Bridgeport Fittings, Inc. ("Bridgeport") appeals from the entry of judgment in favor of Arlington Industries, Inc. ("Arlington") after a jury trial on Arlington's suit for infringement of its United States Patent No. 5,736,674 ("'674 patent"). Arlington Indus., Inc. v. Bridgeport Fittings, Inc., No. 3:99-CV-1857 (M.D.Pa. June 25, 2002). Bridgeport also appeals from the entry of an amended permanent injunction. Arlington Indus., Inc. v. Bridgeport Fittings, Inc., No. 3:99-CV-1857 (M.D.Pa. July 24, 2002). Bridgeport argues that the district court erroneously construed three claim limitations and that these errors rendered the court's jury instructions prejudicially erroneous. Because in each of the constructions it advances on appeal, Bridgeport improperly seeks to import limitations from the specification into the claims and otherwise fails to persuade the court that the jury instructions contained prejudicial error, we affirm both the judgment and the amended permanent injunction.

In addition to other products, Bridgeport and Arlington competitively make and supply electrical box extenders. In 1995 or 1996, Thomas J. Gretz ("Gretz"), an employee of Arlington, and other officials from Arlington met with William McShane ("McShane"), a representative of McShane Industries, a third competitor. McShane held two issued United States Patents, Nos. 5,042,673 ("'673 patent") and 5,117,996 ("'996 patent"), both directed to electrical box extensions. The McShane box extenders were described in the McShane patent specifications as comprising a "generally tubular body . . . sized to receive an electrical device . . . and to be received within an existing electrical box." '673 patent, col. 2, ll. 28-33; '996 patent, col. 2, ll. 28-33. Embodiments of the patented box extensions disclosed in the '673 and '996 patents are shown below. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

McShane Industries sold what it characterized as a commercial embodiment of the '673 patent under the designation "DADJ." The DADJ device is depicted below. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The Arlington officials met with McShane to review McShane's DADJ box extender as a possible addition to Arlington's product line. Arlington ultimately chose not to distribute the DADJ product.

On December 27, 1996, Gretz filed with the United States Patent and Trademark Office ("PTO") application serial number 08/773,413 ("`413 application"), the patent application that ultimately issued as the `674 patent. This application was a continuation-in-part of an application Gretz submitted on June 20, 1996. Each of the applications is directed to an electric box extender, which is used to bring electrical boxes flush with newly installed sheet stock in a renovated building. As described, the box extender is provided with "wings" that "tend to flex inwardly" when the box extender is inserted into an existing electrical box. `674 patent, col. 4, l. 36. This flexing allows the box extender to "accommodat[e] a wide range of sizes of electrical boxes." Id. at col. 4, ll. 38-39. An embodiment of the patented box extender is depicted below. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The "Background of the Invention" section of both specifications discussed prior art box extenders in general terms, including the type of box extender typified by the DADJ. Gretz did not, however, disclose either the McShane patents themselves or the details of the DADJ box extender to the PTO.

Both co-pending applications were rejected as obvious over a combination of references including United States Patent No. 4,134,636 ("Kleinatland"). Gretz then submitted an amendment with accompanying argument in both cases distinguishing, inter alia, Kleinatland. After the Examiner indicated that the pending claims in the `413 patent were allowable, Gretz expressly abandoned the earlier application. The Examiner then allowed all claims of the `413 application, and, on April 7, 1998, it issued to Gretz as the `674 patent. The `674 patent was then assigned by Gretz to Arlington.

The claims at issue are reproduced below, with the disputed limitations underlined.

1. A plastic electrical box extender for extending existing electrical boxe [sic] so that the extended electrical box is flush with a surface comprising:

a face member having an outward side and an inward side and an axis through the center of said face member;

at least two flanges extending from and coplanar with said face member;

a plurality of inwardly extending flat wings each having a base and with a base edge, an inward edge opposite said base edge, and two side edges opposite one another with said wings extending from the inward side of said face, said wings symmetrically disposed about said axis through the center of said face member, said wings and said face forming a generally tubular body portion sized to receive an electrical device, said wings capable of flexing about said base toward said axis through said face thereby allowing said tubular body portion to be received within existing electrical box; and

apertures through at least two of said flanges, said apertures capable of passing screws therethrough for securing said electrical box extender to said existing electrical box.

15. A one-piece plastic electrical box extender for extending existing electrical boxe [sic] so that the extended electrical box is flush with a newly installed surface comprising:

a face member having an outward side and an inward side and an axis through the center thereof;

at least two flanges extending from and coplanar with said face member;

a plurality of sides extending from said inward side of said face member, each having a base end with a base edge and an inward edge opposite said base edge, two side edges with said sides symmetrically disposed about said axis through the center of said face member, said sides and said face member forming a generally tubular body portion sized to receive an electrical device, said sides capable of flexing about said base toward said axis through said face member thereby allowing said tubular body portion to be received within an existing electrical box;

corners connecting said side edges of said sides which are adjacent one another with said sides furthermore capable of having their corners removed to increase the degree of flexing about said base toward said axis through said face member to increase the amount of flexibility; and

apertures through at least two of said flanges, said apertures capable of passing screws therethrough for securing said electrical box extender to said existing electrical box.

On October 21, 1999, Arlington filed a patent infringement suit against Bridgeport, alleging that Bridgeport's box extender product, depicted below, infringed the `674 patent. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

In its answer, Bridgeport counterclaimed for a declaratory judgment of invalidity, of unenforceability based on inequitable conduct, and of noninfringement.

In early December 2000, when Gretz was asked during a deposition whether the wings of the DADJ box extender "flexed about the base" as that expression was used in the `674 claims, Gretz stated that the wings "bell[ied] in" or "flex[ed] inwardly." Trial Tr., Vol. 1 at 158-59. Later that month, while the Bridgeport suit was pending, Arlington submitted to the PTO a request for ex parte reexamination of the `674 patent. The prior art references Arlington supplied with the request included the McShane patents, a brochure describing the DADJ device, and the Kleinatland patent that had been applied during prosecution. An exemplar of the DADJ box extender itself was not submitted to the PTO with the request for reexamination. The PTO denied Arlington's request on March 16, 2001, ruling that the prior art did not "raise a substantial new question of patentability," because it "fails to disclose any features which have not been previously considered in the examination of patent application No. 08/773,413."

On October 25, 2001, the district court handed down its Markman Order, in which it largely adopted Arlington's proposed construction of the disputed claim limitations. This ruling was the basis for the claim construction portions of the jury instructions, to which Bridgeport objected. Trial Tr., Vol. 5 at 7. After a four-day trial, the jury returned a verdict for Arlington. It found literal infringement of Claims 1 and 15 of the `674 patent by Bridgeport. The jury also rejected Bridgeport's assertions that Claim 1 and Claim 15 were invalid as indefinite, as anticipated by the DADJ box extender, or as obvious over a combination of, inter alia, any two of the DADJ device, the McShane `673 patent, and Kleinatland. Finally, the jury found that neither Gretz nor Arlington's patent attorney had committed inequitable conduct in prosecuting the `674 patent. The district court entered a final judgment on the jury's verdict on June 25, 2002, and issued an amended permanent injunction on July 24, 2002. Bridgeport timely appealed the final judgment, the denial of its motion for partial summary judgment of claim interpretation, and the order entering the amended permanent injunction. Bridgeport does not challenge the jury findings on the issues of invalidity and inequitable conduct.

We have jurisdiction under 28 U.S.C. §§ 1292(c)(1) and 1295(a)(1).

Claim construction is a question of law that this court reviews de novo. Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1456 (Fed.Cir.1998) (en banc).

Federal Circuit law applies to our review of jury instructions involving issues of patent claim construction. See Tex. Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1201 (Fed.Cir.2002). The standard of review for jury instructions is prejudicial legal error. Id. "To prevail, the party challenging the jury instruction `must demonstrate both that the jury instructions actually given were fatally flawed and that the requested instruction was proper and could have corrected the flaw.'" Id. (quoting Biodex Corp. v. Loredan Biomedical, Inc., 946 F.2d 850, 862 (Fed.Cir.1991)). "An erroneous instruction regarding claim interpretation that affects the jury's decision on infringement is grounds for a new trial." Ecolab, Inc. v. Paraclipse, Inc., 285 F.3d 1362, 1373 (Fed. Cir.2002).

A. "Capable of Flexing"

In its instructions to the jury, the district court construed the claim limitation "capable of flexing" in Claims 1 and 15 to mean "a generalized combination of cantilever bending and bowing about the general area of the base or base end." Trial Tr., Vol. 5 at 120. Bridgeport takes issue with the district court's construction. It argues that the court erred in construing the "capable of flexing" limitation of Claims 1 and 15 to encompass anything more than "cantilever flexing." It is unclear precisely what "cantilever flexing" means as used by Bridgeport, but we infer from Bridgeport's arguments that it uses the phrase to mean a specific type of flexing: bending of a rigid wing or side solely about its base end connection, without other bending or bowing.

Claim terms are presumed to be given their ordinary and customary meaning. Johnson Worldwide Assocs., Inc. v. Zebco Corp., 175 F.3d 985, 989 (Fed.Cir.1999). The context of the surrounding words in a claim also must be considered in determining the ordinary and customary meaning of a disputed claim limitation. Brookhill-Wilk 1, LLC, v. Intuitive Surgical, Inc., 334 F.3d 1294, 1300 (Fed.Cir.2003). The intrinsic record, comprising the claims, the written description, and the prosecution history if in evidence, "must be examined in every case to determine whether the presumption of ordinary and customary meaning is rebutted." Tex. Digital Sys., 308 F.3d at 1204. A patentee may rebut this presumption by "defin[ing] claim terminology in a manner inconsistent with its ordinary meaning," Biovail Corp. Int'l v. Andrx Pharms., Inc., 239 F.3d 1297, 1301 (Fed.Cir.2001), or by disclaiming a particular interpretation of a claim term during prosecution, Biodex, 946 F.2d at 863.

We begin with the ordinary and customary meaning of the terms used in the claim limitation at issue. Johnson Worldwide, 175 F.3d at 989. The verb "flex" is defined as "to bend." Random House Unabridged Dictionary 733 (2d ed. 1993). The ordinary and customary meaning of "capable of flexing" is thus simply "capable of bending." Neither party disputes that the ordinary and customary meaning of the term "flexing" encompasses not only cantilever flexing but also bowing or bending. Bridgeport concedes that "the term `flexing,' standing alone, would connote a wide range of movements." Accordingly, we begin with the broad meaning of the term, and look to the intrinsic record to determine whether anything in that record overcomes the presumption that "flexing" has this broad ordinary meaning. See Tex. Digital Sys., 308 F.3d at 1204.

Bridgeport first turns to the language of the claims themselves. The "capable of flexing" limitation describes the wings (in Claim 1) or sides (in Claim 15), which are said to be "capable of flexing about said base toward said axis through said face." `674 patent, col. 6, ll. 47-49; col. 8, ll. 26-27. In Claim 15, this limitation is followed by the following claim language: "corners connecting said side edges of said sides... with said sides furthermore capable of having their corners removed to increase the degree of flexing about said base toward said axis through said face member." Id. at col. 8, ll. 31-35. Bridgeport's view of the effect of the removable corners limitation of Claim 15 on the proper construction of the "capable of flexing" limitation differs from that of the district court.

The district court based its construction of "flexing" in part on the fact that in Claim 15, the corners of the box extender "need not be removed," and that, if the corners are not removed, "it is fair to say that ... cantilever flexing cannot exist." Arlington Indus., Inc. v. Bridgeport Fittings, Inc., No. 3:CV-99-1857, slip op. at 8-9 (M.D.Pa. Oct. 25, 2001) ("Memorandum"). The court noted that Claim 15 provides that the corners are capable of being "removed to increase the degree of flexing." `674 patent, col. 8, ll. 33-34. This indicated to the district court that "the corners can be removed ... [to] provide cantilever flexing where there was none before." Memorandum, slip op. at 9.

Bridgeport argues that corners "removed to increase the degree of flexing" mandate a different construction of "flexing." According to Bridgeport, the reference to an "increase" in the degree of flexing means only a change in the amount of flexing, not in the type of flexing as the district court held. Bridgeport notes that "nothing in the claim language or the specification suggests bowing." It also speculates that the corners depicted in the embodiment to which Claim 15 pertains are "specially designed to fold in on themselves to permit cantilever flexing `about said base.'" It is unclear what basis Bridgeport has for this speculation. There is no reference in the specification to corners that are specially designed to permit folding. The specification states that the corner walls "may be trimmed away to create an extender with wings," and Figure 7 depicts "cut lines" that may be employed for this purpose, but there is no indication that the corner walls are designed to facilitate folding in an uncut state, as Bridgeport suggests. We see no reason to read the words "increase [in] the degree of flexing" in the manner argued by Bridgeport, and decline to find error in the district court's construction of "flexing" on this basis.

Bridgeport next contends that the meaning of "flexing" must be limited to cantilever flexing because the written description "teaches only cantilever flexing." Bridgeport draws our attention to the "Description of the Invention" section, where the "flexing of the wings" is described as a "cantilever bending." `674 patent, col. 4, ll. 1-2. Bridgeport equates this "cantilever bending" with "cantilever flexing," and essentially invites us to import a limitation from the preferred embodiments to restrict the meaning of a claim term. We have consistently warned against this approach to claim construction, which is seldom justified. See Amgen, Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313, 1325 (Fed.Cir.2003) ("Because the claims are best understood in light of the specification of which they are a part ... courts must take extreme care when ascertaining the proper scope of the claims, lest they simultaneously import into the claims limitations that were unintended by the patentee."); CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366 (Fed.Cir.2002) (holding that the presumption of ordinary meaning cannot be rebutted "simply by pointing to the preferred embodiment or other structures or steps disclosed in the specification or prosecution history"). There is no indication in the written description of the `674 patent, for example, that Gretz "acted as his own lexicographer and clearly set forth a definition of the disputed claim term." CCS Fitness, 288 F.3d at 1366. Nor do we discern therein any express disclaimer of a particular meaning of "flexing." See id. at 1367; Biodex, 946 F.2d at 863. We accordingly decline Bridgeport's invitation to restrict the meaning of the claim term based on the description of the preferred embodiments.

a. Kleinatland

Turning to the prosecution history of the `674 patent, Bridgeport argues that Gretz disclaimed non-cantilever flexing in arguments he made to distinguish prior art references. Bridgeport principally relies on the argument distinguishing Kleinatland that Gretz submitted in the course of prosecuting the `674 patent. As depicted in the figure below, Kleinatland disclosed an electrical outlet box 12, together with an extension plate 14 having "a center tubular flange portion sized for telescopic engagement with [the] outlet box." Kleinatland, col. 2, ll. 49-50. In the tubular flange portion, the side walls were connected to one another through their entire length. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The Examiner rejected Gretz's application as obvious over a combination of references, including Kleinatland. In his response, Gretz distinguished the Kleinatland box extender in the following terms:

There is no teaching in Kleinatland et al providing an electrical box extender for extending electrical boxes so that the extender is flush with a newly installed surface. There are no plurality of inwardly extending flat wings which flex about a face to allow the extender to be received within an existing electrical box. In Kleinatland et al the box is a relatively rigid telescoping member.

Bridgeport sees in this a redefinition of "flexing" to exclude bowing or bending.

In the course of prosecuting a patent application, a patentee may redefine a claim term. Biovail, 239 F.3d at 1301 ("[W]e review both the specification and applicable prosecution history to determine whether the patentee defined claim terminology in a manner inconsistent with its ordinary meaning."); Hockerson-Halberstadt, Inc. v. Avia Group Int'l, Inc., 222 F.3d 951, 955 (Fed.Cir.2000) ("The court... must examine a patent's specification and prosecution history to determine whether the patentee has given the term an unconventional meaning."). An amendment or argument made in the course of prosecution may also serve as a disclaimer of a particular interpretation of a claim term. Ekchian v. Home Depot, Inc., 104 F.3d 1299, 1304 (Fed.Cir.1997) ("[S]ince, by distinguishing the claimed invention over the prior art, an applicant is indicating what the claims do not cover, he is by implication surrendering such protection."); Southwall Techs., Inc. v. Cardinal IG Co., 54 F.3d 1570, 1576 (Fed.Cir.1995) ("The prosecution history limits the interpretation of claim terms so as to exclude any interpretation that was disclaimed during prosecution.... Claims may not be construed one way in order to obtain their allowance and in a different way against accused infringers."). We are not persuaded that Arlington's arguments distinguishing Kleinatland operate either as a redefinition of the claim term "flexing" or as a disclaimer of any particular interpretation of that term.

As shown above, Kleinatland discloses an electrical outlet box and extension plate. The extension plate "is a rigid molded plastic insulator member having a center flange portion sized for telescopic engagement with [the] outlet box." Kleinatland, col. 4, ll. 59-61. The extension plates described in the Kleinatland specification are shaped to be accommodated within the outlet boxes. There is no indication in Kleinatland that the "rigid" extension plates are intended to flex, bow, bend, or give in any manner to accommodate differently sized outlet boxes. Gretz's description of the Kleinatland device as "a relatively rigid telescoping member" was thus factually accurate and does not amount to a redefinition of "flexing" or a disclaimer of any part of the ordinary meaning of that term. Nor did Gretz's reference to the lack in Kleinatland of "inwardly extending flat wings which flex about a face to allow the extender to be received within an existing electrical box" operate as a redefinition or disclaimer. As noted above, there is no indication in Kleinatland that the extension plates flexed or yielded at all, and that is the gravamen of the argument. The reference to the "wings" set forth in one of the Arlington embodiments simply emphasizes the lack of any reference to flexing in Kleinatland, and cannot be taken as a disclaimer of everything other than cantilever flexing of separate wings, as Bridgeport would have it.

b. Reexamination Request

The arguments made by Arlington in its reexamination request relating to the McShane patents and the DADJ device described in the McShane brochure also fall short of a redefinition or disclaimer. As in the Kleinatland device, the box extenders described in the McShane documents include a tubular body "sized ... to be received within an electrical box." `673 patent, col. 7, ll. 32-33; `996 patent, col. 6, l. 68 — col. 7, l. 1. There is no indication in any of the McShane references that the tubular body was designed to flex or yield in order to be accommodated within differently sized electrical boxes. Indeed, McShane provided for different box extension sizes and shapes to fit different boxes: "[t]he electrical box extension of the present invention can be manufactured for use with an electrical box of any shape or dimensions." `673 patent, col. 6, ll. 35-37. The specification shows examples of such specially manufactured box extensions for use with cylindrical or multi-outlet boxes. Id. at col. 6, ll. 37-48. The tubular body of the McShane patents is also designed specifically to avoid contact with the box. It has a taper to allow it to be "easily ... inserted" into the electrical box in a manner that makes the body "less likely to be obstructed by rough edges, indentations, out-of-line-knockouts, broken-off or short screws, or the like." Id. at col. 4, ll. 29-34; `996 patent, col. 4, ll. 21-24. In the request for reexamination, Arlington asserted that "[t]here is no flexing of the sides or walls involved in the McShane box extenders and they are not capable of flexing about the base." It also stated that "the four sides or walls are substantially rigid." Both McShane patents state that the tubular body can be made of plastic. `673 patent, col. 9, l. 2; `996 patent, col. 7, l. 37. However, Arlington's characterizations of the McShane references are accurate in light of the lack of any discussion of flexing of the tubular body in those references, and we see therein no disclaimer or redefinition affecting the scope of the claim term "flexing." Gretz's testimony that the DADJ box extender "bell[ied] out" or "flexed inwardly" to some degree when it was subjected to pressure may have prompted Arlington to request reexamination of the `674 patent in light of the written references relating to the DADJ device, namely the McShane patents and brochure. Although Gretz's testimony was before the district court during the Markman hearing, it is of little consequence in the claim construction analysis. That analysis must be based primarily on the record established at the time the patent was granted. See Tex. Digital Sys., 308 F.3d at 1202 ("When a patent is granted, prosecution is concluded, the intrinsic record is fixed, and the public is placed on notice of its allowed claims."); Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed.Cir.1996) ("The claims, specification, and file history, rather than extrinsic evidence, constitute the public record of the patentee's claim, a record on which the public is entitled to rely."). Moreover, Gretz's testimony does not appear to be inconsistent with Arlington's characterization of the box extender disclosed in the McShane references. Arlington characterized the sides or walls of the McShane devices as "substantially rigid." The fact that the sides of the device are caused to bow when sufficient pressure is applied does not mean that they are not "substantially rigid." Indeed, the fact that the DADJ device introduced into evidence at the trial broke when a witness tried to cause it to flex proves the very point that the sides of the DADJ device were "substantially rigid." Trial Tr., Vol. 4 at 95.

Finally, Bridgeport argues that the broad construction of the "capable of flexing" limitation must be erroneous, because it renders the claims invalid in light of prior art box extenders. Specifically, Bridgeport argues that the district court's construction renders the `674 claims obvious over a combination of the McShane patents and Kleinatland. Arlington counters that because Bridgeport has not challenged the jury's finding that the `674 patent is not invalid in view of those prior art references, Bridgeport has waived any argument that the content of the references limit the terms of the claims.

Arlington's waiver argument is unavailing. The McShane and Kleinatland prior art references are of record in the prosecution history and may be consulted in the process of claim construction for what they indicate about the state of the prior art. See Tate Access Floors, Inc. v. Interface Architectural Res., Inc., 279 F.3d 1357, 1371 n. 4 (Fed.Cir.2002) ("Prior art cited in the prosecution history falls within the category of intrinsic evidence."). However, Bridgeport does not establish that the district court's construction of "capable of flexing" renders any claim of the `674 patent invalid. Its invalidity argument is chiefly based on a repetition of the assertion we have not found persuasive, namely that the DADJ device "did in fact flex or `bow' when pressure was applied to the sides." Bridgeport also asserts that the McShane `673 patent discloses an embodiment, allegedly depicted in Figure 7 of that patent, that "is used to function in the same way as the `674 patent — i.e., to flex to fit a variety of different shaped and sized electrical boxes without using multiple screws and brackets." However, that embodiment simply indicates that "such boxes can be provided alone, or together with or assembled with electrical box extensions of the present invention for use in new construction or the like." `673 patent, col. 7, ll. 10-13. There is no indication that this embodiment contemplates the use of a single box extender with electrical boxes of different shapes and sizes, as Bridgeport contends.

The district court's construction of "flexing" as "a generalized combination of cantilever bending and bowing about the general area of the base or base end" is consistent with the ordinary and customary meaning of "flexing," which is "bending." Because Bridgeport has not established that the district court's construction is in error based on the claim language, the written description, the prosecution history, or the prior art, we decline to disturb that construction on appeal.

B. "Base"/"Base End"

The district court construed the term "base" in Claim 11 to refer to "the general area that lies beneath the face member of each wing or side where the flexing occurs, rather than a structurally distinguishable finite section of each wing." Trial Tr., Vol. 5 at 120. Similarly, the court construed the term "base end" in Claim 15 to mean "a general area of each side, rather than a structurally distinguishable finite section of each side." Id. Bridgeport maintains that the meaning of these terms is facially unclear, and advocates recourse to the specification to supplement the allegedly deficient claim language. Bridgeport contends that resort to the specification establishes that the "only legitimate construction of `base end'" is "a structurally distinguishable, finite section of each of the plurality of the wings or sides."

We do not accept Bridgeport's premise that the use of "base" or "base end" "`so deprive[s] the claim of clarity' as to require resort to the other intrinsic evidence for a definite meaning." CCS Fitness, 288 F.3d at 1367 (quoting Johnson Worldwide, 175 F.3d at 990). Bridgeport simply asserts that the terms are unclear; it does not explain why one of ordinary skill in the art would not be able to understand what was signified by those terms. Because we do not find the terms at issue unclear, we decline to resort to the specification to supplant the ordinary meaning of those terms, and find no error in the district court's construction.

C. "Corners"

Finally, Bridgeport challenges the district court's construction of the "corners" limitation of Claim 15. This limitation appears in the following context: "corners connecting said side edges of said sides which are adjacent one another with said sides furthermore capable of having their corners removed to increase the degree of flexing about said base toward said axis through said face member to increase the amount of flexibility." `674 patent, col. 8, ll. 31-36. The district court construed this limitation as follows:

The word corners in Claim 15 refers to the meeting place of two lines or surfaces. They are capable of being removed to increase flexing. The corners can be removed but need not be removed. The corners are removable, but there is no requirement that the corners contain cut lines, score lines, or other physical features indicating that the corners are intended to be removed.

Trial Tr., Vol. 5 at 120. Bridgeport maintains, again on the basis of an embodiment described in the specification, that "[t]he only correct construction ... is that the corners must be specially constructed to flex or to be cut or removed to `increase the degree of flexing.'" Bridgeport's attempt to read in limitations from the specification, and its general assertion that the district court's construction causes validity problems and is thus erroneous, fail for the same reasons described above. We find no error in the district court's construction of this limitation.

The district court did not err in its construction of the claim limitations on appeal. The jury instructions incorporating those constructions were thus not prejudicially erroneous. Likewise, we find no error in the entry of the amended permanent injunction. The district court's judgment and the amended permanent injunction are accordingly

AFFIRMED.

DYK, Circuit Judge, concurring in part and dissenting in part.

I agree that the judgment should be affirmed. However, I respectfully dissent from the majority's construction of "capable of flexing" limitation in claim 1 of the `674 patent. The majority's claim construction allows the patentee to claim the very device disclaimed in the course of patent prosecution, a result plainly at odds with our precedent. However, I would affirm on the ground that the appellant did not preserve this error in its objection to the jury instruction.

The contested limitation of claim 1 of the `674 patent requires that the device have a plurality of wings "capable of flexing about said base." In its request for reexamination, Arlington distinguished claim 1 of the `674 patent from the McShane patents and the DADJ device described in the McShane brochure (which was also before the examiner) as follows:

There is no flexing of the wings involved in the McShane box extenders. The wings or sides are not capable of flexing about the base. The four wings of the electric box extender are substantially rigid. There is no adjustability by flexing of the wings designed into the McShane box extenders to allow the tubular body portions to be received in an existing electrical box. The McShane box extenders telescope into the electrical box without engaging the sides or walls.

(emphasis added). The examiner relied on this statement to deny the request for reexamination, finding that the newly cited McShane references did not "raise a substantial new question of patentability" as to the claims of the `674 patent.

The doctrine of prosecution disclaimer "limits the interpretation of claims so as to exclude any interpretation that may have been disclaimed or disavowed during prosecution in order to obtain claim allowance." Omega Eng'g, Inc. v. Raytek Corp., 334 F.3d 1314, 1324 (Fed.Cir.2003) (quoting Standard Oil Co. v. Am. Cyanamid Co., 774 F.2d 448, 452 (Fed.Cir.1985)). Arguments made in a reexamination proceeding will constitute a disclaimer of claim scope if they are "clear and unmistakable statements of disavowal." See Cordis Corp. v. Medtronic Ave, Inc., 339 F.3d 1352, 1358 (Fed.Cir.2003).

Here, Arlington has unmistakably disclaimed box extenders like those disclosed in the McShane patents and the McShane brochure featuring the DADJ device because they are not capable of flexing about the base as claim 1 requires. The question then is—what is the extent of this disclaimer? In other words, "[w]e must determine what one of ordinary skill in the art would believe to have been disclosed" by the disclaimed references. Pall Corp. v. PTI Tech., Inc., 259 F.3d 1383, 1393 (Fed.Cir.2001), vacated on other grounds, 535 U.S. 1109, 122 S. Ct. 2324, 153 L. Ed. 2d 152 (2002).

At trial there was undisputed testimony from both the inventor and Arlington's expert that the DADJ product, a device embodying the McShane patents and disclosed in the McShane brochure, was capable of bowing. Gretz, the inventor of the `674 patent, testified that when the DADJ product was inserted, "[t]he body portion bowed in." Likewise, despite the statements in Arlington's request for reexamination regarding the inflexibility of the McShane device, Arlington's expert testified that the DADJ product was capable of bowing:

[I]f you apply pressure to the end walls of the DADJ product it does flex, does it not?

You can see ... it flex to a modest degree, yes.

[W]hen you pressed on the walls of the DADJ there was some cantilever bending, correct-isn't that correct?

No, sir, there is not. There is a slight bowing that takes place.

Tr., Vol. 2 at 102-03. The majority notes that there is no indication in the McShane references themselves that the disclosed extender boxes or the DADJ product were designed to flex. Ante at 1329. But even if this is true, the uncontradicted testimony of Arlington's experts establishes that one of skill in the art would recognize that the DADJ device embodying the references that were before the examiner was capable of bowing. Under these circumstances, Arlington disclaimed a device that merely bowed, but did not flex about the base. See Pall, 259 F.3d at 1383 (holding that the relevant inquiry is what one of ordinary skill in the art would believe to have been disclosed by the prior art). Therefore, the claim term, "capable of flexing about said base" must be construed to require something more than bowing. Rather, it must be construed to require flexing about the base.

The district court instructed the jury that the "capable of flexing" limitation in claims 1 and 15 means "a generalized combination of cantilever bending and bowing about the general area of the base or base end." Trial Tr., Vol. 5 at 120. While confusing, the instruction appears to permit the jury to find that the "capable of flexing" limitation is satisfied by mere bowing. To that extent the instruction was erroneous.

Bridgeport objected to the instruction, but its proposed alternative did not, in fact, eliminate the reference to bowing; rather it simply offered a more precise definition of "cantilever bending," namely that "[t]he flexing of each of the wings about its respective base end is a cantilever flexing." However, the error lay not in the failure to provide a more precise definition of cantilever bending, but in including the reference to bowing. "`To prevail the party challenging a jury instruction must demonstrate both that the jury instructions actually given were fatally flawed and that the requested instruction was proper and could have corrected the flaw.'" Texas Digital Sys., Inc. v. Telegenix, Inc., 308 F.3d 1193, 1216 (Fed. Cir.2002) (quoting Biodex Corp. v. Loredan Biomedical, Inc., 946 F.2d 850, 862 (Fed.Cir.1991)) (emphasis added). Consequently, Bridgeport has failed to carry its burden on appeal to show prejudicial error. See Id.

I concur in the result.