This is an appeal from the decision of the Patent and Trademark Office (PTO) Board of Appeals (board) affirming the examiner's final rejection under 35 U.S.C. Sec. 103 of claims 15-22 of appellant's application serial No. 54,757, filed July 3, 1979, entitled "Herbicidal Esters of 4-Tri-fluoromethyl-3'-carboxy-4'-nitro diphenyl ethers." We affirm.

* Background

Appellant claims novel compounds showing herbicidal activity and methods of controlling weeds with the claimed compounds. The compounds claimed belong to a class of herbicides known as diphenyl ethers. Although certain diphenyl ethers have been shown useful as herbicides, the effectiveness of a particular diphenyl ether is difficult to forecast from the substituent groups bonded to the phenyl rings of the ether. Applicant discloses its compounds as effective both as preemergence and postemergence herbicides, as selectively effective, and as useful in conjunction with other selective herbicides.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

wherein X is a hydrogen atom, a halogen atom, a trifluoromethyl group, a (C1 -C4 ) alkyl group, or a cyano group; Y is a hydrogen atom, a halogen atom, or a trifluoromethyl group; and Z is (C2 -C6 ) alkenyl, (C2 -C6 ) haloalkenyl, (C3 -C6 ) alkynyl, (C3 -C6 ) haloalkynyl, (C8 -C12 ) arylalkynyl substituted with a halo, methyl or nitro, (C4 -C7 ) cycloalkenyl or (C3 -C6 ) epoxyalkyl).

16. The compound of claim 15 wherein Y is hydrogen and X is halogen.

17. The compound of claim 16 wherein Z is (C2 -C6 ) alkenyl.

18. The compound of claim 16 wherein Z is allyl and X is chloro.

19. A herbicidal composition which comprises a compound according to claims 15, 16, 17 or 18 and an agronomically acceptable carrier.

20. The composition of claim 19 which additionally comprises a surfactant.

21. A method of controlling weeds which comprises applying to the surface of the growth medium prior to the emergency of the weeds from the growth medium the compound of claim 15 in an agronomically effective amount.

22. A method of controlling weeds which comprise applying to weed seedlings the compound of claim 15 in an agronomically effective amount.

The references relied upon by the examiner are:

Bayer et al ("Bayer-1") 3,928,416 Dec. 23, 1975

Bayer et al ("Bayer-2") 3,798,276 March 19, 1974

Theissen 3,907,866 Sept. 23, 1975

The Bayer-1 patent discloses herbicidal compounds of the formula:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

wherein Z is hydroxy, alkoxy, alkyl, halo, amino, alkylthio, cyano, carboxy, carbalkoxy, carboxyalkyl, carbalkoxyalkyl, alkanoyloxy, or carbamoyloxy.

The Bayer-2 patent, the parent of Bayer-1, discloses herbicides of the formula given above wherein Z is an alkoxy group, an alkyl group, alkoxyalkoxy, hydroxyalkoxy, halo, alkylamino, dialkylamino, alkylthio, carboxy, carbalkoxy, carboxyalkoxy, carbalkoxyalkoxy, carboxyalkyl, carbalkoxyalkyl, N,N-dialkylureido, alkanoylamide, or a carbalkoxyamino.

The Theissen patent discloses herbicides of the formula:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

wherein X is halogen, n is 1-3, and R is alkyl (C4 -C12 ), cycloalkyl (C3 -C11 ) or unsaturated hydrocarbyl (C3 -C10 ).

As can be seen, the claimed compounds and the prior art are structurally similar. Johnson differs from Theissen as it contains a CF3 (trifluoromethyl group) at the para position of the ether where a C1 (chloride) radical is found in Theissen. Johnson differs from Bayer-1 and -2 in that it discloses alkenyl esters (compounds having a -COOZ radical wherein Z is, among other things, a C 2-6 alkenyl) while Bayer discloses saturated alkyl esters (compounds having a -COOR radical where R is C1-14 alkyl). Bayer-1 specifically discloses a methyl ester.

The examiner originally rejected the claimed matter, in relevant part, as unpatentable over Bayer-1 under 35 U.S.C. Sec. 103, and unpatentable over Bayer-2 in view of Theissen and Bayer-1 under 35 U.S.C. Sec. 103. The examiner based these rejections on Bayer et al's showing of nitro diphenyl ethers with a carboalkoxyl substituent; Theissen's showing that "R" of the -COOR radical may be alkenyl; and the Bayer-1 teaching of the equivalence of CF3 to halogen substituents (C1 being a member of the halogens) on the ring of the diphenyl ether.

Appellant responded to the examiner's action and filed a Rule 132 affidavit by Dr. Yih, an employee of the true party in interest in this appeal, Rohm and Haas Company. The affidavit set forth a comparative showing of appellant's compound with a representative Theissen compound. Tests had been run which indicated that appellant's compound produced "highly significant and unexpected differences." Dr. Yih concluded from the tests that the Johnson compound is significantly and unexpectedly more active than the tested Theissen compound in general herbicidal activity.

Considering appellant's response and the Yih affidavit, the examiner issued a final rejection under 35 U.S.C. Sec. 103 over the teachings of the Bayer et al and Theissen patents, indicating that the claimed compounds were prima facie obvious. The examiner's reasoning regarding the showing of obviousness and the attempted rebuttal of that conclusion is as follows:

Theissen makes obvious the alkenyl esterification of the Bayer et al herbicides and conversely the Bayer et al patents make obvious the substitution in the Theissen compounds of the para halogen by a trifluoro group and the herbicidal activity of compounds so substituted expected.

The declaration, addressing itself to the latter, is not seen to overcome this rejection. The differences in activity are seen as but expected variations, the trifluoromethyl group generally imparting enhanced activity and the positioning of the same * * * pointed to [ ] by Bayer et al. Further the testing of Bayer et al * * * is seen to teach alkenyl esters of these trifluoro-methyl substituted herbicides, applicants arguments notwithstanding, albeit among numerous substituents. Thus as the point of novelty of the claimed compounds over the Bayer et al compounds appear to reside in the unsaturated ester moity, taught specifically by Theissen and disclosed by Bayer et al, a showing of unexpected activity of the claimed compounds over the Bayer et al alkyl ester counterparts is required to overcome this rejection. [Emphasis supplied.]

The examiner conceded in his answer to appellant's brief to the board that the 132 affidavit overcame that part of the 103 rejection holding that the Bayer et al patents made obvious the substitution in the Theissen compounds of the para halogen by a trifluoromethyl group. The examiner argued, however, that the paramount consideration supporting obviousness was that Theissen makes obvious the alkenyl esterification of the Bayer et al herbicides, and that a showing of unexpected activity over Bayer et al saturated esters was necessary. Applicant argued that no such showing was needed because an applicant relying upon a comparative showing to rebut prima facie obviousness must compare his claimed invention only with one of equally close prior art references, citing In re Merchant, 75 F.2d 865, 197 U.S.P.Q. (BNA) 785 (CCPA 1978) and In re Holladay, 584 F.2d 384, 199 U.S.P.Q. (BNA) 516 (CCPA 1978).

On appeal and reconsideration, the board agreed with the examiner's conclusion of prima facie obviousness. Regarding the attempted rebuttal, the board noted that the question was not which compound was the closest prior art, but that the "references together singly teach a different substituent from appellant's at one position" of the basic diphenyl ether molecule. Thus the board held it incumbent upon appellant to compare his claimed compound to one of Bayer differing only in the ester moity.

Applicant bases his appeal on the statement in In re Holladay, 584 F.2d at 386, 199 USPQ at 518 that "[w]here two pieces of prior art are in fact equally close to the claimed invention, there is no logical reason for requiring an applicant to make a comparison with one instead of the other. On the other hand, practical considerations favor allowing the applicant to choose between them." Applicant argues that because Bayer and Theissen are equally close to Johnson, the prior art in each instance differing from the claimed compound at only one position, the applicant here was free to make a comparative showing over either Bayer or Theissen. The showing of unexpected results over one prior art reference, according to appellant, therefore is necessarily sufficient to overcome the examiner's finding of obviousness. (We note that the examiner never found that the prior art references were equally close and that the Solicitor argued in this brief that Bayer et al is the closest prior art. The Solicitor concedes, however, that the examiner never made explicit to applicant that Bayer was the closest prior art. Because of our reasoning which follows, we need not consider whether the prior art references are equally close or whether the examiner's correspondence was a model of clarity.)

We find appellant's interpretation of Holladay unpersuasive and hold that the board did not err in its conclusion of obviousness. The problem of obviousness considered by the Patent and Trademark Office, and to which we address ourselves here, arises under section 103 of the Patent Act. It is a problem of patent law and not of chemistry. In re Papesch, 315 F.2d 381, 386, 137 U.S.P.Q. (BNA) 43, 47 (CCPA 1963). Thus, the requirement of unobviousness in the case of chemical inventions is the same as for other types of inventions. A patent will not be granted where the invention, when compared to the prior art, is considered to have been obvious to a person having ordinary skill in the art at the time the invention was made. 35 U.S.C. Sec. 103. This is the mandate to which the examiner, the board, and this court must look.

The procedural devices of the examiner's prima facie case and the applicant's rebuttal were formed to clearly allocate the burdens of going forward and of persuasion as between the examiner and applicant. In re Piasecki, 745 F.2d 1468, --- USPQ ---- (Fed.Cir.1984). Rebuttal is merely a showing of facts supporting a conclusion of unobviousness. In re Heldt, 433 F.2d 808, 811, 167 U.S.P.Q. (BNA) 676, 678 (CCPA 1970). When evidence is submitted on rebuttal, the decision maker must consider all the evidence anew, still judging the invention under the Sec. 103 criteria. In re Rinehardt, 531 F.2d 1048, 1052, 189 U.S.P.Q. (BNA) 143, 147 (CCPA 1976). It stands to reason that a fact finder is entitled to his own ideas, within reason, as to what evidentiary facts will persuade him of unexpected results. Thus whether rebuttal evidence is sufficient to persuade the examiner that unexpected results exist is an evidentiary matter left for the trier of fact. Here, the applicant produced no evidence that Bayer and Theissen taught the same invention; it is apparent that the examiner had to find the showing insufficient.

Appellant argues that the reasoning of Holladay somehow changes the standard, that a showing of unexpected results over one of two equally close prior art references somehow automatically hurdles the standard dictated by the patent statute. This is not the case and is a misconstruction of Holladay and cases which have followed.

In Holladay the board held that a patent applicant confronted with an obviousness rejection may successfully rebut using a comparative showing only by testing over the prior art reference relied upon by the examiner. The Court of Customs and Patent Appeals (CCPA) held that this conclusion was incorrect; an applicant must compare to the closest prior art, even if that art was not relied on by the examiner. The court continued that where two references were equally close to the claimed invention there is no reason for insisting that applicant compare with one reference, the one relied upon by the examiner, instead of another. Noting that on occasion an examiner may rely on a paper patent which was never reduced to practice and is unavailable for testing, the court reasoned that "[w]here the applicant uncovers a piece of prior art actually used in the real world and establishes that its teachings are equal to the relevant disclosure in a 'paper patent' relied upon * * *," it would be unfair to insist on a comparison with the unavailable invention. Holladay, 584 F.2d at 386-87, 199 USPQ at 516 (emphasis supplied). Thus Holladay is premised on the notion that the teachings of the closest prior art references are sufficiently similar so that the testing of one compound showing unexpected results would provide the same information as to the relevant teachings of the other equally close references.

In In re Payne, 606 F.2d 303, 203 U.S.P.Q. (BNA) 245 (CCPA 1979), a case considering claimed pesticide compounds, the CCPA again addressed the use of comparative showings and again held that an applicant need not test compounds taught in every reference. "However, where an applicant tests less than all cited compounds, the test must be sufficient to permit a conclusion respecting the relative effectiveness of applicant's claimed compounds and the compounds of the closest prior art." Payne, 606 F.2d at 316, 203 USPQ at 256. See In re De Blauwe, 736 F.2d 699, 705, 222 U.S.P.Q. (BNA) 191, 196 (Fed.Cir.1984) ("an applicant relying on comparative tests to rebut a prima facie case of obviousness must compare his claimed invention to the closest prior art.")

Appellant's argument clearly cannot prevail in view of Holladay and Payne. Here appellant has neither established nor asserted that the teachings of Theissen are so parallel to those of Bayer that it has shown sufficient testing to permit a conclusion respecting the relative effectiveness of Johnson over Bayer. Appellant seems to think itself legally entitled to a patent issued in ignorance of that seemingly important fact. (Indeed in its argument before the board and in its disclosure, appellant insists that one cannot make generalizations concerning effectiveness because similar compounds may perform very differently.) Appellant seems to want the examiner and the board to put on blinders. Because the affidavit did not show relative effectiveness over all compounds of the closest prior art, the board correctly held that the comparative showing did not provide an adequate basis to support a legal conclusion of unobviousness.

AFFIRMED.