concurring-in-part.
I join Judge Lourie’s opinion except for Parts II-A-2 and II-B-2. Respectfully, I conclude that the district court adopted the correct claim construction of “release means.” The only issue in dispute is the identification of the corresponding structure for the release means. The district court concluded that the corresponding *565structure was “chemical bonds and linkages.” I agree. The specification is clear: “[t]he rate of healing can be ... accelerated by attachment of a treating material, either mechanically or by chemical bond, to the inner surface of the device,” which includes a “method of medicine release by chemical bond.” '760 patent, col. 221. 4-7. This passage directly associates the claimed “release means” with the chemical bond structure, which is sufficiently specific to satisfy § 112 ¶ 6. See, e.g., Med. Inst. & Diagnostics Corp. v. Elekta AB, 344 F.3d 1205, 1213-14 (Fed.Cir.2003) (explaining that § 112 ¶ 6 requires only “some link between a generic structural reference and a claimed function” understandable to a person of skill in the art). I simply cannot fathom what more the patentee must do “to link or associate structure to function” so as to provide “sufficient specificity.” Op. at 561-62. By limiting the structure to “hydrolyzable bonds,” my colleagues punish the patentee for providing a detailed description of his preferred embodiment.
My colleagues’ erroneous construction of the “release means” limitation is all the more puzzling because it is unnecessary to resolve this case. As Parts II-A-1 and II-B-1 convincingly explain, the clear prosecution disclaimer of “devices” other than “sheets” mandates reversal of the infringement verdict. Parts II-A-2 and II-B-2 are thus entirely dicta. For these reasons, I decline to join Parts II-A-2 and II-B-2.