This appeal is from that portion of the July 30, 1976, decision of the Patent and Trademark Office (PTO) Board of Appeals (board) rejecting claims 2-5, 51, and 52 in application serial No. 171,736, filed August 13, 1971, entitled “Anorectic Adamantane Derivatives and Method of Using Same.” The board rejected the claims under 35 U.S.C. § 103 on new grounds, as provided in 37 CFR 1.196(b), as obvious from Brake1 in view of Narayanan,2 Bernstein et al.,3 and Bernstein.4 We affirm the rejection of composition claim 52 and reverse the rejection of method claims 51 and 2-5.
The Invention
The invention pertains to a method, as defined in claims 51 and 2-5, of curbing appetite in animals by administering certain adamantane compounds.5 The invention also pertains to the unit dosage form of a composition for curbing appetite comprising such an adamantane compound and a pharmaceutically acceptable carrier as defined in claim 52.
In the specification, appellant identifies his claimed compounds as follows:
*83Independent claim 51 defines the “method of curbing appetite in an animal which comprises administering to the animal an amount effective to curb appetite of a compound” of the above formula.
The References
Brake describes a process for improving the yield of a-methyl multicyclic methyla-mines, one of which is a-methyl-l-adaman-tanemethylamine, illustrated as:
and is described as being useful as an antiviral agent in animals.
Narayanan teaches adamantyl sulfonam-ide compounds, useful as antimicrobial agents, e. g., as antiviral agents, of the formula:
*84
wherein R and R1 each is hydrogen, halogen, lower alkyl, phenyl or phenyl-lower alkyl, R2 is hydrogen or lower alkyl, R3 is hydrogen, lower alkyl, lower alkoxy, halogen or halo-lower alkyl and n is 0,1 or 2, and salts thereof.
Narayanan also teaches the use of his cpm-pounds in dosages corresponding to those of appellant.
Bernstein et al. pertains to adamantyl biguanides of the formula:
and to acid-addition salts thereof.
In Formula I, R and Ri each is hydrogen, halogen, lower alkyl, phenyl or lower alkoxy, Rj, R8 and R4 each is hydrogen, lower alkyl or phenyl-lower alkyl and n is 0 or 1.
These compounds are hypoglycemic agents effective in reducing blood sugar content in mammals.
The compounds of the Bernstein patent are illustrated by the following formula:
and to acid-addition and quaternary ammonium salts thereof.
These compounds are adamantyl derivatives of phenothiazines, therapeutically active as central nervous system depressants.
The Rejection
The examiner rejected appellant’s claimed composition and method as obvious under 35 U.S.C. § 103 in view of the teaching in Brake of administering to animals structurally similar adamantane derivatives “analogous” to those claimed. The Bernstein and Narayanan patents were cited to show similar compounds in the art. The examiner reasoned that the composition claim would have been obvious from the prior art because the respective compounds differ merely by a methylene group, i. e., the instant compounds have at least an ethylene link between the adamantane ring and the amine, whereas the prior art compound has a methylene link. This “minor molecular modification” was further asserted to be made obvious by the Bernstein and Narayanan patents, which disclose lower al-kylene links between adamantane and other moieties and are directed to pharmaceutical uses.
The board treated the examiner’s rejection as relying upon Brake alone and as citing the Bernstein and Narayanan patents to show the state of the art. The board did not sustain the rejection of claims 2-5, 51, and 52 as obvious from Brake alone because Brake’s failure to disclose an amount of his compound effective as an antiviral agent renders unobvious the administration of “adjacent homologs of Brake’s compound ‘in an amount effective to curb appetite’ * * Similarly, the board did not agree that appellant’s composition in an “appetite curbing amount” would have been obvious from Brake alone.
Under 37 CFR 1.196(b), the board made a new ground of rejection under 35 U.S.C. § 103 for obviousness from Brake in view of the Bernstein and Narayanan patents. The board agreed with the examiner that appellant’s compounds having an ethylene linkage would have been obvious in view of Brake’s corresponding adjacent homolog (methylene linkage). Relative to the method claims, the board found sufficient motivation in the prior art to administer Brake’s compound and adjacent ethylene “homo-logs” as antiviral agents, and concluded *85that administering appellant’s compounds in appetite-curbing amounts would have been obvious from Brake and Narayanan since the amounts suggested by Narayanan to achieve antiviral effects encompass the amounts intended and claimed by appellant.
The Arguments
Appellant contends that, after refusing to sustain the examiner’s rejection on the basis of Brake along, the board erred in rejecting the method claims by considering Narayanan in addition to Brake. Appellant argues that Narayanan’s reference to dosage for treating viral infection is an improper basis for rejection. It is urged that the board mistakenly assumed that appetite-suppressant effects of appellant’s compounds would be readily recognized from treating virus-infected animals with a related compound. It is also urged that the board ignored differences in treatments for viral infection and obesity, and that therefore Narayanan’s dosage cannot be said to result in effective anorexia. Relative to the claimed composition, appellant states that there is an appreciable difference between the structure of the compounds of the claim and the prior art compounds, and that the former would not have been obvious because the motivation to make the required structural variation is absent.
The solicitor responds by arguing that in the absence of comparative evidence of any unexpected difference in the properties of appellant’s and Brake’s compounds, the compounds of the claim would have been obvious from and unpatentable over the structurally closely related compound disclosed by Brake. It is argued that Brake and Narayanan render obvious appellant’s pharmaceutical carrier and “unit dosage form.” As to the method claims, the solicitor contends that Narayanan discloses adamantyl compounds as antiviral agents in dosages that correspond to and would suggest similar and inherently appetite-curbing amounts of the Brake antiviral compound. The solicitor supports the board position that because appellant’s compounds are homologous and there is sufficient motivation in the prior art to administer Brake’s compound as an antiviral agent, appellant’s different purpose does not render the method claims unobvious.
OPINION
We note at the outset that the ethylene linkage of appellant’s compound closest to the prior art ( (j-(l-adamantyl)- a-methyle-thylamine) is referred to by the examiner as “analogous” to the methylene linkage of Brake’s a-methyl-l-adamantanemethyl-amine and by the board as a “homolog.” Since the appellant has not challenged either of these classifications, we proceed on the assumption that he accepts the inference that his compounds, whether homologs or analogs, would be expected to have similar properties to the prior art compound. Whether the adamantyl compounds in question are properly classified according to the usual definitions of “homolog” and “analog,” we shall not consider inasmuch as appellant has not argued the point.
The solicitor has taken the position that absent comparative evidence demonstrating any unexpected difference in the properties of the compounds, the claimed composition would have been obvious from and unpatentable over the structurally closely related compound disclosed in Brake. On the other hand, appellant contends that the presence of the ethylene rather than the methylene group constitutes “an appreciable difference in the claimed compound and the prior art compounds,” and relies on In re Taborsky, 502 F.2d 775, 183 USPQ 50 (Cust. & Pat.App.1974) for support of his argument that without some teaching of motivation to make the required molecular variation, a finding of obviousness based on structural similarity is improper.
Regarding this issue of structural similarity, we agree with the solicitor and the PTO position. The examiner noted the difference of a mere methylene group between the compound of the claim and the prior art compounds, cited the Bernstein and Naray-anan references showing the state of the art as prior art knowledge of use of lower alkylene links between adamantane and *86other moieties, and concluded that “this minor molecular modification would clearly be obvious to the pharmaceutical chemist.” We do not accept appellant’s contention that the adjacent alkylene link in question constitutes an “appreciable difference” in the compounds. We think that a person skilled in chemical and/or pharmaceutical arts would not hesitate to extend the alky-lene linkage of the prior art compound. Further, we note that appellant’s compound closest to the prior art and its synthetic preparation are disclosed in Narayanan as one of a group of compounds for producing his adamantyl sulfonamide. This leaves no room for doubt that the prior art knowledge renders appellant’s compound structurally similar and provides sufficient motivation to make it.
Moreover, appellant has no basis for relying on Tabor sky, supra. Unlike the present case, the prior art of record in Taborsky expressly limited the scope of “halogen” to exclude appellant’s claimed fluorosalieylani-lide compounds and stated “ ‘several disadvantages in practice ’ ” of free salicylani-lides. 502 F.2d at 781, 183 USPQ at 55 (emphasis supplied). Appellant here has shown no such reason to preclude the conclusion that appellant’s compounds are structurally similar to the prior art compounds.
Confronted with PTO evidence of obviousness, appellant has offered no evidence of unobviousness, as by showing an actual difference in properties between his compounds and the prior art compounds'. In re Hoch, 428 F.2d 1341, 57 CCPA 1292, 166 USPQ 406 (1970). Appellant merely shows that his novel compounds are appetite suppressants whereas the reference compounds are not so known. Further, appellant has not indicated whether his compounds are antiviral, as is Brake’s prior art compound. Presented with such an absence of comparative or other evidence with respect to the properties of the compounds and the claimed composition, we hold that composition claim 52 would have been obvious from and unpatentable over the prior art.
Regarding method claims 51 and 2-5, the solicitor agrees with the board that:
* * * the compounds of claim 51 are obvious from and unpatentable over the corresponding Brake compound and the Narayanan disclosure of a dosage which corresponds to appellant’s disclosed appetite curbing dosage (therefore, inherently appetite curbing). [Emphasis added.]
We cannot accept this conclusion. The issue here is whether the claimed method of curbing appetite would have been obvious. That appellant’s “amount effective to curb appetite” corresponds to or inheres in Na-rayanan’s amount “to combat microbial infestation” does not persuade us of the obviousness of appellant’s method. As this court said in In re Naylor, 369 F.2d 765, 768, 54 CCPA 902, 905-06, 152 USPQ 106, 108 (1966):
[Inherency] is quite immaterial if, as the record establishes here, one of ordinary skill in the art would not appreciate or recognize that inherent result. * * *
* * * we find nothing in the record which would afford one of ordinary skill reason to anticipate that a trial * * * [of the combined prior art teachings] would be successful in producing the polymer recited in the claims.
The Patent Office has failed to show a reasonable expectation, or some predictability, that Brake’s compound would be an effective appetite suppressant if administered in the dosage disclosed by Narayanan. The mere hindsight assertion that corresponding dosages render appellant’s method obvious is untenable. Prior to appellate’s disclosure, none of the adamantane compounds in any of the references before us suggested a use, much less a dosage, for curbing appetite. What we said in In re Spormann, 363 F.2d 444, 448, 53 CCPA 1375, 1380, 150 USPQ 449, 452 (1966), relative to inherency applies equally here:
As we pointed out in In re Adams, 356 F.2d 998, 53 CCPA 996, 148 USPQ 742 [(1966)], the inherency of an advantage and its obviousness are entirely different questions. That which may be inherent is not necessarily known. Obviousness cannot be predicated on what is unknown.
*87Accordingly, the decision of the board is affirmed as to claim 52 and reversed as to claims 51 and 2-5.
MODIFIED.
. U.S. Patent No. 3,489,802, issued Jan. 13, 1970, on application serial No. 610,779, filed Jan. 23, 1967.
. U.S. Patent No. 3,501,511, issued Mar. 17, 1970, on application serial No. 661,781, filed Aug. 21, 1967.
. U.S. Patent No. 3,270,036, issued Aug. 30, 1966, on application serial No. 493,899, filed Oct. 7, 1965.
. U.S. Patent No. 3,320,249, issued May 16, 1967, on application serial No. 470,930, filed July 9, 1965.
. Adamantane is the trivial name assigned to tricyclodecane. Its structural formula can be represented in any of the following ways:
*83
or their pharmaceutically acceptable acid addition salts, wherein:
Rj ■ H, lower alkyl, aralkyl, aralkyl substituted with NH2, OH, OCHj, halogen, alkyl, NO^; phenoxyalkyl or phenoxyalkyl substituted with NH2, OH, OQij, halogen, alkyl, or N02; acyl such as formyl or acetyl.
* H, lower alkyl, COO-lower alkyl, aralkyl, aralkyl substituted with NH2, OH, OCHj, halogen, alkyl, NO^; phenoxyalkyl or phenoxyalkyl substituted with NH2, OH, OCHj, halogen, alkyl, or N02; acyl such as formyl or acetyl.
R^ and R^ can be joined together to form, with the nitrogen, a heterocyclic
Rj ■ H, lower alkyl, or alkynyl
R4 - H, lower alkyl, or alkynyl
Rj « H, OH, halogen, or lower alkyl
R^ « H, OH, halogen, or lower alkyl
R^ and R^ together may represent a carbonyl oxygen.
R^ « H, lower alkyl, halogen, hydroxy, alkoxy, amino or substituted amino, trifluoromethyl, sulfamyl, nitro, phenyl
R8* V ’S.O* Rn- R12 are any of n » 0 to 4 m « 0 to 4