB of the accused products therefore could not be accepted as accurate; and |
| (i) that Dr. O"Driscoll"s even higher test results of the Mn of the PIB samples could safely be accepted as more accurate than his own. |
| 22. Dr. Billmeyer gave the false testimony aforesaid either knowing it to be untrue or being reckless as to whether it was true or false. |
| 23. In either case, the testimony was fraudulent within the meaning of Rule 1733. |
[3] The particulars of Dr. Billmeyer"s fraudulent testimony are set forth by Imperial at paragraphs 25 to 50 of its Statement of Claim. The particulars are treated under six headings, which identify six grounds of fraud, namely:
| Ground 1: Dr. Billmeyer"s testimony in Canada was false in that contrary to his Canadian testimony, the Mn values he used in constructing his preliminary calibration curve came from Lubrizol"s HSGPC, not VPO. |
| Ground 2: Dr. Billmeyer"s testimony in Canada was false in that contrary to his Canadian testimony Dr. Billmeyer"s final calibration curve did not match VPO results from the anchor samples. |
| Ground 3: Dr. Billmeyer"s testimony in Canada was false in that Dr. Billmeyer gave this Court Waters 200 GPC test results on the Exxon PIB samples that were overstated by seven and a half per cent. This overstatement was the basis of this Court"s finding of infringement. |
| Ground 4: Dr. Billmeyer agreed on April 17, 1990 that Dr. O"Driscoll"s test results should be adjusted to give dramatically higher results without any further tests and thereafter Dr. Billmeyer supported O"Driscoll"s "universal calibration" methods and O"Driscoll"s higher results without any scientific basis for doing so. |
| Ground 5: Dr. Billmeyer"s testimony in Canada was false when he testified that he had followed the Exxon VPO test method "exactly" to determine that Exxon"s method seriously understated the Mn of the PIB. |
| Ground 6: Dr. Billmeyer"s testimony in Canada was false when he testified that the European samples could not be distinguished from Parapol 2200. |
[4] With respect to the mandate which Imperial asserts Lubrizol gave Dr. Billmeyer, the allegations appear at paragraphs 17 to 20 of the Statement of Claim:
| 17. The Ohio evidence shows that on January 25, 1990 Dr. Billmeyer was instructed by Lubrizol"s outside counsel, Mr. John Fairweather ("Fairweather"), to "set up" a Waters 200 GPC located at Lubrizol to produce test results that "matched" test results that Lubrizol said had previously been obtained from its High Speed GPC ("HSGPC"). The Lubrizol methodology using the HSGPC was not a test method called for by the patent. It was plain and obvious to Dr. Billmeyer that calibrating the Waters 200 GPC to match the HSGPC would pre-determine that he would get the same test results as had allegedly been obtained by Lubrizol on the HSGPC, and that this approach would not provide any independent analysis of the "true" Mn of the PIBs in question. |
| 18. Nevertheless, between January and May 1990, as hereinafter particularized, Lubrizol offered and Dr. Billmeyer accepted a mandate that evolved step-by-step ultimately to include the following assignments: |
| (a) to set up the Waters 200 GPC to match Lubrizol"s HSGPC test results; |
| (b) to testify falsely to this court that his final calibration curve for the Waters 200 GPC "matched" VPO data which in fact he did not have until after his calibration was finalized; |
| (c) to testify falsely to this Court that he had followed a Waters 200 GPC calibration procedure that he had not in fact followed; |
| (d) to testify falsely to this Court respecting the "true Mn" of certain PIBs when he had failed to follow the correct scientific calibration procedure for the Waters 200 GPC as he had described it to this Court; |
| (e) to testify in support of the accuracy of even higher Mn results produced by Dr. O"Driscoll, without Dr. Billmeyer possessing any independent study, knowledge or scientific bases to evaluate the O"Driscoll testing method; |
| (f) to testify falsely that he had conducted tests using Exxon"s test method AM-S 320.06 "exactly" (except as to solvent); |
| (g) to endeavour to demonstrate that the AM-S 320.06 test method "understated" the Mn of the PIB as compared to his Waters 200 GPC results; and |
| (h) to testify falsely to this Court that Parapol 2200, Parapol 2225, and ECA 10788 exhibited the same Mn and Mw/Mn characteristics, and to testify that these PIBs were indistinguishable, and could therefore be "averaged" to obtain a "true" Mn for the purpose of establishing infringement. |
| 19. Dr. Billmeyer accepted the original and successive stages of the Mandate knowing that the Mandate did not call for results determined by an independent and proper scientific inquiry but to "match" results needed by Lubrizol to prove its case. Particulars of the communication of pre-determined results presently known to Imperial Oil include: |
| (a) On February 1, 1990 Dr. Billmeyer met at Lubrizol with Mr. Joe Bauer, Lubrizol"s inside counsel, Fairweather, and some Lubrizol employees, and was advised |
| (i) that in the event the results obtained by Dr. Billmeyer on the Waters 200 GPC were not the same as results from Lubrizol"s HSGPC, Lubrizol "would be in some trouble. We need to ensure same results ." (italics added); |
| (ii) that Lubrizol had already obtained test results from Dr. O"Driscoll, who had calibrated his GPC using a different method called "Universal Calibration", (which Dr. Billmeyer did not consider a method of the Meinhardt patent) and that Dr. O"Driscoll"s results matched those generated by Lubrizol"s own HSGPC, and Lubrizol informed Dr. Billmeyer that they "like[d] them"; and |
| (iii) that in light of these events, Dr. Billmeyer was to run some preliminary tests on the Waters 200 GPC and "to operate with a minimum of paperwork". |
| (b) Lubrizol gave Dr. Billmeyer HSGPC results on the anchor samples which he was to match in calibrating the Waters 200 GPC. |
| (c) Lubrizol informed Dr. Billmeyer before he began testing that "establishing PIB Mn of 2700 would also settle the issue of ECA 12819 infringement in our favour"; |
| (d) Dr. Billmeyer was told that the higher the Mn of the PIB the better Lubrizol"s chances of prevailing against Imperial Oil; |
| Instead of ignoring or putting aside Lubrizol"s directions aforesaid, Dr. Billmeyer set about fulfilling Lubrizol"s Mandate putting aside what he knew or should have known to be the true facts, as hereinafter described. |
| 20. In the period February to April 1990, Dr. Billmeyer carried out the Mandate by: |
| (a) calibrating his Waters 200 GPC to match results from Lubrizol"s HSGPC on the two anchor samples and failing to calibrate the Waters 200 GPC to match to proper VPO results. Indeed, on February 15, 1990 Dr. Billmeyer calibrated the Waters 200 GPC without the benefit of any VPO test results at all in respect of his anchor samples contrary to his own evidence of correct procedure; |
| (b) failing to verify that VPO and HSGPC data that were provided to him by Lubrizol were either correct or complete; |
| (c) failing to further adjust his previously finalized Waters 200 GPC calibration curve when he later received VPO data even though the Waters 200 GPC Mn value on the critical Amoco H-1500 anchor sample (2625) was 7.5% higher than the VPO Mn value (2445) for that sample; |
| (d) failing to conduct VPO tests in accordance with AM-S 320.6; and |
| (e) participating with Lubrizol in a decision to boost Dr. O"Driscoll"s test results dramatically to give higher results, without the presence of Dr. O"Driscoll, by scientifically improper manipulations of Dr. O"Driscoll"s existing data. |
Facts
[5] I now turn to the facts relevant to the determination of the issues which are before me. In order to properly understand the issues, a review of the action commenced by Lubrizol, which culminated in Cullen J."s decision, is essential. In dealing with Lubrizol"s action, I will also refer to and discuss proceedings commenced by Lubrizol in the United States. These proceedings are known as the Ohio I and Ohio III proceedings.
[6] Lubrizol commenced an action against Imperial on March 16, 1987, for infringement of Canadian Patent 1,094, 044 ("the patent")3. The patent was applied for on February 24, 1978 and was issued on January 20, 1981, for "the invention of carboxylic acid acylating agent, derivatives thereof, concentrates and lubricating compositions containing the same, and processes for their preparation". Put in simpler terms, the patent was for the invention of dispersement molecules which form part of the additives to lubricating oils for engines. Without the additives, ordinary petroleum coming out of the ground would be inadequate in the modern day gasoline-driven engine. The patent claims at issue were claims number 1, 2, 3, 35, 41, 44, 50, 51, 52 and 57. However, as held by the Federal Court of Appeal4, the monopoly claimed by the patentee is found in claim 1 of the patent, on which the other claims "either expressly or derivatively depend". The numerical parameters of the patent appear clearly from claim 1, which I now reproduce:
| A lubricating composition comprising a major amount of oil lubricating viscosity and a minor amount of one or more carboxylic derivative compositions produced by reacting at least one substituted succinic acylating agent with a reactant selected from a group consisting of (a) amine characterized by the presence of at least one H-N < group, (b) alcohol, (c) reactive metal or reactive metal compound, and (d) a combination of two or more of any of (a) through (c), the components of (d) being reacted with said one or more substituted succinic acylating agents simultaneously or sequentially in any order, wherein said substituted succinic acylating agents consist of substituent groups and succinic groups wherein the substituent groups are derived from polyalkene, said polyalkene being characterized by a Mn5 value of 1300 to about 5000 and a Mw/Mn6 value of about 1.5 to about 4, said acylating agents being characterized by the presence within their structure of an average of at least 1.3 succinic groups for each equivalent weight of substituent groups. (Emphasis added) |
[7] At the relevant time, Lubrizol and Imperial"s sub-division Paramins were the leading competitors in the sale of lubricating oil additives and were the two largest manufacturers in the lubricating oil additive industry. Cullen J. indicated in his Reasons that it was essential to place the patent in its proper context for a proper understanding thereof. At pages 165 and 166, the learned trial judge makes the following remarks:
| [8] ... At present, the market for motor oil additives is in a state of great flux. Lubricants marketed in North America are subject to standards promulgated by the American Petroleum Institute ("API"). On March 4, 1988, the API promulgated new standards for motor oil. These new "SG" standards were more demanding, and were designed for the needs of contemporary passenger vehicles. The improvement in the standards was the most significant upgrading in the history of the passenger motor car oil additive industry. To meet these new high standards, base oils are not changed, but rather motor oil additives are used. The challenge was to develop an additive whose treat rate met the new standards at the lowest possible cost. |
| [9] In essence, reformulation was required of virtually every motor oil, and this created a good deal of concern and some turmoil in the market place. It also created an opportunity for the additive supplier who could meet the new standards at a low cost to gain and keep a significant market share. Purchasers of lubricating oil additives are generally reluctant to change suppliers but when new standards are issued there is potential for loyalties to shift dramatically. |
[8] At pages 171 and 172 of his Reasons, Cullen J. incorporates a direct quote from Lubrizol"s Memorandum of Fact and Law, setting forth a basic explanation of the relevant chemistry. That chemistry is, it goes without saying, the same chemistry necessary for a proper understanding of the issues before me. I will, therefore, reproduce here that part of Cullen"s Reasons where he quotes from Lubrizol"s Memorandum of Fact and Law:
| 16.01 The basic chemistry involved in the preparation of the patented lubricating compositions and concentrates can be described below: |
| (a) Polybutene or PIB (an acronym for polybutene) is a polymer made by joining together (or "polymerizing") many butene molecules. |
| (b) PIB is reacted with maleic anhydride (MA), often in the presence of chlorine, to produce PIBSA (an acronym for polyisobutenyl succinic anhydride). Unreacted PIB (i.e. PIB which does not react with MA) and unreacted MA remain. The unreacted MA is stripped out of the mixture containing PIBSA, but the unreacted PIB remains in the mixture. |
| (c) i. PIBSA is what is referred to as an acylating agent. |
| ii. An acylating agent is a chemical which has an acyl group in its structure. |
| iii. The acylating agent or PIBSA has within its structure both a "substituent group" (a "polybutenyl" or "PIB" group) and a succinic (SA) group. It is sometimes referred to as a "substituted succinic acylating agent" because polybutene is substituted for one of the hydrogens in succinic anhydride. |
| iv. The patent calls, among other things, for the substituted succinic acylating agent or PIBSA to have present within its structure an average of at least 1.3 succinic (SA) groups for each equivalent weight of substituent (or polybutenyl) groups. |
| v. This ratio is often referred to as the "succination ratio" (S.R.) |
| (d) The reaction to form PIBSA is known as succination because it results in the formation of a succinic (SA) group in conjunction with the PIB group as part of the PIBSA. |
| (e) The next step is that PIBSA is reacted with a pollymine (PAM) or trishydroxymethyl amino methane (THAM) to produce PIBSA/PAM or PIBSA/THAM. This is a dispersant. It is a "carboxylic" derivative compound because it is a compound derived from a carboxylic compound, namely PIBSA - which has carboxylic groups. This reaction is known as acylation because the PIBSA "acylates" the PAM or THAM i.e., introduces an acyl group into their structure. |
| (f) The unreacted PIB remains an unreacted PIB and does not participate in the acylation reaction. |
| (g) Finally the PIBSA/PAM and in some cases the PIBSA/THAM or PIBSA itself is combined with other additives to form concentrates which in turn are incorporated in lubricating compositions most usually PCMOs and HDs. |
| "Determination of Succination Ratio |
| 17. 01 The succination ratio is determined by the following succination ratio formula: |
| SR = Mn of PIB x SAP No. of PIBSA |
| 112,200-(98 x SAP No. of PIBSA |
| 17.02 The Sap No. of the PIBSA is: |
| 17.03 (a) S.R. is directly dependent on the Mn of the PIB |
| (b) If Mn is understated, say by 10%, then S.R. is understated by 10%. |
| 17.04 (a) S.R. is also almost directly dependent on the Sap No. of the PIBSA |
| (b) If Sap No. is understated - S.R. is understated. |
| 17. 05 (a) S.R. is also almost inversely dependent on the AI of the PIBSA |
| (b) If AI is overstated - S.R. is understated." |
| Ex. P14-3, Tab 45, 47 (Formulae) |
| [37] The particular parameters mentioned in the claim are: |
| "i. A higher than went before number average molecular weight " specifically an Mn of 1300 to 5000; |
| "ii. A moderate degree of molecular weight distribution " specifically an Mw/Mn ratio of 1.5 to 4.0; |
| "iii. A significant degree of over-succination " specifically a succination ratio (SA to reacted PIB) of at least 1.3." |
[9] As Appendix B to his Reasons, Cullen J. attached a glossary of terms, which can be found as Appendix A to my Reasons.
[10] At issue before Cullen J., were two groups of Imperial products. First, those products derived from a starting PIB called ECA 4600, having an Mn of about 1300. Secondly, those products derived from a starting PIB which Imperial referred to as ECA 10788, Parapol 2225 and Parapol 2200. This PIB had been described by Imperial as "polybutene of 2225 molecular weight".
[11] The issues which Cullen J. had to decide were the validity of the patent and the infringement thereof. With respect to the validity issue, Imperial had challenged the validity of the patent on a number of grounds, including obviousness, anticipation, insufficient disclosure and claims broader than the invention/lack of utility. With respect to the infringement issue, Imperial conceded, subject to its arguments regarding the validity of the patent and subject to a license allegation, that its products derived from the ECA 4600 PIB did infringe the patent. The true debate before Cullen J., as it turned out, was in respect of the Imperial products derived from the PIB referred to by Imperial as ECA 10788, Parapol 2225 and Parapol 2200. Those products are set forth in a chart which appears at page 188 of Cullen J."s Reasons, and which I reproduce below for ease of reference:
[12] More particularly, the debate before Cullen J. was concerned with those Imperial products referred to as ECA 10444, ECA 10271 and ECA 12819 which had been made, used and/or sold by Imperial in Canada. The starting PIB for those products was ECA 10788, Parapol 2200 or Parapol 2225. I should note that the specifications for ECA 10788 required an Mn by VPO7 of between 2150 and 2300, with a target value of 2225.
[13] The infringement issue was whether the PIBSAs from which PIBSA/PAMs ECA 10444, ECA 10271 and ECA 12819 were made, had succination ratios ("S.R.") within the parameters of the patent claims. The answer to that question depended, in part, on the Mn of the starting PIB. At paragraph 4 of its Statement of Claim, Imperial puts the matter as follows:
| 4. The critical issue at trial with respect to infringement was whether the succination ration of the dispersants in question was within the Meinhardt claim limitation. This, in turn, depended in part on the "number average molecular weight" (Mn) of the starting PIB supplied to Imperial Oil by Exxon Chemical Company ("Exxon"). The PIB was used by Imperial Oil to manufacture the dispersants that were incorporated into the various additive packages and finished oils accused of infringement. |
[14] On the evidence before him, Cullen J. made the following findings which are relevant for the present purposes8:
| 1. Parapol 2200, Parapol 2225 and ECA 10788 "were essentially all the same" product. |
| 2. The true Mn value of ECA 10788, Parapol 2225 and Parapol 2200 "averaged at least 2500". |
| 3. The nominal value of 2200-2225 of Parapol 2200, Parapol 2225 and ECA 10788, as per specifications, had been determined by a VPO test method, AMS 320.06, which understated the true Mn of the PIB at issue. |
| 4. The S.R. of Imperial"s dispersants exceeded 1.3 and therefore, fell within the parameters of the patent claims. |
| 5. On a purposive construction of the patent, strict compliance with the words "at least 1.3" was not essential to the invention. |
[15] Imperial appealed Cullen J."s decision and the Court of Appeal dismissed the appeal. In so doing, the Court of Appeal upheld the trial judge"s finding that ECA 10788, Parapol 2225 and Parapol 2200 were the same product. The Court of Appeal also upheld the trial judge"s finding that the true Mn of ECA 10788, Parapol 2225 and Parapol 2200 "averaged at least 2500". The Court of Appeal also upheld the trial judge"s finding that "on a purposive construction of the patent, it is my belief that strict compliance with the words "at least 1.3' was not essential to the invention".
[16] Before discussing Dr. Billmeyer"s evidence before Cullen J., it is also important to discuss what has been referred to as the Ohio I proceedings and the events which unfolded prior to the commencement of the Canadian trial in May 1990. In 1984, an action was commenced by Lubrizol against Exxon Corporation9 in the United States District Court of Ohio, whereby Lubrizol alleged infringement of, inter alia, U.S. patent 4,234,435 ("the U.S. patent")10.
[17] On June 17, 1988, Lubrizol obtained a preliminary injunction against Exxon, enjoining it from manufacturing and selling a product referred to as ECA 10444. As a result of the injunction, Exxon began selling in the U.S. a product referred to as ECA 12819. Lubrizol then moved to add ECA 12819 to the list of Exxon products which were the subject of its Ohio action.
[18] On October 30, 1988, the Ohio I action was settled. The District Court made a final order, on consent, enjoining all of the Exxon products listed in Lubrizol"s action, except for ECA 12819, in respect of which Lubrizol granted to Exxon a limited license. Exxon was granted a paid up license under the U.S. patent to make and sell ECA 12819 by reason of representations made by Exxon that the average S.R. of ECA 12819 was 1.18. As a result of the license, Exxon continued to sell ECA 12819 in the U.S.
[19] When Lubrizol commenced its action in Canada, it sought no order in respect of ECA 10444 and ECA 12819 for the following reasons. Firstly, it was not aware of the existence of ECA 10444. And secondly, ECA 12819 was not in existence at that time. It was only in June 1988 that Lubrizol first learned that ECA 10444, manufactured in Sarnia by Imperial, was being sold in Canada as an additive to meet the new SG standards. According to the specifications for the product, ECA 10444 was made from PIBSA 54, which, in turn, was made from Parapol 2200. The specifications for PIBSA 54 reveal a target Sap. number of 54.3 and a typical active ingredient ("A.I.") level of 80%. The formula for the S.R., which can be found in the glossary of terms, Appendix A to my Reasons, and also in the quote taken from pages 171 and 172 of Cullen J."s Reasons11 is the following:
| SR = SR of PIB x (measured Sap. No.) / AI) |
| 112200-(98 x (measured Sap. No.) / AI) |
[20] Applying this formula to the parameters specified for ECA 10444 and assuming an Mn of 2225, a S.R. of 1.43 is obtained. By reason of this number, Lubrizol applied for an interlocutory injunction and on January 12, 1989, Reed J. granted the injunction, enjoining Imperial from manufacturing and selling ECA 10444 in Canada.
[21] ECA 12819 was first manufactured in Canada in the summer of 1988. At that time, Imperial did not sell the product in Canada, but sold it only to Exxon for the U.S. market. As I indicated earlier, the consent order putting an end to the Ohio I proceedings, excepted ECA 12819 from the injunctive relief.
[22] Imperial began selling ECA 12819 in Canada following Reed J."s order of January 12, 1989. The specifications for this product indicated that it was made from PIBSA 48 which, in turn, was made from ECA 10788. The specifications for PIBSA 48 called for a Sap. number of 48.6 and a typical A.I. level of 86%. Lubrizol took the position that the A.I. level of 86% was overstated and, as a result, this became one of the issues for determination by Cullen J. who ultimately, on the evidence of Dr. O"Driscoll, held that the A.I. was overstated. The A.I. was an important issue, in that the level thereof affects the S.R. and is one of the three parameters revealed by the patent claim, i.e. the Mn of the PIB, the Sap. No. of the PIBSA and the A.I. of the PIBSA. An increase of the A.I. of the PIBSA means a decrease of the S.R., all other values being constant.
[23] The Mn of the PIB (Parapol 2200 and ECA 10788) used by Imperial to make ECA 10444, ECA 10271 and ECA 12819 was of great importance because of its impact on the numerical parameters of the claim. This is why Lubrizol sought and obtained on September 16, 1988, an order directing Imperial to produce the specifications for the starting PIB and all information regarding the Mn and the Mw/Mn ratio for the PIB. The specifications produced by Imperial for ECA 10444 and ECA 10271 indicated that these products were made from PIBSA 54 which, in turn, was made from Parapol 2200. With respect to ECA 12819, the specifications provided that it was made from PIBSA 48 which, in turn, was made from ECA 10788. These specifications were provided as attachments to an affidavit dated September 23, 1988, sworn to by Mr. William Levy. Mr. Levy was examined on discovery and on June 21, 1989, he testified that ECA 10444 and ECA 12819 were made from the same PIB, i.e. Parapol 2200 and ECA 10788. The specifications for the starting PIB, ECA 10788, provided for an Mn in the range of 2150 to 2300, with a target of 2225, as determined by a test method referred to as AMS 320.06. In December 1989, Mr. Levy confirmed that ECA 10788 and Parapol 2200 were the same product. Again, in February 1990, Mr. Levy indicated that Parapol 2200 and ECA 10788 were one and the same.
[24] At a meeting held in Toronto on April 17, 1990, between counsel for Lubrizol and Imperial, Imperial"s counsel were advised that samples of Parapol 1300 and Parapol 2200 had been analysed by Lubrizol"s experts and that it was anticipated that results would show an Mn of about 1300 for Parapol 1300 and an Mn of about 2500 for Parapol 2200. Counsel for Lubrizol indicated to counsel for Imperial that samples of Parapol 1300 and Parapol 2200 would be provided to them upon demand. This verbal offer was subsequently confirmed in writing. In any event, neither the verbal nor the written offers were taken up by Imperial. As a result, Imperial did not test the samples of Parapol 1300 and Parapol 2200 which were tested by Drs. O"Driscoll and Billmeyer.
[25] Paragraph 11 of an Agreed Statement of Facts, filed as Exh. P-7 during the May 1990 trial, provides as follows:
| 11. (a) Another starting polybutene (PIB) is ECA 10788. |
| (b) ECA 10788 has been called Parapol 2200. |
| (c) It has also been called Parapol 2225. |
| (d) It has been described as "polybutene of 2225 molecular weight". |
[26] Paragraph 13 of that same Agreement provides for the following:
| 13. The following PIBSA based materials have been made, used and/or sold by the Defendant in Canada since January 20, 1981 starting from ECA 10788: |
| (a) Borated PIBSA/PAM dispersants: |
[27] As I indicated earlier, Dr. O"Driscoll testified that the AI level of ECA 12819 had been overstated by Imperial. However, Lubrizol also had doubts regarding the Mn of the starting PIB for ECA 10444 and ECA 12819. Initial investigation and tests by Lubrizol regarding the Mn of the starting PIB led it to the conclusion that the Mn thereof was higher than that represented by Imperial. As a result, Lubrizol retained experts to determine the Mn of the starting PIB. I now turn to the evidence of Dr. Billmeyer.
[28] The trial before Cullen J. began on May 7, 1990 and concluded on June 29, 1990 (29 days). Dr. Billmeyer gave his evidence in chief (and was cross-examined thereon) on May 15 and 16, 1990, and gave his reply evidence (and was cross-examined thereon) on June 15, 1990. Before testifying, Dr. Billmeyer"s affidavit, dated April 23, 1990, was filed. A copy of that affidavit was provided by counsel for Lubrizol to counsel for Imperial on April 27, 1990. In his 14-page affidavit, Dr. Billmeyer arrived at the following conclusions:
| (1) Exxon"s VPO test method, AMS 320.06, should not be used to determine the Mn of PIBs whose Mn exceeds 800, because the results will, in all likelihood, "be highly inaccurate". Specifically, Dr. Billmeyer opined that Exxon"s test method, AMS 320.06, understated the Mn of the starting PIB. |
| (2) the true Mn of Parapol 1300 was above 1300 and the true Mn of Parapol 2200 was approximately 2500. |
[29] In paragraphs 12 and 13 of his affidavit, Dr. Billmeyer gives a brief outline of polymers and molecular weight. This outline includes a discussion of "average molecular weight" and a discussion of the various ways of calculating it, namely, by way of the number average molecular weight (Mn) and the weight average molecular weight (Mw). In paragraphs 18 to 23, Dr. Billmeyer discusses the several analytical techniques by which Mn and Mw/Mn can be determined, namely, gel permeation chromatography (GPC) and vapour phase osmometry (VPO). At paragraphs 24 to 41, Dr. Billmeyer discusses Exxon"s/Imperial"s VPO method, AMS 320.06, and states his opinion that Exxon"s GPC results are unreliable and understate the true Mn of the polymers tested.
[30] The crucial paragraphs of Dr. Billmeyer"s affidavit, for the present case, are paragraphs 42 to 54, where Dr. Billmeyer sets out and explains the method he chose to determine the Mn and Mw values of Imperial"s polybutenes. These paragraphs also provide, it goes without saying, Dr. Billmeyer"s results regarding the relevant Mn and Mw values. Paragraphs 42 to 54 of Dr. Billmeyer"s affidavit read as follows:
| 42. I was asked to determine Mn and Mw values for Paramins polybutenes duplicating, as closely as possible, methods and instruments that would have been available as early as 1977-1978 when the Patent was first applied for. |
| 43. For this work I used a Waters 200 GPC instrument, and columns having the following gel pore size designation: 103A, 500A, 500A and 60A. With this instrument, I measured column and instrument efficiency and ascertained that the instrument was operating properly following the procedures in ASTM D3536-76, a GPC standard published in 1976 (Exhibit B2). |
| 44. Following the procedure described in ASTM D3536-76, a first calibration curve was drawn using mean molecular weight (the square root of the product of Mn and Mw) for fractionated polybutene standards. Six such fractions were used. The Mn values used in this first calibration were based on three point VPO determinations following a procedure similar to D3592-77 and the Mw was determined using GPC with a viscosity detector. |
| 45. As expected, my first calibration curve gave Mn values lower than Mn determined by VPO using the three point method. Consequently, I adjusted the calibration curve to give Mn values agreeing with true Mn measured by VPO and also adjusted the calibration curve to ensure that Mw values were also correctly determined. Exhibit B4 is a copy of the final calibration curve. |
| 46. After the calibration was completed, I had the samples of Paramins polybutene analysed on the Waters 200 instrument, producing GPC tracings. Exhibit B5 is the copy of the GPC tracings generated during the course of the analysis. |
| 47. I read each such tracing and entered the data into my computer which I programmed to calculate Mn and Mw according to ASTM D-3536-76.. I have checked one such calculation by hand to ensure that there were no errors in the computer program. Exhibit B6 is a copy of the printouts showing the various calculations. |
| 48. Exhibit B7 is a table summarizing the results that I have obtained. |
| 49. In order to test the hypothesis that values of Mn obtained by VPO using the AMS 320.06 method understate the true values of Mn, I obtained VPO values for samples of Parapol 2200, measured on a Wescan Vapor Pressure Osmometer under my supervision. The method of AMS 320.06 was used exactly, with one exception: the solvent specified in AMS 320.06, benzene, can no longer be used in the United States because it has been found to be carcinogenic. Instead, I used toluene as the solvent at a temperature of 50oC. It is my opinion that this change had no significant effect on my results. To calibrate the instrument, I used the common calibration standard benzyl and the two calibration standards specifically mentioned in AMS 320.06, tetracosane and squalane. The three gave results in good agreement, and an average for all three was used to obtain the values of Mn by this method. Exhibit B8 is a copy of the VPO worksheets. The results of these measurements appears in Exhibit B7. |
| 50. The results obtained by VPO measurement using the method of AMS 320.06 understate the true values of Mn for Parapol 2200 in 13 out of 15 cases. The average understatement is 4%. This confirms that the AMS 320.06 procedure understates Mn. |
| 51. Further, 11 out of 15 samples measured by AMS 320.06 had MN values above the upper specification limit of 2300, and none had an Mn value as low as the specification target value of 2225. Therefore, the VPO studies also show that, even by their own VPO method of obtaining values of Mn, Paramins consistently uses polybutenes that have Mn not only above the target value but outside the range of its specifications. |
| 52. On the basis of my GPC analyses, the true Mn of Parapol 1300 samples ranged between a minimum of 1269 (as opposed to a specified minimum of 1200 or 1275) and a maximum of 1393 (as opposed to a specified maximum of 1375) with an average of 1322 (as opposed to the nominal or target value of 1300). This average is in excess of the specified target value. |
| 53. On the basis of my GPC analyses the true Mn of Parapol 2200 samples ranged between a minimum of 2195 (as opposed to a specified minimum of 2100) and a maximum of 2776 (as opposed to a specified maximum of 2300) with an average of 2457 (as opposed to the nominal or target value of 2225). This is significantly and substantially in excess of both the target value and the specified range. On the assumption that the average of the true Mn of the analysed samples corresponds to the specified target value, the specified target value of 3225 corresponds to a true Mn of 2500. |
| 54. On the basis of my results, it is my opinion that the true Mn of Parapol 1300 is above 1300, and the true Mn of Parapol 2200 is approximately 2500. |
[31] In his affidavit, Dr. Billmeyer deals only with the Mn of Parapol 2200. No mention is made nor is there any reference to Parapol 2225 and ECA 10788, which have been referred to in these proceedings as the "European samples". Before Cullen J., Dr. Billmeyer testified that he had tested 16 samples of Parapol 2200 purchased in the U.S. market, and had determined their Mn. He also testified that he had tested one sample of ECA 10788 and one sample of Parapol 2225 obtained from Europe, and had also determined their Mn. Dr. Billmeyer analysed the European samples after the filing of his affidavit on April 27, 1990. He testified before Cullen J. regarding the European samples and no objection was taken in regard thereto. Dr. Billmeyer also testified that the Mn values of the European samples fell within the range of the Mn values that he had determined for Parapol 2200.
[32] Imperial"s Rule 1733 proceedings are concerned only with that part of Dr. Billmeyer"s evidence which relates to the Mn of the starting PIB. I now turn my attention to the specific reasons given by Cullen J., in making his findings regarding infringement, insofar as they pertain to Dr. Billmeyer"s evidence.
[33] Under the general title of Infringement of the Patent?, Cullen J. posed three questions which are relevant for our purposes. The first question was: Are products variously called ECA 10788, Parapol 2225 and Parapol 2200 really one product? In answering that question in the affirmative, the trial judge referred to the discovery evidence of William Levy, an Imperial officer who testified on its behalf. During the course of his evidence, Mr. Levy clearly stated that ECA 10788, Parapol 2200 and Parapol 2225 were "synonyms"12. In answering the question in the affirmative, Cullen J. also relied on the evidence of Drs. O"Driscoll and Billmeyer. At page 186 (para. 84), he puts the matter as follows:
| 84. ... Analysis by Drs. O"Driscoll and Billmeyer of ECA 10788 and Parapol 2225 established that its Mn values fell within the range of Mn values they had determined for Parapol 2200, and that the true Mn values of ECA 10788, Parapol 2200 and Parapol 2225 averaged at least 2500. Paramins in their specifications for PIBSA 54 and PIBSA 51 state they are prepared from Parapol 2200. (Exhibit P-14-3, Tab 38 - PIBSA 54 specification, p. 3; Exhibit P-14-3, Tab 48 - PIBSA 51 specification p. 3). The PIBSA 48 specification (prepared after the U.S. injunction) states that it is "prepared from ECA 10788". (Exhibit P-14-4, Tab 50 - PIBSA 48 specification p. 3). Mr. Dixon, called by the defendants, has not in any of his evidence stated that there was any difference in the Mn of the products. |
In reaching his conclusion on this point, Cullen J. also indicated that Imperial had failed to call any evidence with a view of distinguishing Parapol 2200, Parapol 2225 and ECA 10788.
[34] The second question posed by Cullen J. was: What is the true Mn of ECA 10788, Parapol 2225 and Parapol 2200? The learned judge answered that question, at p. 186 (para. 85) in the following words:
| 85. Having determined that these products were essentially all the same, the next question to be answered concerned the true Mn of ECA 10788, Parapol 2225 and Parapol 2200. This was a very difficult question to answer and was further complicated by the fact the defendant was not cooperative in providing the plaintiffs with the necessary samples. Here again it should be emphasized that the defendant called no analysis of retained material, nor did it take advantage of an offer made to it of analyzing the samples used by Drs. O"Driscoll and Billmeyer. Paramins" specifications show the nominal Mn of the product at 2200-2225 molecule weight. It was, however, determined by VPO test method AMS 320.06 which understates the true Mn of commercial polybutene at issue here. The analysis of O"Driscoll and Billmeyer underwent searching cross-examination but their conclusion was unassailed - the true Mn value of ECA 10788/Para 2225/Para 2200 averaged at least 2500. |
[35] The above passage makes it clear that in coming to a conclusion regarding the second question, Cullen J. relied on the evidence of both Drs. O"Driscoll and Billmeyer. He did not deal specifically with the evidence of these witnesses, simply concluding that he accepted their conclusion that the true Mn of the starting PIB "averaged at least 2500". The judge also found that Imperial"s VPO test method AMS 320.06 understated the true Mn of the PIB. Finally, the judge made an adverse inference against Imperial when he stated that Imperial had not cooperated in providing Lubrizol with relevant samples. To this, he added that Imperial had called "no analysis of retained material, nor did it take advantage of an offer made to it of analyzing the samples used by Drs. O"Driscoll and Billmeyer".
[36] The last question posed by Cullen J. was one regarding the S.R. That question, he answered as follows, at p. 186 (para. 86):
| 86. Unreacted PIB is excluded when calculating the succination ratio and as the Federal Court of Appeal in Imperial Oil Limited and its subdivision Paramins v. The Lubrizol Corporation and Lubrizol of Canada Limited (supra) ruled: "It was manifestly dictated by the words of the patent itself." Based on the finding that the average number of Mn of the PIB used in Paramin"s dispersant was at least 2500, it follows that the succination ratio of all of the defendant"s dispersants exceeded 1.3 and fall precisely into the parameters of the claim specified in the plaintiff"s patent. On a purposive construction of the patent, it is my belief that strict compliance with the words "at least 1.3" were not essential to the invention. I have concluded that unreacted PIB is to be excluded in calculating the succination ratio. It is therefore my understanding of the situation, based on the evidence presented, that all of the materials the defendants manufactured were at least 1.3 succination, or SA groups for each equivalent weight of the substituent or PIB group, once all the relevant factors were taken into consideration. |
[37] I will return to Dr. Billmeyer"s evidence and, more particularly, to his oral evidence when discussing Imperial"s allegations of fraud. However, I will now continue this discussion by reference to the Ohio III proceedings. Following Cullen J."s judgment rendered on September 17, 1990, Lubrizol commenced the Ohio III proceedings in the United States District Court of Ohio. These proceedings were concerned with allegations of infringement regarding the U.S. patent and allegations regarding a breach of the final order on consent made on October 30, 1988.
[38] Early on in the Ohio III proceedings, Lubrizol sought a preliminary injunction with respect to the sale by Exxon in the U.S. of ECA 12819 and, in opposition thereto, Exxon filed a Memorandum dated November 15, 1990. The Memorandum, filed two months after Cullen J."s decision, refers to the Canadian proceedings and particularly to the evidence of Dr. O"Driscoll and that of Dr. Billmeyer. At footnote 18, found at page 12 of the Memorandum, counsel for Exxon make the following statement:
| Dr. O"Driscoll"s high results are not surprising given that he failed to make a fundamental measurement. The GPC instrument is not as good at detecting the presence of low molecular weight molecules as high ones, and unless a basic correction for this phenomenon is made, GPC results can be 15 to 25 percent too high... Dr. O"Driscoll failed to make any measurements to take this into account and made no such correction... Nor is it surprising that Dr. Billmeyer found the average of the Mn of his samples of Parapol 2200 to be 2457, given that the product is not tested for Mn routinely but that Exxon"s analysis in 1988 showed that the Mn had risen to about 2400. |
[39] The information appearing in footnote 18 of the Exxon Memorandum, is corroborated by the sworn statement of Mr. Gregory C. Giffin, the Technical Manager for Exxon Chemical Company at its Bayway, New Jersey, manufacturing plant. In his affidavit, sworn November 15, 1990, Mr. Giffin makes the following assertion at para. 6, regarding the Mn of Parapol 2200:
| We did not then, and never have since, established [sic] any molecular weight specification for Parapol 2200. Early tests of this product led us to conclude that, under operating conditions then prevailing at the Baytown PIB reactor, the number average molecular weight ("Mn") of the product was around 2200. This was the origin of the name. But because molecular weight was not (and is not) a specification, we did not and do not test Parapol 2200 for number average molecular weight as part of the manufacturing process. (Some time after Parapol 2200 went into regular production, and certainly by 1988, it became apparent that the typical Mn of the product had risen to about 2400 as a result of changes in the operating conditions of the reactor, not a deliberate decision.) |
[40] In its Memorandum, Exxon discusses, at pages 10 to 13, the Canadian trial and Cullen J."s judgment. At pages 11 and 12 thereof, we find the following comments regarding the evidence of Drs. O"Driscoll and Billmeyer:
| Second, Lubrizol"s experts asserted that the "true" Mn of the PIB was substantially higher than 2225 " indeed, that it was above 2454, the minimum number that under the product specifications would give Imperial"s ECA 12819 a succination ratio of 1.3 or more. These experts testified that they tested Parapol 2200, which, according to Lubrizol, was the same PIB as ECA 10788. (Howard Decl., " 22). Both Lubrizol experts also tested two samples of what was purported to be ECA 10788 acquired in Europe (Howard Decl., " 23), although there was no evidence that any PIB manufactured in Europe had ever been used to manufacture ECA 12819 in Canada. (Howard Decl., " 23). |
| Lubrizol"s two experts obtained substantially different results on the same samples, having used different methods to calibrate their GPC instruments. (Howard Decl., "" 22-24). Dr. Billmeyer, using a method generally akin to the patent, found an average Mn of 2457 on the Parapol 2200 he tested. (Howard Decl., " 22). (This was the same Dr. Billmeyer whose expert report in Ohio I attached the AI of ECA 12819 but made no criticism of the PIB Mn.) Dr. O"Driscoll"s results for the same material, based on a different calibration technique called the universal calibration curve " a method that he admitted was not specified in the patent " were about 10% higher than those of Dr. Billmeyer; he found an average Mn of 2709. (Howard Decl., " 22-24). As to the European PIB, Dr. Billmeyer found that these samples had an average Mn of 2288 (Howard Decl., " 23) " a number which, besides being close to the target stated in the specification sheet, produces a succination ratio of only 1.20. Using his different calibration method, Dr. O"Driscoll"s average Mn result for the same two samples was 2500, again about 10% higher that those of Dr. Billmeyer. (Howard Decl., " 23). As noted, Lubrizol contended that Parapol 2200 and ECA 10788 were the same PIB, so it argued that its experts" findings for the Mn of Parapol 2200 were really the Mn values for ECA 10788 as well. (Howard Decl., " 22). |
[41] Then, at p. 29 of its Memorandum, Exxon makes the following submissions regarding the weight given by Cullen J. to the evidence of Dr. O"Driscoll and Dr. Billmeyer:
| Of Lubrizol"s two experts in Canada, only Dr. Billmeyer calibrated his GPC using generally the patent"s method13; the other, Dr. O"Driscoll, used a different approach called the universal calibration method. Both Dr. O"Driscoll and Dr. Billmeyer admitted in testimony in Canada that Dr. O"Driscoll"s calibration method was not the method of the patent. (Howard Decl., " 23). Under U.S. patent law, therefore, a U.S. court presented with the question of the Mn of the starting PIB would err as a matter of law if it credited Dr. O"Driscoll"s Mn results. |
| This is, however, exactly what the Canadian court did. In its discussion of the Mn of the starting PIB, the Canadian court repeatedly cited the results of Dr. O"Driscoll"s GPC testing; Dr. O"Driscoll is mentioned by name no fewer than seven times. (Reasons at 34-38). Furthermore, the court"s critical finding " that the Mn of the starting PIB was at least 2500 " was clearly based on Dr. O"Driscoll"s testimony. This conclusion could not have been based on Dr. Billmeyer"s GPC analysis, because he found only an average Mn of 2457 for Parapol 2200 and an average Mn of 2288 for the purported ECA 10788 made in Europe. Instead, the Canadian court must have relied on the GPC analysis of Dr. O"Driscoll, who found an average Mn for Parapol 2200 of 2709, and an average Mn for European ECA 10788 of 2500. |
| The Canadian court thus concluded, sub silentio, that it was proper under Canadian law to measure Mn by a GPC method other than that specified by the "435 patent. Under U.S. patent law, that was clear error; only Mn determined using a GPC calibrated by fractionated polymer may be credited. |
[42] In the Ohio III proceedings, Dr. Billmeyer filed two expert reports, one in July 1994 and the other in March 1995. Neither report dealt with his evidence in Canada. However, Dr. Billmeyer had signed affidavits in support of a motion for summary judgment filed by Lubrizol early on in the proceedings, and in those affidavits, he referred to his 1990 Canadian evidence.
[43] Dr. Billmeyer was deposed by Exxon counsel primarily in connection with the substance of his two expert reports. He was deposed on July 11 and 12, 1994, and again on March 29, 30 and 31, 1995. During the course of his deposition, Dr. Billmeyer was questioned in regard to the evidence he gave in Canada in May 1990. The present Rule 1733 proceedings stem from a number of answers given by Dr. Billmeyer in response to questions regarding his Canadian evidence. In discussing the Imperial allegations of fraud, I will be returning to the answers given by Dr. Billmeyer during the course of his deposition in the United States.
[44] Dr. Billmeyer also gave evidence in the present proceedings, by way of commission evidence taken from December 8 to December 18, 1997. I will also return to this evidence when discussing Imperial"s allegations of fraud.
[45] During the course of the trial before me, two expert witnesses were called to testify with respect to Dr. Billmeyer"s evidence in 1990. Lubrizol called Dr. Alfred Rudin, a retired "distinguished professor of chemistry" at the University of Waterloo. Dr. Rudin retired in 1992 and continued to supervise graduate research until 1997. Dr. Rudin is an expert in the field of polymer chemistry and polymer characterization. Dr. Rudin"s affidavit, dated April 7, 1998, was marked as Exh. D-60. The opinion put foward by Dr. Rudin, for consideration by this court, is the following:
| 1. The method used by Dr. Billmeyer to determine the Mn of the starting PIB was scientifically sound and Dr. Billmeyer correctly related his method and work to the court. |
| 2. The VPO calibration factors used by Dr. Billmeyer were justified and Dr. Billmeyer"s April 1990 comparisons supported his calibration curve. |
[46] Imperial called as an expert witness John Henderson Knox, emeritus professor and honorary fellow of the University of Edinburgh. Professor Knox is an expert in the theory of liquid chromatography, of which, he stated, GPC was a part. Professor Knox filed two affidavits. The first one is dated December 30, 1996, and the second one, entitled "Supplementary Affidavit", was sworn to on January 8, 1998. Professor Knox"s supplementary affidavit was marked as Exhibit P-40. Professor Knox"s opinion is that in many respects, Dr. Billmeyer"s evidence in May 1990 was either intentionally misleading or reckless as to its truth or falsity.
[47] I should also point out that Professor Rudin swore to a rebuttal affidavit on April 7, 1998. This affidavit was filed as a response to Professor Knox"s supplementary affidavit, dated January 8, 1998, and to the affidavit of Dr. Meyer Ezrin, dated January 7, 1998. I need not refer any further to Dr. Ezrin, as he was not called by Imperial to testify. Professor Knox also signed a rebuttal affidavit, but he was not called to testify in regard thereto.
[48] Before turning to Imperial"s specific allegations of fraud against Dr. Billmeyer and Lubrizol, I must say a few words regarding the applicable law.
The Relevant Principles of Law
[49] Rule 1733 is an exception to the general rule that there is to be finality to litigation. This rule is embodied in the well-known principle of res judicata. Parties involved in litigation must bring all issues to the fore; they cannot simply "pick and choose" the issues which suit them. If an issue could have been raised by a party, then that party cannot, in subsequent litigation, except under special circumstances, raise that issue. In Maynard v. Maynard , [1951] S.C.R. 346 at pp. 356-59, the Supreme Court of Canada made its own the following passage from Hoystead v. Commission of Taxation, [1926] A.C. 155 at 170:
| Parties are not permitted to begin fresh litigations because of new views they may entertain of the law of the case, or new versions which they present as to what should be a proper apprehension by the Court of the legal result either of the construction of the documents or the weight of certain circumstances. |
| If this were permitted litigation would have no end, except when legal ingenuity is exhausted. It is a principle of law that this cannot be permitted, and there is abundant authority reiterating that principle. |
[50] On the same point, the words of Osborne J. in International Corona Resources Ltd. v. LAC Minerals Ltd. (1989), 66 O.R. (2d) 610 (H.C.), at p. 624, are apposite:
| It is important to recognize that trials are conducted within the adversary system: what witnesses are called and what questions they are asked is a matter left to counsel. This does not excuse fraud, perjury or deliberate non-disclosure. However, it must be borne in mind that counsel make decisions, both strategic and tactical, as to what approach will be taken to a witness who is being cross-examined. The approach taken by LAC"s counsel at trial, in the circumstances, led to the issues which are now before me being left twisting in the wind. That was a decision of trial counsel and a decision for which trial counsel should not now be criticized. Nevertheless, having chosen not to cross-examine McKinnon on matters pertaining to the West Newman claims, LAC cannot now complain about the absence of evidence at trial on those matters. |
[51] There also cannot be any doubt, on the authorities, that a prior judgment of the court will only be set aside in the clearest of cases. In Saywack v. Canada (M.E.I.), [1986] 3 F.C. 189 (C.A.), Stone J. for the Federal Court of Appeal, made the following remarks at pp. 197-98:
| The Rule [Rule 1733] must be seen as exceptional. It purports to permit relief in an action or proceeding subsequent to its disposition by solemn pronouncement of the Court even though that relief would be at variance or even wholly contrary to that pronouncement. Yet, if it covers an application the Court may grant relief. Obviously, a case would have to be a clear one before the Court will be induced to act under the Rule. Otherwise, the finality of judgments would be imperilled and that would be bad. |
[52] It goes without saying that this Court, like all courts, will tread lightly when faced with a Rule 1733 motion. Great care will be exercised in deciding whether a prior judgment should be set aside for fraud. In MacDonald v. Pier, [1923] S.C.R. 107, Duff J. for the Supreme Court of Canada, at pp. 120-21, made the following remarks:
| [...] The principle [i.e. that there is jurisdiction in the court to entertain an action to set aside a judgment obtained by reason of perjury] I conceive to be this; such jurisdiction exists but in the exercise of it the court will not permit its process to be made use of and will exert the utmost care and caution to prevent its process being used for the purpose of obtaining a re-trial of an issue already determined, of an issue which transivit in rem judicatam under the guise of impugning a judgment as procured by fraud. Therefore the perjury must be in a material matter and therefore it must be established by evidence not known to the parties at the time of the former trial. |
[53] For a party to succeed on a Rule 1733 motion, the following elements must be established to the satisfaction of the court:
| 1. That a false representation has in fact been made; |
| 2. that the false representation was made either (i) knowingly, without an honest belief in its truth, or (ii) recklessly, careless of whether it be true or false. |
[ See Nesbitt Thompson & Co. v. Pigott, [1941] S.C.R. 520 per Taschereau J. at 530; Vale et al. v. Sun Life Assurance Company of Canada (1998), 39 O.R. 444 (General Division) at 453, var"d [1999] O.J. 1507 (C.A.); and Derry v. Peek (1889), 14 App. Cas. 337 (H.L.), per Herschell L.J. at 374.]
[54] In McLaughlin v. Colvin, [1941] 4 D.L.R. 568 (Ont. C.A.), aff"d [1942] 3 D.L.R. 292 (S.C.C.), the Ontario Court of Appeal, at p. 583, explained recklessness in giving evidence as follows:
| [...] I cannot formulate the principle more succinctly or in better terms than as it appears in 23 Hals. (2nd ed.), pp. 45-6, para. 63: |
| It is well established that a misrepresentation which was founded on a belief in its truth, if that belief really existed, and was genuinely and honestly entertained, is not deprived of its character of innocence by reason of the mere fact that the belief resulted from want of care, skill, or competence, or lapse of memory, though such conduct, in other aspects, may have been of a most culpable character. Negligence is not dishonesty; ... Carelessness or stupidity in arriving at a genuine conviction ought not to be, though it frequently has been, confounded with that moral recklessness or callousness, already referred to, which prompts the putting forward of a misrepresentation as to which the representor has no belief at all. |
[55] With respect to the pleadings put forward by the party which seeks to set aside a prior judgment of the court, the allegations of fraud must be set out in clear and unequivocal terms. In Jonesco v. Beard, [1930] A.C. 298 (H.L.), Buckmaster L.J., at p. 300, put the matter as follows:
| It has long been the settled practice of the Court that the proper method of impeaching a completed judgment on the ground of fraud is by action in which, as in any other action based on fraud, the particulars of the fraud must be exactly given and the allegations established by the strict proof such a charge requires. |
Also, in Wilde v. Gibson, [1848], 73 H.L. 191, Cottenham L.C., at 205, said:
| The result appears to be to be, first, that the plaintiff, having rested his case in the bill upon imputations of direct personal misrepresentation and fraud, cannot be permitted to support it upon any other ground; secondly, that the evidence, at the most, proves only constructive notice of the fact upon the non-communication of which the plaintiff founds his claim for setting aside his completed purchase; and that nothing short of positive knowledge can be sufficient for that purpose. The case alleged is not proved,, and the case proved is not alleged; and if it has been, would not have been sufficient to support the decree. |
[56] More recently, Dubé J. of this Court, in Commercial Union Assurance Co. plc et al. v. M.T. Co. and Wood (1996), 107 F.T.R. 291 at p. 294, reiterated the principle as follows:
| [...] Fraud is a conclusion of law and may not be inferred. As referred to by Wetston, J., of this court in Smith-Kline Beecham Animal Health Inc. et al. v. Interpharm Inc. et al. (1993), 67 F.T.R. 146; 52 C.P.R. (3d) 400 (T.D.), the requirements of pleading fraud are commented on in Williston and Rolls, The Law of Civil Procedure (Toronto: Butterworths, 1970), vol. 2, at p. 659: |
| Fraud is a conclusion of law; it is wholly immaterial and insufficient to allege that an instrument has been obtained by fraud unless the things done constituting the fraud are stated on the face of the statement of claim. An allegation of fraud will not be inferred but must be stated expressly. |
[57] The most recent statement of the applicable test when an unsuccessful litigant seeks to set aside a judgment on the basis of fraud is that of Osborne J. in International Corona Resources Ltd., supra. At pp. 620-23, Osborne J. sets out those principles which, in his view, are relevant in deciding whether a decision ought to be set aside:
| The judgment of Estey J.A. in 100 Main Street East Ltd. v. Sakas (1975), 8 O.R. (2d) 385, 58 D.L.R. (3d) 161 (C.A.), seems to me to be an appropriate point of commencement, as most regard 100 Main Street East Ltd. as the leading authority, or at least one of the leading authorities, in the test to be applied when an unsuccessful litigant seeks to set aside a trial judgment on the basis of fraud. |
| In 100 Main Street East, the defendant Sakas, and his partner Leaukus (who was not a defendant), misled the court on an issue which, by anyone"s definition of the term, was material. The defendant Sakas, the vendor, had refused to close on a sale of his half-interest on a parcel of land. Sakas took the position at trial that he could not close because his partner (Leaukus (who held title to the property in trust for the two-man partnership) would not provide a deed for the purchaser. That fact gave rise to the trial issue of why Leaukus would not provide the required deed. The defendant Sakas took the position that he did all he could to acquire the deed from his partner Leaukus. Sakas neglected to tell the court that after signing the agreement of purchase and sale on which the plaintiff sued, and before trial, Sakas had made a deal with his partner Leaukus to sell the land to Leaukus for more than Sakas had agreed to sell it to the plaintiff. Although the evidence is unclear, it would appear that Sakas and Leaukus took the position in giving their evidence that the deed required was not given because Leaukus did not want to end up as a partner of one Frisina, the principal of 100 Main Street East. |
| Estey J.A. concluded that the failure to disclose the Sakas/Leaukus agreement was improper and deliberate, and not inadvertent. That having been done, Estey J.A. stated the test for setting aside a trial judgment on the ground of fraud. At p. 389 O.R., p. 165 D.L.R., after referring to Re Cook and Morley and Forster, [1950] O.W.N. 739 (H.C.J.); Hip Foong Hong v. H. Neotia & Co., [1918] A.C. 888, and Johnston v. Barkley (1905), 10 O.L.R. 724 (Div. Ct.), Estey J.A. said: |
| These authorities make it clear that the new evidence of fraud must relate to the "foundation" of the decision or be "material" to the claim or defence, but need not necessarily amount to a "determining factor in the result". |
| Estey J.A. then went on to refer to the dissenting judgment of Duff J. in MacDonald V. Pier, supra, where Duff J. said [at pp. 680-1 D.L.R., pp. 120-1 S.C.R.]: |
| ... there is jurisdiction in the court to entertain an action to set aside a judgment on the ground that it has been obtained through perjury. The principle I conceive to be this; such jurisdiction exists but in the exercise of it the court will not permit its process to be made use of and will exert the utmost care and caution to prevent its process being used for the purpose of obtaining a re-trial of an issue already determined, of an issue which transivit in rem judicatam under the guise of impugning a judgment as procured by fraud. Therefore the perjury must be in a material matter and therefore it must be established by evidence not known to the parties at the time of the former trial. |
| It seems to me that the word "material" means that which goes to the foundation of the decision or which goes to the crux of a central issue before the court. "Material" obviously means something different from and, in this context, more than "relevant"; it means more than evidence which would be an appropriate topic for cross-examination on credibility, but collateral in the sense of being divorced from any of the main issues in the case in a trial. |
| I think it is important that in 100 Main Street East, supra, Estey J.A. specifically recognized that circumstances will vary. Estey J.A. said, at p. 390 O.R., p. 166 D.L.R.: "Each judgment must of course be viewed in the light of the facts in issue to which the judgment has responded." |
| Estey J.A. concluded that Meek v. Fleming, [1961] 3 All. E.R. 148, a case involving grossly misleading evidence of the defendant, did not add a requirement that the fraud or perjury alleged probably tipped the scales or may reasonably have tipped the scales in favour of the successful trial litigant. |
| In considering the appropriate test to be applied in a motion to set aside a trial judgment in the face of new evidence disclosing fraud, I start by briefly considering what I view to be the underlying principles. |
| The court will not allow its process to be abused by permitting a successful litigant to benefit from his or her own misconduct. |
| Balanced against the court"s right and duty to remedy abuses in appropriate circumstances is the public and private interest in there being certainty and finality in litigation. Each case must be assessed on its own merits, particularly having in mind the nature of the fraud involved, its perpetrator, the evidence in each case, and the judgment sought to be set aside. |
| The cases suggest that there is a distinction between the fraud of a party and the fraud of a witness. 100 Main Street East, Hip Foong Hong, Meek v. Fleming, MacDonald v. Pier, Re Cook and Morley, Johnston v. Barkley, Industrial Development Bank v. Bertolo, all, supra, and Wentworth v. Rogers (No. 5) (1986), 6 N.S.W.L.R. 534 (C.A.), are all cases involving fraud on the part of a successful party or fraud on the part of a witness participated in by the successful party. |
| Taking into account what I consider to be the essential underlying principles and the cases referred to by counsel, I think that the following propositions are evident: |
| (1) The fraud alleged must be proved on a reasonable balance of probability. The more serious the fraud alleged, the more cogent the evidence going to establish it will have be to meet the civil onus of proof. The reasonable balance of probability is not an inflexible standard of proof. |
| (2) The proved fraud must be material, that is, it must go to the foundation of the case. |
| (3) The evidence of fraud must not have been known at the time of trial to the party seeking to rely upon it on a motion to set aside a trial judgment. |
| (4) The unsuccessful trial party is exposed to a test of due or reasonable diligence. This is clear from cases such as, MacDonald v. Pier, supra; Johnston v. Barkley, supra, and Industrial Development Bank v. Bertolo, supra. In my view, the onus is on the moving party to establish due diligence. Evidence cannot be stockpiled during the litigation to be taken from inventory after an unsuccessful trial or appeal: see Becker Milk Co. Ltd. v. Consumers" Gas Co. (1974), 2 O.R. (2d) 554 at p. 558, 43 D.L.R. (3d) 498 at p. 502 (C.A.). |
| (5) If the fraud alleged is that of a non-party, and if the successful party at trial is not connected with the fraud alleged, the tests to which I have referred must be more stringent than for the fraud of a party. It is not, however, necessary for me to set out the added burden to be placed upon a moving party seeking a new trial in the face of the fraud of a non-party, as I have concluded that LAC has not established a right to success when its case is measured against the standards imposed in cases involving fraud of a party. |
| (6) The test imposed upon the unsuccessful trial party to obtain relevant evidence " that is, evidence going to establish fraud " with due diligence, is objective. The questions to be asked are what did the moving party know, and what ought the moving party to have known? |
| (7) Delay will defeat a motion to set aside a trial judgment under rule 59.06. I refer in this regard to cases where the evidentiary burden has been met and the due diligence test passed, but where there is unreasonable delay in bringing or pursuing the motion to set aside. Johnston v. Barkley, supra, is ample authority for the proposition above referred to. |
| (8) Relief under rule 59.06 is discretionary14. The conduct of the moving party is relevant. |
| (9) At the end of the day, the central question to be answered is as stated in Wentworth v. Rogers (No. 5), supra, at p. 539: |
| ... it must be shown, by the party asserting that a judgment was procured by fraud, that there has been a new discovery of something material, in the sense that fresh facts have been found which, by themselves or in combination with previously known facts, would provide a reason for setting aside the judgment. |
[58] Osborne J."s second proposition is that the fraud must be material, i.e. it must go to the foundation of the case. In 100 Main Street East Ltd. v. Sakas (1975), 8 O.R. (2d) 385 (Ont. C.A.), Estey J.A. (as he then was) discusses the materiality of the impugned testimony in the following terms, at p. 389-90:
| The important distinction between the consequences in law of the discovery of new evidence generally, and new evidence or circumstances revealing fraud, is the subject of an observation by Lord Buckmaster in Hip Foong Hong v. H. Neotia & Co., [1918] A.C. 888 at p. 894: |
| ... the applicant must go further and show that the [new] evidence was of such a character that it would, so far as can be foreseen, have formed a determining factor in the result. Such considerations do not apply to questions of surprise, and still less to questions of fraud. A judgment that is tainted and affected by fraudulent conduct is tainted throughout, and the whole must fail; |
| These authorities make it clear that the new evidence of fraud must relate to the "foundation" of the decision or be "material" to the claim or defence, but need not necessarily amount to a "determining factor in the result". Duff, J., in a dissenting opinion in MacDonald v. Pier , [1923] S.C.R. 107 at pp. 120-1, [1923] 1 D.L.R. 670 at pp. 680-1, [1923] 1 W.W.R. 376, in discussing this principle, stated: |
| ... there is jurisdiction in the court to entertain an action to set aside a judgment on the ground that it has been obtained through perjury. The principle I conceive to be this: such jurisdiction exists but in the exercise of it the court will not permit its process to be made use of and will exert the utmost care and caution to prevent its process being used for the purpose of obtaining a re-trial of an issue already determined, of an issue transivit in rem judicatam under the guise of impugning a judgment as procured by fraud. Therefore the perjury must be in a material matter and therefore it must be established by evidence not known to the parties at the time of the former trial. |
| The process appears to be one of balancing the need for the Courts to protect the integrity of their process against fraudulent practices before the Court, on the one hand, against the practical requirement of certainty and finality in litigation on the other. The latter principle is expressed by Lord Justice James in Flower v. Lloyd (1978), 10 Ch.D. 327 at pp. 333-4: |
| Where is litigation to end if a judgment obtained in an action fought out adversely between two litigants sui juris and at arm"s length could be set aside by a fresh action on the ground that perjury has been committed in the first action, or that false answers had been given to interrogatories, or a misleading production of documents, or of a machine, or of a process had been given? There are hundreds of actions tried every year in which the evidence is irreconcilably conflicting, and must be on one side or other wilfully and corruptly perjured. In this case, if the Plaintiffs had sustained in this appeal the judgment in their favour, the present Defendants, in their turn, might bring a fresh action to set that judgment aside on the ground of perjury of the principal witness and subornation of perjury; and so the parties might go on alternately ad infinitum . There is no distinction in principle between the old Common Law action and the old Chancery suit, and the Court ought to pause long before it establishes a precedent which would or might make in numberless cases judgments supposed to be final only the commencement of a new series of actions. Perjuries, falsehoods, frauds, when detected, must be punished and punished severely; but, in their desire to prevent parties litigant from obtaining any benefit from such foul means, the Court must not forget the evils which may arise from opening such new sources of litigation, amongst such evils not the least of which being that it would be certain to multiply indefinitely the mass of those very perjuries, and falsehoods, and frauds. |
[59] Later on, at p. 395, Estey J.A. completes his remarks on materiality by stating:
| What a thorough cross-examination of these witnesses by counsel armed with knowledge of the 1969 agreement might reveal is not for us to speculate. It may be that that important process will reveal the date of the 1969 agreement to be different from that asserted on the document itself. It may be that substance can be given to what now must be speculation as to the purpose lying behind the course of conduct by the actors in these events. In any case, a reviewing tribunal should not construe these events in favour of persons who have intended to mislead the Court or at the very least, their adversary. By not disclosing the 1969 agreement, the appellant may have been protecting the valuation evidence tendered on his behalf to the effect that the value of the land had not increased between the date of the 1968 agreement and the trial. By their explanation of the appellant"s failure to complete the sale to the respondent Sakas and Leaukus may have intended to keep open the way to completion of their private sale under the 1969 agreement. We need not, and should not, speculate as to all the possibilities of intent on the part of the parties to the undisclosed agreement. In the circumstances, the principle of law preserving the sanctity of the principle of res judicata must give way to the principle recognizing the right and duty of the Court to protect its own processes so as to ensure that litigants do not profit from their improper conduct. This conclusion flows inexorably from the evidence and is reached reluctantly in view of the serious difficulties facing the respondent in seeking to achieve a different outcome on a second trial. |
| In my view, it is not necessary to determine at this stage of the process that the disclosure of this additional evidence will affect the result or "may reasonably have affected the result", as Holland, J., put it. It is only sufficient to conclude that the evidence withheld relates to the material issues arising from the appellant"s failure to perform the 1968 contract and the damages, if any, flowing therefrom. In all the circumstances, the failure of the appellant and his partner to disclose the 1969 agreement must, in my view, be considered to be improper and deliberate and not the result of mere inadvertance. The result reached by Holland, J., is therefore correct and the appeal should be dismissed with costs. |
[60] With these principles in mind, I now turn to Imperial"s allegations of fraud against Dr. Billmeyer and Lubrizol.
The Allegations of Fraud
[61] I begin with the allegations of fraud made against Lubrizol. These allegations appear in paragraphs 17 to 20 of Imperial"s Statement of Claim. Simply put, these allegations are to the effect that Lubrizol, through its outside counsel, John Fairweather, instructed Dr. Billmeyer to set up a Waters 200 GPC in order to obtain results which had been pre-determined by Lubrizol from its high speed GPC. The allegations are also to the effect that sometime between January and May 1990, Lubrizol offered and Dr. Billmeyer accepted a mandate to set up the Waters 200 GPC to match Lubrizol"s high speed GPC test results and to testify falsely in the ways particularized by Imperial in paragraphs 17 to 20 of its Statement of Claim.
[62] In making its allegations, Imperial has not only "accused" Dr. Billmeyer of testifying falsely, but has also "accused" Mr. John Fairweather, then outside counsel to Lubrizol, and Mr. Joe Bauer, general counsel for Lubrizol, of having "mandated" Dr. Billmeyer to conduct tests that were not scientific and to give false testimony to this Court. In my view, there is not a shred of evidence to support Imperial"s allegations of fraud against Lubrizol, nor is there any shred of evidence in regard to the allegations made against Mr. Fairweather and Mr. Bauer. Both impressed me as being extremely honest and credible witnesses. A careful review of the totality of the evidence, including the notes taken at the relevant time by Dr. Billmeyer, John Fairweather and Joe Bauer, clearly demonstrates that no mandate was ever given by Lubrizol to Dr. Billmeyer pursuant to which he would falsely testify in regard to his test results, his calibration curve, both preliminary and final, the Mn of the starting PIBs, the accuracy of the results obtained by Dr. O"Driscoll, the accuracy of Exxon"s VPO test method, AMS 320.06, and in regard to the Mn and Mw/Mn characteristics of Parapol 2200, Parapol 2225 and ECA 10788, and whether these PIBs were distinguishable or not. I can state without hesitation that if Imperial had any evidence to support its allegations, that evidence never made it to the courtroom.
[63] It is interesting to note that in part 6 of its Closing Arguments, entitled "Lubrizol Involvement in Connection to Dr. Billmeyer"s Evidence", Imperial makes submissions that are directed to Lubrizol"s and Dr. Billmeyer"s failure to comply with guidelines concerning the testimony of experts. Whether Dr. Billmeyer and Lubrizol have failed in the respects alleged by Imperial, I need not decide. However, it is clear from Imperial"s submissions that it has not made any submissions in support of the allegations appearing at paragraphs 17 to 20 of its Statement of Claim.
[64] I now turn to Imperial"s allegations of fraud against Dr. Billmeyer. I begin with ground No. 6, pursuant to which Imperial alleges that Dr. Billmeyer falsely testified that the European samples (Parapol 2225 and ECA 10788) could not be distinguished from the Parapol 2200 samples. Imperial"s allegations in support of ground No. 6 appear at paragraphs 47 to 50 of its Statement of Claim:
| 47. In his evidence-in-chief, Dr. Billmeyer told this Court that he was not able to distinguish the two European samples (tested after his expert evidence had been delivered) from the Parapol 2200 samples as listed in Exhibit B7A to his affidavit. |
| 48. Accordingly, Dr. Billmeyer stated that the conclusions in paragraph 54 of his affidavit were unaffected and that, in his opinion, the true Mn of the PIB used by Imperial Oil was approximately 2500. |
| 49. When cross-examined with respect to the conclusion, Dr. Billmeyer referred to Mn and the ratio of Mw to Mn and testified that the ratio for the two European samples was 2.5 to 2.7, and consequently fell within the range of Parapol 2200 samples, which he testified was 2.2 to 3.7. |
| 50. In his subsequent deposition testimony in Ohio, Dr. Billmeyer conceded that the Mw/Mn range for the Parapol 2200 samples was in fact 2.9 to 3.7, a range that excluded the two European samples, namely ECA 10788 and Parapol 2225. |
[65] It goes without saying that both Lubrizol and Dr. Billmeyer deny these allegations. At paragraphs 34 to 38 of its Statement of Defence, Lubrizol states:
| 34. Dr. Billmeyer initially measured the Mn and Mw of 16 samples of Exxon PIB identified as Parapol 2200 (as well as samples of an Exxon PIB identified as Parapol 1300). The results of testing these samples were exhibited to his expert affidavit. He subsequently tested samples of two other Exxon PIBs, one called Parapol 2225 and one called ECA 10788, the "European Samples". He reported the results of the tests of these products at the trial. |
| 35. Dr. Billmeyer testified at the Canadian trial that he could not distinguish the European samples from the Parapol 2200 samples on the basis of his measurement of their respective Mn values. That testimony was honestly given and was correct. |
| 36. On cross-examination, he was asked about the Mw/Mn values. Having given the Mn values in his evidence in chief, he read the Mw values off the GPC charts which were in evidence. His statement of his recollection of the Mw/Mn ratio range was qualified by his statement that he would want to check the mathematics and, in any event his testimony was honestly given. |
| 37. Counsel for Imperial stated during his cross-examination that Imperial"s advisors had calculated the average Mw and accordingly they already knew the individual values. If Imperial considered the issue to be material, it could have calculated the Mn/Mw values and put these to Dr. Billmeyer, or put them forward through its own witnesses. Counsel for Imperial in referring to the calculation of Mw/Mn ratios and comparison of them with the Exxon PIB specifications stated: "We can do the exercise ourselves". Imperial elected not to pursue this issue. |
| 38. In any event, as referred above, Imperial admitted that Parapol 2200, Parapol 2225 and ECA 10788 were the same product. |
As to Dr. Billmeyer, he states the following at paragraphs 23 to 26 of his Response:
| 23. The Intervenor initially measured the Mn and Mw of 16 samples of Exxon PIB identified as Parapol 2200 (as well as samples of an Exxon PIB identified as Parapol 1300). The results of testing these samples were exhibited to his expert affidavit. He subsequently tested samples of two other Exxon PIBs, one called Parapol 2225 and one called ECA 10788, the "European Samples". He reported the results of the tests of these products at the trial. |
| 24. The Intervenor testified at the Canadian trial that he could not distinguish the European samples from the Parapol 2200 samples on the basis of his measurement of their respective Mn values. That testimony was honestly given and was correct. |
| 25. On cross-examination, the Intervenor was asked about the Mw/Mn values. Having given the Mn values in his evidence in chief, he read the Mw values off the GPC charts that were in evidence. His statement of his recollection of the Mw/Mn ratio range was qualified by his statement that he would want to check the mathematics and, in any event his testimony was honestly given. |
| 26. Counsel for Imperial stated during his cross-examination that Imperial"s advisors had calculated the average Mw and accordingly they already knew the individual values. |
[66] The debate herein stems from an answer given by Dr. Billmeyer during his cross-examination before Cullen J. that the Mw/Mn ratio of the Parapol 2200 sample was 2.2 to 3.7 when, in fact, the true ratio was 2.9 to 3.7. The specific question and answer which led to this issue are as follows:
| Q. Well you notice that the Mn"s you said were not significantly different but the Mw"s were? |
| A. Well, if you look at the ratio Mw over Mn for 3435, which was identified to me as 10788, that turns out to be 2.5 and that is within the range of the Mw over Mn"s that I found for the samples that were identified to me as Parapol 2200's. And that range as I reported when I ... as an added remark when I gave my direct testimony was from 2.2 to 3.7. 2.5 falls within that range. For the other, identified to me as Parapol 2225, I obtained a 2.7 for the ratio weight average to number average and that also falls within the range 2.2 to 3.715 |
Thus, according to Imperial, since the Mw/Mn ratio of the European samples was outside the range of the Mw/Mn ratio of the Parapol 2200 samples, the European samples were clearly "distinguishable" from the Parapol 2200 samples.
[67] On the evidence, it is clear to me that Dr. Billmeyer made no attempt to mislead the Court. When he answered that the Mw/Mn ratio for the Parapol 2200 samples was 2.2 to 3.7, he simply made an error. I am of this view for the following reasons.
[68] Firstly, Dr. Billmeyer stated in unequivocal terms that he could not distinguish Parapol 2225 and ECA 10788 from Parapol 2200 on the basis of their Mn. During his cross-examination at trial in 1990, Dr. Billmeyer gave the following answers to counsel for Imperial:
| Q. When you were asked this morning by Mr. Wells about the samples that were added to your Affidavit this morning and which you identified in your Exhibit D7-A as ECA 10788, and Parapol 2225, I understand that you had just been told that these were 10788 and 2225? |
| Q. You said that based on these number average molecular weights that you reported there, that you could not distinguish between the two samples that you added this morning and the Parapol 2200? |
| A. By that, of course, I meant that the molecular weights I obtained fell within the range of molecular weights for other samples that I had been told were Parapol 2200's.16 |
[69] With respect to the Mw of the European samples, Dr. Billmeyer conceded, again in cross-examination, that the Mw thereof were lower than the Mw of the Parapol 2200 samples. This is what he said:
| Q. Then on B6-A I have got 5985 and 5992, significantly lower weight average molecular weights? |
| Q. And based on that information you can say, sir, they are not the same class of materials as the Parapol 2200? |
| A. All I can say is that they appear to have a significantly lower weight average molecular weights.17 |
[70] Why did Dr. Billmeyer make an error when he answered that the range of the Parapol 2200 samples was 2.2 to 3.7? In my view, that mistake arose because Dr. Billmeyer failed to remember, when giving the impugned answer in cross-examination, that his range of 2.2 to 3.7 applied not only to the Parapol 2200 samples, but also to the Parapol 1300 samples. Consequently, the value of 2.2 came from his testing of Parapol 1300 samples, while the high value of 3.7 came from Parapol 2200 samples. The other Parapol 1300 and Parapol 2200 samples had Mw/Mn ratios somewhere between 2.2 and 3.7. I therefore accept Lubrizol"s submission that Dr. Billmeyer "clearly misrecollected that it [the Mw/Mn range of 2.2 to 3.7] included the Parapol 1300's as well as the Parapol 2200's".
[71] It is worthwhile to point out that Professor Knox, who was called by Imperial as an expert witness, conceded during his cross-examination that Dr. Billmeyer had not falsely testified on this point. Professor Knox gave the following answer:
| Yes, well I think I have to agree with you that Dr. Billmeyer didn"t clearly say that they were indistinguishable. He said that they were indistinguishable on a certain basis of the Mn"s, so, presumably then, I have to agree that he did not make a serious misrepresentation personally.18 |
[72] I am also of the view, in any event, that Imperial was not misled by Dr. Billmeyer"s testimony. During his testimony before me, Mr. Deeth, counsel for Imperial during the 1990 trial, made it clear that his side had recalculated the Mw/Mn ratios prior to making their closing arguments to Cullen J. In my view, Imperial knew or had to know what the correct Mw/Mn ratios were. However, for reasons which are of no concern to me, Dr. Billmeyer was not challenged regarding his erroneous answer. It is also significant to note that Imperial admitted before Cullen J. that Parapol 2200, Parapol 2225 and ECA 10788 were the same product. I have already referred to Cullen J."s Reasons in that regard and in particular, I have referred to Cullen J."s reliance on the evidence of Mr. William Levy.
[73] For the sake of completion, I should refer to that part of Dr. Billmeyer"s U.S. deposition where are found the answers which have given rise to this specific ground of fraud:
| " At page 711 [Ex. P18 p. 1442] |
| Q. Doctor, that testimony wasn"t true, was it? |
| A. I don"t know why not. That was the way I remembered the situation, and the best of my recollection, it was stated correctly. |
| " At pages 712-713 [Ex. P18 p. 1442] |
| Q. Now, doctor, with that document in front of you, can you confirm to me that the samples designated Parapol 2200 had Mw over Mn ranges from 2.9 to 3.7, and I will represent to you, doctor, that we have checked those against the original Waters 200 reports in your Canadian affidavit and what you have listed there is your handwriting is accurate for each of those samples, although rounded to two digits as opposed to the three that are stated in the Waters 200 report. |
| So, doctor, I ask you again, isn"t it correct to say that your testimony from page 184 of May 16, 1990 in the Canadian trial was in fact false? |
| A. At this point in time, I don"t know what I based that testimony on, but I can"t argue with the numbers that you point out here except to say that there were several estimates of Mw over Mn available for these samples, one more is listed on this page, it is O"Driscoll GPC results, and those numbers are somewhat lower, although I don"t see them going down to, what was it, 2.2 that I had indicated as the bottom of the range. |
| I note also that in the series listed at the bottom, which is the Parapol 1300-type materials, that there are some as low as 2.2, there is one as low as 2.2.19 |
[74] It is clear from these questions and answers that Dr. Billmeyer never admitted that he had given false testimony, in the sense that he would have made a deliberate attempt to mislead Cullen J. when he testified in 1990. As I have already made it clear, it is my view that Dr. Billmeyer"s "incorrect" answer was simply that. He made a mistake. No more, no less. Consequently, Imperial has failed to convince me that Dr. Billmeyer falsely testified that the European samples could not be distinguished from the Parapol 2200 samples.
[75] I now turn to ground No. 5 of Imperial"s allegations of fraud, i.e. that Dr. Billmeyer"s testimony in Canada was false when he testified that he had followed the Exxon VPO test method "exactly" to determine that Exxon"s method seriously understated the Mn of the PIB. Imperial"s allegations, in respect of this ground of fraud, appear at paragraphs 42 and 46 of its Statement of Claim:
| 42. Dr. Billmeyer testified that AM-S 320.06 was an Exxon test method for determining the Mn of PIBs. Before this Court, Dr. Billmeyer said he had determined the Mn of PIB samples using AM-S 320.06 and had demonstrated, he said, that the Exxon understates the true values of Mn for PIBs as compared to his Waters 200 GPC tests. Dr. Billmeyer testified at trial here that he used AM-S 320.06 "exactly" with the exception that he substituted toluene for benzene as the solvent because benzene was no longer used in the United States. |
| 43. In fact, AM-S 320.06, as Dr. Billmeyer knew, required |
| (a) the use of benzene as a solvent, and Dr. Billmeyer knew that benzene was available and could be used; |
| (b) employment of the correct sample and calibrant concentrations; |
| (c) the preparation of a log plot; and |
| (d) the construction of a calibration graph. |
| 44. Dr. Billmeyer complied with none of these requirements of the AM-320.06 method, and he admitted in Ohio that he did not peform the AM-S 320.06 test "exactly" or at all. |
| 45. Accordingly, Dr. Billmeyer had no basis to testify that AM-S 320.06 understated the Mn of the PIB because he did not run the method. |
| 46. Moreover, had Dr. Billmeyer calibrated the Waters 200 GPC to match VPO results on the anchor samples " as he testified falsely to this Court that he did " Exxon"s test method AM-S 320.06 as actually run by Dr. Billmeyer, would have over stated (not understated the Mn of the PIB when compared to his Waters 200 GPC results. This would have had the impact of reversing the conclusion of this Court that the Exxon Certificates of Quality Assurance on the PIB supplied to Imperial Oil understated the Mn of the PIB " rather it would have concluded the Mn on the Certificates of Quality Assurance was over stated. |
[76] In its Statement of Defence, Lubrizol states at paragraph 33:
| 33. Dr. Billmeyer testified with respect to Exxon"s VPO test method honestly and in accordance with his knowledge and understanding at the time. Imperial has not alleged or shown any material difference between Dr. Billmeyer"s Canadian trial testimony and the facts as alleged by Imperial. |
In his Further Response, Dr. Billmeyer states at paragraph 22 thereof:
| 22. The Intervenor testified with respect to Exxon"s VPO test method honestly and in accordance with his knowledge and understanding at the time. |
[77] Before Cullen J., Dr. Billmeyer made the following assertion:
| As I have already explained, Your Lordship, I used the method of AMS 320.06 exactly, but with one exception: "... the solvent specified in AMS 320.06, benzene, can no longer be used in the United States because it has been found to be carcinogenic. Instead, I used toluene as a solvent at a temperature fo 50o C..."20 |
As I have already highlighted, the trial judge accepted Dr. Billmeyer"s evidence that Exxon"s VPO method understated the true Mn of the PIBs at issue.
[78] Simply put, Imperial submits that Dr. Billmeyer did not, in effect, use Exxon"s VPO method but that rather, the results he reported were produced by Lubrizol which used its version of ASTM D-2503 to analyse the Parapol 2200 samples. Imperial also asserts, at paragraph 348 of its Closing Arguments:
| 348. The calibration plot from Dr. Billmeyer"s so-called AMS 320.06 testing also revealed that Dr. Billmeyer"s plots were done according to Lubrizol"s modified version of ASTM D-2503. Indeed, the plots are labelled 2503. |
[79] Lubrizol points out that Dr. Billmeyer"s use of ASTM D-2503 was not a secret, in that Dr. Billmeyer, at paragraph 24 of his affidavit of April 1990, made it clear that Exxon"s VPO method AMS 320.06 was, in his view, a virtual copy of ASTM D-2503-67, which first came into use in 1966. As Exh. B-1 to his affidavit, Dr. Billmeyer attached a copy of ASTM D-2503-67.
[80] With respect to whether or not Dr. Billmeyer was correct in testifying that he had used "exactly" Exxon"s VPO method, I am satisfied that if he did not use it "exactly", he truly believed that he had. Consequently, Imperial"s allegation of fraud on this point cannot succeed.
[81] I agree entirely with the submission which Lubrizol makes at paragraph 531 of its Closing Brief, that looking at the evidence as a whole on this point, Dr. Billmeyer did not simply make a bald assertion that he had followed AMS 320.06 "exactly". Rather, he spelled out in detail the procedure which he used, "making it abundantly clear what he meant and stated this as an opinion".
[82] At paragraph 44 of its Statement of Claim, Imperial alleges that Dr. Billmeyer did not abide by the four requirements of AMS 320.06 and that he admitted that fact in the Ohio III proceedings.
[83] Firstly, in the Ohio III proceedings, Dr. Billmeyer gave the following answer to Exxon"s counsel regarding AMS 320.06:
| " Q. So, are you now telling me, doctor, that when you gave that testimony to the Canadian court, that it was false, that in fact AM-S 320.06 was not followed, with one exception, namely, the substitution of toluene or benzene as solvent? |
| A. No, I don"t believe that was false testimony. At the time I made that testimony, that was the truth as far as I was aware of it. Now we"re talking about statements that come three or five years later, and in that period of time, I have had the opportunity to look more carefully at 320.06, and I have concluded that there are inconsistencies in it which make it very difficult if not impossible for someone using a modern GPC instrument to follow it, quote, exactly, unquote."21 |
It appears clearly from Dr. Billmeyer"s answer that he did not believe that his Canadian evidence constituted false testimony. He indicated that when he testified in Canada, he believed that his answer was truthful. He went on to explain that in the light of further reading and additional knowledge concerning AMS 320.06, he could not then say that he followed the method "exactly".
[84]
With respect to the solvent he used, toluene, Dr. Billmeyer clearly stated at paragraph 49 of his affidavit, that he had used toluene and not benzene because benzene "can no longer be used in the United States because it has been found to be carcinogenic". That point is not now an issue before me.
[85] With respect to the employment of the correct sample and calibrant concentrations, Dr. Billmeyer did set out, in the second page of Exh. A to his affidavit, the concentrations of the samples he used. He further described the calibration steps and described the concentration of the calibrant benzyl used as .0847 g per 50 ml. of solution. AMS 320.06, in its description of calibration, refers specifically to a Mecrolab VPO instrument for steps 6.3 to 6.13. At paragraph 6.2, AMS 320.06 states that if an instrument other than a Mecrolab is used, then the manufacturer"s instructions are the ones that are to be followed. The concentrations used by Dr. Billmeyer were, no doubt, in accordance with the Wescan instrument manual.
[86] With respect to the preparation of a log plot, again, AMS 320.06 makes specific reference to a Mecrolab VPO instrument, regarding steps 6.3 to 6.13. Paragraph 6.2, as I have just stated, makes it clear that if another instrument is used, that instrument"s manufacturer"s instructions are the ones to be followed. Since Dr. Billmeyer used a Wescan instrument, he followed the procedures provided for in the Wescan manual and consequently, in my view, in accordance with paragraph 6.2 of AMS 320.06. It is also relevant to note that ASTM D-2503, which is virtually identical to AMS 320.06, makes no reference to a log plot. Lastly, Dr. Billmeyer described in clear terms to the Court the calibration factors obtained for each of the three calibrants he used.
[87] With respect to the construction of a calibration graph, Dr. Billmeyer used calibration graphs, namely a benzyl calibration graph, a tetracosane calibration graph and a squalane calibration graph (Exh. D-1, Tabs 40, 79 and 49).
[88] Lubrizol points out that Imperial"s VPO expert, Dr. Kennedy, was present in the courtroom when Dr. Billmeyer gave his evidence before Cullen J. Mr. Deeth, then counsel for Imperial, testified before me that he could not recall his experts advising him that Dr. Billmeyer had not followed "exactly" AMS 320.06 and that he could not recall being advised that the method used by Dr. Billmeyer would yield a result different from that of AMS 320.06. As Lubrizol points out, at paragraph 535 of its Closing Brief:
| 535. The AM-S 320.06 procedure was both in evidence and Mr. Deeth and his expert could follow it line by line and where Dr. Billmeyer departed from it, if he did, Mr. Deeth and his expert Dr. Kennedy could have noted it right away. |
[89] During his commission evidence in these proceedings, Dr. Billmeyer testified that he had never intended to mislead Mr. Justice Cullen when he testified before him in 1990:
| Q. Well now, was it your intention to hide from the Court any of the things which you had done in preparing the AM-S 320.06 procedure? |
| Q. And, insofar as the allegation in ground 5, namely which is a general statement, that your testimony was false when you testified you followed the Exxon VPO test method exactly to determine that Exxon"s method seriously understated the Mn of the PIB. Insofar as that allegation is concerned, or insofar as it is given more flesh in paragraph 43(a), (b), (c) and (d) of the Statement of Claim, where the allegations have been made that you deviated from Imperial"s standard of exactitude, did you, in fact, disclose in your trial evidence what you were doing? |
| Q. So that, with respect to the question of use of benzene as a solvent, did you disclose that in your Canadian trial evidence? |
| Q. With respect to employment of correct sample and calibrant concentrations, did you disclose your sample and calibrant concentrations? |
| Q. And, with respect to the preparation of a log plot, did you disclose that you were using a calibration factor instead? |
| Q. And, with respect to the construction of a calibration graph, did you feel that you were using one? |
[90] It is also relevant to this discussion that in January 1993, Mr. Bialo, counsel for Exxon, retained an analytical laboratory to carry out VPO tests as per Exxon"s method AMS 320.06. As the laboratory did not have a Mecrolab instrument, but a Wescan instrument, it was agreed between the laboratory and counsel for Exxon that the laboratory would conduct the tests in accordance with a procedure spelled out by counsel in a letter which was marked as Exh. D-34. The letter reads, in part, as follows:
| "I have asked you to perform three separate sets of measurements. In set 1, for each of the 10 samples you will make 5 determinations of Mn. using solutions with concentrations spaced across the range of 10 to 20 mg. of sample per gram of solvent, as we discussed, all in strict and exact accordance with a three-page typewritten procedure entitled "Molecular Mass of Polybutenes by Vapor Pressure Osmometry." We reviewed that procedure during our discussion on January 12 and I left a copy with Mr. Petro, your Technical Advisor. We have agreed that you will follow literally the second clause of step 6.2; that is, because you will be using a Wescan Model 233 VPO instrument, not a Mecrolab, you will follow the instructions recommended by Wescan for steps 6.3 through 6.12. |
When Mr. Bialo refers to a typewritten procedure entitled "Molecular Mass of Polybutenes by Vapor Pressure Osmometry", he is referring to Exxon"s method AMS 320.06.
[91] I agree with Lubrizol"s submission that what Mr. Bialo and the laboratory agreed to do was exactly what Dr. Billmeyer did. The variations to the method made by Dr. Billmeyer and the laboratory retained by Mr. Bialo were necessary by reason of the use of a Wescan instrument. In my view, what Exh. D-34 and surrounding circumstances show is that, at the very least, Dr. Billmeyer did not deliberately attempt to mislead Cullen J. when he testified in respect of AMS 320.06
[92] In support of its position, Imperial called Professor Knox. I agree entirely with Lubrizol that Professor Knox"s affidavit evidence "dealt with semantics, not substance". At paragraph 30 of his first affidavit, Professor Knox makes the following statement:
| It thus appears that had Dr. Billmeyer calibrated his GPC as he affirmed in his Canadian testimony, he would have concluded that the Exxon AMS 320.06 (VPO) test, as he carried it out, gave high rather than low Mn values when compared to GPC. |
This, as counsel for Lubrizol submits, is not an indictment of Dr. Billmeyer"s evidence regarding his use of the AMS 320.06 method. The point made by Professor Knox in paragraph 30 of his first affidavit is that if Dr. Billmeyer had calibrated his GPC as he testified he had in 1990, he would have concluded that Exxon"s AMS 320.06 test gave higher rather than lower Mn values when compared to GPC. Professor Knox makes no adverse comment in his first affidavit regarding Dr. Billmeyer"s use of AMS 320.06.
[93]
In his supplementary affidavit, at paragraph 52, Professor Knox raises a number of questions regarding Dr. Billmeyer"s use of AMS 320.06. During Professor Knox"s testimony, Lubrizol objected to paragraph 52 and I advised counsel that they should raise the objection during their final arguments. After consideration of these arguments, I am of the view that paragraph 52 of Professor Knox"s supplementary affidavit is simply argumentative, it does not contain the substance of an opinion. Consequently, that paragraph will be completely disregarded.
[94] Lubrizol called Dr. Rudin, who testified that Dr. Billmeyer"s evidence with respect to his use of AMS 320.06 was not misleading. In paragraphs 51 and 52 of his rebuttal affidavit, Dr. Rudin sets out in clear terms the factual basis for his opinion. I have no hesitation in accepting Dr. Rudin"s opinion on this point. I wish to note that Dr. Rudin was not cross-examined on this point, and that his qualifications to testify as an expert were, at no time, challenged by Imperial.
[95] I am therefore of the view that the allegation of fraud against Dr. Billmeyer on this point has not been proven.
[96] I now turn to ground No. 4. Imperial alleges that Dr. Billmeyer agreed on April 17, 1990, that Dr. O"Driscoll"s test results should be adjusted to give dramatically higher results without any further tests, and thereafter, Dr. Billmeyer supported Dr. O"Driscoll"s "universal calibration" method and Dr. O"Driscoll"s higher results without any scientific basis for doing so. Paragraphs 37 to 41 of Imperial"s Statement of Claim read as follows:
| 37. On April 15, 1990 Dr. Billmeyer"s draft affidavit recorded Dr. O"Driscoll"s test results as averaging an Mn of 2453. |
| 38. Two days later, on April 17, Dr. Billmeyer decided with Lubrizol that Dr. O"Driscoll should adjust his method to give higher numbers. Dr. O"Driscoll was not even present at this meeting. |
| 39. Dr. O"Driscoll thereafter raised his average by over 10% from 2453 on April 15 to 2709 (without additional testing) in his expert affidavit sworn on April 20, 1990, as submitted to this Court. |
| 40. At trial before this Court, Dr. Billmeyer testified that he had reason to believe that Dr. O"Driscoll"s test results, achieved by use of the universal calibration method, were more reliable than his own. |
| 41. Dr. Billmeyer admitted in Ohio that his support at trial for Dr. O"Driscoll"s results was not based on any independent scientific study, knowledge or investigation. |
[97] In answer to Imperial"s allegations, Lubrizol makes the following statement at paragraph 32 of its Statement of Defence:
| 32. Dr. Billmeyer was not responsible for the test results reported upon by Dr. O"Driscoll at the Canadian trial. To the extent he commented upon them prior to the trial, or in his trial testimony, he did so honestly and based upon his knowledge, experience and expertise in the particular area of science. |
[98] Ground No. 4 has two prongs. The first one is that Dr. O"Driscoll had initially recorded an average Mn of 2453 for the Parapol 2200 samples, and that at a meeting held on April 17, 1990, a meeting at which Dr. O"Driscoll was not present, Dr. Billmeyer and representatives of Lubrizol decided that Dr. O"Driscoll"s numbers should be raised, and that, in effect, these numbers were raised without further testing. The second prong is that Dr. Billmeyer, without having made any independent scientific study, investigation, and without knowledge, testified before Cullen J. that, in his view, Dr. O"Driscoll"s test results, obtained by use of the universal calibration method, were more reliable than his results.
[99] In my view, Imperial"s allegations are without merit. With respect to the first prong of ground No. 4, i.e. the decision to raise Dr. O"Driscoll"s numbers, it is clear from the evidence that that decision was made by Dr. O"Driscoll, and by him alone. Although Dr. Billmeyer"s notes do not indicate that Dr. O"Driscoll was present at the meeting, it is not now disputed by Imperial that Dr. O"Driscoll was indeed present at the meeting. Mr. Bauer"s evidence on that point is clear; Dr. O"Driscoll was present at the meeting. Mr. Bauer also testified that it was his recollection that Dr. O"Driscoll, during the meeting, discussed his decision to use a cut-off of 200, rather than use the default calculation of 100 supplied by the computer. Dr. O"Driscoll testified before Cullen J. that his decision to use a cut-off of 200 stemmed, in part, from an article written by Exxon employees in the Encyclopedia of Polymer Science . Dr. O"Driscoll was cross-examined by Mr. Deeth on this specific point. Mr. Deeth"s cross-examination of Dr. O"Driscoll on this point can be found in Dr. O"Driscoll"s trial evidence, vol. 7, pp. 129-136. Dr. O"Driscoll was then re-examined by Mr. Wells, and that can be found at vol. 7, pp. 156-160. There is reference in Dr. Billmeyer"s notes (Exh. D-1, Tab 98, at pp. 5-6) to Dr. O"Driscoll"s decision to use a cut-off of 200 as a result of his reading the article in the Encyclopedia of Polymer Science .
[100] Consequently, with respect to the first prong of ground No. 4, Imperial has not proved that Dr. Billmeyer agreed to adjust Dr. O"Driscoll"s test results so as to give higher results without further tests. I now turn to the second prong, i.e. that Dr. Billmeyer supported Dr. O"Driscoll"s "universal calibration" method without any scientific basis for doing so.
[101] It should be remembered that Dr. Billmeyer was qualified as an expert in the area of polymer science and more particularly, in the area of the measure unit of molecular weight and molecular weight distribution of polymers by a variety of means, including GPC and VPO. No objection was made before Cullen J. regarding Dr. Billmeyer"s qualifications to give the opinions that he gave, including his comments regarding Dr. O"Driscoll"s GPC testing of the Parapol 2200 samples and the European samples.
[102] It should also be remembered that a section of Dr. Billmeyer"s textbook on polymer science gave a description of the universal calibration method. At trial before Cullen J., Dr. Billmeyer was cross-examined in regard thereto. All of this to say that there could not and cannot be any doubt concerning Dr. Billmeyer"s qualifications to give the opinions he gave before Cullen J.
[103] The basis of Imperial"s attack is that Dr. Billmeyer testified in the Ohio III proceedings that his support for Dr. O"Driscoll"s "universal calibration" method was not based on any independent scientific study, knowledge or investigation. Dr. Billmeyer also stated during the course of his U.S. deposition that:
| In my testimony to the Canadian court, I was giving my belief and impressions and to the best of my knowledge at the time, these were correct.23 |
[104] Before Cullen J., Dr. Billmeyer, both in his evidence in chief and in his rebuttal evidence, was called upon to comment on the test method used by Dr. O"Driscoll. Initially, Dr. Billmeyer stated that Dr. O"Driscoll"s results were more representative of the true Mn value of the samples. He indicated that his results understated the true Mn. For all intents and purposes, Dr. Billmeyer was not cross-examined on that evidence. As I indicated earlier, he was asked a few questions regarding chapter 7 of his textbook, where he comments on the universal calibration method. In his rebuttal evidence, Dr. Billmeyer responded to criticisms made of Dr. O"Driscoll"s method by Drs. Kennedy and Porter, experts for Imperial. In giving that evidence, Dr. Billmeyer, in my view, made it clear that he was quite familiar with Dr. O"Driscoll"s test method and the literature concerning that method (See Exh. P-15, Tab 4, Billmeyer evidence, pp. 691-700, 733-38. See also Evidence Vol. 18, pp. 26-38).
[105] Dr. Billmeyer"s notes are also instructive on this subject. More particularly, his notes of January 4, 1990 (p. 2) and of February 1, 1990, demonstrate his knowledge of the work that Dr. O"Driscoll was doing in preparation for the trial and his knowledge of Dr. O"Driscoll"s method.
[106] After careful consideration of all the evidence on this point, including Dr. Billmeyer"s deposition in the Ohio III proceedings, his testimony before Cullen J., his commission evidence in 1997 and his qualifications, I cannot but conclude that Dr. Billmeyer was indeed qualified to give an opinion with respect to Dr. O"Driscoll"s universal calibration method and regarding the results obtained by him. Consequently, it cannot be said, in my view, that Dr. Billmeyer"s opinion "was not based on any independent scientific study, knowledge or investigation". I am satisfied that Dr. Billmeyer, in commenting upon Dr. O"Driscoll"s method before Cullen J., had no intention of misleading nor did he, in fact, mislead Cullen J. Imperial therefore fails in respect of ground No. 4. I now turn to ground No. 3.
[107] Pursuant to ground No. 3, Imperial alleges that Dr. Billmeyer"s testimony in Canada was false, in that he gave the Court Waters 200 GPC test results on the Exxon PIB samples that were overstated by seven and a half percent. This overstatement was the basis of this Court"s finding of infringement. The specifics of ground No. 3 are set out at paragraphs 34 to 36 of Imperial"s Statement of Claim:
| 34. Dr. Billmeyer admitted in Ohio that his calibration curve produced results for the Amoco H-1500 anchor sample that were 7-1/2% higher than the VPO value for that sample which Dr. Billmeyer received after the Waters 200 GPC calibration curve was finalized. Dr. Billmeyer suppressed this fact when he testified to this Court. |
| 35. If Dr. Billmeyer had in fact set out to match the VPO data from his anchor samples when he adjusted his final calibration curve, as he testified to this Court in 1990, he would have produced Mn values by the Waters 200 GPC on the Parapol 2200 and European samples that were 7-1/2% lower than those he described to Mr. Justice Cullen. The impact of this misrepresentation is shown on this chart: |
| |
Parapol 2200
Samples
|
Parapol 2200
Samples plus
European samples
(ECA 10788 /
Parapol 2225)
|
European Samples
only (ECA 10788 /
Parapol 2225)
|
| Dr. Billmeyer"s
Canadian testimony
(Exhibit B7A)
|
2457
|
2438
|
2289
|
| Dr. Billmeyer"s
corrected Mn
values (if matched
to VPO on the
Amoco H-1500
anchor)
|
2289
|
2270
|
2131
|
| 36. The impact on the succination ratios of the PIBSA for the accused products if the corrected Mn"s are used, is respectively: for PIBSA 54, SR = 1.25; for PIBSA 51, SR = 1.21; and for PIBSA 48, SR = 1.12. These average values are below 1.3. Accordingly, had Dr. Billmeyer given truthful evidence, there would have been no finding of infringement for those PIBSAs. |
[108] Lubrizol"s answer, as it appears from paragraph 31 of its Statement of Defence, is as follows:
| 31. As to paragraphs 32 to 36 of the statement of claim, the 2445 VPO Mn value for Amoco H-1500 PIB was a single point VPO Mn value. As stated in his evidence, Dr. Billmeyer relied upon 3-point VPO Mn values. As referred to above, Dr. Billmeyer used the 3-point VPO Mn value for that sample of 2622. The allegation that his test results were overstated by 7 " percent is based on a false premise. |
[109] Imperial"s ground No. 3, as it is pleaded in paragraphs 34 to 36 of its Statement of Claim, has not been proven. Consequently, that ground must fail.
[110] To properly understand ground No. 3, it is necessary to refer to paragraphs 32 and 33 of Imperial"s Statement of Claim, which provide as follows:
| 32. Dr. Billmeyer testified in Ohio that after his calibration curve was finalized, he received VPO data on the anchor samples from Lubrizol. According to Dr. Billmeyer, the VPO value for the Amoco H-1500 anchor sample was 2445. In 1993, Dr. Billmeyer told Lubrizol"s Canadian counsel, Ridout & Maybee, for purposes of answering questions by Lubrizol"s U.K. lawyers, Allen & Overy, that the VPO value for that anchor was 2445. In 1995, Dr. Billmeyer testified in his Ohio deposition that it was 2445, both on examination by opposing counsel and on examination by Lubrizol"s own counsel. Dr. Billmeyer had the opportunity weeks later to correct his testimony, before signing the transcript, and, although he made changes he made no changes on this point, although he did change other testimony. |
| 33. In his Ohio testimony, Dr. Billmeyer admitted that his Waters 200 GPC Mn value for the critical Amoco H-1500 anchor sample (2625) was 7.5% higher than the VPO value (2445) for that sample. Despite this discrepancy, Dr. Billmeyer said he declined to make any further adjustments to his Waters 200 GPC calibration curve in order to match the VPO value. Accordingly, Dr. Billmeyer presented this Court with test results using a Waters 200 GPC calibration curve that, contrary to his testimony in this Court, was not matched to VPO results but was, in fact, inflated in Lubrizol"s favour by 7.5%. |
[111] The basis of Imperial"s ground No. 3 is that Dr. Billmeyer obtained a VPO Mn value of 2445 for the Amoco H-1500 anchor sample, and that his Waters 200 GPC Mn value for that sample was 2625, i.e. 7.5% higher than the VPO value of 2445. Imperial alleges that notwithstanding this 7.5% discrepancy, Dr. Billmeyer refused to adjust his Waters 200 GPC calibration curve in order to match the GPC value to the VPO value. As a result of his refusal to adjust downward his calibration curve, Dr. Billmeyer obtained Mn values for the Parapol 2200 samples and the European samples that were inflated by 7.5%. The effect of Dr. Billmeyer"s overstatement, according to Imperial, is shown on the chart which appears at paragraph 35 of Imperial"s Statement of Claim. Hence, with higher Mn values for the Exxon products, the S.R.s of the PIBSAs were accordingly higher and, in the event, were found to infringe by Cullen J. According to Imperial, had Dr. Billmeyer presented to the Court the true Mn values, the S.R.s for PIBSA 48, PIBSA 51 and PIBSA 54 would have all fallen below 1.3. Thus, Imperial states that if Dr. Billmeyer had testified in a truthful manner, Cullen J. would not have made a finding of infringement in regard to these PIBSAs.
[112] The evidence has clearly shown that the Mn VPO value of 2445 for the Amoco H-1500 anchor sample was a single point VPO, which Dr. Billmeyer did not use. Rather, Dr. Billmeyer relied upon 3-point VPO Mn values, which in regard to the Amoco H-1500 anchor sample was 2622 (See Exh. D-1, Tab. 19). The premise of Imperial"s ground No. 3, i.e. that the VPO Mn value for the Amoco H-1500 anchor sample was 2445, is, no doubt, unfounded. The result obtained by Dr. Billmeyer for the Amoco H-1500 sample, by way of the 3-point method, is 2662 and, accordingly, Imperial"s ground No. 3 fails. I now turn to ground No. 2.
[113] Pursuant to ground No. 2, Imperial takes the position that Dr. Billmeyer"s testimony in Canada was false in that contrary to his Canadian testimony, his final calibration curve did not match VPO results for the anchor samples. The particulars of this ground appear in paragraphs 28 to 33 of the Statement of Claim:
| 28. In his evidence at trial before Mr. Justice Cullen, Dr. Billmeyer stated that he shifted, or adjusted, his preliminary calibration curve to obtain a final calibration curve for his Waters 200 GPC that would "give Mn values agreeing with true Mn measured by VPO". He amplified the description of his adjustment by stating that he shifted the preliminary curve using two "polybutenes not fractionated in the molecular weight range of 1300 and again around 2500, that we had very carefully prepared to be free of impurities and measured by the three point VPO method which was the most accurate measure of number average molecular weight that was available to us". This testimony was false. |
| 29. Further, not only had Dr. Billmeyer not "carefully prepared" the anchor samples to be free of impurities, he had not prepared the anchor samples at all, and he had no information about the anchor samples other than what Lubrizol chose to tell him, which he failed to verify. |
| 30. Dr. Billmeyer admitted in Ohio that, contrary to his testimony before this Court, he had not set out to match his Waters 200 GPC calibration curve to VPO values on the anchor samples. Instead, he adjusted his Waters 200 GPC calibration curve to match results on the anchor samples from Lubrizol"s HSGPC. He admitted in direct conflict with his testimony to this Court, that he did not have VPO results when he prepared his final Waters 200 GPC calibration curve. |
| 31. Dr. Billmeyer used this calibration curve to perform all of the Waters 200 GPC tests on PIB samples he testified about in this Court. |
| 32. Dr. Billmeyer testified in Ohio that after his calibration curve was finalized, he received VPO data on the anchor samples from Lubrizol. According to Dr. Billmeyer, the VPO value for the Amoco H-1500 anchor sample was 2445. In 1993, Dr. Billmeyer told Lubrizol"s Canadian counsel, Ridout & Maybee, for purposes of answering questions by Lubrizol"s U.K. lawyers, Allen & Overy, that the VPO value for that anchor was 2445. In 1995, Dr. Billmeyer testified in his Ohio deposition that it was 2445, both on examination by opposing counsel and on examination by Lubrizol"s own counsel. Dr. Billmeyer had the opportunity weeks later to correct his testimony, before signing the transcript, and, although he made changes he made no changes on this point, although he did change other testimony. |
| 33. In his Ohio testimony, Dr. Billmeyer admitted that his Waters 200 GPC Mn value for the critical Amoco H-1500 anchor sample (2625) was 7.5% higher than the VPO value (2445) for that sample. Despite this discrepancy, Dr. Billmeyer said he declined to make any further adjustments to his Waters 200 GPC calibration curve in order to match the VPO value. Accordingly, Dr. Billmeyer presented this Court with test results using a Waters 200 GPC calibration curve that, contrary to his testimony in this Court, was not matched to VPO results but was, in fact, inflated in Lubrizol"s favour by 7.5%. |
[114] With respect to paragraphs 32 and 33, I have already indicated, in dealing with ground No. 3, that Imperial proceeded on a false premise, i.e. that Dr. Billmeyer obtained an Mn VPO value of 2445 for the Amoco H-1500 anchor sample. Rather, the Mn value obtained by Dr. Billmeyer by VPO was 2622.
[115] In paragraph 45 of his affidavit filed for the 1990 Canadian trial, Dr. Billmeyer stated that he had adjusted his preliminary curve to give Mn values in agreement with Mn values obtained by VPO and that he had also adjusted his curve so as to make sure that Mw values were also properly determined. Dr. Billmeyer attached as Exh. B-4 to his affidavit a copy of his final calibration curve24. During his examination in chief, he explained the adjustments he had made, i.e. by taking his whole curve and shifting it linearly right to left. He explained that the basis for this shift were "measurements of some polybutenes not fractioned in the molecular weight range of 1300 and around 2500". Dr. Billmeyer then went on to indicate that the Mn values obtained for these polybutenes had been measured by 3-point VPO which, in his view, "was the most accurate of measure of number average molecular weight that was available to us".
[116] Later on in his examination in chief, Dr. Billmeyer testified with regard to impurities which might be found in the polybutenes. In answer to questions posed to him by Mr. Wells, counsel for Lubrizol, Dr. Billmeyer made the following remarks:
| Q. Did you do anything to address the issue of things that were not polymer? |
| A. "It is possible to examine the GPC trace itself and make some judgment as to whether and approximately how much impurity there may be present. In most polybutenes we can see this, and when we come to look at the GPC traces I can probably point out some of this evidence out to My Lordship. But I was a very careful to select polybutenes in which I could see no evidence of such impurities for the purpose of calibrating the curve."25 |
[117] During his cross-examination by Mr. Deeth, Dr. Billmeyer provided information to the Court with regard to the two anchor samples which, in his words, he "used most extensively in adjusting my first calibration curve"26. He further indicated, in answer to Mr. Deeth"s questions, that one of the anchor samples had a molecular weight level of about 1300 and that the other had a molecular weight level of about 2500. Later on during his cross-examination by Mr. Deeth, Dr. Billmeyer again confirmed that he had moved his calibration curve so that his GPC values would match the values obtained by VPO27. It was only after he had moved the calibration curve, as he indicated above, to match the Mn values obtained by VPO, that Dr. Billmeyer proceeded to "run" the Parapol samples28.
[118] Again, under cross-examination by Mr. Deeth, Dr. Billmeyer had occasion to inform the Court with respect to his knowledge of the preparation of the polymers to be free of impurities. This is what he said:
| Q. We. will have a look at your GPC procedure this afternoon, but just to finish off this paper, page 172 you have summary of this chapter. At page 172? |
| Q. In the second paragraph from the bottom of the page, you refer to VPO and say: ". . . As experience with vapor-phase osmometry grows, most users come to the realization that many of the difficulties experienced with it stem from deficiencies in the samples and not the method . . ." And that is the point you were making in your Affidavit about the need to remove the low molecular weight impurities from the samples? |
| Q. The technique for removing low molecular weight impurities from the sample was something that could be easily done? |
| A. No, it turns out to be very difficult with most polymers to get these low molecular weight impurities out. |
| Q. For polybutene how would you do it? |
| A. I am not an expert in purifying polybutene, so I really cannot answer that qestion. |
| Q. How about polyethylene? You have published a number of papers on that. |
| A. Well polyethylene usually is at sufficiently high in molecular weights that one is not too concerned with them. |
| No, I cannot say I am an expert on exactly how you would treat any polymer to remove the low molecular weight impurities. |
| Q. Well, in your photographs yesterday, you referred to the fact that when you run your GPC that if you set up GPC columns correctly that the impurities come out after all of the polymers come out? |
| Q. That would be one way of separating them? |
| A. Yes, that certainly could be done. There is such a thing as preparative GPC operation in which you can selectively remove certain portions of the material.29 |
[119] During his commission evidence, Dr. Billmeyer flatly denied that his 1990 evidence on this point was false. He also testified that he had, in fact, calibrated his final curve to match VPO results. He made the following statement: "My final calibration curve did match VPO results for those whole polymer standards"30. Both in cross-examination and in re-examination, Dr. Billmeyer was thoroughly examined on his evidence in 1990 before Cullen J., and more particularly, with respect of the adjustments he claims to have made to his final curve so as to match VPO Mn values.
[120] Before me, in these Rule 1733 proceedings, Professor Knox made a number of concessions which, in effect, constitute a retraction of a substantial part of the position he initially took. In particular, he conceded that Dr. Billmeyer"s use of high speed GPC Mn and Mw data for his two whole polymer samples, on February 15, 1990, while waiting for the VPO results, was proper in the circumstances. He also conceded that in referring to samples 21140 and 2117231 as the samples he had used "most extensively", Dr. Billmeyer was probably referring to the April 1990 samples as the other samples which he had used. Professor Knox then conceded that he was satisfied that up to April 10, 1990, all of Dr. Billmeyer"s work was scientifically sound. This is how he put the matter:
| Q. ... if you were looking for evidence of people trying to cook the books, you wouldn"t expect them to be lopping off points that would help, cooking them in that direction, would you? |
| A. Well, I don"t think at this point in time there was any intention to cook the experimental evidence, at any rate. The experimental evidence was coming in, and it was good data that was coming in. So I think up to, say, the 10th of April, there was this good evidence, and I think it was the assessment of this evidence in the rounds which went wrong subsequently.32 |
[121] With respect to Dr. Billmeyer"s assertion that he had verified the purity of his samples, Professor Knox gave the following answer to a question posed by Mr. Plumley, counsel for Lubrizol:
| Q. Well, now, you agreed with me that he checked the purity of these samples before using them? |
| A. I agree with that, yes.33 |
Professor Knox then conceded that the H-1300 sample and the H-1900 sample were appropriate substitutes for 21140 and 2117234.
[122] Professor Rudin, who testified on behalf of Lubrizol, took the view that Dr. Billmeyer had adjusted his preliminary calibration curve so as to obtain a final curve that would yield Mn values in agreement with Mn values obtained by 3-point VPO. In Professor Rudin"s opinion, the results obtained by Dr. Billmeyer are clear evidence that his calibration work was proper35. Dr. Rudin also took the view that "the overall agreement between the Waters 200 and the multi-point VPO Mn values is satisfactory..." I have no hesitation in accepting, in its entirety, Dr. Rudin"s opinion on this point.
[123] In paragraph 399 of its Closing Arguments, Lubrizol submits that in paragraphs 135 to 240 of its closing arguments, Imperial has made an attempt to recast its allegations against Dr. Billmeyer. I agree entirely with Lubrizol that Imperial should not be permitted to add these allegations. In any event, I am of the view that these allegations are also without merit. In paragraph 29 of his rebuttal affidavit (Ex. D-61), Dr. Rudin makes the following statements:
| In summary, Dr. Billmeyer"s February 1990 calibration work was directed toward having the Waters 200 GPC Mn measurements match multi-point VPO measurements. His follow-up work in March and April 1990 was explicitly directed to the same goal and confirmed his achievement of that goal. There is no basis for rejecting Dr. Billmeyer"s verification of the Waters 200 GPC calibration curve, in April 1990. |
[124] It is also useful to read paragraphs 5 and 6 of Dr. Rudin"s rebuttal affidavit, where he says:
| 5. Professor Knox has described the calibration of the Waters 200 GPC carried out by Dr. Billmeyer in 1990 as "an effective calibration" (suppl. affidavit, para. 7) method. Dr. Billmeyer described his procedure as an empirical method (Ex. D-1 Tab 25 p. 3). I do not perceive a scientifically legitimate basis for drawing a distinction between "empirical" and "effective" calibration method. |
| 6. An empirical method is one which is guided by observation of experimental data and the experience of the person using the method rather than being based directly upon scientific theory. The validity of an empirical method does not depend upon its being derived from first principles, but upon its ability to predict results that can be verified by some independent means. Unless it can be shown that the results achieved by an empirical method are incorrect, it is irrelevant to criticize the steps followed in reaching the final result. |
[125] The gist of Imperial"s "new" allegations is that Dr. Billmeyer testified in 1990 that he used Mn values obtained by VPO to adjust his curve. I agree with Lubrizol that this is not what he said. What he said is what he relates in paragraph 45 of his affidavit:
| ... Consequently, I adjusted the calibration curve to give Mn values agreeing with true Mn measured by VPO and also adjusted the calibration curve to ensure that Mw values were also correctly determined... |
[126] Before Cullen J., Dr. Billmeyer testified as follows:
| Q. ... You used what you were satisfied would give you the true Mn. |
| Q. Then you moved your curve to match that sample? |
| Q So that the number average molecular weight that you got from the GPC calibration curve would match the number that you got from VPO? |
[127] Counsel for Lubrizol submitted, and I agree with the submission, that the goal of Dr. Billmeyer"s VPO work in March and April 1990 was to ensure that his Waters 200 GPC calibration would give Mn values that matched those obtained by multi-point VPO. To conclude with Imperial"s "new" allegations, I can do no better than refer to paragraphs 399 to 407 of Lubrizol"s closing arguments, and say that I agree with these submissions. In particular, I agree with the submissions that are found at paragraphs 401, 402 and 403:
| 401. If one looks beyond the exercise in semantics set out in Imperial"s submissions in order to ascertain if any assertions of substance have been made, it is apparent that Imperial recognizes the focus of Dr. Billmeyer"s work and of his 1990 testimony was that he calibrated the Waters 200 GPC so that it would give measurements of Mn for polybutenes that would agree with multi-point VPO values that could be obtained for those polybutenes (e.g. Imperial submissions, para. 176). |
| 402. In terms of semantics, Imperial goes to great lengths to attempt to leave the impression that Dr. Billmeyer suppressed knowledge of his interim use of HSGPC Mns and Mws while constructing his calibration curve. This is despite the evidence of all experts that the data used to create iterations was not important, but rather the data used for evaluating the performance of the calibration was what mattered. |
| 403. In terms of substance, Imperial does not put forward cogent evidence attempting to prove that Dr. Billmeyer calibrated the Waters 200 GPC to match HSGPC Mns. The reason is clear; he did not attempt to do so and he did not do so. |
[128] In my view, it is clear that ground No. 2 is without merit. There is no evidence to support Imperial"s allegation that Dr. Billmeyer tried to mislead Cullen J. when he testified that he used VPO values in adjusting his final calibration curve. I now turn to ground No. 1.
[129] Imperial"s ground No. 1 allegation is that Dr. Billmeyer falsely testified that he used VPO in constructing his preliminary calibration curve. Paragraphs 25 to 27 of the Statement of Claim provide the factual basis for ground No. 1:
| 25. Throughout his testimony before this Court Dr. Billmeyer laid great stress on his view that VPO was an "absolute" or "primary" method to test the Mn, whereas GPC was only as accurate as its calibration by VPO. |
| 26. Dr. Billmeyer testified in this Court that he constructed a preliminary curve based on narrow fractions whose Mn"s were determined by VPO. When cross-examined on this point during the trial in this Court, Dr. Billmeyer stated several times that he had determined the Mn of the narrow fractions by VPO. Some weeks later Dr. Billmeyer returned before Mr. Justice Cullen for reply evidence and repeated the substance of this testimony. |
| 27. In his Ohio testimony, Dr. Billmeyer conceded that he had in fact used Mn and Mw values from Lubrizol"s HSGPC to construct his Waters 200 GPC calibration curve. He admitted that his testimony before Mr. Justice Cullen on this point was false. |
[130] At paragraph 44 of his April 1990 affidavit, Dr. Billmeyer stated:
| 44. Following the procedure described in ASTM D3536-76, a first calibration curve was drawn using mean molecular weight (the square root of the product of Mn and Mw) for fractionated polybutene standards. Six such fractions were used. The Mn values used in this first calibration were based on three point VPO determinations following a procedure similar to D3592-77 and the Mw was determined using GPC with a viscosity detector. |
[131] In his evidence before Cullen J., Dr. Billmeyer explained, in the following terms, what he had stated in paragraph 44 of his affidavit:
| The first curve that I plotted as I have just read was taken for fractionated samples and obtained by plotting the mean molecular weight. |
| The square root of Mw times Mn and this is a convenient thing to do because as we will see when we look at GP curves which I anticipate we will in a few moments, the curves rise to a peak and then drop off again and the mean molecular weight is very close to the peak value. |
| For the six fractionated polybutenes I used, I was able to plot a first curve. But since those are fractions containing a relatively narrow distribution of molecular weights their curves are relatively narrow and when one applies that calibration curve to ordinary or normal polybutenes with a somewhat broader curve and asks about the number average molecular weight is not at the peak of the curve. Then one gets an answer which is too low and it is necessary to shift the curve in some way so that it gives correct values for Mn for polybutenes are of interest to us.37 |
[132] Both Dr. Knox and Dr. Rudin testified that they were satisfied, after a careful examination of all of Dr. Billmeyer"s work and notes, that he had used Mn values based on three-point VPO in constructing his preliminary calibration curve. During his testimony, Professor Knox conceded that his earlier view that Dr. Billmeyer had not used VPO values to construct his preliminary calibration curve was a mistaken view. This is what Professor Knox stated in answers to questions posed by Mr. Plumley:
| Q. Well, now, as you can see, I have got a lot of notes here, but I am not sure how much of this stuff we have between us, so I would like to ... 19 is sort of a wrap-up paragraph, so let"s look at that one. Now, you see the first sentence, you say: |
| " ... VPO measurements were not used to fix either the initial calibration curve or the final calibration curve ..." |
| Let"s take it a step at a time. Are you still saying that with respect to the initial calibration of the curve? |
| A. I think we established this morning that the initial calibration curve, the one that he started from before he started doing these adjustments, actually came from February 6th. |
| I was very firmly of the opinion previously that he wasn"t, and I think you convinced me now that it probably was. |
| Q. So we will take [sic] initial calibration curve off the table, shall we? |
| A. I think we have to take that out.38 |
[133] I need not say more on this ground as, in my view, the evidence is overwhelming that Dr. Billmeyer did use Mn values based on three-point VPO in constructing his preliminary curve. As an example, reference should be made to Dr. Billmeyer"s notes of March 13, 1990, where the following entry appears:
| "Cal for general curve shape, run fr stds & plot Vr vs. MwxMn1/2, per D 3536. (Used VPO Mn & HSGPC visc det Mw as approxn.) But know this applies to narrow MWD fractions. For broader, this overstimates Mn. So drew instead a curve for the cal stds to get (best) agreet w their VPO Mn"s (OK since fractionated, no peculiar low stuff that cd make VPO"s low)" |
| (1) Results: (4 1990s) std, Mn from (200 cal. curve, HS, VPO); |
| (2) Curve checked w 2 1978 stds, results Mn"s. Second adj for whole polys with Mn and Mw/Mn in range of interest. Used L spls in Mn, range nr 260039 |
[134] Finally, Dr. Rudin stated before me, as he had done so in his affidavit at paragraphs 9 to 15, that in constructing his February 6, 1990 preliminary calibration curve, Dr. Billmeyer had used VPO Mn"s for the six fractions. Dr. Rudin also stated that Dr. Billmeyer"s February 15, 1990 preliminary calibration curve was a modified iteration of the February 6 curve. At paragraph 15 (a) of his April 7, 1998 rebuttal affidavit, Dr. Rudin states:
| In summary, Dr. Billmeyer constructed his February 6, 1990 preliminary calibration curve so as to provide Mn measurements of the fraction standards which match multi-point VPO Mn measurements of those standards. He refined the left end, or high molecular weight end, of that curve on February 15, 1990. Professor Knox believed that "there was no continuity between his February 6, 1990 work and his subsequent work of February 15, 1990. |
[135] I have no reason not to accept Dr. Rudin"s opinion, nor do I have any reason not to accept Professor Knox"s evidence that his earlier view was wrong. Consequently, Imperial has not succeeded in proving ground No. 1.
[136] I have come to the conclusion that Imperial has not demonstrated fraud or recklessness on the part of Dr. Billmeyer, and consequently, Imperial"s Rule 1733 motion will be dismissed. However, before concluding, there are a number of issues which I ought to address.
[137] The first issue is the materiality of Dr. Billmeyer"s evidence in 1990 before Cullen J. It cannot be disputed, nor was it, that the Mn of ECA 10788 and Parapol 2200 was material to Cullen J."s decision. However, I agree with Lubrizol that what was material were the results of Dr. Billmeyer"s tests and, on the evidence before me, it appears that his results were correct. In that respect, the evidence of Dr. O"Driscoll cannot be forgotten. Dr. O"Driscoll testified before Cullen J., who found him to be a credible and reliable witness and, accordingly, accepted his evidence. It should also not be forgotten that Dr. O"Driscoll"s results showed Mn values higher than those put forward by Dr. Billmeyer. As I pointed out in my summary of the relevant facts, Exxon, in the Ohio III proceedings, took the position that Cullen J."s finding that the Mn of the starting PIB was at least 2500, was based on Dr. O"Driscoll"s testimony. At page 29 of its Memorandum, Exxon made the following submission:
| . ... Furthermore, the Court"s critical finding " that the Mn of the starting PIB was at least 2500 " was clearly based on Dr. O"Driscoll"s testimony. This conclusion could not have been based on Dr. Billmeyer"s GPC analysis, because he found only an average Mn of 2457 for Parapol 2200 and an average Mn of 2288 for the purported ECA 10788 made in Europe. Instead, the Canadian court must have relied on the GPC analysis of Dr. O"Driscoll, who found an average Mn for Parapol 2200 of 2709, and an average Mn for European ECA 10788 of 2500. |
[138] With respect to Exxon"s Memorandum in the Ohio III proceedings, I also wish to recall that Exxon took the position that Dr. Billmeyer"s results for Parapol 2200, i.e. an average Mn of 2457, were in line with its analysis in 1988 which showed that the Mn of Parapol 2200 "had risen to about 2400".
[139] Earlier on, I quoted paragraphs 5 and 6 of Dr. Rudin"s rebuttal affidavit which were to the effect that Dr. Billmeyer had used an empirical method in obtaining his results and that the validity of such a method depended upon its ability to predict results that could be verified by independent means. Dr. Rudin went on to state that unless it could be shown that the results obtained by the empirical method were incorrect, it was irrelevant "to criticize the steps followed in reaching the final results". Dr. Rudin"s comments go to grounds of fraud Nos. 1 and 2. On my understanding of the evidence, I can only agree with Dr. Rudin.
[140] In closing on the subject of the materiality of Dr. Billmeyer"s alleged misrepresentations, I wish to state that I agree with the submission which Lubrizol makes at paragraph 637 of its Closing Arguments:
| 637. Imperial are left with submitting that the alleged mis-statements if exposed at the trial might have raised a possible issue of credibility that might have been significant. This is completely contrary to the requirement of materiality on the authorities. |
[141] The second point which I wish to address is the credibility of Professor Knox who testified as the only expert witness on behalf of Imperial. My overall assessment of Professor Knox is that his evidence should not be given much weight and, in fact, I did not place, for all intents and purposes, any weight on his evidence. Professor Knox did not really deny that he did not have much expertise in the practical use and calibration of a GPC instrument. With respect to VPO, he gained some experience as a result of being retained by Exxon"s U.K. attorneys in connection with litigation between Lubrizol and Exxon in that country. However, his practical experience with VPO was very limited. During his testimony, he conceded that he did not have enough skill to calibrate VPO instruments.
[142] In reading Dr. Knox"s affidavits, one has the impression that it is a lawyer"s brief signed by a scientist. This, in my view, appears quite clearly when one reads Professor"s Knox"s first affidavit. Paragraphs 44, 45 and 46 of the first affidavit, which fall under the title "Overall Conclusion", are illustrative of my point:
| 44. Dr. Billmeyer misrepresented his experimental data to the Canadian court. He did not carry out his measurements as he said he did in his sworn testimony. |
| 45. Dr. Billmeyer"s misrepresentations led Mr. Justice Cullen to the view that the Mn value of the PIBs used in the production of the PIBSAs was above 2500 when in fact the best estimate available from the data that Dr. Billmeyer revealed in the Ohio III depositions was about 2130. |
| 46. According to my considered view, the true Mn of the Succination Ratio of the material, as calculated using Dr. Billmeyer"s data presented in Ohio III when treated according to the methods sworn in the Canadian Affidavit, is below 1.3, and is a material departure from what Dr. Billmeyer presented to Mr. Justice Cullen. Not only did Dr. Billmeyer demonstrate in his Canadian testimony a reckless disregard for the truth, but the recklessness related to key findings on infringement, as noted above. If I have correctly understood the relevant findings of fact in the judgment of Mr. Justice Cullen, and I believe that I have, the products in contention would not have been considered to infringe the Meinhardt patent had Dr. Billmeyer truthfully presented his experimental data. |
[143] Professor Knox appears to have devoted his skills to an analysis of Cullen J."s Reasons in the light of Dr. Billmeyer"s testimony in the Ohio III proceedings and in the Canadian trial, with a view to deciding whether or not Dr. Billmeyer had perjured himself or had testified recklessly. In my view, Professor Knox ought to have spent more time examining Dr. Billmeyer"s work in an objective manner. As Lubrizol points out, Professor Knox"s mandate was to discover if Dr. Billmeyer "had told lies" when he testified before Cullen J. On at least one occasion, I indicated to Professor Knox that it was my task to decide whether Dr. Billmeyer had lied or had testified recklessly. At paragraphs 311 and following of its Memorandum, Lubrizol makes a number of submissions concerning Dr. Knox, and I believe that these submissions are correct. In particular, Lubrizol submits that Professor Knox"s testimony was deceptive in regard to Imperial"s ground of fraud No. 3. Lubrizol"s submission on this specific point appears at paragraphs 316 to 318 of its Closing Arguments:
| 316. In fact, it was apparent from his cross-examination that parts of his evidence in chief were knowingly deceptive. |
| 317. He filed his first affidavit believing from the cross-examination of Mr. Bialo that the VPO value of 2445 was a 3-point VPO and hence one which Dr. Billmeyer would have had regard to. He said that he knew that Dr. Billmeyer had said he used 3-point VPO, not single point VPO in his calibration. He fell into the same error that Dr. Billmeyer was led into by Mr. Bialo"s questioning (even though mr. Bialo knew full-well that the number was a single point VPO). There was information in the Ohio transcript from which Prof. Knox could have determined that the 3-point VPO value was 2622. It was there if he had looked to determine the truth rather than to pick up inconsistencies as he said was his mandate. |
| 318. However, when he came to make his second affidavit he knew this was wrong. He knew that the whole basis upon which he had placed his first affidavit in respect of Ground 3 of the allegations of fraud was without foundation and yet he structured his second affidavit to make it appear that there was no change in his position. In fact, in para. 9 of his Second Affidavit [Ex. P-40] he said that he had no reason to change the opinion he expressed in para. 18 of his First Affidavit [Ex. P-39] although he tried to hide this by re-phrasing it as a miscalibration. All of this was pointed out in the cross-examination in Vol. 11. |
[144] I have carefully reviewed both of Professor Knox"s affidavits and, in particular, the paragraphs referred to by Lubrizol, and I fail to understand why he maintained the position he had taken in regard to the Mn VPO values of the two Amoco samples. I therefore agree with Lubrizol that when he made his second affidavit, Professor Knox must have known that the position he had taken in his first affidavit was not correct.
[145] Consequently, in the circumstances, I cannot say that Professor Knox approached his work as an unbiased scientific advisor. He appears to have become a "partisan" of the Plaintiff"s cause. In my view, and I say this out of deference and respect for Professor Knox, he was given an impossible mission when he was asked to discover whether Dr. Billmeyer had lied to the Court in 1990.
[146] On the other hand, I was very much impressed by Dr. Rudin"s evidence. He remained a man of science throughout; he did not attempt, at any time, to advocate either Lubrizol"s or Dr. Billmeyer"s cause. He carefully examined Dr. Billmeyer"s work to determine whether credible scientific work had been performed. In the end, his conclusion was that Dr. Billmeyer had performed honest and credible scientific work. As I have made it clear throughout my Reasons, I have accepted Dr. Rudin"s opinion on most issues.
[147] A few words about the "method of the patent" argument which Imperial and Professor Knox raised. As Cullen J. accepted Dr. O"Driscoll"s "universal calibration" method as well as Dr. Billmeyer"s method, it could not be argued, in my view, that Dr. Billmeyer"s testimony carried more weight before Cullen J. because Dr. Billmeyer had used the "method of the patent". Cullen J. made no distinctions, in his Reasons, between the methods used by Dr. O"Driscoll and Dr. Billmeyer. He found both methods to be acceptable in making his findings with regard to the Mn of Imperial"s products. Therefore, as Lubrizol asserts in its Closing Arguments, the Mn determinations for the purposes of the patent claims are not limited to determinations by a particular GPC procedure, nor to determinations by GPC rather than another procedure, such as VPO. This matter was decided by Cullen J. and he was confirmed on this point by the Federal Court of Appeal. It is now res judicata .
[148] In view of my conclusion on the merits of the case, I need not address some of the issues raised by Lubrizol in its Statement of Defence, namely, abuse of process, lack of due diligence and delay in bringing this Rule 1733 motion. However, I wish to quote at length the view of the case put forward by Mr. Hughes, counsel for Dr. Billmeyer, in his Closing Arguments. This view, on my understanding of the evidence, is indeed a fair assessment of what has happened. At pages 8 to 11 (paragraphs 10 to 25), Mr. Hughes makes the following submissions:
| 10. Dr. Billmeyer"s evidence at the 1990 trial took the form of an affidavit, supplemented by verbal testimony in chief, cross-examination and re-examination. His evidence ranged through a number of topics, including: |
| % Qualifications and Experience |
| % Polymer and Molecular Weight |
| % Methods for Determining Mn and Mw/Mn |
| % Paramins AMS 320.09 [320.06] is inappropriate |
| % Paramins GPC Determination |
| % Independent Determination of Mn and Mn Values for Paramins |
| Affidavit of Fred W. Billmeyer, Jr. " Joint Document Book, Vol. 1, Tab 31 |
| 11. The portions of that affidavait that are important to the issue here are the last topic, set out in paragraphs 42 to 53 forming the basis for the conclusions expressed in paragraph 54. |
| (8) Billmeyer Affidavit (paras. 42 to 54) |
| 12. Imperial"s only expert witness critical of Dr. Billmeyer"s work, Dr. Knox, clearly understood the procedure followed by Dr. Billmeyer and had no difficulty in following what he did in using an "effective calibration" or "curve shifting procedure". He stated this in paragraphs 11 and 12 of his first Affidavit: |
| 11. Dr. Billmeyer described in Canada the importance of adopting and executing the correct calibration procedure for the Waters GPC. As described to Mr. Justice Cullen, Dr. Billmeyer carried out the calibration of the Waters 200 instrument by what may be termed the "effective calibration" or the "curve shifting procedure". This methodology yields a calibration curve which is required in order to convert the output of the GPC into Mn (number average molecular weight), Mw (weight average molecular weight) and Mw/Mn (polydispersity) values. In this procedure an initial calibration curve is obtained using fractionated PIB standards having a narrow Molecular Weight ("MW") range. There are a number of well-known reasons why such a calibration curve may not give correct Mn values when applied to samples having a wide MW range (large Mw/Mn values). The initial curve may therefore have to be adjusted in such a way that the final curve gives the true Mn values of specifically chosen unfractionated "anchor samples" and whose Mns are known, and whose MW ranges are similar to those of the unknown to be tested. |
| 12. The calibration procedure as described by Dr. Billmeyer to this Canadian Court thus involved an initial calibration using narrow MW range standards followed by an adjustment using the two "anchor samples". In contrast to his deposition in Ohio, as discussed below, Dr. Billmeyer swore in Canada that both stages of the calibration procedure were based upon Mn values determined by Vapour Pressure Osmometry ("VPO"). |
| (9) Knox, Exhibit p. 39, paras. 11 and 12 |
| 13. The criticism levelled by Dr. Knox is based on what he states in the earlier part of paragraph 12, above, namely, what appear to him as discrepancies between what Dr. Billmeyer said in the 1990 trial and the Ohio III. A view which he admitted, he expressed as an individual and not as an expert. |
| (10) Knox, Oral Testimony, April 30, 1999, p.24, I. 4 " p.26, I. 14 (Tab 14) |
| 14. Dr. Knox conceded that Dr. Billmeyer was a very skilled and able scientist who fully understood what he was doing. |
| (10) Knox, April 30, 1999, page 8, I. 17 to p. 9, 1. 7 |
| 15. Similarly, Dr. Rudin, the Lubrizol expert, in reviewing Dr. Billmeyer"s evidence, understood that Dr. Billmeyer had set out to construct an "empirical calibration" and that he "had successfully accomplished his task", that he was "scientifically sound" and had "properly reported" his results. |
| (11) Rudin (1st Affidavit) Ex. D-60, paras 12 and 13 |
| 16. As stated by Gordon Henderson: |
| "The judge, who rarely will have any training in engineering or science as a profession, will need to come to terms with this technology and to decide on a judicious basis, as between competing experts, who is right or who is wrong." |
| (2) Henderson, op. cit.40 |
| 17. At the 1990 trial, there was no competing expert challenging Dr. Billmeyer. Imperial, simply filed "certificates" provided by Exxon as to Mn without any proof as to their truth. Justice Cullen commented on this failure by Imperial and affirmed the credibility of Dr. Billmeyer in his reasons. |
| (12) Reasons, Cullen J. " Joint Book, Vol. 1, Tab 9, pp. 30-31 |
| 18. Imperial now complains that, had Dr. Billmeyer disclosed more about the effective or empirical method he used, they "might" have adduced some evidence that "might" have influenced the Trial Judge. |
| 19. There is no doubt that scientists such as Dr. Knox and Dr. Rudin knew what Dr. Billmeyer had done and how he did it. They could have provided Imperial, if asked, evidence, or assistance as to running their own tests or probing to whatever degree they thought adequate concerning what Dr. Billmeyer had done. In the 1990 trial Imperial called Drs. Kennedy and Porter as expert witnesses for its case. Neither of them attempted to contradict Dr. Billmeyer. Exxon, in its "Jihad of litigation" around the world, had countless experts at its disposal none of whom came forward to contradict Dr. Billmeyer. |
| 20. Skilled lawyers retained by Imperial now seek to pursuade [sic] this the [sic] Court that had they had a disclosure of get [sic] more "maybe" the Trial Judge "might" have done something different on the issue of infringement. |
| 21. Dr. Billmeyer in the 1990 trial, clearly did not refuse to answer any question put to him or to produce any document requested of him at trial. Imperial"s counsel had ample opportunity to ask whatever questions and request whatever documents in order to fill out whatever detail Imperial thought the Court may require. |
| 22. From Dr. Billmeyer"s point of view, how much more was needed? Nothing as far as a scientist was concerned. |
| 23. From the Trial Judge"s point of view how much more was needed? Nothing, since Imperial failed to produce any expert witness giving any opinion whatsoever to contradict Dr. Billmeyer, nor were any tests performed to verify or contradict those of Dr. Billmeyer. |
| 24. From a practical point of view how much more from Dr. Billmeyer is enough? Just as a judge does not need to voice all his or her hesitations and all verifications before he or she writes the final decision, neither does a scientist in giving evidence. |
| (13) Mailhot "the Fact Finding Judicial Process and Solomon"s Story Revisited", Cambridge Lectures 1989, Les Éditions Yvon Blais Inc. 87 at 92-95. |
| 25. It was not fraudulent for Dr. Billmeyer to give his evidence as he did. |
Conclusion
[149] I have found, on the merits, that Imperial has not proved its grounds of fraud against Lubrizol and Dr. Billmeyer. I am satisfied, on the whole of the evidence, that any discrepancies which exist between Dr. Billmeyer"s evidence in the United States and his Canadian work and evidence, do not result from an attempt on his part to mislead the Court. Imperial"s Rule 1733 motion will therefore be dismissed.
[150] There cannot be any doubt, in view of my findings, that Lubrizol is entitled to its costs. The only issue is on what basis and how much. With respect to Dr. Billmeyer, the issue is whether, as an intervenor, he should get costs and, if so, like Lubrizol, on what basis and how much. I invite counsel to contact me through the Registry by July 7, 2000, in order to fix a hearing date to determine these issues.
JUDGE
O T T A W A, Ontario
June 12, 2000.
__________________
1 Lubrizol Corp. et al. v. Imperial Oil Ltd. et al, 39 F.T.R. 161.
2 Former Rule 1733 is now Rule 399(2) of the Federal Court Rules, 1998 , which reads as follows: 399 (2) On motion, the Court may set aside or vary an order (a) by reason of a matter that arose or was discovered subsequent to the making of the order; or (b) where the order was obtained by fraud.
These proceedings were commenced on January 2, 1997, by way of a notice of motion brought pursuant to Rule 1733. By order of Cullen J., dated June 9, 1997, a trial of the issues was ordered and, accordingly, Imperial filed a statement of claim on June 23, 1997.
3 The patent named Norman Meinhardt and Kirk Davis as inventors. This is why the patent is sometimes referred to as the Meinhardt patent.
4 The Court of Appeal"s decision, to which I shall be referring shortly, is reported at 45 C.P.R. (3d) 449.
5 Mn stands for number average molecular weight. The Mn of a mixture of molecules is the total mass of the sample divided by the number of molecules in the sample.
6 Mw stands for weight average molecular weight. In calculating the weight average molecular weight, the weight of all molecules in a polymer sample having a given molecular weight is multiplied by that molecular weight, and the sum of these products is divided by the weight of the entire sample. Mw/Mn is the quotient obtained by dividing weight average molecular weight by number average molecular weight.
7 VPO stands for vapour phase osmometry. I have attached as Appendix B to my Reasons a copy of paragraphs 18 to 23 of Dr. Billmeyer"s affidavit, where he explains VPO and gel permeation chromatography (GPC), the two methods by which Mn and Mw can be determined.
8 The issues relating to the validity of the patent are not relevant for the determination of the issues before me.
9 At the relevant time, Exxon owned 69.6% of the shares of Imperial, a publicly traded company.
10 This patent, sometimes referred to as the Meinhardt patent, is, for all intents and purposes, identical to the Canadian patent.
11 This can be found at pages 9, 10 and 11of these Reasons.
12 In referring to Mr. Levy"s evidence, the trial judge mistakenly referred to a part of the transcript which was not relevant to the issue under discussion. That point was raised before the Court of Appeal, which agreed with Lubrizol"s submission that Cullen J. "clearly intended to recite questions 6109, 6140 and 6141 (Appeal Book, Vol. 4, pp. 267-8)". [Court of Appeal decision, pp. 469-470].
13 The "method of the patent" is an issue which Imperial has raised in these proceedings. I will deal with that issue later on in my Reasons.
14 Rule 59.06(2) of the Ontario Rules of Civil Procedure provides as follows: Setting Aside or Varying: (2) A party who seeks to, (a) have an order set aside or varied on the ground of fraud or of facts arising or discovered after it was made; (b) suspend the operation of an order; (c) carry an order into operation; or (d) obtain other relief than that originally awarded, may make a motion in the proceeding for the relief claimed.
15 Reference: Trial Evidence, May 16, 1990, pp. 184-185, Ex. P-15, J.B. Vol. 2, Tab 3, pp. 659-660.
16 Ex. P-15, Tab 3 Billmeyer Ev. p. 656; Deeth Ev. Vol. 17 pp. 31-32.
17 Ex. P15, Tab 3 pp. 657-59, esp. at p. 659.
18 Knox Ev. Vol. 12 pp. 209-211.
19 Ex. P18 Tab 5, Billmeyer U.S. Deposition Vol. 5 pp. 1441-1443.
20 Trial Evidence, May 16, 1990, p. 39, Ex. P-15, J.B. Vol 2, Tab 3, p. 514.
21 Ex. P-18 Tab 5, pp. 1451-1452 (U.S. Transcript, pp. 749-750).
22 Ex. P-16, Tab 3, Billmeyer Comm. Ev. pp. 124-126 (p. 915).
23 Ex. P-18, Tab 4, pp. 1404-1405 (Deposition transcript, pp.s 628-629).
24 Ex. P-15, Tab B1 Billmeyer affidavit p. 351 para. 45.
25 Ex. P-15 Tab B3 Billmeyer Trial Ev. pp. 498-499.
26 Ex. P-15 Tab B3 Billmeyer Ev. p. 588.
27 Ex. P-15 Tab B3 Billmeyer Ev. pp. 633-634.
28 Ex. P-15 Tab B3 Billmeyer Ev. p. 636.
29 Ex. P-15, Tab B3, pp. 569-570.
30 Commission Ev. Vol. 3 pp. 51-52.
31 This is the Amoco H-1500 anchor sample.
32 Trial Ev. Vol. XI pp. 52-53.
33 Trial Ev. Vol. XII pp. 42-43.
34 Trial Ev. Vol. XIII p. 15 / Vol. XIV pp. 101 and 123.
35 Trial Ev. Vol. XXI pp. 28, 37-37.
36 Ex. P-15, p. 614.
37 Ex. P-15, Tab B-3, Trial Ev. p. 489-490.
38 Trial Ev. Vol. XI pp. 169-170.
39 Lubrizol Closing Arguments, p. 114.
40 Gordon Henderson "Patent Law of Canada", 1994 Carswell, Toronto pp. 2-3.