This is an appeal from a decision by the United States Patent and Trademark Office (PTO) Board of Patent Interferences (board) awarding priority of invention to Armitage, the senior party-applicant,1 on the ground that the junior party-patentee Thomson et al.2 (Thomson) had not demonstrated an actual reduction to practice predating Armitage’s effective filing date. We affirm.
Background
The Counts
The subject matter in interference concerns a particular class of ethylene-carbox-ylic acid copolymers and a process for their synthesis. There are nine phantom counts involved, but the following three are illustrative:
Count 1
A homogeneous, compositionally uniform, random copolymer of ethylene and from about 0.1 to about 35 percent by weight based on the copolymer of an (X, jj-ethylenically unsaturated carboxylic acid selected from the group consisting of acrylic acid, methacrylic acid and crotonic acid, which copolymer has a standard melt flow rate value by ASTM D-123857T in the range from about 0.01 gram per ten minutes at Condition E to about 20 grams per 10 minutes at Condition D. Count 5
A copolymer according to Count 3 [wherein the carboxylic acid is acrylic acid] in the form of a thin sheet consisting essentially thereof.
Count 7
Process of making homogeneous, com-positionally uniform, random copolymers of ethylene and an acid comonomer as hereinafter defined, said copolymers containing from about 0.1 to about 35 percent by weight based on the copolymer of the said acid and having a standard melt flow value by ASTM D-1238-57T of from about 0.01 gram per ten minutes at Condition E to about 20 grams per ten minutes at Condition D by continuously feeding, to a homogeneous reaction mixture in a reaction zone maintained in a steady state at a pressure of at least 1000 atmospheres and a temperature in the range from about 90° to about 300°C., feed materials comprising ethylene, a free radical initiator, and an acid comonomer selected from the group consisting of acrylic acid, methacrylic acid and crotonic acid, said acid comonomer being fed in a ratio of one part thereof to an amount of ethylene in the range from about 10 to about 10,000 parts by weight, copolymerizing ethylene and the acid comonomer in the reaction mixture while thoroughly admixing the reaction mixture in the reaction *1034zone and maintaining therein the concentrations of ethylene, acid comonomer and resulting copolymer product substantially constant and polymerizing only part of the ethylene and acid comonomer present in the reaction zone, and continuously withdrawing an amount of the whole resulting reaction mixture containing un-reacted ethylene, unreacted acid como-nomer, and the resulting copolymer product from the reaction zone at a rate equivalent to the rate of feeding the feed materials to the reaction zone.
Certain terms used to describe the copo-lymers of the counts are highly significant to the present case. According to the record, a copolymer chain characterized as “random” consists of monomeric units joined together without a set order, i. e., the occurrence of a particular type of monomer (here, either ethylene or a carboxylic acid) is strictly a matter of chance. The record also indicates that in a “compositionally uniform” copolymer substantially all of the constituent macromolecules have substantially the same chemical composition.
Unfortunately, the record is not entirely clear on the meaning of “homogeneous.” The Thomson specification states that a “homogeneous” copolymer “is discerned as a single physical phase in solid state.” Yet according to uncontradicted expert testimony,3 copolymers of ethylene and unsaturated carboxylic acids are semi-crystalline, i. e., consist of at least two phases (crystalline and amorphous), and, therefore, cannot be “homogeneous” if Thomson’s definition of that term is strictly applied. Therefore, to avoid an internal inconsistency in the language of the counts, we will interpret “homogeneous” here as denoting polymeric material that does not separate when extruded into phases which are discernable to the unaided eye. We believe this construction is reasonable in light of the expert testimony quoted above in note 3, and is consistent with the ordinary meaning one skilled in the relevant art would lend to the language of the counts.4
The Proceedings Below
Thomson, as the junior party, introduced testimony and related exhibits to show an actual reduction to practice of the subject matter of the counts before June 26, 1961, the filing date of the grandparent of Armi-tage’s involved application. To substantiate a prior actual reduction to practice, Thomson submitted pages from various notebooks wherein were recorded the results of test runs made before 1961 with the miniplant (2-liter) and semiplant (50-liter) reactors, in addition to memoranda and progress reports from the same period referring to these runs. The recorded data include the results of physical analyses conducted by an independent laboratory to ascertain, among other properties, the tensile strength, melt index, elongation, low-temperature brittle point, rigidity, hardness and density of the end products obtained from the test runs. However, Thomson present*1035ed no evidence of testing prior to 1961 for the composition or homogeneity of the copo-lymeric material produced in either type of reactor. Instead, Thomson relied primarily on testimony by himself and his coinventor Waples, and by two Dow research supervisors, Klumb and Landry, to show that adequate stirring of the polymerization mixture, reflected in the absence of a runaway reaction or a temperature gradient within the reactor, assured the production of random copolymers that were both composi-tionally uniform and homogeneous.
The Board’s Opinion
The board ruled that Thomson could not sustain his burden of proving that the copo-lymers were “inherently homogeneous and compositionally uniform” without presenting data from contemporaneous tests for these physical properties. In addition, the board gleaned from the testimony of Thomson, Waples, and others evidence that the copolymers in fact did not come within the counts. Specifically, the board looked to three principal areas for support of its conclusion that Thomson had not reduced the invention of the counts to practice before 1961:
(1) observations that some of the films produced from the copolymer materials contained gels;
(2) reports by Waples that the stirring mechanism in the mini-plant autoclave reactor did not achieve proper back mixing; and
(3) a report by Waples describing an extraction test of an ethylene acrylic acid (EAA) copolymer which separated into two fractions with substantially different acrylic acid contents.
OPINION
Evidentiary Standard
Before reaching the merits of Thomson’s priority case, we must dispel the apparent confusion over the standard of proof applicable to evidence of an alleged actual reduction to practice. In its opinion, the board stated initially that Thomson had to rely on “the doctrine of inherency,” since “the copo-lymers were not tested for homogeneity or compositional uniformity at the time they were made.” Yet the board concluded thereafter that “such testing is necessary to clearly establish that these properties would inevitably result from the process for making these copolymers,” as required (according to the board) by the “doctrine of inher-ency.” The board thus presented Thomson with a dilemma: he must invoke the “doctrine of inherency” since he has no direct evidence of the properties at issue, but he cannot satisfy the requirements for “inher-ency” without such evidence.
A basic error in the board’s approach stems from its premise that the so-called inherency doctrine is applicable here at all. This is not a case where the junior party must demonstrate that the limitation is “inherent” in the explicit teachings of his specification, so that one of average skill would necessarily be led to the invention of the counts by those teachings. See, e. g., Reed v. Tornqvist, 58 C.C.P.A. 864, 436 F.2d 501, 168 U.S.P.Q. 462 (1971); Dreyfus v. Sternau, 53 C.C.P.A. 1050, 357 F.2d 411, 149 U.S.P.Q. 63 (1966); Brand v. Thomas, 25 C.C.P.A. 1053, 96 F.2d 301, 37 U.S.P.Q. 505 (1938). Rather, Thomson must prove by a preponderance of the evidence that he actually reduced the invention of the counts to practice, since Thomson’s application was copending with the application which matured into the Armitage patent. Frilette v. Kimberlin, 56 C.C.P.A. 1242, 412 F.2d 1390, 162 U.S.P.Q. 148 (1969). A prerequisite element of Thomson’s proofs in this context is a showing by a preponderance of all the evidence, direct and circumstantial alike, that copolymers produced under his direction prior to Armitage’s 1961 filing date satisfied all the limitations of the counts.5
*1036 Thomson’s Evidence
Thomson argues that Klumb corroborated testimony by Waples to the effect that adequate stirring was reflected in the “reasonably close” temperature uniformity achieved throughout the reactors, indicative of a uniform concentration of catalyst (and hence, a uniformity of reaction rate) conducive to the formation of compositionally uniform, homogeneous copolymers. Thomson also relies on testimony by Landry concerning the absence of decomposition, which would be expected to occur if localized concentrations of catalyst due to inadequate stirring had resulted in a runaway reaction, and consequent pressure build-up, in portions of the reactor.
However, the board emphasized several reports authored by Waples between 1961 and 1963 that contained, in the board’s opinion, admissions of continued problems with mixing in the miniplant reactor. While the statements cited by the board arguably are open to other interpretations,6 we cannot say that the board’s reading of them is in error. At the very least, the contemporaneous reports weaken the credibility of present testimony on Thomson’s behalf that inadequate stirring was no longer a problem in Thomson’s EAA polymerization system by the time Armitage filed his application.
As evidence that the copolymers synthesized under his direction before 1961 met the limitations of the counts, Thomson calls attention to test data in the Thomson patent indicating that the EAA copolymers disclosed therein displayed lower plastic recovery (PR) values7 than control samples of known, heterogeneous resin blends. Thomson argues that such “low” PR values are the hallmark of homogeneous, composition-ally uniform EAA copolymers generally, and he offers an affidavit by Tandy, a duPont research chemist, in support of this contention. However, in a 1965 report Wa-ples considered whether certain test data indicated that “plastic recovery measurements tell us anything other than molecular weight distribution estimates.” Under “Conclusions,” the report states that “plastic recovery * * * may be a key measurement in determining homogeneity of copo-lymer samples.” (Emphasis added.) Apparently, even six or more years after the alleged reduction to practice, the validity of PR measurements as indicators of homogeneity was still a topic of discussion at Dow. This interpretation of the 1965 report finds support in the uncontradicted testimony by Dr. Wunderlich, Armitage’s expert witness, which casts serious doubt on any correlation drawn from the data in the Thomson patent between comparatively low PR values and either copolymeric homogeneity or compositional uniformity.
A similar lack of solid scientific grounding undermines Thomson’s attempt to explain away statements by Klumb and others that films and sheets drawn from Thomson’s copolymers contained gels, evidence interpreted by the board to indicate nonhomogeneity. Referring to other testimony that films drawn from his copolymers were “very clear,” Thomson urges that this absence of “milkiness” from the films evidences homogeneity, and that the observed gel particles could have been caused by contamination or oxidative crosslinking in the copolymer product. However, there is no question that the presence of gel particles is associated with nonhomogeneity, and we are not convinced by speculation on alternative explanations for gel formation that the board erred in finding that Thomson’s copolymers lacked homogeneity.
*1037Finally, Thomson characterizes as misplaced the board’s reliance on extraction-test data contained in a 1963 Waples report. These data show that an EAA co-polymer extracted with hexane separated into two fractions, the denser of which comprised some 92% of the charged copolymer. The plastic-rich, denser layer had a substantially higher acrylic acid content (11.2%) than the lighter fraction (4.7%), which included the remaining 8% of the charged resin. Thomson takes issue with the board’s conclusion that the extracted “copolymer material can [not] be said to be com-positionally uniform * * * since both fractions do not have ‘substantially the same chemical composition.’ ” However, in arguing that extraction of over 90% of the copolymer into a single layer is itself evidence of compositional uniformity, Thomson makes no reference to what one skilled in the art would consider “substantially uniform.” In this regard, we note Dr. Wunderlich’s uncontradicted testimony that the extraction results summarized above do not correspond to those predicted for a random, composi-tionally uniform EAA copolymer having the same acrylic acid content as the plastic actually tested.8 Accordingly, we are not convinced that the board’s finding of substantial nonuniformity, based on the difference between acrylic acid content of the two extraction fractions, is erroneous.
We hold that Thomson has failed to show by a preponderance of the evidence that he or his co-workers synthesized copolymers of the counts before 1961. Therefore, the decision of the board is affirmed.9
AFFIRMED.
. Serial No. 164,074, filed July 19, 1971, which is a continuation-in-part of serial No. 349,759, filed March 5, 1965, which is a continuation-in-part of serial No. 119,265, filed June 26, 1961. Armitage was accorded benefit of both serial No. 349,759 and serial No. 119,265.
. U.S. 3,520,816, “Copolymers of Ethylene,” issued July 21, 1970, on serial No. 787,236, filed December 26, 1968, which is a continuation-in-part of serial No. 276,085, filed April 26, 1963.
. Dr. Bernard Wunderlich, an authority on polymers who was called by Armitage as an expert witness, testified on direct examination as follows:
A 30 It says here [in the Thomson patent], “By ‘homogeneous’ is meant that the copo-lymer material is discerned as a single physical phase in the solid state,” that this I think is a poor definition because the copolymers which are described in the patent are partially crystalline, and partially crystalline means that they consist at least of two phases, one phase being the crystalline, the other phase being the amorphous.
So any of these materials could in the strict sense not be called homogeneous.
♦ * * sH * *
Q 33 Well, as you understand from reading this patent, the use of “homogeneous” in the
patent, do you believe that the definition that you have given was what was intended here?
A No.
Q 34 Did you reach an understanding of what you thought was intended?
A What was intended here is probably that the material is something which will not macroscopically fall apart. It is not quite clear — this is why I picked on this particular term “homogeneous.” It probably meant that when you put it through an extruder or so, it will still be a mass * * *.
. For example, our interpretation effectively obviates the issue of whether “homogeneous” must be read as “substantially homogeneous," as Thomson contends, since a polymer either separates into discernable phases upon extrusion or it does not.
. Since we agree with the board that Thomson did not sustain his burden of proof in this regard, we do not consider whether the record evidences contemporaneous recognition by Thomson of the invention. See Langer v. Kaufman, 59 C.C.P.A. 1261, 465 F.2d 915, 175 U.S. P.Q. 172 (1972); Heard v. Burton, 51 C.C.P.A. 1502, 333 F.2d 239, 142 U.S.P.Q. 97 (1964).
. For example, in an August 6, 1962 report, Waples stated that an “undesirable flow pattern [in the miniplant] was particularly noticeable when making ethylene copolymers with highly reactive monomers, such as styrene and methyl methacrylate.” (Emphasis added.) It is not clear from the report whether Waples was referring to mixing problems encountered with the denoted monomers, rather than with the miniplant generally.
. According to the Thomson specification, the PR value of a polymer is directly related to the increase in diameter of an extruded rod of the polymer as it emerges from a “plastometer” orifice of standard size.
. Waples testified that the difference in composition between the two fractions could be attributed to the “heat ageing” of the lighter, solvent-rich fraction, which remained in the settling tank of the extraction apparatus longer than the plastic-rich fraction. Since no mention of this possibility appears in the 1963 report of the extraction results, we feel that Wa-ples’ testimony is illuminated by hindsight and cannot be used nunc pro tunc as support for an actual reduction to practice. See, e. g., Clauss v. Foulke, 54 C.C.P.A. 1514, 379 F.2d 586, 154 U.S.P.Q. 85 (1967).
. Before the PTO, Thomson was denied the benefit of his 1963 application’s filing date. However, the board did not consider the issue of support for the counts in the 1963 application, noting that Armitage’s effective filing date was earlier still and that Thomson, therefore, had to prove a prior actual reduction to practice. Accordingly, the question of benefit is moot.