Full opinion text
RADER, Circuit Judge. In this case, the same patent, U.S. Patent No. 4,935,507 (the '507 patent), occasions litigation in both the United States District Court for the Eastern District of Virginia and the United States District Court for the Northern District of Illinois. The Virginia District Court granted the motion of Lupin Ltd. and Lupin Pharmaceuticals Inc. (collectively Lupin) for summary judgment of noninfringement. In the other case, the Illinois District Court denied a preliminary injunction to Abbott Laboratories, the exclusive licensee of the '507 patent, based on the claim construction from the Eastern District of Virginia. Because the Eastern District of Virginia correctly construed the claims of the '507 patent and correctly discerned no genuine issues of material fact on literal infringement of claims 2-5 or infringement by equivalents of claims 1-5, this court affirms its partial summary judgment of noninfringement. Likewise, this court affirms the Northern District of Illinois’ denial of Abbott’s motion for a preliminary injunction, based in large part on the same correct claim construction. I. Abbott Laboratories, the exclusive licensee of the '507 patent, markets crystalline cefdinir according to the '507 patent under the trade name Omnicef. The Virginia case arose when Lupin sought a declaratory judgment of noninfringement against Abbott Laboratories and Astellas Pharma Inc., the owner of the '507 patent (collectively Abbott). The Food and Drug Administration had previously approved Lu-pin’s Abbreviated New Drug Application (ANDA) to market a generic version of Omnicef. Lupin’s generic product contains almost exclusively the Crystal B form of crystalline cefdinir (cefdinir monohydrate), whereas Abbott’s Omnicef product contains the Crystal A form of crystalline cefdinir (cefdinir anhydrate). Further, Lupin makes its products with processes other than those claimed in the '507 patent. For these reasons, Lupin brought the Virginia action to clarify that its proposed product would not infringe a valid patent. Abbott counterclaimed for infringement. The Eastern District of Virginia construed the claims, Lupin Ltd. v. Abbott Laboratories, 484 F.Supp.2d 448 (E.D.Va.2007) (Lupin CC Order), and ultimately granted-in-part Lupin’s motion for summary judgment of noninfringement, as to both literal and equivalent infringement for claims 2-5 and as to equivalent infringement for claim 1, Lupin Ltd. v. Abbott Labs., 491 F.Supp.2d 563 (E.D.Va.2007) (Lupin SJ Order). The parties stipulated to the dismissal without prejudice of the remaining claims (invalidity) and counterclaims (literal infringement of claim 1). In the Illinois action, Abbott sued San-doz, Inc. and Sandoz GmbH (collectively Sandoz), Teva Pharmaceuticals USA, Inc. and Teva Pharmaceuticals Industries, Ltd. (collectively Teva), Ranbaxy Laboratories, Ltd., Ranbaxy, Inc., Par Pharmaceutical Companies, Inc., and Par Pharmaceutical (all defendants, collectively, Sandoz and Teva) for infringement of the '507 patent. Like Lupin, Sandoz and Teva had previously filed ANDAs, seeking to market generic versions of Omnicef. Abbott sought a preliminary injunction in the Illinois case. For purposes of that motion, the parties agreed to adopt the Eastern District of Virginia’s claim construction from the Lupin case. Abbott Labs. v. Sandoz, Inc., 486 F.Supp.2d 767 (N.D.Ill.2007) (Sandoz PI Order). Despite this agreement, the parties to the Sandoz case disagreed as to how to interpret some of the Eastern District of Virginia’s constructions, necessitating some clarification by the Northern District of Illinois. 486 F.Supp.2d at 770-71 (disputing “Crystal A,” “peaks,” and “about,” and seeking construction of “powder X-ray diffraction pattern,” which the Eastern District of Virginia had not defined). Ultimately, the Northern District of Illinois, based on the claim construction from Virginia, denied the preliminary injunction. Both cases arrived at this court on appeal. This court heard the cases together and decides them together with this decision. II. The '507 patent has five claims, all of which Abbott asserts against Lupin as well as Sandoz and Teva. Claim 1 claims crystalline cefdinir, using its chemical name, and defining its unique characteristics with powder X-ray diffraction (PXRD) angle peaks: 1. Crystalline 7-[2-(2-a minothiazol-4-yl)-2-hydroxyiminoacetamido]-3-vinyl3-cephem.-4-carboxylic acid (syn isomer) which shows the peaks at the diffraction angles shown in the following table in its powder X-ray diffraction pattern: diffraction angle (°) about 14.7° about 17.8° about 21.5° about 22.0° about 23.4° about 24.5° about 28.1° '507 patent, col.16 11.13-27. In contrast, claims 2-5 claim crystalline cefdinir, without any PXRD peak limitations, but with descriptions of processes used to obtain the crystalline cefdinir. Claims 2 and 5 are independent: 2. Crystalline 7-[2-(2-a minothiazol-4-yl)-2-hydroxyiminoacetamido] -3-vinyl-3-cephem-4-carboxylic acid (syn isomer) which is obtainable by acidifying a solution containing 7-[2-(2-aminothia-zol-4~yl)-2-hydroxyiminoacetamido]-3vinyl -3-cephem-4-carboxylie acid (syn isomer) at room temperature or under warming. 5. Crystalline 7-[2-(2~aminothiazol-4-yl)-2-hydroxyiminoacetamido]-3-vinyl - 3-cephem-4-carboxylic acid (syn isomer) which is obtainable by dissolving 7-[2-(2-aminothiazol-4-yl)-2 -hydroxyiminoacetamido]-3-vinyl-3-cephem-4-carboxylic acid (syn isomer) in an alcohol, continuing to stir the solution slowly under warming, then cooling the solution to room temperature and allowing the solution to stand. Id. at col. 16 11.29-34, 43-50. These claims use PXRD as a way to claim the structure and characteristics of the unique crystalline form. PXRD is a method for identifying and distinguishing different crystalline compounds. The method beams X-rays toward a powdered chemical. The method then measures the ways the rays reflect or bend upon contact with the chemical. The diffraction angles and intensities vary with the type and purity of the test compound. A graph then plots the diffraction angle on one axis and the intensity on another. These graphs yield a unique “fingerprint” for each crystalline form of a chemical. A more sensitive form of X-ray diffraction is single crystal X-ray diffraction (SCXRD). As this name suggests, this method uses only a single crystal as a sample. SCXRD does not detect intensity, but produces a more precise diffraction angle measurement. The '507 patent was not the first cefdinir patent. Rather, Astellas’ prior art U.S. Patent No. 4,559,334 (the '334 patent) describes the discovery of cefdinir as a compound demonstrating high antimicrobial activity. '334 patent, col. 11 11.18-24. The '334 patent expired on May 6, 2007. The '507 patent claims priority to Japanese Patent Application No. 62-206199 (the JP '199 application), which claimed two crystalline forms of cefdinir, “Crystal A” and “Crystal B.” The JP '199 application claimed Crystals A and B very specifically, defining Crystal A by three infrared (IR)-absorption wavelengths and sixteen PXRD angles and intensities. In contrast, Crystal B featured five IR-absorption wavelengths and twenty-one PXRD angles/intensities. Despite using the JP '199 application for priority, the '507 patent’s specification differs significantly. Specifically, Abbott (actually Fujisawa Pharmaceutical Co., Ltd., Astellas’ predecessor in interest) jettisoned the Crystal B disclosure found in the JP '199 application and crafted broader claims in its prosecution of the '507 patent. Because the JP '199 applications defines Crystal A and Crystal B physiochemically rather than structurally, the forms actually represent subgenuses of crystalline cefdinir. Thus Crystals A and B comprise crystalline forms of varying structures, which in the context of this case means varying levels of hydration. The Eastern District of Virginia construed the claim terms “crystalline,” “shows,” “peaks,” and “about” as follows: 1) “crystalline” means “Crystal A as outlined in the specification”; 2) “shows” requires the display of a powder X-ray diffraction pattern which demonstrates the existence of the relevant peaks to a scientifically acceptable degree of certainty either visually or by other appropriate means of data display; 3) “peaks” is the plural of “peak;” a “peak exists at a powder X-ray diffraction angle that corresponds to an intensity measurement greater than measurements attributable to “noise” if that angle is immediately preceded by and followed by powder X-ray diffraction angle with a lower intensity measurement; “noise” refers to those portions of a PXRD pattern produced by intrinsic measurement error, and which cannot be associated with a scientifically significant quantity of the material which is the subject of the PXRD test; 4) “about” encompasses measurement errors inherently associated with powder X-ray diffraction testing. Lupin CC Order; 484 F.Supp.2d at 459, 466. The Eastern District of Virginia also concluded that claims 2-5 were product-by-process claims. Id. Later the district court concluded that the process terms of claims 2-5, indicated by the phrase “obtainable by,” limit the claims to the specified processes and process steps. In reaching that conclusion, the trial court followed this court’s opinion in Atlantic Thermoplastics Co. v. Faytex Corp., 970 F.2d 834 (Fed.Cir.1992). Lupin SJ Order, 491 F.Supp.2d at 567-68; Lupin Ltd. v. Abbott Labs., No. 3:06-CV-400 (E.D.Va. May 10, 2007) (Lupin PbyP Order). In the Lupin appeal, Abbott challenges only the Eastern District of Virginia’s constructions of “crystalline” and “obtainable by.” III. Evaluation of a summary judgment of nonmfringement requires two steps: claim construction, which this court reviews without deference, Cybor Corp. v. FAS Technologies, Inc., 138 F.3d 1448, 1451 (Fed.Cir.1998) (en banc), and comparison of the properly construed claims to the accused product, process, or composition of matter, which in the context of summary judgment also occurs without deference, see Ormco Corp. v. Align Technology, Inc., 498 F.3d 1307, 1312 (Fed.Cir.2007). Although infringement by equivalency is a question of fact, this court may affirm summary judgment “where no reasonable fact finder could find equivalence.” Sage Prods., Inc. v. Devon Indus., Inc., 126 F.3d 1420, 1423 (Fed.Cir.1997) (citing Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17, 39 n. 8, 117 S.Ct. 1040, 137 L.Ed.2d 146 (1997)). A. Claim Construction Because the claims define the patent right, see Innova/Pure Water, Inc. v. Safari Water Filtration Systems, Inc., 381 F.3d 1111, 1115 (Fed.Cir.2004), naturally “the claims themselves provide substantial guidance as to the meaning of particular claim terms.” Phillips v. AWH Corp., 415 F.3d 1303, 1314 (Fed.Cir.2005) (en banc). But the claims “must be read in view of the specification, of which they are a part.” Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed.Cir.1995) (en banc), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996). A patent’s specification provides necessary context for understanding the claims, and “is always highly relevant to the claim construction analysis.” Phillips, 415 F.3d at 1315 (quoting Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed.Cir.1996)). While equally true in a general sense, sometimes the specification offers practically incontrovertible directions about claim meaning. For example, inventors may act as their own lexicographers and give a specialized definition of claim terms. See id. at 1316. Likewise, inventors and applicants may intentionally disclaim, or disavow, subject matter that would otherwise fall within the scope of the claim. See id. When consulting the specification to clarify the meaning of claim terms, courts must take care not to import limitations into the claims from the specification. This court has recognized the “fine line between” the encouraged and the prohibited use of the specification. Comark Commc’ns, Inc. v. Harris Corp., 156 F.3d 1182, 1186 (Fed.Cir.1998). When the specification describes a single embodiment to enable the invention, this court will not limit broader claim language to that single application “unless the patentee has demonstrated a clear intention to limit the claim scope using ‘words or expressions of manifest exclusion or restriction.’ ” Liebel-Flarsheim Co. v. Medrad, Inc., 358 F.3d 898, 906 (Fed.Cir.2004) (quoting Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1327 (Fed.Cir.2002)). By the same token, the claims cannot “enlarge what is patented beyond what the inventor has described as the invention.” Biogen, Inc. v. Berlex Labs., Inc., 318 F.3d 1132, 1140 (Fed.Cir.2003) (quoting Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed.Cir.2001)). Thus this court may reach a narrower construction, limited to the embodiment(s) disclosed in the specification, when the claims themselves, the specification, or the prosecution history clearly indicate that the invention encompasses no more than that confined structure or method. See Liebel-Flarsheim, 358 F.3d at 908. Along with the specification, the prosecution history is “intrinsic evidence” of the meaning of the claims, because it “provides evidence of how the [United States Patent & Trademark Office (PTO) ] and the inventor understood the patent.” Phillips, 415 F.3d at 1317. Although often producing ambiguities occasioned by ongoing negotiations between the inventor and the PTO, “the prosecution history can often inform the meaning of the claim language by demonstrating how the inventor understood the invention and whether the inventor limited the invention in the course of prosecution, making the claim scope narrower than it would otherwise be.” Id. “[C]lear and unmistakable” statements during prosecution may also disavow claim scope. Computer Docking Station Corp. v. Dell, Inc., 519 F.3d 1366, 1374 (Fed.Cir.2008) (quoting Purdue Pharma L.P. v. Endo Pharms., Inc., 438 F.3d 1123, 1136 (Fed.Cir.2006)). Again owing in part to the inherent ambiguities of prosecution history, the doctrine of prosecution disclaimer only applies to unambiguous disavowals. See id. at 1375. 1. “crystalline” The Eastern District of Virginia’s construction of “crystalline” in claims 1-5 as “Crystal A” included the important caveat “as outlined in the specification.” Lupin CC Order, 484 F.Supp.2d at 459. Although the Eastern District noted the parties agreed that “crystalline” ordinarily means exhibiting “uniformly arranged molecules or atoms,” id. at 454, the court relied on the language of the claims themselves, the specification, and the prosecution history to arrive at the more specific meaning recited in the specification. The '507 specification states that “Crystal A of the compound (I) [cefdinir] shows its distinguishing peaks” at the seven particular PXRD angles enumerated in claim 1. '507 patent col.l 11.51-62. Indeed, the phrase “Crystal A of the compound (I)” appears throughout the written description, and the patent offers the following definition: “any crystal of the compound (I) which shows substantially the same diffraction pattern [as in the table in col.l/claim 1] is identified as Crystal A of the compound (I).” Id. at col.l 1.67-col.2 1.2. As the Eastern District correctly concluded: Had Astellas intended, in the chart found in column 1, to distinguish Crystal A from other forms of crystalline cefdinir that also fall within the scope of claim 1, it would have listed, at a minimum, an eighth peak associated only with Crystal A. However, by listing in column 1 only the same seven ‘distinguishing’ peaks featured in Claim 1, Astellas confirmed that Crystal A was synonymous with the invention listed in Claim 1. Lupin CC Order, 484 F.Supp.2d at 456-57. The problem, within the confines of claim 1, is that defining “crystalline” as “Crystal A, ” where “Crystal A” incorporates the seven PXRD peak limitations, arguably renders the remainder of that claim redundant. To distinguish the invention, however, the specification refers several times to “Crystal A of the compound (I) of the present invention,” see, e.g., '507 patent, col.2 11.15-17, and offers no suggestion that the recited processes could produce non-Crystal A compounds, even though other types of cefdinir crystals, namely Crystal B, were known in the art. As noted earlier, the Crystal B formulation actually appears in the parent JP '199 application. Thus, Abbott knew exactly how to describe and claim Crystal B compounds. Knowing of Crystal B, however, Abbott chose to claim only the A form in the '507 patent. Thus, the trial court properly limited the term “crystalline” to “Crystal A.” The trial court’s definition correctly identifies claim l’s literal scope. Unlike claim 1, claims 2-5 do not recite the seven PXRD peaks expressly associated with Crystal A in the '507 specification. Nonetheless, the Eastern District of Virginia limited “crystalline” to “Crystal A” in these claims as well, The trial court gave two reasons for this limitation. First, “[t]he process steps detailed in those claims [claims 2-5] correspond with the processes for making Crystal A disclosed in the specification under the heading ‘The Process For Preparing Crystal A of The Compound (I).’ ” Id. at 457 (quoting '507 patent, col.2 11.13-14). Second, the parent JP '199 application recited these steps “to distinguish between preparations of Crystal A and Crystal B.” Id. (citing JP '199 application, col.6 11.1-25). In limiting “crystalline” to “Crystal A” in claims 1-5, the Eastern District of Virginia did not improperly import the preferred embodiment into the claims. Initially, Crystal A is the only embodiment described in the specification. As discussed above, the specification’s recitation of Crystal A as its sole embodiment does not alone justify the trial court’s limitation of claim scope to that single disclosed embodiment. See Liebelr-Flarsheim, 358 F.3d at 906 (“[T]his court has expressly rejected the contention' that if a patent describes only a single embodiment, the claims of the patent must be construed as being limited to that embodiment.”). In this case, however, the rest of the intrinsic evidence, including the prosecution history and the priority JP '199 application, evince a clear intention to limit the '507 patent to Crystal A as defined by the seven PXRD peaks in the specification and in claim 1. Initially, the Eastern District of Virginia properly considered the JP '199 application as relevant objective evidence of the inventor’s knowledge at the filing of the '507 patent. While statements made during prosecution of a foreign counterpart to a U.S. patent application have a narrow application to U.S. claim construction, Pfizer Inc. v. Ranbaxy Labs. Ltd., 457 F.3d 1284, 1290 (Fed.Cir.2006), in this case the JP '199 application is part of the prosecution history of the '507 patent itself. Indeed the '507 patent claims priority from the JP '199 application. Furthermore, the trial court did not rely on attorney argument or amendments during a foreign prosecution as in Pfizer, but consulted only the contents of the foreign priority application. The JP '199 application strongly suggests that the '507 patent intentionally excluded Crystal B compounds. As discussed above, the JP '199 application establishes unequivocally that Abbott knew and could describe both Crystal A and Crystal B. Abbott could have retained the disclosure of Crystal B to support the broader claims of the '507 patent, but instead disclosed and claimed A alone. Furthermore, the prosecution history of the '507 patent shows a clear and intentional disavowal of claim scope beyond Crystal A. Co-inventor Takao Takaya, who prepared samples according to Examples 14 and 16 of the prior art '334 patent and a sample of “Crystal A of the present application,” offered a declaration that Crystal A was more stable than the prior art samples from the '334 patent. An analytical chemist, Yoshihiko Okamato, corroborated this evidence. J.A. 501-04. Beyond these declarations, the applicant specifically limited the invention to Crystal A: “the method of preparation of the crystalline form of the presently claimed compounds is not considered the heart of the present invention. The crystalline form of the compound represents the inventive concept hereof, and it is clear that [the '334 patent] does not anticipate or suggest said crystalline form.” J.A. 511 (Response to Office Action of May 11, 1989, received October 27,1989, at 6). Given the exclusive focus on Crystal A in the specification as well as the prosecution history of the '507 patent, the Eastern District of Virginia properly limited “crystalline” in claims 1-5 to “Crystal A.” 2. proper interpretation of product-by process claims This court addresses Part III.A.2 of this opinion en banc, which addresses the proper interpretation of product-by-process claims in determining infringement. Claims 2-5 of the '507 patent begin by reciting a product, crystalline cefdinir, and then recite a series of steps by which this product is “obtainable.” The Eastern District of Virginia correctly categorized claims 2-5 as product-by-process claims. On appeal, Abbott argues that the Eastern District erred in construing the process steps of claims 2-5 under the rule in Atlantic Thermoplastics, 970 F.2d at 846-47, that “process terms in product-by-process claims serve as limitations in determining infringement,” rather than in accordance with Scripps Clinic & Research Foundation v. Genentech, Inc., 927 F.2d 1565, 1583 (Fed.Cir.1991) (“[T]he correct reading of product-by-process claims is that they are not limited to product prepared by the process set forth in the claims.”). This court takes this opportunity to clarify en banc the scope of product-by-process claims by adopting the rule in Atlantic Thermoplastics. In Atlantic Thermoplastics, this court considered the scope of product-by-process claim 26 in the patent at issue: “[t]he molded innersole produced by the method of claim 1.” 970 F.2d at 836. The patentee urged that competing, indistinguishable innersoles made by a different method nonetheless infringed claim 26. Id. at 838. This court rejected the patentee’s position. This court in Atlantic Thermoplastics construed product-by-process claims as limited by the process. Id. at 846-47. This rule finds extensive support in Supreme Court opinions that have addressed the proper reading of product-by-process claims. See Smith v. Goodyear Dental Vulcanite Co., 93 U.S. 486, 493, 23 L.Ed. 952 (1877) (“The process detailed is thereby made as much a part of the invention as are the materials of which the product is composed.”); Goodyear Dental Vulcanite Co. v. Davis, 102 U.S. 222, 224, 26 L.Ed. 149 (1880) (“[T]o constitute infringement of the patent, both the material of which the dental plate is made ... and the process of constructing the plate ... must be employed.”); Merrill v. Yeomans, 94 U.S. 568, 24 L.Ed. 235 (1877); Cochrane v. Badische Anilin & Soda Fabrik, 111 U.S. 293, 4 S.Ct. 455, 28 L.Ed. 433 (1884) (BASF); The Wood-Paper Patent, 23 Wall. 566, 90 U.S. 566, 596, 23 L.Ed. 31 (1874); Plummer v. Sargent, 120 U.S. 442, 7 S.Ct. 640, 30 L.Ed. 737 (1887); Gen. Elec. Co. v. Wabash Appliance Corp., 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402 (1938); see also Atl. Thermoplastics, 970 F.2d at 839-42 (discussing each of these cases). In these cases, the Supreme Court consistently noted that process terms that define the product in a product-by-process claim serve as enforceable limitations. In addition, the binding case law of this court’s predecessor courts, the United States Court of Customs and Patent Appeals (see In re Hughes, 496 F.2d 1216, 1219 (CCPA 1974) (acknowledging that “true product claims” are “broader” in scope than product-by-process claims)), and the United States Court of Claims (see Tri-Wall Containers v. United States, 187 Ct.Cl. 326, 408 F.2d 748, 751 (1969)), followed the same rule. This court’s sister circuits also followed the general rule that the defining process terms limit product-by-process claims. See, e.g., Hide-Ite Leather v. Fiber Prods., 226 F. 34, 36 (1st Cir.1915) (“It is also a well-recognized rule that, although a product has definite characteristics by which it may be identified apart from the process, still, if in a claim for the product it is not so described, but is set forth in the terms of the process, nothing can be held to infringe the claim which is not made by the process.”); Paeco, Inc. v. Applied Moldings, Inc., 562 F.2d 870, 876 (3d Cir.1977) (“A patent granted on a product claim describing one process grants no monopoly as to identical products manufactured by a different process.”). Indeed, this court itself had articulated that rule: “For this reason, even though product-by-process claims are limited by and defined by the process, determination of patentability is based on the product itself.” In re Thorpe, 777 F.2d 695, 697 (Fed.Cir.1985) (emphasis added). The Supreme Court has long emphasized the limiting requirement of process steps in product-by-process claims. In BASF, the Court considered a patent relating to artificial alizarine. Specifically, the patent claimed “[a]rtificial alizarine, produced from anthracine or its derivatives by either of the methods herein described, or by any other method which will produce a like result.” 111 U.S. at 296, 4 S.Ct. 455 (quoting U.S. Pátent Reissue No. RE 4,321). In turn, the specification generally described a method for making artificial alizarine involving anthracine or its derivatives. Alizarine had been in use for thousands of years as a red textile dye, traditionally extracted from, madder root. Pure alizarine has the chemical formula C14H804, but “artificial alizarines” available in the market at the time of the litigation varied from almost completely pure alizarine, to combinations of alizarine and anthrapurpurine, to pure purpurine containing no alizarine whatsoever. Id. at 309-10, 4 S.Ct. 455. The defendant’s product contained approximately sixty percent anthrapurpurine. Thus both alizarine and artificial alizarines were known in the prior art. The Supreme Court clearly articulated some of the scope and validity problems that arise when process limitations of product-by-process claims are ignored: [The defendant’s product] is claimed by the plaintiff to be the artificial alizarine described in No. 4,321, and to be physically, chemically, and in coloring properties similar to that. But what that is is not defined in No. 4,321, except that it is the product of the process described in No. 4,321. Therefore, unless it is shown that the process of No. 4,321 was followed to produce the defendant’s article, or unless it is shown that that article could not be produced by any other process, the defendant’s article cannot be identified as the product of the process of No. 4,321. Nothing of the kind is shown. * * * If the words of the claim are to be construed to cover all artificial alizarine, whatever its ingredients, produced from anthracine or its derivatives by methods invented since Graebe and Liebermann invented the bromine process, we then have a patent for a product or composition of matter which gives no information as to how it is to be identified. Every patent for a product or composition of matter must identify it so that it can be recognized aside from the description of the process for making it, or else nothing can be held to infringe the patent which is not made by that process. Id. at 310, 4 S.Ct. 455 (emphasis added). After BASF, the Supreme Court continued to emphasize the importance of process steps in evaluating the infringement of product-by-process claims. See, e.g., Plummer, 120 U.S. at 448, 7 S.Ct. 640 (“[W]hatever likeness that may appear between the product of the process described in the patent and the article made by the defendants, their identity is not established unless it is shown that they are made by the same process.”); Gen. Elec. Co., 304 U.S. at 373, 58 S.Ct. 899 (“[A] patentee who does not distinguish his product from what is old except by reference, express or constructive, to the process by which he produced it, cannot secure a monopoly on the product by whatever means produced.” (footnote omitted)). Thus, based on Supreme Court precedent and the treatment of product-by-process claims throughout the years by the PTO and other binding court decisions, this court now restates that “process terms in product-by-process claims serve as limitations in determining infringement.” Atl. Thermoplastics, 970 F.2d at 846-47. As noted earlier, this holding follows this court’s clear statement in In re Thorpe that “product by process claims are limited by and defined by the process.” 777 F.2d at 697. More recently, the Supreme Court has reiterated the broad principle that “[e]ach element contained in a patent claim is deemed material to defining the scope of the patented invention.” Warner-Jenkinson, 520 U.S. at 19, 117 S.Ct. 1040. Although Wamer-Jenkinson specifically addressed the doctrine of equivalents, this rule applies to claim construction overall. As applied to product-by-process claims, Wamer-Jenkinson thus reinforces the basic rule that the process terms limit product-by-process claims. To the extent that Scripps Clinic is inconsistent with this rule, this court hereby expressly overrules Scripps Clinic. The dissenting opinions lament the loss of a “right” that has never existed in practice or precedent — the right to assert a product-by-process claim against a defendant who does not practice the express limitations of the claim. This court’s en banc decision in no way abridges an inventor’s right to stake claims in product-by-process terms. Instead this decision merely restates the rule that the defining limitations of a claim — in this case process terms — are also the terms that show infringement. Thus this court does not question at all whether product-by-process claims are legitimate as a matter of form. The legitimacy of this claim form was indeed a relevant issue in the nineteenth century when Ex parte Painter, 1891 C.D. 200, 200-01 (Comm’r Pat. 1891), and some later cases were before the Commissioner of Patents. However, this court need not address that settled issue. The issue here is only whether such a claim is infringed by products made by processes other than the one claimed. This court holds that it is not. The jurisprudence of the Court of Customs and Patent Appeals — a court with virtually no jurisdiction to address infringement litigation — can shed little light on the enforcement of the only claim limitations that an applicant chooses to define the invention. Indeed, this court’s venerable predecessor expressed its ambivalence towards the relevant infringement analysis: The policy of the Patent Office in permitting product-by-process type claims to define a patentable product, where necessary, has developed with full cognizance of the fact that in infringement suits some courts have construed such claims as covering only a product made by the particular process set forth in the claim and not to the product per se. In re Bridgeford, 53 C.C.P.A. 1182, 357 F.2d 679, 683 n. 5 (1966). The reference to “some courts” in this prior citation, as this court notes en banc, includes the United States Supreme Court and every circuit court to consider the question, including this circuit. See also Jon S. Saxe & Julian S. Levitt, Product-by-Process Claims and Their Current Status in Chemical Patent Office Practice, 42 J. Pat. Off. Soc’y 528, 530 (1960) (“[P]roduet-by-process claims have met with a most strict interpretation in the courts in infringement proceedings .... [T]he courts uniformly hold that only a product produced by the claim-designated process may be held to infringe the claim.”) (citing Gen. Elec. Co., 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402 and BASF, 111 U.S. at 310, 4 S.Ct. 455). Product-by-process claims, especially for those rare situations when products were difficult or impossible to describe, historically presented a concern that the Patent Office might deny all product protection to such claims. See In re Butler, 17 C.C.P.A. 810, 813, 37 F.2d 623 (1930) (“Process claims are valuable, and appellant thinks he is entitled to them; but it is submitted that he should not be limited to control of the process when the article which that process produces is new and useful.”). In the modern context, however, if an inventor invents a product whose structure is either not fully known or too complex to analyze (the subject of this case — a product defined by sophisticated PXRD technology — suggests that these concerns may no longer in reality exist), this court clarifies that the inventor is absolutely free to use process steps to define this product. The patent will issue subject to the ordinary requirements of patentability. The inventor will not be denied protection. Because the inventor chose to claim the product in terms of its process, however, that definition also governs the enforcement of the bounds of the patent right. This court cannot simply ignore as verbiage the only definition supplied by the inventor. This court’s rule regarding the proper treatment of product-by-process claims in infringement litigation carries its own simple logic. Assume a hypothetical chemical compound defined by process terms. The inventor declines to state any structures or characteristics of this compound. The inventor of this compound obtains a product-by-process claim: “Compound X, obtained by process Y,” Enforcing this claim without reference to its defining terms would mean that an alleged infringer who produces compound X by process Z is still liable for infringement. But how would the courts ascertain that the alleged infringer’s compound is really the same as the patented compound? After all, the patent holder has just informed the public and claimed the new product solely in terms of a single process. Furthermore, what analytical tools can confirm that the alleged infringer’s compound is in fact infringing, other than a comparison of the claimed and accused infringing processes? If the basis of infringement is not the similarity of process, it can only be similarity of structure or characteristics, which the inventor has not disclosed. Why also would the courts deny others the right to freely practice process Z that may produce a better product in a better way? In sum, it is both unnecessary and logically unsound to create a rule that the process limitations of a product-by-process claim should not be enforced in some exceptional instance when the structure of the claimed product is unknown and the product can be defined only by reference to a process by which it can be made. Such a rule would expand the protection of the patent beyond the subject matter that the inventor has “particularly point[ed] out and distinctly claim[ed]” as his invention, 35 U.S.C. § 112 ¶ 6. Thus, the Eastern District of Virginia correctly applied the rule that the recited process steps limit the product-by-process claims 2-5 for any infringement analysis. 3. “obtainable by” In this case, Abbott’s plain language argument, that “obtainable by” introduces an optional process, even if “obtained by” would introduce limiting process steps, is also unavailing. The BASF case, an analogous situation to this case, controls. As noted above, the Supreme Court in BASF considered the following claim language: “Artificial alizarine, produced from anthracine or its derivatives by either of the methods herein described, or by any other method which will produce a like result.” 111 U.S. at 296, 4 S.Ct. 455 (emphasis added). The patentee argued that even though the defendant did not make artificial alizarine by “either of the methods herein described,” the claim should capture the product because of the “or by another method” language. Id. at 309, 4 S.Ct. 455. The Supreme Court refused to attach importance to those expansive words: “No. 4,321 furnishes no test by which to identify the product it covers, except that such product is to be the result of the process it describes.” Id. at 305, 4 S.Ct. 455. Abbott’s claims 2-5, like those in BASF and like product-by-process claims in general, do not furnish any test by which to identify the cefdinir crystals except that they are the result of their respectively claimed processes. As per BASF, Abbott’s claim cannot capture a product obtained by or obtainable by processes other than those explicitly recited in the claims. If this court were to strip the process elements from the claims, as Abbott would urge, for infringement purposes, there would then be nothing to differentiate independent claim 2 from independent claim 5. After all, if those claims are not bound by the process terms but only “define” the basic cefdinir compound, then each of the claims recite the same thing, over and over again. Though Abbott argues that it merely intends to give meaning to the word “obtainable,” it instead seeks to have this court render meaningless the explicit process limitations that the applicant chose to define its invention. The intrinsic evidence in this case further rebuts Abbott’s contention that its claims are not limited to those products actually obtained by the processes recited. In column 2 of the '507 patent, under the title heading “The Process for Preparing Crystal A of the Compound (I),” the patentee used specific language to describe the very two processes that are mirrored in claims 2 and 5. '507 patent col.2 11.13-51. This language is not open-ended, nor does it constitute a mere description of the product by reference to the manner in which it can be made, as Abbott argues. By drafting claims 2 and 5 to incorporate these specific processes, Abbott made a conscious choice to place process requirements on its claimed product. If Abbott had wanted to obtain broader coverage for crystalline cefdinir devoid of any process limitations, as it seeks to do here, it could have simply done so (if indeed, as it argues, it is really the product that is the heart of the invention, not the process). But it did not. The crystals of claims 2 and 5 are simply not identifiable other than by the processes disclosed in column 2. This court must enforce the ways and terms that a party chooses to define its invention. The prosecution history also does not support Abbott’s contention that “obtainable by” offers merely an optional set of definitional process conditions. During prosecution, Abbott faced obviousness rejections based on application claims 6-9, which were process claims that mirrored the very process limitations of issued claims 2-5. The PTO refused to issue the claims until one set of duplicates was can-celled. Abbott’s action in cancelling claims 6-9 demonstrates its acquiescence to the PTO’s view that the process elements of claims 2-5 are critical parts of those claims. In addition, in a response to the PTO’s office action, Abbott chose to differentiate a cited § 103 reference, Takaya, on the basis that Abbott’s claimed processes are different. For these reasons, the applicant’s statement in the file wrapper that “the method of preparation ... is not considered the heart of the present invention” should not be afforded undue gravitas. The process limitations cannot be haphazardly jettisoned for an infringement analysis when they were so important in the patentability analysis. In sum, a patentee’s use of the word “obtainable” rather than “obtained by” cannot give it a free pass to escape the ambit of the product-by-process claiming doctrine. Claims that include such ambiguous language should be viewed extremely narrowly. If this court does not require, as a precondition for infringement, that an accused infringer actually use a recited process, simply because of the patentee’s choice of the probabilistic suffix “able,” the very recitation of that process becomes redundant. This would widen the scope of the patentee’s claims beyond that which is actually invented — a windfall to the inventor at the expense of future innovation and proper notice to the public of the scope of the claimed invention. For all the above reasons, the Eastern District of Virginia correctly construed the process limitations beginning with “obtainable by” in claims 2-5 as limiting the asserted claims to products made by those process steps. B. Summary Judgment In the Lupin case, the Eastern District of Virginia granted summary judgment of noninfringement of claims 2-5, both literal and by equivalents, and of claim 1 by equivalents. Lupin SJ Order. Literal infringement of claim 1, i.e., whether Lupin’s generic cefdinir product contains any Crystal A, is therefore not a live issue on appeal. As for claims 2-5, the Eastern District noted that “Abbott and Astellas have conceded that literal infringement of Claims 2-5 cannot be established if the product-by-process analysis is performed pursuant to Atlantic Thermoplastics,” given that “Abbott and Astellas have presented no evidence that Lupin is practicing the process steps set forth in Claims 2-5.” Lupin SJ Order, 491 F.Supp.2d at 568. Because the Eastern District correctly applied the rule from Atlantic Thermoplastics and likewise properly construed the limiting process terms in claims 2-5, only infringement by equivalents of claims 1-5 remains before this court. Infringement analysis under the doctrine of equivalents proceeds element-by-element; a generalized showing of equivalency between the claim as a whole and the allegedly infringing product or process is not sufficient to show infringement. See Warner-Jenkinson, 520 U.S. at 29, 117 S.Ct. 1040 (“the doctrine of equivalents must be applied to individual elements of the claim, not to the invention as a whole”). The primary test for equivalency is the “function-way-result” or “triple identity” test, whereby the patentee may show an equivalent when the accused product or process performs substantially the same function, in substantially the same way, to achieve substantially the same result, as disclosed in the claim. Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 389 U.S. 605, 608, 70 S.Ct. 854, 94 L.Ed. 1097 (1950). But, because “[d]ifferent linguistic frameworks may be more suitable to different cases,” Warner-Jenkinson, 520 U.S. at 40, 117 S.Ct. 1040, the function-way-result test is not the only test for equivalency. Equivalency may also be proven where the differences between the invention as claimed and the accused product or process, are insubstantial. Hilton Davis Chem. Co. v. Warner-Jenkinson Co., 62 F.3d 1512, 1517-18 (Fed.Cir.1995) (en banc), rev’d on other grounds, 520 U.S. 17, 117 S.Ct. 1040, 137 L.Ed.2d 146 (1997). In no case, however, may the doctrine of equivalents ignore the individual claim elements. See Warner-Jenkinson, 520 U.S. at 40, 117 S.Ct. 1040 (requiring “a special vigilance against allowing the concept of equivalence to eliminate completely any such [individual] elements”). Because “crystalline” in claims 1-5 is limited to “Crystal A” as defined by the seven PXRD peaks enumerated in claim 1 and in the specification of the '507 patent, the doctrine of equivalents cannot capture crystals that are not themselves equivalent to Crystal A. In turn, the bounds of Crystal A equivalents cannot ignore the limits on Crystal A in the '507 patent, which as discussed above, includes a conscious decision to distinguish Crystal B from the claimed invention. To recall, the applicant removed Crystal B from the U.S. prosecution of the parent JP '199 application. The '507 patent indisputably describes and claims Crystal A, and not Crystal B. The '507 patent, of course, could have claimed the known Crystal B formulation which was known to the inventors because it appeared in their priority JP '199 application. The applicants chose not to claim Crystal B. Thus Crystal B compounds, most relevantly cefdinir monohydrate, fall outside the scope, literal or equivalent, of claims 1-5 of the '507 patent. The parties agree that the “bulk” of Lupin’s cefdinir product is Crystal B, not Crystal A. The degree to which Lu-pin’s product may or may not also contain Crystal A is the central inquiry regarding the alleged literal infringement of claim 1, which is not part of the present appeal. Abbott cannot extend its exclusive right in the '507 claims under the doctrine of equivalents to embrace known but unclaimed subject matter. In other words, Abbott effectively disclaimed Crystal B during prosecution of the '507 patent, by removing the Crystal B disclosure from the parent JP '199 application and emphasizing the sole teaching of Crystal A in communications with the PTO as well as in the '507 specification itself. Abbott cannot now recapture that unclaimed subject matter under the doctrine of equivalents because the Eastern District properly interpreted claims 2-5 to limit “crystalline” to Crystal A. To expand that claim term to embrace Crystal B would ignore the specific claim limitations of the '507 patent. Alternatively this court notes that this case seems to fit within the dedication doctrine that forecloses invocation of the doctrine of equivalents. The patent applicant clearly knew of the Crystal B forms of the claimed invention because it claimed and disclosed them in its Japanese priority application. Yet it declined to claim an embodiment expressly disclosed in its priority document, thus dedicating that embodiment to the public and foreclosing any recapture under the doctrine of equivalents. See Johnson & Johnston Assocs. v. RE. Serv. Co., 285 F.3d 1046, 1054 (Fed.Cir.2002). During prosecution, Abbott chose to eschew Crystal B and focus exclusively on Crystal A compounds. Without a complete record and no arguments about validity before this court on appeal, this court cannot speculate on the reasons for this choice. Nonetheless, the parties hotly contest whether Example 14, which reports obtaining “crystals” not specifically identified or described, and/or Example 16 of the '334 patent enable cefdinir monohydrate, i.e. Crystal B type crystals. Beyond the attempt to reinflate the claims to encompass Crystal B based on mathematical comparisons of the PXRD peak patterns of Crystal A and Crystal B, Abbott also asserts that Lupin effectively admitted infringement by equivalents when it claimed before the Food and Drug Administration that its cefdinir generic was a bioequivalent to Abbott’s Omnicef product. While bioequivalency may be relevant to the function prong of the function-way-result test, bioequivalency and equivalent infringement are different inquiries. Bioequivalency is a regulatory and medical concern aimed at establishing that two compounds are effectively the same for pharmaceutical purposes. In contrast, equivalency for purposes of patent infringement requires an element-by-element comparison of the patent claim and the accused product, requiring not only equivalent function but also equivalent way and result. Different attributes of a given product may thus be relevant to bioequivalency but not equivalent infringement, and vice versa. As the Northern District of Illinois observed in the Sandoz case, “[i]f bioequivalency meant per se infringement, no alternative to a patented medicine could ever be offered to the public during the life of a patent.” Sandoz PI Order, 486 F.Supp.2d at 776. Thus, while potentially relevant, the bioequivalency of an accused product with a product produced from the patent at issue is not sufficient to establish infringement by equivalents. Because Crystal B is not an equivalent of Crystal A, the Eastern District of Virginia did not err in granting summary judgment of noninfringement of claims 2-5, both with respect to literal and equivalent infringement, and with respect to equivalent infringement of claim 1. IV. This court reviews the grant or denial of a preliminary injunction for abuse of discretion. Amazon.com, Inc. v. Barnesand.noble.com, Inc., 239 F.3d 1343, 1350 (Fed.Cir.2001). A district court may enter a preliminary injunction based on its consideration of four factors: “(1) the likelihood of the patentee’s success on the merits; (2) irreparable harm if the injunction is not granted; (3) the balance of hardships between the parties; and (4) the public interest.” Erico Int’l Corp. v. Vutec Corp., 516 F.3d 1350, 1353-54 (Fed.Cir.2008) (quoting PHG Techs., LLC v. St. John Cos., Inc., 469 F.3d 1361, 1365 (Fed.Cir.2006)). Sandoz and Teva’s Omnicef generic products, like Lupin’s, are also at least primarily cefdinir monohydrate, a Crystal B compound. Sandoz PI Order, 486 F.Supp.2d at 769. Before the Northern District of Illinois, the parties to the San-doz litigation disputed whether Sandoz and Teva’s products also contained small amounts of cefdinir anhydrate, i.e., Crystal A, which would fall within the literal scope of claim 1 of the '507 patent. Working primarily from the Eastern District of Virginia’s claim construction, to which the parties to the Sandoz litigation agreed would bind their litigation as well for purposes of the preliminary injunction motion, the Northern District of Illinois denied Abbott’s motion for a preliminary injunction, finding that Abbott was unlikely to prevail on the merits at trial. This court detects no abuse of discretion in the Northern District of Illinois’ preliminary injunction denial. As described above, the '507 patent is properly construed to exclude Crystal B, both as to literal and equivalent infringement. Thus, this court need not delve into the Northern District of Illinois’ clarifications of the Eastern District of Virginia’s claim constructions. The Northern District of Illinois succinctly concluded: “[w]e know that Crystal B was known to the plaintiffs because it had been included in the Japanese '199 patent. Thus we conclude that the plaintiffs deliberately excluded from the definition of Crystal A, cefdinir monohydrate, which is Crystal B.” Id. at 775. As to the alleged presence of small amounts of Crystal A in Sandoz and Teva’s products, Abbott’s evidence did not persuade the Northern District of Illinois. Id. This court perceives that decision as well within the trial court’s discretion. As additional support, the Northern District observed that there was no evidence that any trace amounts of cefdinir anhydrate, i.e. Crystal A, in Sandoz and Teva’s products “could be a contributing factor in the efficacy” and that even “if there is a small amount of cefdinir anhydrate in defendants’ products, we do not conclude that this could cause literal infringement.” Id. While these may be misstatements of the law, because de minimis infringement can still be infringement, see 35 U.S.C. § 271(a); see also SunTiger, Inc. v. Sci. Res. Funding Group, 189 F.3d 1327, 1336 (Fed.Cir.1999) (“If a claim reads merely on a part of an accused device, that is enough for infringement.”), this court need not reach that issue in a preliminary injunction context which affords the trial court broad leeway to discern a “likelihood of success.” Likewise the district court may have overstated the relevance of efficacy, because the '507 patent contains no claim limitations relating to efficacy. But these misstatements were harmless because they merely formed an alternative basis for the Northern District of Illinois’ reasonable assessment of the evidence proffered by Abbott for its preliminary injunction motion. As noted, this court sustains the trial court’s discretion based primarily on its administration of the proper claim construction and its finding that Abbott was not likely to show Sandoz and Teva’s products contained any Crystal A at all. CONCLUSION The Eastern District of Virginia correctly construed the '507 patent’s recitation of “crystalline” in each of the asserted claims as limited to Crystal A, as outlined in the specification. Because Abbott scrubbed all references to Crystal B in the '507 patent’s specification, which were present in the '507 patent’s parent foreign application, Abbott clearly demonstrated its intent to limit the '507 patent to Crystal A. This intent was further underscored by comments made during prosecution. As such, Abbott is unable to recapture Crystal B through broad claim language or under the doctrine of equivalents. The Eastern District of Virginia therefore properly concluded on summary judgment that Lupin’s cefdinir product did not infringe claims 1-5 literally or claims 2-5 by equivalency. Similarly, the Northern District of Illinois did not abuse its discretion in declining to enter a preliminary injunction against San-doz and Teva’s cefdinir products. AFFIRMED COSTS Each party shall bear its own costs. NEWMAN, Circuit Judge, with whom Circuit Judges MAYER and LOURIE join, dissenting from en banc Section III.A.2. The court today acts en banc to overturn a century of precedent and practice, and holds that a new product that is difficult to describe without reference to how it was made, but that is nonetheless a new and unobvious product, cannot be protected as a product if its description is aided by reference to how it was made. Heretofore a new product whose structure was not fully known or not readily described could be patented as a product by including in the product description sufficient reference to how it can be made, to distinguish the new product from prior art products. Patentability was determined as a product, independent of any process reference in the claim, and validity and infringement were based on the product itself. This expedient for patenting products whose structure was not fully known at the time of filing the patent application has been called the “rule of necessity.” It was pragmatic, fair, and just, for it attuned patent law and practice to the realities of invention. Today the court rejects this expedient and discards this practice, ruling that all claims containing a process term under the rule of necessity now must be construed, for purposes of infringement, as limited to use of any process term that was used to assist in defining the product. That is, such a product is not patented as a product, however it is produced, but is limited to the process by which it was obtained. This is a new restraint on patents for new products, particularly today’s complex chemical and biological products whose structure may be difficult to analyze with precision. It is a change of law with unknown consequences for patent-based innovation. The court acts sua sponte, without explanation of what policy is intended to be served by this change, without consideration of the technologies that may be adversely affected by elimination of this expedient, without notice to those whose property rights may be diminished. In so doing, the court departs from statute, precedent, and practice. This change is as unnecessary as it is flawed, gratuitously affecting inventions past, present, and future. I respectfully dissent. DISCUSSION For most product inventions, the process by which the product was made, whether or not the process is itself a patentable invention, is not stated in the product claims. However, as the variety and complexity of invention and technology have increased, various forms of product claims with process terms have been used in specific circumstances, depending on the nature of the invention. The form here at issue relates to product claims for new and unobvious products whose structure is not fully known, and for which process parameters are used to aid in defining the product. This claiming expedient has been recognized since at least 1891. The court today overturns this expedient for all circumstances, brooking no exception. Acting en banc for the purpose, the court rules that if any process term or descriptive aspect is included in a product claim to aid in distinguishing a new product, the claim cannot be infringed by the identical product unless the same process aspect is used in making the accused product. The court holds that it is irrelevant whether the product is new or was known, irrelevant whether the product could have been fully described by its structure at the time of the patent application, irrelevant whether the particular invention is a new product or is actually a process. The court adopts a simplistic universal rule, thereby targeting a small but significant class of inventions. The effect of this decision on innovation in complex fields of science and technology is unknown to the court, for we have had no advice on the consequences of this change of law. My dissent is directed as much to the court’s procedure, as to the substance of the court’s decision. I PROCEDURE The court has given no notice of this impending en banc action, contrary to the Federal Rules of Appellate Procedure and contrary to the Federal Circuit’s own operating procedures. The en banc court has received no briefing and held no argument, although the Federal Rules so require. The communities of inventors, innovators, and the public who may be affected by this change of law have had no opportunity to be heard. The court has received no information concerning the effect on patents that were granted based on this long-established practice, no advice on what kinds of inventions may now lie fallow because they are unprotected. Nor does the court explain its suspension of the standards of judicial process. The Federal Rules have the force of law. 28 U.S.C. § 2072. Federal Rules of Appellate Procedure 34 and 35 are here implicated. Rule 34 provides that “oral argument must be allowed in every case” unless certain specific exceptions exist: Rule 34(a)(2) Standards. Oral argument must be allowed in every case unless a panel of three judges who have examined the briefs and record unanimously agrees that oral argument is unnecessary for any of the following reasons: (A) the appeal is frivolous; (B) the dispositive issue or issues have been authoritatively decided; or (C) the facts and legal arguments are adequately presented in the briefs and record, and the decisional process would not be significantly aided by oral argument. Applying the Rule 34 standards, it is clear that (A) this appeal is not frivolous and (B) the dispositive issue has not been authoritatively decided, for it is currently being addressed en banc. There has been (C) no briefing and no record to the court, and this is not a case in which the decisional process would not be aided by oral argument. The en banc court has heard no argument, and has received neither written nor oral exploration of the diverse aspects of this long-established claiming practice. Federal Rule of Appellate Procedure 35 has also failed of compliance. That rule recognizes the exceptional nature of en banc hearing or rehearing, and identifies the two circumstances warranting such procedure: Rule 35(a)... An en banc hearing or rehearing is not favored and ordinarily will not be ordered unless: (1) en banc consideration is necessary to secure or maintain uniformity of the court’s decisions; or (2) the proceeding involves a question of exceptional importance. When an en banc hearing or rehearing is ordered sua sponte by the court, whether for uniformity of decision or on a question of exceptional importance, the hearing or rehearing must receive the appellate process set by the Rules. I agree that en banc review is appropriate, for this apparent conflict in our precedent has existed since 1992. Now that the court has undertaken to resolve the conflict, the withholding of public notice, or even notice to the parties to this case, is devoid of justification. The question is of importance, but there has been no assertion of urgency sufficient to require bypassing the standard appellate procedures. The breadth of the en banc court’s ruling, the solidity of the prece