MEMORANDUM AND ORDER

WILLIAM K. THOMAS, District Judge.

Westwood Chemical, Inc. is the owner of United States Patents No. 2,742,378 (hereafter '378) entitled "Fillers Having Vinyl Siloxane Groups Bonded to the Surface Thereof and Copolymers with Ethylenically Unsaturated Polymerizable Monomers," and No. 2,841,566 (hereafter '566) entitled "High Polymers with Chemically Bonded Reinforcing and Method of Making Same." Theodore A. TeGrotenhuis is inventor and assignor to Westwood.

Charging infringement of these patents, Westwood Chemical, Inc. (hereafter Westwood) filed seven actions in this court. It sued Ferro Corporation in 1962 and Molded Fiber Glass Body Company (hereafter Molded Fiber Glass) in 1963. It is this latter action that is before the court. A joint infringement action was brought against Owens-Corning Fiberglas Corporation and Johns-Manville Fiber Glass, Inc. in 1963 (later severed into separate actions against each corporation). In 1967 three more infringement actions were filed against Certain-Teed Products Corporation, Union Carbide Corporation, and Dow-Corning Corporation.

Owens-Corning, Johns-Manville, Ferro, and Certain-Teed Products supply Molded Fiber Glass with fiber glass treated with hydrolyzable silanes, principally vinyl silanes (herein called coupling agents). The treated fiber glass is chemically bonded to polyester resins throdgh in situ copolymerization of the coupling agents with unsaturated monomers in the polyester resins. Composite articles of reinforced polyester, e. g. auto bodies or boat hulls are thus formed. Union Carbide and Dow-Corning manufacture the silane coupling agents used by Owens-Corning, Johns-Manville, Ferro, and Certain-Teed.

The present proceeding, stemming from Molded Fiber Glass's motion for summary judgment, raises the affirmative defenses of res judicata and collateral estoppel, based on adjudication of the Owens-Corning case adverse to the plaintiff.

Initially the action against Molded Fiber Glass was stayed pending disposition of the Ferro case. Subsequently, plaintiff moved to vacate that stay order and to consolidate all of its seven infringement actions for trial. Thereafter, upon entering a stipulation with Owens-Corning, Ferro, and Johns-Manville, and later Certain-Teed, plaintiff Westwood withdrew its motion for consolidation. Under the stipulation the Owens-Corning case was to be tried first and its "final decision (including appeals) . . . shall be final and binding upon all parties hereto, the same as if each defendant were a party to said case."

Following a trial in June and early July of 1969, Judge James C. Connell of this court found both patents invalid for obviousness under 35 U.S.C. § 103 and further under 35 U.S.C. § 102(a)(g) and (f). His findings of fact and conclusions of law are reported at 317 F. Supp. 201 (N.D.Ohio 1970). On appeal the Court of Appeals upheld Judge Connell's finding and conclusion of invalidity based on obviousness under Section 103. The Court found it unnecessary to consider the other grounds of invalidity or Judge Connell's judgment of infringement, see 445 F.2d 911 (6 Cir. 1971).

In affirming Judge Connell's judgment, the Court of Appeals for the Sixth Circuit restricted the determination of the invalidity of the two patents to claims "directly in issue" at trial; namely, claims 1, 3, 4, 6-8, 11, 14-19, 21, 23-26, and 28 of the '378 patent and claim 8 of the '566 patent. Thereafter, the Court of Appeals denied plaintiff's petition for rehearing. On February 22, 1972, petition to the Supreme Court for a writ of certiorari was also denied. The Court of Appeals thereupon issued its mandate which was filed in this court on March 3, 1972.

Upon the final decision in the Owens-Corning case judgments were entered in favor of Ferro, Certain-Teed, and Johns-Manville, dismissing these actions in accordance with the foregoing stipulations. The plaintiff appealed these judgments. On May 4, 1973, the Court of Appeals affirmed these judgments of dismissal; its opinion concludes with this language:

If it was Westwood's intent to reserve certain patent claims to assert against the remaining defendants, it should have said so. Clearly, none of the parties contemplated having the issue of validity tried piecemeal, which is the interpretation urged by Westwood. 477 F.2d 1160, 1164.

I.

The invention defined in the counts relates to certain solid "composite articles" comprising textile fibers and a polymeric resin, and to a method of making the articles. The articles are said to possess superior cohesive properties because an organosilicon compound is used as a sort of "molecular cement" between fibers and resin. In other words, each silicon compound molecule is attached by chemical bonds both to a fiber and to the resin which surrounds it.

The opinion then explains how the coupling agent is bonded to the fibers by hydrolysis.

In making the articles, it is recognized that a plurality of hydroxyl groups is normally present on the surface of a textile fiber. Certain "hydroxyl-reactive" organosilicon compounds with one or more carbon-to-carbon double bonds, for example, vinyl silicon trichloride, are allowed to react with the fibers whereby the fiber hydroxyl groups, fiber-OH, are transformed to fiber siloxane groups, fiber-O-$i-, in which at least one free bond to each silicon atom is connected to an organic group with carbon-to-carbon unsaturation.

This carbon-to-carbon unsaturation of the chemically bonded fibers permits interpolymerization.

These modified fibers are then contacted with an olefinically unsaturated liquid, for example, styrene, and the liquid is polymerized to the solid state. The unsaturated groups attached to the modified fibers interpolymerize with the liquid. Thus a plurality of chains comprising a silicon atom and two or more carbon atoms bind fibers and surrounding resin together in the finished composite article.

More briefly but accurately, Judge Connell and the Court of Appeals defined the subject matter of the invention. Judge Connell found:

The substance of this action relates to glass fibers treated with unsaturated organo-silanes for use in reinforcing polyester resins and the resulting polyester resin bodies. 317 F. Supp. at 202.

The Court of Appeals stated:

The substance of the invention claimed by its inventor, T.A. TeGrotenhuis, relates to treating fillers, pigments and fibers with unsaturated organo-silanes for use in reinforcing polyester resins and the resulting polyester resin bodies. The invention claimed by patent '378 is the use of certain hydrolyzable silane chemicals having vinyl groups as coupling agents to reinforce the polyester resins. Patent '566 makes the claim of using hydrolyzable silane chemicals having alkenyl groups other than vinyl, specifically allyl silanes, as the coupling agent. Generally, the claimed objective of both inventions was to provide a resin, such as plastic or glass fibers, with greater strength and water resistance by treatment with the hydrolyzable silane chemicals. 445 F.2d at 912.

The present action charging infringement of the invention of patent-in-suit '378 was stayed pending final adjudication of Westwood Chemical, Inc. v. Owens-Corning Fiberglas Corp. On May 17, 1972, defendant Molded Fiber Glass moved to amend its answer by adding the affirmative defenses of res judicata and collateral estoppel. Leave was later granted. Five days later, on May 23, 1972, defendant Molded Fiber Glass filed its motion for summary judgment. It predicates its motion upon:

(A) Collateral estoppel arising out of "the decisions and judgments rendered by this court and the Court of Appeals in Westwood Chemical, Inc. v. Owens-Corning Fiberglas Corp., 317 F. Supp. 201, mod. and aff'd, 445 F.2d 911," in which the TeGrotenhuis patent No. 2,742,378, which is involved here, Was "held invalid and not infringed as to all claims there in issue," and (B) res judicata arising out of the judgments in the Owens-Corning case, the Ferro case, the Certain-Teed Products case and the Johns-Manville case and "privity which exists in this defendant and the defendants in said action."

With respect to the defense of collateral estoppel the defendant relies on Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, 402 U.S. 313, 91 S.Ct. 1434, 28 L.Ed.2d 788 (1971). In Triplett v. Lowell, 297 U.S. 638, 56 S.Ct. 645, 80 L.Ed. 949 (1936) the Supreme Court had held that a determination of patent invalidity is not res judicata against the patentee in subsequent litigation against a different defendant. Blonder-Tongue overruled Triplett "to the extent that it forecloses an estoppel plea by one facing a charge of infringement of a patent that has once been declared invalid."

Blonder-Tongue does not suggest that a plea of estoppel by an infringement or royalty suit defendant must automatically be accepted once the defendant in support of its plea identifies the issue in suit as the identical question finally decided against the patentee or one of his privies in previous litigation. Rather, the patentee-plaintiff must be permitted to demonstrate, if he can, that he did not have "a fair opportunity procedurally, substantively and evidentially to pursue his claim the first time." Eisel v. Columbia Packing Co., 181 F. Supp. 298, 301 (D.Mass. 1960).

Critically material to the final determination of this case the opinion then declares:

Determining whether a patentee has had a full and fair chance to litigate the validity of his patent in an earlier case is of necessity not a simple matter. In addition to the considerations of choice of forum and incentive to litigate mentioned above, certain other factors immediately emerge. For example, if the issue is nonobviousness, appropriate inquiries would be whether the first validity determination purported to employ the standards announced in Graham v. John Deere Co., [ 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966)]; whether the opinions filed by the District Court and the reviewing court, if any, indicate that the prior case was one of those relatively rare instances where the courts wholly failed to grasp the technical subject matter and issues in suit; and whether without fault of his own the patentee was deprived of crucial evidence or witnesses in the first litigation.

The paragraph concludes:

But as so often is the case, no one set of facts, no one collection of words or phrases, will provide an automatic formula for proper rulings on estoppel pleas. In the end, decision will necessarily rest on the trial courts' sense of justice and equity. 402 U.S. at 332-334, 91 S.Ct. at 1445.

In its written response to the claimed defense of collateral estoppel plaintiff asserts that this defense does not apply because in its present infringement action against defendant Molded Fiber Glass, plaintiff Westwood relies on claims of its '378 patent not adjudicated in the Owens-Corning case. Moreover it attacks the collateral estoppel defense spelled out in Blonder-Tongue, supra. Plaintiff primarily relies on the "second of the two exceptions provided by the Supreme Court." It states that

It will show to this court that the prior courts wholly failed to grasp the technical subject matter and technical issues in the suit, and that he was deprived of crucial evidence and witnesses for this point in the first litigation.

Supporting its motion plaintiff has submitted an affidavit of Dr. Tito T. Serafini, who states that he is "now head of the section dealing with reinforced plastics (whether rubbery or rigid) at the National Aeronautics and Space Administration, Lewis Research Center." Hired by counsel for the plaintiff, he explains that he was "engaged to review the court's opinions, testimony, the exhibits, prior art patents, and the patents of TeGrotenhuis in the case of Westwood Chemical, Inc. v. Owens-Corning Corporation." He continues:

I was charged with the task of determining whether or not testimony presented in the case was true or false, and whether or not the opinions of the court demonstrated an understanding of the technical aspect of the patents in suit and their relation to the prior art.

His affidavit principally contradicts testimony of Dr. Eugene Rochow, professor of silicon chemistry at Harvard University, who appeared as an expert witness for Owens-Corning in the previous trial before Judge Connell. In part, Dr. Serafini's affidavit states:

My study of all the prior patents (patents filed before TeGrotenhuis' date, March 30, 1945, which was granted by the Court of Customs and Patent Appeals) shows that none of these prior patents discloses the combination of reactants required to obtain a permanent chemical bond between any surface and a high polymer, whether it be a resin, a rubber, a drying oil, etc.

Contrary statements by certain witnesses of Owens Corning are absolutely false. The untruthfulness and incredibility of those statements would be apparent to any unbiased chemist with relatively meager training in organic chemistry. For example, I have found considerable false and misleading testimony by defendant's witness, Dr. Rochow, professor at Harvard University.

In a related case in the United States Court of Claims, the United States Government filed a supplemental affidavit of Dr. Serafini dated November 16, 1972. Referring to his previous characterization of Dr. Rochow's testimony in the Owens-Corning case as "false," Dr. Serafini states:

The suggestion that the statements referred to in the affidavit amounted to false testimony did not originate with me but with Mr. TeGrotenhuis.

Dr. Serafini's affidavit continues:

I adopted the affidavit attached to plaintiff's brief as my own without fully appreciating the differences between a false statement and a scientific conclusion upon which chemists might disagree with one another. I considered certain statements or conclusions appearing in the Owens-Corning transcript to constitute false testimony where I did not feel there was support in the chemical literature for the statements or conclusions. Upon further reflection, I feel that I would have more accurately expressed my opinion in the earlier affidavit by using the term "not scientifically established" in the place of the term "false."

He concluded his second affidavit by stating:

While I do not agree with many of the scientific statements appearing in the transcript or in the court opinions in the Owens-Corning case, I must concede there is room for a difference in scientific opinion on these matters among chemists.

Dr. Serafini's affidavit is considered as part of the record in the present case, along with the motions of the parties, oral arguments made at the hearing of July 17, 1972, ensuing depositions and answers to interrogatories which plaintiff propounded to defendant, the oral arguments and oral testimony of chemical experts at the hearings of February 14, 1973 and April 27, 1973; also the entire record in the Owens-Corning case has been reviewed.

Upon this total record this court makes its final rulings in this case only after it believes that it has sufficiently mastered the applicable essentials of the subject matter to understand the issues. Therefore, it respectfully overrules plaintiff's motion that the court appoint a technical advisor. Since the closing of the evidentiary record both sides have filed proposed findings of fact and conclusions of law. At all stages of these proceedings counsel have filed numerous briefs, the last of which was filed on August 21, 1973.

II.

The defense of res judicata will first be considered and decided. Both in their briefs and in their arguments at the oral hearing of July 17, 1972, counsel for the defendant, in pressing the res judicata defense, relied primarily on Kessler v. Eldred, 206 U.S. 285, 27 S.Ct. 611, 51 L.Ed. 1065 (1907). Both Kessler and Eldred manufactured electric cigar lighters. Eldred, who owned the Chambers patent, sued Kessler for infringement. Kessler prevailed, a decree of noninfringement being issued in Kessler's favor. Subsequently, Eldred charged a vendee-merchant of Kessler's lighters with infringement of Eldred's patent. Kessler sought to enjoin Eldred from prosecuting any of Kessler's customers for alleged infringement of the Chambers patent. Further explaining Kessler v. Eldred, Rubber Tire Co. v. Goodyear Co., 232 U.S. 413, 417, 34 S.Ct. 403, 405, 58 L.Ed. 663 (1914), states:

Kessler, being defeated in the circuit court, appealed to the Circuit Court of Appeals for the Seventh Circuit. Answering questions certified by that court, this court held that the decree in the suit of Eldred v. Kessler had the effect of entitling Kessler to continue the business of manufacturing and selling throughout the United States the same lighter he had theretofore been manufacturing and selling, without molestation by Eldred through the patent which he held; and that the decree also had the effect of making a suit by Eldred against any customer of Kessler for the alleged infringement of the patent by use or sale of Kessler's lighters a wrongful interference with Kessler's business, with respect to which he was without adequate remedy at law.

Based on Kessler and other cases cited at the oral hearing, counsel for defendant Molded Fiber Glass urged "that under well established principles of res judicata, Westwood is barred from asserting against Owens-Corning or any of Owens-Corning's customers any claim which was asserted or any claim which could have been asserted against Owens-Corning in the earlier case." Counsel extended the same argument to embrace the other suppliers of Molded Fiber Glass, all of whom, defense counsel asserted, "have been exonerated by the judgments and stipulations which have been entered in the litigation which I have described."

Counsel for the plaintiff, on the other hand, distinguishes Kessler v. Eldred from the present case. In a dialogue at the July 17, 1972 hearing, the court asked counsel for plaintiff:

The Court: But the Supreme Court says if the manufacturer has been exonerated, then the customer is likewise exonerated.

Mr. TeGrotenhuis: I agree, but this manufacturer — the customer is not the customer of the product that was exonerated.

The Court: Well, in other words, you are saying there has to be the precise product, and then this case applies?

Mr. TeGrotenhuis: Yes; as I pointed out before, if Molded Fiber Glass bought the tanks from Owens-Corning, they could use them and sell them and I can't do a thing; but they are not buying the composite articles. They're only buying material as one supplier. . . .

Though he did not then cite Rubber Tire Co. v. Goodyear, supra he did so in a post-hearing brief. It is evident in retrospect that in this dialogue with the court, plaintiff's counsel was correctly arguing Rubber Tire's restriction of the breadth of Kessler. Rubber Tire Company owned the Grant patent which comprised (1) a channel or grove with tapered sides, (2) a rubber tire adapted to fit into the channel, and (3) independent retaining wires which fasten the elements in a particular position. Goodyear produced a similar tire. Sued for infringement of the Grant patent, Goodyear obtained a deYcree of invalidity in the Sixth Circuit. Goodyear Tire Rubber Co. v. Rubber Tire Wheel Co., 116 F. 363 (1902). (Eventually, in a further infringement action involving other parties, the Second Circuit upheld the patent's validity and the Supreme Court affirmed Diamond Rubber Co. v. Consolidated Rubber Tire Co., 220 U.S. 428, 31 S.Ct. 444, 55 L.Ed. 527 (1911)).

Goodyear also produced rubber which it sold to A. Doherty, who maintained a shop in New York City where he assembled rubber-tired vehicle wheels by combining rubber purchased from Goodyear, and wire and channel obtained from other parties.

Rubber Tire sued Doherty for infringement in the Circuit Court for the Southern District of New York. Goodyear then instituted a suit in the Southern District of Ohio to restrain Rubber Tire Company from prosecuting suits for infringement against Goodyear's customers. The trial court granted a preliminary injunction which was sustained on appeal to the Sixth Circuit. On certiorari the Supreme Court reversed, Rubber Tire Co. v. Goodyear Co., supra, 232 U.S. at 418-420, 34 S.Ct. at 405 distinguishing Kessler v. Eldred:

ash; as an article of lawful commerce, and it continues only so long as the commodity to which the right applies retains its separate identity. If that commodity is combined with other things in the process of the manufacture of a new commodity, the trade right in the original part as an article of commerce is necessarily gone.

could demand protection for its trade rights in the commodities it produced. But it had no transferable immunity in manufacture. The decree gave it no privilege to demand that others should be allowed to make and sell the patented structure in order that it might have a market for its rubber. 232 U.S. at 419, 34 S.Ct. at 405.

Applying Rubber Tire to the present facts, this court finds that Owens-Corning can protect its trade rights under its decree of invalidity of the TeGrotenhuis patents. However, Owens-Corning cannot transfer this trade right to Molded Fiber Glass Body Company. Furthermore, the trade rights of Owens-Corning continue only so long as the fiber glass treated with the coupling agent "retains its separate identity." Thus, when Molded Fiber Glass combines the treated fiber glass with other things in the manufacture of reinforced plastics, "the trade right in the original part as an article of commerce is necessarily gone." Moreover, since Molded Fiber Glass does sell treated fiber glass in the same form in which it is purchased from Owens-Corning, this court need not determine whether a vendor-vendee relationship establishes privity under traditional res judicata principles.

It is, therefore, concluded that this court erred in its conditional ruling rendered at the oral hearing of July 17, 1972, when it determined that "under all of these principles of res judicata . . . as a matter of law the defense is a valid and proper one." The defendant's motion for summary judgment based on the affirmative defense of res judicata is denied.

III.

Another hearing was held on February 14, 1973. Preparatory to this further hearing this court wrote counsel, in part, as follows:

The parties are requested to be ready to present testimony that will define the chemical terms as used in the unadjudicated claims. These claims appear to be, limited necessarily to patent '378, claims 2, 5, 9, 10, 12, 13, 20, 22, 27. Expert testimony is also sought that will aid the court in determining whether one or more of the unadjudicated claims are embraced within, or separate and distinct from, the inventions measured by the adjudicated claims.

The letter recalled:

In the hearing dialogue of July 17, 1972, the court and counsel discussed whether or not testimony should be taken on the subject of the "similarities or differences of these several claims" (R. 30, 72) characterizing it as "a rather mixed question of law and fact."

The paragraph then stated:

Research confirms the tentative conclusion that the subject is a "mixed question of law and fact." See Indiana General Corp. v. Lockheed Aircraft Corp., 408 F.2d 294 (9th Cir. 1968). . . .

Counsel was then informed in the next paragraph:

The adjudication of this separate issue on the merits is directed pursuant to Rule 42(b) FRCivP in conjunction with the defenses of res judicata and collateral estoppel (under Westwood Chemical, 445 F.2d 911) but particularly in connection with collateral estoppel. Irrespective of the aptness of the defense of res adjudicata . . . it is concluded that Blonder-Tongue Laboratories, Inc. applies to the claims adjudicated in 445 F.2d 911; and its doctrine of collateral estoppel therefore would also apply to any unadjudicated claim that is embraced within an adjudicated claim.

The last paragraph enlarged the scope of the proceedings from that time on. No longer was the issue concerning the affirmative defense of collateral estoppel restricted to a Rule 56 (summary judgment) question. Thereupon the proceedings became, and have since continued to be, an adjudication on the merits of the separate issue of whether or not, weighing all the facts, any of the unadjudicated claims of patent-in-suit '378 is embraced within any of the adjudicated claims. The concept of one claim being embraced within another claim is equated with language from Shatterproof Glass Corp. v. Guardian Glass, 462 F.2d 1115, 1125 (6 Cir. 1972) as a situation in which unadjudicated claims are "merely restatements of adjudicated claims, or . . . they contain no significant variations therefrom."

Further analysis confirms the correctness of the conclusions that Blonder-Tongue's doctrine of collateral estoppel applies "to any unadjudicated claim that is embraced within an adjudicated claim." Blonder-Tongue, supra, 402 U.S. at 333, 91 S.Ct. at 1445 indicates that a defendant, asserting a collateral estoppel plea, must identify "the issue in suit as the identical question finally decided against the patentee or one of his privies in previous litigation." It is noteworthy that the Court is here speaking of the necessity of the defendant identifying "the issue in suit as the identical question finally decided . . . in previous litigation." Therefore, when "the identical question" involved in the adjudication of one claim of a patent is likewise involved in the construction of an unadjudicated claim the defense of collateral estoppel applies to the unadjudicated claim as well as the adjudicated claim. In arriving at this teaching of Blonder-Tongue, significance is attached to the point that the Court speaks of "the identical question," and not the identical claim or claims of the patent-in-suit.

The objectives of Blonder-Tongue are several. Restriction of repeated, and sometimes conflicting, adjudications in different courts of the validity of the same patent is intended to avoid the costly economic consequences of adhering to the mutuality of estoppel principle of Triplett v. Lowell, supra and to save "some judicial time if even a few relatively lengthy patent suits may be fairly disposed of on pleas of estoppel." Blonder-Tongue also recognizes "that the uncritical acceptance of the principle of mutuality of estoppel expressed in Triplett v. Lowell is today out of place." While he seeks to accomplish these goals, the second court is nonetheless directed by Blonder-Tongue to act with a "sense of justice and equity." In exercising this balancing act the second court necessarily must allocate and quantify the burdens of proof between the parties.

Without expressly labeling the burdens of proof, Blonder-Tongue nevertheless distributes these burdens between the defendant, who asserts the defense of collateral estoppel based on a previous finding that the patent-in-suit is invalid, and the patentee-plaintiff, who is suing the defendant as the alleged infringer. The defendant has the burden of identifying "the identical question finally decided against the patentee . . . in previous litigation." On the other hand, the patentee-plaintiff must "demonstrate, if he can, that he did not have a fair opportunity procedurally, substantively, and evidentially to pursue his claim the first time."

The inquiry cannot stop here. Application of these opposing burdens of proof requires that each burden be quantified. Though Blonder-Tongue is silent on this matter, this void must be filled. Unless it is understood what amount of proof is needed to sustain these opposing burdens, it is impossible for "the court in the second litigation [to] decide in a principled way whether or not it is just and equitable to allow the plea of estoppel in the case before it," Blonder-Tongue, supra, 402 U.S. at 334, 91 S.Ct. at 1445.

Impelled by these considerations it is determined that the defendant who asserts the plea of estoppel must prove by a preponderance of all the evidence that the action in which he is charged with infringement presents "the identical question finally decided against the patentee . . . in previous litigation." This "identical question" is easily identified if the entire patent (including all its claims) was in issue and declared invalid in the first litigation, see Blonder-Tongue, remand opinion of Judge Hoffman, 334 F. Supp. 47, 51.

However, where, as here, certain claims of the patent-in-suit were not adjudicated in the first litigation, a plea of estoppel would nevertheless be established if the defendant is able to prove by a preponderance of all the evidence that "the identical question" is contained in the adjudicated and unadjudicated claims.

Once the defendant has established his burden the plea of estoppel will terminate the second litigation unless the patentee-plaintiff can then prove his burden which only then arises. However, the burden then imposed upon the patentee-plaintiff is not the advantageous presumption of patent validity that he possessed on the first trial. Blonder-Tongue, supra makes no suggestion that the second district judge should retry the case. Indeed, United States v. Utah Constr. Co., 384 U.S. 394, 422, 86 S.Ct. 1545, 1560, 16 L.Ed.2d 642 (1966) a case that applied collateral estoppel to an issue resolved in an administrative proceeding, held that there was

neither need nor justification for a second evidentiary hearing on the matters already resolved as between these two parties.

Nor can the Supreme Court contend that the second district judge should sit as an appellate court with the power to reverse the previous judgment if plain error is present. In Blonder-Tongue on remand, 334 F. Supp. at 50, Judge Hoffman gives a general direction for the second court who is asked to deny a collateral estoppel defense. He rules out mere disagreement with the first court —

. . . this goal [elimination of "repetitive litigation of the same patent"] cannot be achieved if mere disagreement in the second court, supervenes the defense.

From this he says

It follows that the second court must defer unless it appears from the face of the prior opinions [necessarily considered in light of the case record] that the first court completely missed the point.

whether the opinions filed by the District Court and the reviewing court, if any, indicate that the prior case was one of those relatively rare instances where the courts wholly failed to grasp the technical subject matter and issues in suit.

The application of this test must be manageable. To defeat collateral estoppel the prior courts' opinions must reveal, when tested against the record of that case (as supplemented by additional evidence if justice and equity warrant it), that the prior courts "wholly failed to grasp the technical subject matter and issues" in the prior suit. The substantial evidence test, the standard in judicial review of administrative proceedings is determined to be both appropriate and fair to employ here. If the total record contains substantial evidence (i. e., evidence upon which reasonable minds may differ) to support the prior courts' judgment of invalidity, as expressed in the prior opinions, then it has not been demonstrated that the courts "wholly failed to grasp the technical subject matter and issues in suit." But if the total record lacks such substantial evidence then the patentee-plaintiff should not be estopped by the prior adjudication of patent invalidity.

IV. A.

At the hearing of February 14, 1973, before any testimony was taken the question arose as to which side should proceed first. This court had not then resolved the burden of proof questions, as it now has. However, the order of proof there followed, with defendant's expert testifying first, conforms to the burden now imposed on the defendant. To establish its right to the defense of collateral estoppel the defendant has the burden of proving that the unadjudicated claims in issue are embraced within the claims of patent-in-suit '378, adjudicated in the Owens-Corning case.

Early in the hearing Mr. TeGrotenhuis and his counsel announced their voluntary relinquishment of all remaining claims of patent-in-suit '378 other than method or process claims 5 and 10 and composite article or product claims 12 and 13. Likewise, the sole remaining claim of '566 has been relinquished.

To compare the four unadjudicated claims of patent-in-suit '378 that remain in issue and similar adjudicated patent claims of '378, defendant Molded Fiber Glass called as an expert witness Mr. Samuel Sterman, who has a B.Sc. in chemistry and has been employed by Union Carbide Corporation since 1940. Plaintiff Westwood called as an expert witness, Dr. Guido Stempel, with a Ph.D. in chemistry, whose last employment was research center administrator of the General Tire Rubber Co. Mr. Sterman in the last 15 or 20 years has been involved primarily in silicon chemistry with "the specific subject of the synthesis, evaluation of coupling agents in many different composite systems." Under his direction and supervision, and individually, he has made silanes and polyester resins.

Dr. Stempel, after teaching organic chemistry at Carnegie Institute of Technology, went with General Tire Rubber Co. There, among other responsibilities, he supervised the product development laboratory "which included all the plastics" and he has been concerned with "polymerization of polyesters" since 1946.

Claims 5 and 15

Each expert witness compared unadjudicated claim 5 with adjudicated claim 15, the claim regarded by defendant as most similar to claim 5. The texts of these two claims are reproduced side by side:

Claim 5

A method of forming a composite article which comprises wetting an inorganic solid with olefinic polymerizable liquid capable of setting to the solid state and having groups which react with olefine groups, said inorganic solid having attached to the surface thereof vinyl siloxane portions of hydrolysis products of a vinyl halosilane having only vinyl and halogen groups attached to silicon.

A method of forming a composite article which comprises wetting an inorganic solid with an olefinic polymerizable liquid capable of setting to the solid state, and allowing said liquid in contact with said solid to solidify, said inorganic solid normally having hydroxyl groups on the surface thereof but being modified by having attached to the surface thereof vinyl siloxane portions of hydrolysis products of a vinyl halosilane having only vinyl and halogen groups attached to silicon, the vinyl group being attached through a silicon oxygen linkage.

Language in claim 5, which does not appear in claim 15 is single underlined. Language in claim 15 which does not appear in claim 5 is double underlined. Language not underlined is identical.

The phrase "olefinic polymerizable liquid" appears in both claims and the phrase "having groups which react with olefine groups," as seen, appears only in claim 5. The words "olefine" and "olefinic" appearing in these claims are repeatedly referred to by the expert witnesses. Comprehension of the meaning and properties of an olefine becomes essential. Present chemical texts use the word "alkene" interchangeably with "olefine." Referring to "the alkene" (olefine) family of hydrocarbons, "C_nH_2n," one text states:

The alkenes are characterized by the presence of a double bond between two carbon atoms. . . . Compounds in which there are one or more double bonds (or triple bonds) are referred to as unsaturated compounds. The unsaturation here is with respect to valence and denotes a state of affairs where all the valences of the carbon atom have not been used for combination with other elements. The names of unsaturated hydrocarbons end in the termination ene. Thus, alkene is the family name and ethylene (ethene is the I.U.C. name), (C₃H₆) propene, and butene (C₄H₈) are the names of the first three members of the series.

The need for a more extended analysis of the term "olefinic" became especially evident in the following cross-examination of Mr. Sterman by plaintiff's counsel — Mr. TeGrotenhuis, the inventor.

Q I mean that those double bonds in an olefinic compound are not in a benzene ring or a derivative of the benzene ring which would be a phenyl group.

A If we have a phenyl group bonded to the reference group, just that and nothing else, then there is no active polymerizable double bond in that phenyl group.

Q It's not an olefine?

A It's a matter of semantics on whether that's an olefine or not. It's not an active polymerizable olefine.

First not answering either "yes" or "no," Mr. Sterman then volunteered a compound negative pregnant. However, a study of relevant chemical literature indicates that Mr. Sterman's answer should not be read as an affirmative statement that a phenyl group is an olefine. The generally accepted view is that "alkenes (olefines)" include only those carbon-carbon double-bond-containing molecules which belong to the class of hydrocarbons known as "aliphatic" compounds.

Aliphatic compounds are open-chain compounds and those cyclic compounds that resemble the open-chain compounds . . . alkanes, alkenes, alkynes, and their cyclic analogs are all members of the aliphatic class. [Emphasis supplied.]

While the phenyl group about which Mr. Sterman spoke does contain double bonds, the phenyl group is not aliphatic; on the contrary, the phenyl group belongs to the other broad class of hydrocarbons known as the aromatic compounds. The aromatic compounds include "benzene and compounds that resemble benzene in chemical behavior." Morrison and Boyd, at 311.

Chemists distinguish between aliphatic and aromatic hydrocarbons because of the significantly different reactive properties of the two groups:

Aliphatic hydrocarbons . . . undergo chiefly addition and free-radical substitution; addition occurs at multiple bonds, and free-radical substitution occurs at other points along the aliphatic chain. In contrast, . . . aromatic hydrocarbons are characterized by a tendency to undergo ionic substitution. Morrison and Boyd, at 311.

Hence, the "aliphatic" or "aromatic" nature of each molecule will generally determine its reactive properties. In this opinion it will, therefore, be understood that the term "olefine" is used solely to refer to aliphatic compounds which contain reactive carbon-carbon double bonds, and not to aromatic compounds.

It is this inherently reactive property of the olefine carbon-carbon double bonds with a free valence available to react and combine with other elements, to which the following testimony of Mr. Sterman is obviously directed. With reference to the phrase `having groups which react with olefine groups," which appears in claim 5, but not in claim 15, Mr. Sterman testified:

To me, in my opinion, this phrase has no additional meaning because it is inherent in any olefinic polymerizable liquid. All such materials have groups which react with olefine groups. This is inherent in the definition of olefinic polymerizable liquid.

Moreover, as this testimony implies, and as other testimony of Mr. Sterman expressly says, it is his opinion that both claim 5 and claim 15 read only on "addition polymerization" and not on "condensation polymerization."

Dr. Stempel, plaintiff's expert, was of the opinion that claim 5's phrase, "having groups which react with olefine groups," indicates "polymerization working with the olefine groups," that is, by "addition polymerization." But it was his opinion that claim 15 "does not specify the manner of polymerization." He concludes that claim 15's phrase, "olefinic polymerizable liquid" can read on condensation polymerization to a solid state of polyester resins containing nonreactive olefinic groups. He charted two possible condensation polymerization reactions, each containing an unreactive olefinic double bond. However, he concedes that he does not know of anyone "who has used and carried out the method of claim 15 of the '378 patent to commercially make articles embodying any of the olefinic polymerizable liquids shown" on his two charts.

Thus, the issue over the indentity or difference of claim 5 and 15 develops into the question of whether claim 15 reads on condensation polymerization. Both witnesses agree that claim 5 reads only on addition polymerization.

In Organic Chemistry the topic "Polymerization" commences with subsection 8.21, "Free-radical polymerization of alkenes." It introduces the subject by diagramming the formation of polyethylene by the polymerization of many ethylene units treated under pressure with oxygen. It goes on to say:

The formation of polyethylene is a simple example of the process called polymerization: The joining together of many small molecules to make very large molecules. The compound composed of these very large molecules is called a polymer (Greek: poly + meros, many parts). The simple compounds from which polymers are made are called monomers ( mono = one).

The text continues:

The particular kind of polymerization undergone by ethylene, in which many molecules of monomer are simply added together, is called addition polymerization. (Later on, in Sec. 29.5, we shall encounter condensation polymerization, in which monomer molecules combine with the loss of some simple molecules, usually water.) [Sec. 29.5 diagrams the condensation polymerization reactions that form the giant molecules of Nylon 66, Dacron and Glyptal.]

The text then makes this significant point about addition (free-radical) polymerization:

Addition polymerization is one of the most important reactions of alkenes and conjugated dienes alike — and, indeed, of compounds of all kinds that contain carbon-carbon double bonds.

Having in mind that addition polymerization is "one of the most important reactions . . . indeed, of compounds of all kinds that contain carbon-carbon double bonds," it seems evident that the following opening language in both claims 5 and 15 is clearly referring to addition (free-radical) polymerization.

A method of forming a composite article which comprises wetting an inorganic solid with an olefinic polymerizable liquid capable of setting to the solid state, . . . .

In an effort to meet Mr. Sterman's testimony that Claim 15 and Claim 5 state the same invention, Mr. TeGrotenhuis put this question to Mr. Sterman on cross-examination:

Q Assume that we have polyester that has sterically hindered olefine groups and we polymerize that on a substrate. We could form a polymer, could we not?

A By olefine polymerization?

Q No, I didn't say. We polymerize it by condensation.

A Oh. All right.

Q You agree?

A Hypothetically, yes.

Q Now, if we add that substrate here, with a monomeric solid that had vinyl siloxane groups thereon, we would have everything that this claim calls for, wouldn't we?

A Yes, we would.

Part of my problem here is that having read your disclosure and the specification of the patent several times, I found no discussion or mention of this concept [condensation polymerization and steric hindrance] that you have brought in here. And I'm trying to respond properly to the hypothetical statement you are making, and then I try to read the claim in the light of both the hypothetical statements and what I remember of having read the patent very thoroughly, and I see no basis for that in the patent. But I do see that your statement can be interpreted in the light of the claim and I'm sort of torn between the two issues here.

The witness Sterman later again refers to the omission from both the claims and the specification of any mention of either condensation polymerization or steric hindrance.

Mr. Sterman: . . . The language [of claims 5 and 15] in blue lines is identical. Now, this means in Claim 5 that the olefinic polymerizable liquid capable of setting to a solid state could, as you said for 15, be capable of setting to a solid state through condensation polymerization.

Q Yes, agreeable.

A There is no limitation in five that says —

Q Agreeable.

A that is not in the specification. I am bringing it in because it further makes 5 show no significant difference from 15. The condensation polymerization could happen in 5 if you want to assume that, just as it could in 15, and 5 has the phrase that this polymer has groups which react with olefines which are residual olefinic groups which could be the same sterically hindered that are not specified, and this is the same language that's in 15. I see no significant difference.

To attempt to show that claim 15 is broader than claim 5, and thereby distinguishable, the plaintiff contends that claim 15 reads on condensation polymerization and steric hindrance. Yet the polymerization mentioned in claim 15 (and in other claims of the patent) is nowhere identified as condensation polymerization, nor is steric hindrance mentioned in claim 15 or in any other claim. Thus while going beyond the face of claim 15 to support its position, plaintiff moved at the trial to strike that testimony of witness Sterman in which he construes claim 15 in light of the '378 patent specification. That motion was overruled at the hearing of April 27, 1973. Relying on the following legal analysis, this court reaffirms that ruling and hereafter consults the specification to determine whether claim 15 reads on condensation polymerization.

United States v. Adams, 383 U.S. 39, 49, 86 S.Ct. 708, 713, 15 L.Ed.2d 572 (1965) states that it is fundamental that "claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention." After noting this language, Noll v. O. M. Scott Sons Co., 467 F.2d 295, 298-299 (6 Cir. 1972) proceeds to find "the limitation-construction distinction to be unworkable," referring to Graver Tank Mfg. Co. v. Linde Air Prod. Co., 336 U.S. 271, 277, 69 S.Ct. 535, 93 L.Ed. 672 (1949) ( Graver I) and 339 U.S. 605, 608, 70 S.Ct. 854, 94 L.Ed. 1097 (1950) ( Graver II). Chief Judge Phillips reads the two Graver cases as holding that

overly broad claims may not be narrowed to the limitations set forth in the specification, but narrow claims may be broadened to embrace "equivalent" subject matter.

In this case resort to the specification is not made to narrow by its limitations an overly broad claim 15. Indeed, claim 15 has been ruled invalid but not because it was overly broad.

Applying the previously stated general rule of Adams, supra and Tilotson, supra there is apt and ample reason to construe claim 15 in the light of the specification. A word or phrase in a patent claim may be defined in a specification. Thus, the word "pigments," as used in the specification and claims is defined at Column 8, Line 17 of the specification. Similarly, in Column 7, Line 68 the specification of '378 particularizes the type of polymerization intended by the invention.

The free-radical mechanism of polymerization is preferred, but the ionic, or acidic polymerization mechanisms such as utilized by the so-called alfin catalytic process of polymerization or the Friedel-crafts process of polymerization wherein a Friedel-crafts or alfin catalyst or even a sodium catalyst is used in place of the free-radical producing alkyl peroxide is also applicable.

As previously seen, free-radical polymerization is the most common type of addition polymerization. Ionic or acidic polymerization mechanisms are also a subset of addition polymerization. Designating addition polymerization as the preferred polymerization mechanism thus defines the type of polymerization intended by claim 15 and the other patent claims. Identification of addition polymerization as the patent's preferred mechanism of polymerization is repeatedly revealed in the specification. At 11 points the specification refers to "free-radical polymerizable," "polymerizable by free-radical initiators," "addition polymerizable," "free-radical initiators," and "free-radical mechanisms."

Other words and phrases in the specification consistent only with addition (free-radical) polymerization are enumerated in the specification. There are 24 different references to "unsaturated," "unsaturated groups," "polymerizable unsaturated," or "unsaturated linkages." At six points appear the words "olefin," "olefinic," "mono-olefinic," "di-olefinic," "olefinic compound," and "olefinic groups." There are seven references to "polymerizable olefinic," "olefinic polymerizable," and "polymerizable mono-olefinic." There are ten references to "double bonds," "carbon-to-carbon double bonds," "two carbons connected by double bonds," as well as two references to chemical formulae indicating carbon-to-carbon double bonds. Finally, there are numerous references to specific olefinic compounds, such as vinyl silane, and to specific free-radical catalysts, such as hydrogen peroxide.

While the specification makes these multiple references to free-radical or addition polymerization there is not a single reference to condensation polymerization. It is, therefore, concluded that claim 15 construed by its own terms, and also construed in the light of the specification, clearly contemplates polymerization only by addition polymerization. Claim 15, therefore, can only read on addition polymerization. It may not read on condensation polymerization.

The specification provides a wholly independent reason why condensation polymerization is not taught by '378. The sine qua non of the invention disclosed in the specification is a chemical bonding between fibers or pigments treated with vinyl silane coupling agents and liquid polymerizable olefinic monomer. It is undisputed in the record, and indeed both parties agree, that chemical bonding between the vinyl siloxane and the liquid monomer can only be achieved by means of addition polymerization (free-radical polymerization). But if the invention of '378 could be achieved by means of condensation polymerization it would not make any difference whether the glass fibers or pigments were treated with a hydrolyzable silane, such as vinyl silane containing olefinic double bonds, or with a hydrolyzable silane like phenyl trichlorosilane (phenyl silicon trichloride) which is not an olefine. Chemical bonding by co-polymerization with the hydrolyzable silanes can occur only during addition polymerization (free-radical polymerization). During condensation polymerization no chemical bond is formed with the double bonded carbon of the vinyl silane. Hence, condensation polymerization eliminates the need for vinyl silane — yet vinyl silane is an indispensable element of the invention of '378.

Plaintiff now argues that claim 15 reads on condensation polymerization and "therefore, Claim 15 in its language describes composite articles known in the prior art which could render Claim 15 invalid for obviousness. . . ." However, the Owens-Corning record offers no support for this argument.

Throughout the Owens-Corning record there are very few references to any specific type of polymerization. Although the fact of polymerization was very much at issue, there is very little discussion about the type of polymerization which was allegedly taking place. This paucity of discussion supports the inference that everyone assumed that the type of polymerization under consideration was addition polymerization. After all, it is addition polymerization that produces the chemical bonding between the vinyl groups on the surface of the treated glass fibers and the double carbon bonds of the olefinic monomers. It was this subject around which the evidence and arguments revolved.

Moreover, there is no indication in the record that the prior courts ever considered the prior art for any purpose other than its relevancy to the chemical bonding between vinyl-silane-treated glass fibers and monomers (i. e. addition polymerization). The record shows that the prior courts thought that Westwood's patent calling for the chemical coupling of glass fibers to monomer with vinyl silane was obvious to one skilled in the prior art. The reason the patent was held to be obvious has nothing to do with whether or not the chemical coupling proceeded via any particular type of polymerization.

It is concluded, therefore, that claim 5 is merely a restatement of claim 15. Each claim describes the same invention.

Claims 10 and 11

A comparison will next be made of unadjudicated patent claim 10 and adjudicated claim 11, which latter claim defendant Molded Fiber Glass asserts is the adjudicated claim most similar to unadjudicated claim 10. Each of these claims is a method claim for forming a composite article. The first part of each claim is similar except for the underlined words, "pulverulent solids" in 10, and "pigments" in 11.

Claim 10

A method of forming an article comprising a member of the group consisting of fibers and inorganic pulverulent solids which normally have hydroxyl groups on the surface thereof but which are bonded together by a resinous polymer, which comprises contacting surface portions of said group member with a vinyl chlorosilane to modify surface portions thereof. . . .

A method of forming an article comprising a member of the group consisting of fibers and inorganic pigments, bonded together by a resinous polymer, said method comprising wetting said group member which normally carries hydroxyl groups on the surface thereof, but which has surface portions thereof modified by having vinyl groups attached thereto through silicon of a silicon-oxygen linkage. . . .

This is a distinction without a difference. At Column 8, Line 17 of the specification "pigments" are defined broadly enough to include "pulverulent solids." Hence, claim 10's description of a group "consisting of fibers and inorganic pulverulent solids" is clearly included within and not different from claim 11's description of a "group consisting of fibers and inorganic pigments."

There are other differences in language between the two claims. However, both expert witnesses, by their testimony, indicate that only two other differences are possibly significant. Discussion of the remaining language of claims 10 and 11 will, therefore, be concentrated on consideration of the two differences stressed by the expert witnesses.

Following the previously quoted portions, each claim contains the following comparable language:

Claim 10

. . . wetting the treated surface of said group member with a polymerizable liquid comprising a free-radical catalyzed monomeric polymerizable liquid olefine having two sets of double bonds, and causing polymerization of said olefin in contact with said group member to procure a solid body.

. . . said method comprising wetting said group member . . . with a liquid comprising a liquid polymerizable olefinic monomer of said polymer, and causing polymerization of said monomer to the solid state while in contact with said group member. . . .

It will be seen that claim 10 includes the language "a free-radical catalyzed monomeric polymerizable liquid olefine having two sets of double bonds," and said language does not appear in claim 11.

The discussion of claims 5 and 15 and the conclusions there reached apply equally here, and need not be repeated. The mechanism of polymerization intended by the patent claims, and specifically identified in the specification, is free-radical catalyzed polymerization, a type of addition polymerization. Hence, to mention the term "comprising a free-radical catalyzed monomeric polymerizable liquid olefine" in claim 10 adds nothing to claim 11's language "wetting said group member . . . with a liquid comprising a liquid polymerizable olefinic monomer of said polymer." This language of claim 11, like the more specific language of claim 10, reads only on addition polymerization.

Concerning the previous language quoted from claim 10, Mr. Sterman was asked by defendant's counsel what that means. He answered:

Well, it's almost self evident. We have a liquid polymerizable olefine with two sets of double bonds.

Now, may I be specific by example?

Q Please.

A This might be, for example, something like butadiene or any other compound with two double bonds. It might be styrene, as some people define the conjugated double bond system in styrene. These are two typical examples of materials that would fit that definition. And there is also the statement that it is free-radical catalyzed, which presumes there is something of the nature of an organic peroxide present.

Q What does that organic peroxide do?

A This is merely a means to initiate solidification through reaction of the double bond.

Mr. Sterman was then asked:

Q Now, let's go to Claim 11 and refer to the expression, "said method comprising wetting said modified group member with a liquid polymerizable olefine monomer of said polymer."

Would you please explain the meaning of that expression to us?

A Well, the wetting portion is the same as in 10 and I don't believe that is the issue here.

The "liquid comprising the liquid polymerizable olefinic monomer" — again, liquid olefinic monomer is very general nomenclature, and we can illustrate by example.

Again, styrene is very commonly used or, again, butadiene is very commonly used.

Q Does that expression as used in Claim 11 include butadiene and styrene?

A Yes, I believe, in my opinion, it does.

In the mechanism of addition polymerization, on which both claims 10 and 11 read, an olefinic monomer containing two double carbon bonds may be used to practice the invention of each of these claims. Mr. Sterman gave as examples butadiene or styrene "as some people define the conjugated double bond system in styrene." The method taught by claim 10 limits the liquid monomer to a monomer having two sets of double bonds, while the liquid monomer taught by the method of claim 11, contains one set of double bonds, or two or more sets of double bonds. Plaintiff points out that claim 10 could not read on alkenes (olefines) as liquid monomers that contain only one set of double bonds, for example, ethylene, propene, butene. But this is not deemed material. There is nothing in the record to indicate that the invention disclosed by the patent in any way depends upon the use of a liquid monomer with a particular number of sets of carbon-carbon double bonds. Moreover, since ethylene, propene, and butene are gasses at standard temperature and pressure, the patent term "liquid monomer" does not read on them.

It is, therefore, concluded that the inclusion in claim 10 of language not contained in claim 11; namely, "a free-radical catalyzed monomeric polymerization of liquid olefine having two sets of double bonds," does not define an invention in claim 10 that is separate and apart from the invention defined in claim 11.

Claims 12 and 3, 13 and 7

The remaining unadjudicated claims are product claims 12 and 13. Claim 12 is identical with adjudicated claim 3, except that claim 12 describes "an article of manufacture comprising an inorganic pigment . . .," and claim 3 describes "an article of manufacture comprising a member of a group consisting of fibers and inorganic pulverulent solids . . ." Claim 13, a dependent claim, states merely "the article according to Claim 12 wherein the polymer is a polymer of a liquid comprising a monomeric styrene." Similar claim 7, previously adjudicated, describes

"an article of manufacture comprising (1) a member of the group consisting of fibers and inorganic pulverulent solids . . . and (2) a contacting in situ polymerized solid polymer of a styrene, said solid polymer being bonded to said fibers by polymerization of said styrene with said unsaturated aliphatic groups, said surface layer being different from the body of said member of said group."

As seen, claims 12 and 13 are articles of manufacture "comprising an inorganic pigment," while claims 3 and 7 are articles of manufacture "comprising a member of the group consisting of fibers and inorganic pulverulent solids." Column 8, Line 17 of the specification defines the term "pigments" to include

Powders, fillers and pulverulent inorganic solids generally regardless of whether they have as does carbon black and titanium dioxide, high coloring value, or have like silica and glass, no appreciable coloring value.

Thus, it is possible that the term "pigments" may include some materials not encompassed within the terms "fibers and pulverulent solids."

However, this difference loses any significance in view of the basis of the prior courts' determination of obviousness. The prior courts found that the '378 patent-in-suit was obvious because the use of vinyl silanes as coupling agents had been disclosed in the prior art. This finding was not dependent in any way on whether or not the vinyl silanes were used to treat fibers as opposed to pulverulent solids or other pigments. It is manifest that the adjudication of obviousness of the vinyl silane coupling agents that invalidated claims 3 and 7 relating to "fibers and pulverulent solids," equally invalidates claims 12 and 13, involving "pigments."

Though procedurally premature unless and until the question of infringement is reached, the record fails to show that the defendant Molded Fiber Glass used any "pigments" other than fiber glass in making its composite articles. In this proceeding plaintiff was given opportunity to discover evidence that defendant used other pigments but produced no such proof.

Hence, the issue determined in the adjudication of claims 3 and 7 is the identical issue that is presented by claims 12 and 13. In addition, there is nothing in this record to suggest that the gist of the invention of patent-in-suit '378; namely, the chemical bonding of vinyl silanes to different materials, in any way depends on the type of "pigments," as distinguished from "fibers and pulverulent solids," which are used.

It is, therefore, determined and concluded that claim 12 does not define an invention separate and apart from claim 3; and that claim 13 does not define an invention separate and apart from claim 7.

Upon all the evidence it is determined and concluded that by a preponderance of all the evidence defendant Molded Fiber Glass Company has shown that unadjudicated claims 5, 10, 12, and 13 present questions of fact identical to the questions presented in the adjudicated claims; that each merely restates, without significant language differences, an adjudicated claim; and that none of these four unadjudicated claims defines an invention separate and apart from the inventions defined in the adjudicated claims.

Defendant Molded Fiber Glass has, therefore, established the affirmative defense of collateral estoppel, under Blonder-Tongue. At this point, the burden of proof, as previously explicated, shifts to plaintiff Westwood to seek to establish a pertinent Blonder-Tongue exception.

B.

At the hearing of February 14, 1973, counsel for the plaintiff forcefully insisted that plaintiff

. . . should have been entitled to try to prove either (A) that the court below did not understand the invention when it made the rulings it did — . . . or (B) in the alternative, to show that we did not have a fair trial — . . . either one of which are the two exceptions set forth in Blonder-Tongue.

This court is deciding anew the collateral estoppel issue, laying aside its ruling of July 17, 1972 sustaining this defense. However, it has concluded that Blonder-Tongue does not contemplate nor require a retrial of the issue of patent validity. To permit such a retrial, albeit under the guise of a Blonder-Tongue exception, would defeat the purpose of Blonder-Tongue.

The bifurcated claim of plaintiff's counsel, just quoted, echoes the original response to the defendant's motion for summary judgment in which plaintiff's counsel stated,

plaintiff will rely primarily on the second of two exceptions provided by the Supreme Court. It will show to this court that the prior courts wholly failed to grasp the technical subject matter and technical issues in the suit, and that he was deprived of crucial evidence and witnesses for this point in the first litigation.

As the proceedings have developed, and as the plaintiff has submitted his side of the case, plaintiff has not stressed the point "that he was deprived of crucial evidence and witnesses . . . in the first litigation." This procedural claim of lack of a full and fair trial was one of its principal grounds of appeal (1) from Judge Connell's findings and conclusions, (2) in its petition for rehearing of the Court of Appeals' invalidity affirmance, and (3) in its petition for certiorari to the Supreme Court. This ground of appeal is typified by the fourth question presented in its certiorari petition to the Supreme Court. It argued that

the trial record demonstrates that the petitioner (patentee's assignee) was denied fair opportunity to produce rebuttal evidence to sustain validity of the patent and . . . cross-examination of alleged infringer's principal expert was unduly restricted.

It is concluded and determined that the procedural claim of lack of a full and fair trial, one of the Blonder-Tongue exceptions, has been fully and openly adjudicated in the Owens-Corning case. Plaintiff is not entitled to raise again this possible Blonder-Tongue exception; and, therefore, this part of plaintiff's case against collateral estoppel is denied.

The plaintiff's burden of proof, formulated in this proceeding, and fashioned to fit the purpose and guidelines of Blonder-Tongue, will be applied to the remaining claimed Blonder-Tongue exception, i. e. lack of a substantive fair trial. To defeat collateral estoppel under this claim the prior courts' opinions must reveal, when tested against the record of the Owens-Corning case (as supplemented by additional evidence warranted by justice and equity) that the prior courts "wholly failed to grasp the technical subject matter and issues" in the Owens-Corning case. If the total record lacks substantial evidence to support the prior courts' judgment of invalidity, as expressed in the prior opinions, then the plaintiff has sustained its burden of proof.

In making its determination this court considers both Owens-Corning's voluminous record, which has been reviewed, and the record in this case ( Molded Fiber Glass), including the testimony of expert witnesses Sterman and Stempel. In the interest of justice and equity, the court considers as supplements to the Molded Fiber Glass record (1) the affidavits of Dr. Tito T. Serafini of June 29, 1972 (submitted by the plaintiff) and November 16, 1972 (submitted by the defendant), and (2) the deposition testimony and interrogatory answers of the defendant, secured by plaintiff in the course of discovery granted to the plaintiff. Reference to the Owens-Corning transcript of testimony will be cited as "Tr O-C."

Blonder-Tongue particularizes the substantive fair trial exceptions to the application of collateral estoppel.

For example, if the issue is non-obviousness, appropriate inquiries would be whether the first validity determination purported to employ the standards announced in Graham v. John Deere Co., [ 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966)]; whether the opinions filed by the District Court and the reviewing court, if any, indicate that the prior case was one of those relatively rare instances where the courts wholly failed to grasp the technical subject matter and issues in suit; . . . . 402 U.S. at 333, 91 S.Ct. at 1445.

In the Owens-Corning case the "first validity determination" patently "purported to employ the standards announced in" Graham. On appeal the Court of Appeals concluded that "the District Court did an excellent job in determining the state of the prior art in a highly complex and technical area."

In its review the Court of Appeals considered the findings of the district court as to teachings of nine prior art patents: Four prior art patents which were considered by the Patent Office in studying TeGrotenhuis's applications are (1) Johannson Patent No. 2,436,304 granted on application made June 11, 1943; (2) Johannson Patent No. 2,557,786, entitled to a filing date of June 11, 1943, since it was granted on a divisional application of Johannson '304; (3) Patnode Patent No. 2,306,222 granted on application made November 16, 1940; and (4) Kropa Patent No. 2,465,731 granted on application made August 1, 1944. Three prior art patents, not considered by the patent examiner but deemed significant by the Court of Appeals, particularly the first one, are (5) Steinman Patent No. 2,513,268 granted on application filed December 30, 1944; (6) Biefeld Patent No. 2,392,805 granted on application filed October 11, 1943, and (7) Hyde Patent No. 2,390,370 granted on application filed October 11, 1943. Two remaining patents considered by the trial court as pertinent prior art, but questionably viewed as proper prior art by the Court of Appeals include (8) Steinman Patent No. 2,563,288 granted on application filed November 13, 1945; and (9) Witt Patent No. 2,649,396 granted on application filed March 17, 1949. After considering these patents the Court of Appeals concluded that the trial court's "determinations on the status of the prior art are not clearly erroneous and this Court is bound by them."

Centering its attack on the charge that the prior courts "wholly failed to grasp the technical subject matter" in the Owens-Corning case, plaintiff focuses attention on several findings of the trial court. (Without each time repeating the case citation it should be understood that Judge Connell's 108 findings of fact and 22 conclusions of law comprise his opinion reported at 317 F. Supp. 201. Any findings or conclusions that are hereafter cited will merely be referred to by number.) Westwood contends that the prior courts, relying heavily on allegedly inaccurate testimony of defendant's expert witness Dr. Eugene Rochow, misinterpreted the state of the prior art and failed to grasp the unique aspects of the TeGrotenhuis patents '378 and '566.

Johannson '876; Johannson '304

Westwood's chief contention is that Dr. Rochow's discussion of the chemical properties of phenyl trichlorosilane misled the court into holding that the use of vinyl silanes was disclosed in the Johannson '786 patent and, hence, was obvious in the prior art. (See 445 F.2d at 915, 916 which adopts Finding No. 100.) Dr. Rochow testified that phenyl trichlorosilane which had been used in the Johannson '786 patent could be considered a vinyl silane capable of polymerization. He stated,

. . . when [Johannson] used this [phenyltrichlorosilane], he was using as a coupling agent a material which reacted with the adsorbed water on the glass surface and established an organosilane layer which later could bond with the overlying material which he added on top of that coupling layer. [Tr O-C, p. 797.]

Indeed, Dr. Rochow's discussion of phenyltrichlorosilane seems clearly mistaken or at least misleading. As his starting point, Dr. Rochow refers to a portion of the file history of the TeGrotenhuis '378 patent (page 44) where TeGrotenhuis indicates that styrene is a molecule containing conjugated double bonds. The terms "conjugation" and "conjugated double bonds" describe a particular hydrocarbon configuration in which the carbon atoms are joined together in a double bond-single bond-double bond pattern, as in the structural formula for butadiene:

Although the reasons are not well understood the double bonds in a conjugated configuration are much more reactive (e.g. undergo addition polymerization more readily) than are the double bonds in a non-conjugated configuration. Examples of non-conjugated double bond configuration are:

CH₂=CH-CH₂-CH=CH₂ CH₂=CH-CH₂-CH₃ 1, 4 pentadiene 1 butene

As Mr. Sterman testified, there is a "conjugated double bond system in styrene." However, its configuration is more complicated than that of molecules like butadiene because of the presence of a benzene ring:

Molecules like styrene are said to be "conjugated with the ring." That is, the benzene ring as a whole acts like another double bond and, as in butadiene, confers unusual reactivity upon the double bond (vinyl group) which is located outside the ring. Thus, the vinyl group outside the ring readily undergoes polymerization to form polystyrene. TeGrotenhuis was, therefore, correct in stating that styrene contained a conjugated configuration.

However, Dr. Rochow went a step further in his testimony to state:

. . . if we concede that the double bond of a phenyl group [benzene ring] adjacent to the point of attachment is a double bond for that purpose [i.e. conjugation], in styrene, it must also be a double bond for every other purpose. And if it is a double bond, indeed, then we must see that the phenylsilicon trichloride [phenyl trichlorosilane] used by [Johannson] in many of his examples, does, in fact, constitute a substituted vinyl group; that is, it contains a double bond adjacent to the point of attachment of the benzene ring by the applicant's own argument and his diagrams shown on page 44. [Emphasis added. [Tr O-C, 797.]

The error in Dr. Rochow's testimony lies in his statement that the double bonds in the phenyl group are the same "for every other purpose" as ordinary double bonds. This statement overlooks the generally accepted scientific fact that the phenyl group (benzene ring) is highly stabilized by a property known as resonance. According to resonance theory, the benzene ring does not contain three distinct double bonds; instead, the extra electrons are shared equally by all six carbons in the ring, to produce hybrid one-and-a-half bonds between all six. Because of the unusual stability produced by resonance, the benzene ring itself does not undergo the addition reactions, which are essential for addition polymerization (free-radical polymerization). Thus, benzene rings (phenyl groups) cannot form the chemical linkages which are the sine qua non of the TeGrotenhuis '378 patent.

Since the phenyl group does not undergo addition polymerization, Dr. Rochow was in error when he stated that phenyl trichlorosilane can be considered a vinyl silane for the purpose of polymerization. As the chemical formulae of the two molecules indicate, phenyl trichlorosilane only contains double bonds within its phenyl groups:

Unlike the double bond in vinyl trichlorosilane, these double bonds cannot polymerize. On this point Dr. Steinman in the Owens-Corning trial and present case expert witnesses Sterman and Stempel agree. Thus, phenyl trichlorosilane cannot form the chemical bond with olefine monomer which is essential to the TeGrotenhuis '378 patent.

Therefore, the disclosure in the Johannson '786 patent of the use of phenyl trichlorosilane to treat glass cannot be considered a disclosure of the use of polymerizable vinyl silanes. It follows that in entering Finding No. 100, Judge Connell was misled by the mistaken opinion of Dr. Rochow. Therefore, this finding was unfounded and impermissible. Likewise, the adoption of this finding by the Court of Appeals, see 445 F.2d at 915-916, is undermined.

Westwood also contends that Dr. Rochow's testmony concerning the chemical properties of methyl violet dye misled Judge Connell into holding that the Johannson '304 patent disclosed "all the essential elements of the claims of the patents in suit," see 445 F.2d at 915 which adopts Finding No. 94. In Example No. 2 of the Johannson '304 patent, glass fibers are dipped in methallylsilicon trichloride (an allyl silane) and then coated with methyl violet dye at 180° C. Dr. Rochow testified that

Methallyl silicon trichloride contains a double bond which may react with or add to one of the conjugated double bonds of Methyl Violet.

That there is reaction is indicated by the tests carried out by Johannson which show that the Methyl Violet dye did not wash off the fibers with cold water as it washed off of untreated fibers, so that some form of reaction must have occurred with the underlying layer of methallyl silicon trichloride, and I believe it to be a reaction between the double bonds of the treating agent, that is, methallyl groups on the glass, with conjugated double bonds of the Methyl Violet structure. [Tr O-C 686, 687.]

Dr. Rochow's conclusions regarding methyl violet are contradicted in the June 29, 1972, affidavit of polymer chemist Dr. Tito T. Serafini, which was submitted by Westwood in this case. He states,

Dr. Rochow, however, falsely testified that Methyl Violet has olefinic double bonds [Tr O-C 681-682, DX-L], whereas an organic chemist would know that the double bonds, referred to by Dr. Rochow as olefinic, are quinoid double bonds that are highly stabilized by resonance and that accounts for the color of the dye.

He continues

Dr. Rochow testified that the fact that color was retained, i.e. the dye did not readily wash off of the methallyl silane treated glass, indicated that a reaction took place between the methallyl groups of the silane and the double bonds in the dye. This is clearly another false statement, for even in elementary organic chemistry it is taught that such a reaction would consume some double bonds and the color of the dye would be lost. The fact that color was retained shows just the opposite of Dr. Rochow's testimony.

He then gives his opinion on the color retention.

The only reason that the dye was retained better on the treated than on the untreated glass is the fact that the dye is organic and the treated glass had an organic surface. It is axiomatic that substances of similar nature have an affinity for each other. That there is no reaction is also shown by Johannson's statement in Example 3 that the surface treated with phenylsilane, which has no olefinic unsaturated group, retained the dye much better than the methallysilane treated surface. . . .

Reproduced at the left is DX-L (DX stands for "defendant's exhibit") in the Owens-Corning case. On it Dr. Rochow has diagrammed the structural formula of methyl violet. His formula has been compared with the formula for methyl violet appearing in The Chemistry of Synthetic Dyes Vol. IV. Apart from the inconsequential transposition of one methyl group and one hydrogen atom his drawing of methyl violet structure is accurate. By lead arrows he has located three carbon-to-carbon double bonds. Numbers (1) and (2) are his numbers. [3] has been added for clarification. The unnumbered fourth arrow pointing to a carbon-to-nitrogen double bond is not material to the present question.

The issue really centers on whether or not methyl violet will react (as Dr. Rochow states) at any of the three carbon-carbon double bonds that have been marked by arrow and number. Unlike the carbon-carbon double bonds in phenyl trichlorosilane, these double bonds are not located within a phenyl group. Dr. Rochow insists that they are olefinic carbon-carbon double bonds which can react. Dr. Serafini, on the other hand, asserts that they are "quinoid double bonds that are highly stabilized by resonance and that accounts for the color of the dye." He therefore concludes that these quinoid double bonds, like those in phenyl groups, do not undergo olefinic (addition) reactions.

Dr. Rochow and Dr. Serafini each draw opposite inferences from the fact that the color of the dye does not change or disappear when applied to the treated glass. Condon Meislich, in Introduction to Organic Chemstry, state that

The systems of conjugated double bonds characteristic of quinones are important chromophores ( colorbearers) . . Reduction converts a quinone to a colorless dihydric phenol known as hydroquinone and involves addition of hydrogen to the ends of a conjugated system. [Emphasis added.]

This tends to support Dr. Serafini's opinion that the color retention of methyl violet and similar quinoid dyes depends upon the preservation of the conjugated double bonds. Dr. Rochow explains that the methyl violet color did not wash off the fibers treated with the methallyl silanes because of an assumed olefinic reaction of the quinoid double bonds with the double bonds of the allyl silane coupling agent. This explanation appears to be contrary to the above reference that indicates that color formation and retention depend upon the preservation of the conjugated double bonds. Dr. Rochow's assumed olefinic reaction would destroy those conjugated double bonds and alter the color of the dyes.

There is sufficient challenge of the accuracy of Dr. Rochow's opinions concerning the methyl violet example of Johannson '304, upon which Judge Connell based his Finding No. 94, and to which the Court of Appeals refers, 445 F.2d at 915, to warrant a conclusion that in this respect Finding No. 94 lacks substantiation. This insubstantiation of Finding No. 94 of Judge Connell's findings necessarily undercuts his conclusion that the Johannson patent '304 contans all the essential elements of the TeGrotenhuis patents '378 and '566.

With the insubstantiation of Findings Nos. 94 and 100, to the extent indicated, it follows that Johannson '304 and '786 do not teach the chemical bonding of the treated glass fibers with olefinic monomers, the sine qua non of the TeGrotenhuis patents. Nevertheless, there is substantial evidence in the record that shows that these Johannson patents disclose the rendering of glass fibers hydrophobic and organophilic by treatment of the fibers with hydrolyzable organosilanes.

Biefeld '805 and Hyde '370

Westwood further charges that Dr. Rochow's testimony, along with other allegedly inaccurate evidence in the case, misled Judge Connell into holding that the Hyde '370 and Biefeld '805 patents disclose the use of allyl and vinyl silanes. ( 445 F.2d at 917 adopts Judge Connell's Findings Nos. 89 and 91.) The dispute concerns the interpretation to be given to a common paragraph found in both above-mentioned patents as well as in the Johannson '304 patent and the Hyde '689 patent. That paragraph reads:

By hydrolyzable organo-silicanes, I mean derivatives of SiH4 which contain readily hydrolyzable radicals such as halogens, amino groups, alkoxy, aroxy, and acyloxy radicals, etc., and organic radicals that are joined to the silicon atoms through carbon atoms. Examples of such organic radicals are as follows: aliphatic radicals, such as methyl, ethyl, propyl, isopropyl, butyl, amyl, hexyl, heptyl to octadecyl and higher; alicyclic radicals such as cyclopentyl, cyclohexyl, cyclic radicals such as cyclopentyl, cyclohexyl, etc.; aryl and alkaryl radicals such as mono- and poly-alkyl phenyls as tolyl, xylyl, mesityl, mono-, di-, and tri-ethyl phenyls, mono-, di- and tri-propyl phenyls, etc.; naphthyl, mono- and poly-alkyl naphthyls as methyl naphthyl, di-ethyl naphthyl, tri-propyl naphthyl, etc.; tetra-hydronaphthyl, anthracyl, etc.; aralkyl such as benzyl, phenylethyl, etc.; alkenyl such as methallyl, allyl, etc.; and heterocyclic radicals. The above organic radicals may also, if desired, contain inorganic substituents such as halogens, etc. [Emphasis added.]

Dr. Rochow's testimony (Tr O-C pp. 659-661) supports the following series of facts: Neither Hyde '370 nor Biefeld '805 were considered by the patent examiners, when they granted the TeGrotenhuis '378 patent. These patents deal with the coating of glass with hydrolyzable organo-silanes. Among the hydrolyzable organo-silanes listed in the common paragraph were "alkenyl such as methallyl, allyl, etc." The "etc." indicates that vinyl silanes were included as well. Therefore, both patents disclose the treatment of glass fibers with hydrolyzable organo-silanes, including vinyl and allyl silanes.

Supported by Dr. Serafini's affidavit, Westwood principally disputes inclusion of Biefeld '805 and Hyde '370 as prior art because the common paragraph in the two patents is not adequate disclosure of either allyl or vinyl silanes. It notes that alkenyls are listed among a broad list of organo-silanes; and no special attention is given in either Biefeld '805 or Hyde '370 to the special ability of the alkenyls to polymerize. Westwood further argues that this same paragraph was considered by the patent examiners in the TeGrotenhuis application when they reviewed Johannson '304 and Hyde '689; and that the examiners rejected these patents as relevant prior art because of an inadequate disclosure of either allyls or vinyls. Westwood insists that this rejection buttresses its contention that Biefeld '805 and Hyde '370 are also not pertinent prior art.

The patent examiners' rejection of Johannson '304 and Hyde '689 (a different patent than uncited Hyde '370), though entitled to some weight, does not preclude de novo court consideration of uncited Biefeld '805 and Hyde '370, as pertinent prior art. Judge Connell's findings accurately represent the purposes and attributes of both patents Moreover, his specific findings concerning the inclusion of vinyls within the list of alkenyls are adequately based on the testimony of several witnesses. Thus, Dr. Rochow testified that "the word alkenyl embraces vinyl, allyl, substituted vinyl, substituted allyl, and higher radicals containing double bonds between carbon atoms." (See TR O-C 661.) Dr. Price, an expert witness called by the plaintiff gave the following answers to questions propounded to him on cross-examination:

Q That term alklenyl means alkenyl, does it not, Doctor?

A Yes; and vinyl is the simplest of all of those.

Q Yes. And alkenyl embraces your vinyl and your allyls?

A Yes. (TR O-C, 183)

This testimony together with additional evidence in the record supports all findings relating to Hyde '370 and Biefeld '805; namely, Findings Nos. 88 through 92.

On the total record it is decided that Westwood has not shown that the prior courts lacked substantial evidence for the ultmate conclusion that Hyde '370 and Biefeld '805 are pertinent prior art.

The Prior Art

The affidavit of Dr. Serafini, submitted by the plaintiff, also questions the court's consideration of Steinman '268 as prior art. The affidavit states that the courts failed

to recognize the very great technical difference in operation between the carbon-oxygen-silicon bond of the allylsilicates [Steinman '268] and the direct carbon-silicon bond of the silanes [TeGrotenhuis '378 and '566].

Subsequently Steinman '268 is studied as part of this court's independent condensation of the relevant art that antedates TeGrotenhuis '378. It will then fully appear that the trial court grasped and understood the fundamental teaching of Steinman '268, irrespective of how one may technically evaluate his Finding No. 104 that,

. . . Steinman teaches the treating of glass fibers with allyl silicates; these allyl silicates are not identical with the materials presently used; however, they are structurally similar and function in the same manner as the allyl silanes and the vinyl silanes.

The factual enumeration that follows is supported by substantial evidence in the Owens-Corning record. The facts enumerated considered with the opinions of the prior courts show a lack of substantial evidence that the prior courts "wholly failed to grasp the technical subject matter" of the TeGrotenhuis patents, both '378 and '566. In reaching this conclusion the court gives no weight to the unsubstantiated portions of trial court findings (Nos. 94 and 100) which erroneously find that the two Johannson patents teach that the chemically treated glass fibers disclosed in the patents will react with olefinic monomers.

1. The Patnode, Hyde '370, Biefeld, and both Johannson patents, previously identified, all teach that glass treated with hydrolyzable organosilanes becomes hydrophobic and organophilic; and that the organosilanes form chemical bonds with the glass by hydrolysis. As Judge Connell found, Dr. Rochow discussed the teachings of these patents at a lecture in Cleveland on November 15, 1944, attended by T.A. TeGrotenhuis. Dr. Rochow there

demonstrated the properties of the organosilane compounds, pointing out that organochlorosilanes would hydrolyze and, therefore, when brought into contact with a glass surface, they would react with the water adsorbed on that glass surface, and deposit a highly water repellent film of an organo siloxane attached to the glass, just as the adsorbed water had been. (Finding No. 47)

2. The Steinman '268 patent, dealing solely with allyl silicates, was filed on December 30, 1944 and is pertinent prior art for the TeGrotenhuis '378. As Judge Connell notes on Finding No. 105, this patent clearly discloses Steinman's "double fishhook" or coupling theory as follows:

The improvements experienced by practice of the present invention are believed due, at least in part, to the fact that the unsaturated silicates are capable of co-polymerizing with the resin to become in effect one with the resin surrounding the fiber surfaces. They also are believed to react by hydroylsis with the glass surface to be anchored securely to the fibers. [Col. 4, Line 28]

3. Steinman's "double fishhook" or coupling theory was conceived as early as 1943 and was discussed with several other scientists. (See DX-X; DX-TT; DX-UU; DX-VV; DX-WW; DX-XX; Tr O-C pp. 1098-1101, 1142, 1155-56) (also see Judge Connell's Findings Nos. 23, 24).

4. In 1943 and 1944, Steinman planned a program of experimentation using allyl silicates and allyl silanes as coupling agents to bond glass fibers to polymers. (See DX-UU; DX-VV; DX-WW; DX-XX) (Also see Judge Connell's Finding No. 23).

5. Allyl Silicates were experimented with extensively during 1943 and 1944. (Also see Judge Connell's Findings Nos. 27, 28.)

6. Allyl silanes were produced in experimental quantities; but were probably not applied to glass before 1945.

7. Evidence in the record indicates that the chemical bond on the substrate formed by allyl silicates is weaker than that formed by allyl and vinyl silanes, especially in the presence of moisture or excessive humidity. The explanation for this difference lies in the fact that in the silicate the allyl group is connected to the silicon atom through an oxygen atom, whereas in the silane the allyl group is connected directly to the silicon atom. The oxygen atom in the allyl silicate tends to hydrolyze in the presence of moisture causing the allyl group to break away from the silicon atom. The allyl silane, which contains no oxygen, avoids this problem. Therefore, as Judge Connell pointed out in Finding No. 104, allyl silicates and allyl silanes are not identical.

However, unlike non-olefinic silicates and silanes, both allyl silicates and allyl silanes co-polymerize with olefinic monomers to form an actual chemical bond. In this sense, the two compounds function in the same manner. Indeed, judged by changes made in his various applications in the Patent Office, it would appear that TeGrotenhuis himself as late as 1952 considered alkenyl silicates to be acceptable coupling agents for his invention. For example, in his 1951 application (Serial No. 243,737) TeGrotenhuis stated that the akylenyloxy groups (another name for alkenyl silicates) were among the group of preferred organosilicanes. Obviously, the inventor himself then thought that alkenyl silicates function in a manner similar to that of alkenyl silanes. Thus the evidence shows that, although the difference between the chemical properties of allyl silicates and allyl silanes has been noted this difference is less important than is the fundamental fact that both allyl silicates and allyl silanes form chemical bonds between treated glass fibers and olefinic monomer via addition (free-radical) polymerization. Thus, both allyl silicates and allyl silanes are glass-treating compounds which can function as coupling agents in accordance with Steinman's "double fishhook" theory.

8. The Kropa '731 patent was granted on an application filed on August 1, 1944 and, hence, is prior art for the TeGrotenhuis '378 patent. This "invention relates to allyl chlorosilicanes (allyl silanes) and processes of preparing same." Disclosing how allyl silanes may be used to treat glass, it states

. . . in laminating the polymer may be employed to bind together fabrics or paper compositions of cellulosic fibers, asbestos, glass, polyamides ("nylon"), etc. [Col. 6, line 10]

It further teaches that allyl silanes may be co-polymerized with unsaturated monomers including styrene, methyl methacrylate, etc.:

It is apparent that triallyl silicol would form a dimer by condensation but this dimer could be polymerized because of the unsaturated allyl groups, to form a high polymer or it could be copolymerized with other unsaturated material to form thermoplastic or infusible copolymers. [Col. 4, line 10]

9. Therefore, at the time TeGrotenhuis made his conception on the evening of November 14, 1944, after he attended Dr. Rochow's lecture and later filed his application, the major elements of his invention were already known in the prior art; a) that organo-silanes which include alkenyl (allyl, vinyl, etc.) hydrolyzable silanes bond with glass fibers by hydrolysis; b) that Dr. Steinman's "double fishhook" theory ssdisclosed that a coupling agent was needed chemically to bond glass fibers to polymer; c) that allyl silicates would form such a chemical bond though subject to breaking down in the presence of moisture or excessive humidity; d) that allyl silanes could be made; e) that allyl silanes, unlike allyl silicates, form a bond which can withstand moisture and excessive humidity; and f) that allyl silanes will copolymerize with unsaturated monomers.

Improvement Over Prior Art and Obviousness of TeGrotenhuis '378

In Owens-Corning, 445 F.2d at 917 the Court of Appeals notes,

The District Court concomitant with its consideration of the prior art heard testimony from expert witnesses on the subject of exactly what improvement the claimed TeGrotenhuis inventions made over the prior art. This is the second analytical step in determining obviousness as prescribed in Monroe Auto Equipment Co. v. Heckethorn Mfg. Supply Co., 332 F.2d 406, 411 (6 Cir. 1964).

In this respect the Court of Appeals manifestly was complying with Graham v. John Deere Co., supra as it was also when it determined that

the District Court resolved the legal question of whether the inventions in issue were obvious to one skilled in the art adversely to plaintiff.

The Court of Appeals upholds this conclusion as well as the district court's conclusion that

the teachings of the TeGrotenhuis patents were either entirely or significantly portended by the prior art and little if any contribution was made to the existing art by the TeGrotenhuis inventions.

These conclusions of the trial court and the Court of Appeals give no indication that the courts "wholly failed to grasp the technical subject matter" of the issues prescribed by Graham v. John Deere Co. in considering a claim of obviousness. This court arrives at this conclusion from the following summary findings and conclusions based on substantial evidence found in its review of the Owens-Corning record.

10. The use of allyl silanes as coupling agents was obvious in the prior art. Therefore, TeGrotenhuis's only possible invention, if any, would be the discovery and demonstration that vinyl silanes are different from and superior to allyl silanes. In order to claim this as his invention and to withstand a charge of obviousness, TeGrotenhuis must take advantage of his application filing date of March 30, 1945. During the Owens-Corning trial Westwood conceded that unless it can take advantage of this filing date its patent is invalid for obviousness (Tr O-C p. 13, Finding No. 58). This is true because the disclosures in the Steinman '288 and Witt '396 patents would clearly make the TeGrotenhuis '378 patent invalid for obviousness, if in fact they are pertinent prior art. Steinman '288, which was granted on an application filed on November 13, 1945, expressly teaches the use of allyl silanes as coupling agents to bind fibers and polymers; and Witt '396, which was granted on an application filed on March 17, 1949, teaches the use of both allyl silanes and vinyl silanes as coupling agents.

Therefore, for TeGrotenhuis to prevail against a charge of obviousness the record must show that TeGrotenhuis discovered the superiority of vinyl silanes over allyl silanes and that that discovery can take advantage of the March, 1945 filing date. Specifically, the record must show a) that vinyl silane could be made at the time TeGrotenhuis disclosed it (if not, the original filing date is inapplicable); b) that vinyl silanes are, in fact, superior to allyl silanes; and c) that he disclosed the superiority of vinyl silanes at the time he filed his application. However, there is substantial evidence in the record from which Judge Connell was entitled to conclude, and he did so conclude, that TeGrotenhuis had failed to fulfill any of these three requirements (see Conclusions Nos. 13, 14, 15).

11. Testimony by Dr. Price, Dr. Jellinek, Dr. Rochow, Dr. Kanner, Dr. Gilman, Dr. Van Auken, and Dr. Aufder-march (Tr O-C pp. 239, 621, 900-904, 935-7, 958-60, 968-9, and 1009-1016) (see Judge Connell's Findings Nos. 63-68) tended to establish that vinyl silanes could not be made in March, 1945 when TeGrotenhuis disclosed them. Since a theory cannot be patented and since at least some of an applicant's claims must be operable in order to have a valid patent, the fact that vinyl silanes could not have been produced means that TeGrotenhuis did not have a patentable invention in March, 1945.

12. There is sufficient evidence in the record to justify Judge Connell in concluding, as he did (see Conclusion No. 14), that vinyl silanes are not superior to allyl silanes, but are in fact virtually equivalent in function and structure. Dr. Price, Westwood's expert, stated:

There is very little difference between the relative rates of polymerization of vinyl silanes and allyl silanes. [Tr O-C p. 248]

Dr. Rochow asserted that:

. . . the vinyl groups are not much more reactive to interpolymerization than allyl. [Tr O-C p. 787]

Dr. Biefeld, another expert for the defendant, claimed that the only reason that vinyl silanes have been used more widely in commercial processes is not because they are superior to allyl silanes, but because they are significantly cheaper. (Tr O-C pp. 877-880) These opinions were confirmed by Dr. Jellinek (Tr O-C pp. 1083-1085) and Dr. Phillips (Tr O-C pp. 1106-1108). (See also Judge Connell's Findings Nos. 61 and 75.)

If vinyl silanes are not superior to allyl silanes, TeGrotenhuis has no invention because allyl silanes were clearly disclosed in the prior art. (See Kropa '731.) To avoid invalidity for obviousness a patent must bring about some improvement over the prior art. Therefore, if vinyls are no improvement over allyls the patent is invalid for obviousness.

13. However, if it were to be assumed that vinyl silanes are superior to allyl silanes, TeGrotenhuis's 1945 application must fail under the doctrine laid out by the Court of Appeals in Acme Highway Products Corp. v. D. S. Brown Co., 431 F.2d 1074 (6 Cir. 1970). Acme holds that a continuation-in-part from an earlier application, what the '378 patent claims to be, may contain embellishments or refinements on the original invention, but may not contain new patentable disclosures. 431 F.2d 1081. This rule of patent law was emphasized in an earlier district court opinion in the Second Circuit:

A continuation-in-part application is not entitled to the benefit of the earlier filing date of its parent application where the changes included within the subsequent application are "new matter" which either alters the substance of the invention or makes the composition an invention for the first time. . . . Indiana General Corp. v. Krystinel Corp., 297 F. Supp. 427, 443 (S.D.N Y 1969). [Emphasis supplied.]

TeGrotenhuis '378 was granted on a continuation-in-part application, August 2, 1952 (Serial No. 302,415) from parent application (Serial No. 585,824) filed March 30, 1945. There is evidence in the record that the disclosure of the superiority of vinyl silanes over allyl silanes was not contained in the 1945 parent application. (See Judge Connell's Findings Nos. 61 and 75.) Referring to TeGrotenhuis's 1945 application, Dr. Biefeld stated:

. . . there is no difference taught as to the relative effectiveness of the allyl and vinyl silanes. [Tr O-C p. 894]

This fact is confirmed by the opinion of the United States Court of Customs and Patent Appeals in TeGrotenhuis v. Yaeger, 290 F.2d 951 (1961), an interference proceeding. While that court found that vinyl silanes were disclosed in the 1945 application, it determined that no preference was stated for vinyl silanes over the sixteen other specific alkenyl silanes listed in the patent application. (See 290 F.2d 955-957.) Likewise, TeGrotenhuis's application (Serial No. 243,737) filed on August 25, 1951, expressed a preference for "vinyl and butadienyl groups" and for other "groups with no more than 4 carbon atoms per carbon to carbon double bond present" (among which are allyl groups) but did not claim that vinyls were superior to those other groups. (See PX-7, p. 3) Thus, it was not until August, 1952 when TeGrotenhuis's final application (Serial No. 302,415) was filed that a preference was stated for vinyls over all other silanes. (See PX-9, p. 4.)

As mentioned above, TeGrotenhuis's invention, if any, must be the superiority of vinyl silanes over allyl silanes. Since this crucial disclosure was not made until 1952, there was "an invention for the first time" in 1952, and TeGrotenhuis cannot take advantage of the 1945 filing date. Therefore, Acme precludes the use of the March, 1945 filing date even if Westwood had proved that vinyl silanes are superior to allyl silanes, a fact not shown by substantial evidence.

Upon the entire record it is concluded and determined that the plaintiff has failed to show that either Judge Connell or the Court of Appeals "wholly failed to grasp the technical subject matter." This is true as to the issue of obviousness under Section 103 upon which ground the Court of Appeals sustained Judge Connell's determination of patent invalidity, though it limited his judgment to the patent claims of '378 that were actually in issue. It is likewise true as to all other statutory grounds of patent invalidity relied upon by Judge Connell but not reached by the Court of Appeals.

Having previously ruled that the defendant has sustained its burden of proving the identity of questions between the adjudicated claims and the unadjudicated claims of '378, the affirmative defense of collateral estoppel is justified and it will be applied.

A final comment: The Supreme Court in Blonder-Tongue declared "In the end, decision will necessarily rest on the trial court's sense of justice and equity." It is just and equitable that the defense of collateral estoppel be accepted and effected in this case. Owens-Corning, Johns-Manville, Ferro Corporation, and Certain-Teed Products are the suppliers of treated glass fibers that Molded Fiber Glass Company buys and uses to form composite auto and boat bodies. Each of these companies has the benefit of the judgment of invalidity of TeGrotenhuis patents '378 and '566. It would offend this court's "sense of justice and equity" to deny to Molded Fiber Glass, the customer of these four suppliers, the right to assert the Owens-Corning judgment as a collateral estoppel defense while each of Molded Fiber Glass Company's four suppliers is immune from patent infringement claims under that same judgment.

Judgment is rendered on the merits in favor of the defendant and against the plaintiff.

It is so ordered.