This is a patent interference on a single count which reads:

"The process which comprises polymerizing a mixture of butadiene-1,3 and styrene at a temperature below 0°C. in aqueous emulsion in the presence of methanol, an emulsifying agent, an activating-reducing composition comprising a ferrous compound, and as the catalyst a phenylcyclohexyl hydroperoxide in which the hydroperoxy group is attached to that carbon atom in the cyclohexyl ring which is attached to the phenyl ring." (Emphasis ours.)

In short, the invention is a process of making synthetic rubber and its novelty resides in the catalyst used.

The Board of Patent Interferences awarded priority to Reynolds in a decision of January 5, 1955, from which decision an appeal (No. 6232) was taken to this court. We did not therein pass on the priority issue but sent the case back to the Patent Office to decide another question, the decision of which had been deferred in the Patent Office pending a decision by us on priority, namely, whether the Reynolds application should be stricken from the files because of changes made in it after execution. For our opinion of March 29, 1957, in the prior appeal see Vandenberg v. Reynolds, 242 F.2d 761, 44 CCPA 873. On July 16, 1957, the Commissioner of Patents, acting through First Assistant Commissioner Crocker, held that the alteration of Reynolds' application involved no change of substance and denied Vandenberg's petition to strike, adhering to his decision after two requests for reconsideration, and referred the case to the Board of Patent Interferences. The board, on November 19, 1957, adhered to and repeated its original decision on priority and from that decision the present appeal was taken.

The instant appeal presents two issues: (1) Whether the Commissioner erred in refusing to strike the Reynolds application from the files and (2) whether the board erred in awarding priority to Reynolds.

(1) Alteration of Application

The question here is whether the Commissioner abused his discretion in not striking the Reynolds application under Patent Office Rule 56, 35 U.S.C. Appendix, which provides, inter alia, that "any application altered or partly filled in after being signed or sworn to * * * may be stricken from the files." While Patent Office Rules not contrary to the law have the force and effect of law, it is clear that this rule reposes in the Commissioner a discretion. The decision must stand unless the discretion has been abused.

Reynolds admits that two changes were made in his application after execution and also admits the possibility that a third change was made. The last and one of the former changes, characterized by the Commissioner as "purely formal", were a change from "our" to "my" and the change of "cyclohexylbenzene" to "phenylcyclohexane," which is but another name for the same compound. Vandenberg, at the hearing before us, waived further reliance on these two changes, thus reducing the alteration issue to the significance of a single change, namely, the deletion of the italicized words "low temperature" from the following portion of the general statement of the invention:

"I have now discovered that greatly increased conversion rates are obtained when carrying out emulsion polymerization reactions at low temperatures using redox recipes if the oxidizing component employed is an arylcyclohexyl hydroperoxide, such as is formed upon low temperature reaction of free oxygen with a liquid arylcyclohexane." (Emphasis ours.)

There is in evidence a handwritten note, attached to the executed application by Reynolds, suggesting changes, reading as follows:

  p 1 l 23      change "low-temperature
              reaction of free
              oxygen" to "reacting
              oxygen or oxygen-containing
              gases such as
              air"

  p 10 l 23     change "our" to "my"

Jones, his attorney, testified that he and Reynolds had a `phone conversation about the first proposal and agreed to make the change which was actually made. Jones said:

"Since low temperature was not definite at that point in the specifications it seemed to me somewhat ambiguous or indefinite, and Dr. Reynolds and I decided to omit it entirely."

When Reynolds was testifying the following transpired:

"XQ35. Do you regard the changes which are indicated in this exhibit [his aforesaid note] to be minor changes? A. I regard the first change, page 1 line 23 to be a fairly basic change. I regard page 10 line 23 to be minor."

The Commissioner has held that the omission of the words "low temperature," when considered in the light of the entire specification, was not a change which was material or which involved new matter. He said further that this omission

"* * * made the sentence in which it occurred more nearly consistent with the specific disclosure of the arylcyclohexane oxidation reaction in the specification. The deletion was in the interest of clarity and, but for the manner in which it was made, was desirable."

Criticism of this quotation as a factual statement is conspicuously absent from Vandenberg's brief. Without attempting to controvert the facts stated, Vandenberg says he is at a loss to understand how such a conclusion could be reached and merely alleges by way of a legal conclusion that the change "actually accomplishes a considerable expansion of the invention as originally described" and for this reason is a material alteration.

We are constrained to agree with the Commissioner both as to his conclusion that the omission was not a material change and as to his supporting reasons.

Vandenberg strenuously urges upon us, as he did upon the Patent Office, that Reynolds' testimony that his suggested change — not the one actually made — was "fairly basic" rather than "minor" is a formal admission under oath which is controlling and conclusive, citing a number of cases.

The Commissioner found, and we agree, that Reynolds had not expressed any opinion about the change actually made and we have no idea what he would have said had he been asked about it. In any case he was not asked whether the change he proposed was material or substantial but simply whether he thought it was "minor" and we are unable to attach to his reply that it was "fairly basic," in this context, any clear meaning. The Commissioner further held that Reynolds' opinion was not controlling in the matter.

It is the materiality of the change that should govern the Commissioner's exercise of discretion in striking applications from the files. Materiality is a question of law which must be decided on the facts and a witness' opinion at best can do no more than tend to prove a fact. It cannot bind either the Patent Office or this court on a question of law. It certainly is not controlling as would be admissions in pleadings or stipulations, as suggested by Vandenberg. All we have here is an ordinary answer to a question by a witness asked if he thought a change was minor. We find no parallel in the cases cited wherein, for example, a party admitted that he did not make a certain invention, that reading a specification suggested a certain invention to him, that certain disclosures had been received and like admissions against interest. Such cases are not in point.

Concluding on this alteration issue, we find no error in the Commissioner's decision denying the petition to strike the application from the files. Since we believe his opinion to be unpublished and lest our affirmance be taken as a "liberal" view on what may be done to an application after its execution without incurring a penalty, we repeat the following statement, with full approval, from the Commissioner's opinion:

"It should be emphasized, however, that while the materiality of an alteration of an application may determine whether or not an application shall be stricken, this consideration provides no standard of propriety for an attorney." (Emphasis ours.)

We might add that the costly litigation of issues such as this and the risk of losing patent protection altogether can be avoided by following the obviously safe course of altering first and executing afterward.

So now, after a regrettable delay of more than two years, we reach the question of who, in the eyes of the law, is the first inventor of the subject matter of the count. On January 5, 1955 the Board of Patent Interferences decided it was Reynolds on the basis that he was the first to conceive and the first to constructively reduce to practice by filing his application, serial No. 101,676, filed June 27, 1949.

Vandenberg filed his application, serial No. 134,857, on December 23, 1949, is therefore the junior party and has the burden of establishing priority by a preponderance of the evidence. This he attempts to do by proving an actual reduction to practice between June 2 and 6, 1949, prior to Reynolds' filing date. The board credited Reynolds with a conception date of February 25, 1949 and Vandenberg also attacks that holding as not proved. It is evident that if the board was correct in its holding that Vandenberg failed to establish an actual reduction to practice, that alone disposes of the priority issue. We shall therefore direct our consideration to that point.

Preliminarily, as above indicated, the entire novelty of the count resides in the catalyst used. This catalyst is an oxidizing agent and it is used with the reducing composition of the count in a process which thus employs a combination of reducing and oxidizing agents and is hence known as a "redox" polymerization. Appellant has pointed out with some care that the catalyst called for by the count is one having a very definite structure. In fact, as appellant's witnesses testified, it is a single specific compound, 1-hydroperoxy-1-phenylcyclohexane, which is a phenylcyclohexyl hydroperoxide. There are three other such compounds which apparently can be made, wherein the hydroperoxy substituent is differently positioned, but the limitations of the count with respect to the point of attachment of the hydroperoxy group limit it to the single compound. It seemms obvious, threfore, that an actual reduction to practice of the invention of the count could be established only by proving that that particular compound had in fact been used as the catalyst in a process otherwise conforming to the count, that is to say, that the polymerization reaction was being catalyzed by 1-hydroperoxy-1-phenylcyclohexane.

The only problem in this case revolves around the question of just what Vandenberg used as a catalyst in certain experiments in June, 1949, on which he relies to sustain his burden of proof. In his brief he has adopted the term "PCH" allegedly to designate only the specific compound of the count, 1-hydroperoxy-1-phenylcyclohexane. We shall so use it.1 Taking the facts from appellant's briefs, only insofar as they are admitted therein, it appears that Vandenberg used as catalyst a reaction product prepared in the laboratories of Hercules Powder Company, where Vandenberg worked, by Dr. Taves, a coworker. This material was prepared by blowing oxygen through a mixture of phenylcyclohexane and calcium hydroxide at 90-100°C., the mixture being illuminated with an ultraviolet lamp to initiate the reaction. This oxidation must give one [or more?] of the four isomeric phenylcyclohexyl hydroperoxides. Dr. Taves determined that the hydroperoxide content of the oxidation product was 30.1% and to his own satisfaction he determined that "at least 60% of his hydroperoxide product was PCH." In his Reply Brief appellant appears to contend, for the sake of greater accuracy, that "the exact figure is 56%" PCH. In the same brief the factual situation is nicely summed up as follows. Referring to the "catalyst composition" Vandenberg used in the tests relied on as an actual reduction to practice, the brief says:

"Now as to the contention that Vandenberg did not know the identity of the 30.1% fraction of the material which responded to the iodine test. This, too, is essentially a misstatement of fact. He did know that at least 60% of this material consisted of PCH. 40% of the 30% or a mere 12% of the whole catalyst composition, remained unaccounted for in the analysis, but as shown infra was very probably PCH."

We find appellant using the term "catalyst composition" to refer both to the entirety of what Vandenberg used and to the 30.1% of it which was a hydroperoxide and when we straighten this out it is clear that at least 40%, not 12%, of the hydroperoxide, whatever it may have been, asserted to be a catalyst, was actually an unknown. Deducting the 30.1% from 100%, the remainder, or 69.9%, of the material Vandenberg used is asserted to be unreacted phenylcyclohexane. As to either 40% or 44% of the hydroperoxide fraction (depending on whether we take 60% or 56% to be PCH), we are asked to proceed on the assumption that it was "very probably PCH." We cannot, however, accept mere probability or likelihood as proof of an actual reduction to practice.

What Vandenberg used as a catalyst, therefore, giving him the benefit of every doubt and accepting his contentions, many of which are strongly contested, was the product of the oxidation of phenylcyclohexane, 69.9% of which was unreacted pure phenylcyclohexane, 56% of the remainder being the compound called for by the count and 44% of the remainder being admittedly unknown but possibly other phenylcyclohexyl hydroperoxides not conforming to the count. Thus the proofs are, at best, that 16.8% (56% of 30.1%) of Vandenberg's "catalyst composition" was PCH. This is the reality of the situation behind Vandenberg's oft reiterated statement that the June, 1949, tests were made "using PCH as catalyst."

"No competent evidence by Vandenberg has been adduced to satisfactorily establish that the particular hydroperoxide, the inventive feature of the count, was the polymerization catalyst, bearing in mind that it was the entire liquid reaction mass which was used as the catalyst, and that four hydroperoxides of phenyl cyclohexane are theoretically possible. No evidence whatsoever has been introduced to demonstrate that other than the catalyst liquid reaction mass as a whole promoted the superior results stated to have been obtained by PCH."

We entirely agree. Vandenberg has not sustained his burden of proof. It may very well be that PCH was in fact acting as the catalyst, to the exclusion of anything else in the composition, whatever it contained, but the proofs have not established that with the degree of certainty required in cases of this kind. One attempting to prove an actual reduction to practice of a chemical process must prove every step thereof. Brooker v. Riester, 161 F.2d 745, 34 CCPA 1088.

Appellant has relied heavily on certain language taken from Smith v. Swaine, 127 F.2d 140, 29 CCPA 973. We find no parallel in that case to the situation here. Suffice it to say that there was no question whatever in that case that the material being tested as an insecticide was xanthone. The tests had been rejected below as not proving an actual reduction to practice on the basis of several criticisms and this court, finding them unjustified, reversed. The material being tested as a catalyst here was not PCH. While it may have been present, it cannot have been known with a reasonable degree of certainty that it was responsible for whatever catalysis was produced (and we have assumed, arguendo, that there was such) considering the other possible catalysts which might have been present in the mixture. It was incumbent on Vandenberg to eliminate such other possibilities in order to establish the effectiveness of PCH as a catalyst.

The decision of the Board of Patent Interferences is affirmed.

Affirmed.

MARTIN, J., did not participate in decision.