Appeal No. 2003-1162 Page 9 Application No. 09/468,292 essentially of” in the claims field for patents issued from 1996-current produced over 700 hits (Id.). Appellant attaches three patents using “consisting essentially of” in the context of a Markush group as Exhibits D, E, and F (Id.). The question here is not whether the phrase “consisting essentially of” is acceptable and commonly used terminology for a Markush group. The question is whether the claims under review set out and circumscribe a particular area with a reasonable degree of precision and particularity as required by 35 U.S.C. § 112, ¶ 2. Moore, 439 F.2d at 1235, 169 USPQ at 238. We are cognizant of the fact that the Examiner labeled the language “consisting essentially of” in the rejected claims as “improper Markush group language.” Such a label does not lend clarity to the issue of indefiniteness. It is not the form of the language itself which is important to the question of indefiniteness: In all cases what is required is a determination of whether “one skilled in the art would understand the bounds of the claim when read in light of the specification.” Miles Lab., Inc. v. Shandon, Inc., 997 F.2d 870, 875, 27 USPQ2d 1123, 1126 (Fed. Cir. 1993). In different contexts, the same words may result in different outcomes with respect to indefiniteness. In the case at hand, the Examiner did not rest his case on a mere deviation from the language normally used to set forth Markush groups; the thrust of the rejection was that one of ordinary skill in the art would not recognize what substituents are encompassed by the claims. The Examiner’s analysis is consistent with the law on indefiniteness. The Examiner has correctly interpreted the claim language as being open to substituents other than those explicitly recited in the Markush groups (Answer, pp. 19-20). “By using thePage: Previous 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 NextLast modified: November 3, 2007