MEMORANDUM OPINION AND ORDER
CASTILLO, District Judge.
Schreiber Foods, Inc. brought this patent infringement action against Saputo Cheese USA Incorporated alleging that Saputo sells several imitation cheese products infringing Schreiber’s Reissue Patent No. 35,728 (“Patent 728”). The parties submitted claim-construction briefs1 in which they disagree about the scope and meaning of five terms used throughout Patent 728: “edible mass,” “about 25%,” “about 2 minutes to about 4 minutes,” “temperature of about 190° F. to about 205° F.,” and “principal protein source.” Accordingly, on February 4, 2000, we conducted a Markman hearing, Markman v. Westview Instruments, Inc., 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996), and now construe the disputed terms.
BACKGROUND
When interpreting a patent claim, we look to the intrinsic evidence of record: the patent claims, the specification, and the prosecution history. Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed.Cir.1996). Here, we perform a brief overview of Schreiber’s patent history and the patent at issue. We postpone in-depth scrutiny of these matters for the relevant analysis sections.
A. Patent Prosecution History
Schreiber was not the first to conceptualize (and then patent) the use of dry rennet casein to make imitation cheese. In 1977, Badertscher Patent No. 4,055,555 disclosed a process for solvating dry rennet casein for the purpose of creating curd from milk. (Saputo Ex. A, Patent No. 4,055,555.) Likewise, Middleton Patent No. 4,197,322 disclosed a process for producing “synthetic cheese” from dry rennet casein. (Saputo Ex. D, Patent No. 4,197,-322.) Instead, Schreiber’s invention purportedly improved the prior art by solvat-ing a larger amount of rennet casein as a percentage of the total product. According to Schreiber, the advantage of its process is a cheese analog with rennet casein as the principal, or sole, protein source.
[944]*944Schreiber filed its initial patent application, which disclosed a one-step cheese-making process, on February 28, 1977. The Patent Office immediately rejected all of Schreiber’s claims, so Schreiber abandoned its initial filing, (Saputo Ex. F, Serial No. 877,071 Tab 1, Patent Application of Feb. 18, 1978, at 2), and filed a new application2 in which Schreiber, for the first time, disclosed a two-step process for making imitation cheese. The 1978 application added a two-step process during which an “edible mass” was first formed and then other ingredients added. The patent examiner again rejected Schreiber’s claims. After two more proposed amendments, Schreiber’s application was granted, and patent number 4,444,800 (Patent 800) issued.
In 1987, however, Schreiber filed a reissue application because it believed part of Patent 800 was invalid based on prior art; namely, a cheese produced and sold by Schreiber in May 1976. (Saputo Ex. G, Tab 1, Reissue Declaration at 2.) The 1976 cheese was made pursuant to the two-step process first disclosed in Schreiber’s 1978 application. Because Schreiber did not disclose the two-step process within one year of offering for sale a product made using that process, it appeared that at least some of Patent 800’s two-step claims were barred by 35 U.S.C. § 102(b).
Schreiber’s reissue application proposed amending some existing claims and adding claims 32 through 43.3 The examiner rejected Schreiber’s reissue application because of the 1976 cheese prior art. (Saputo Ex. G, Tab 6, Office Action of Feb.17,1988^ at 5 (“Claims 1-43 are rejected ... as being anticipated by the admitted public sale of an American imitation cheese product ... more than one year prior to filing an application for the presently claimed invention.”).) After further attempts to amend the reissue application failed, Schreiber appealed the preliminary examiner’s decision to the Patent Office Board of Appeals.
The Board rejected all but one of the examiner’s rejections, which had the practical effect of affirming the examiner’s decision. (Saputo Ex. G, Tab 27, Board of Patent Appeals’ Decision.) Specifically, the Board disagreed with the examiner that certain reissue claims contained new matter, (id. at 12-14), but affirmed the examiner’s opinion that Claims 1-43 were unpatentable over the “admitted public sale” of the 1976 cheese, (id. at 4, 21-28). In other words, the Board affirmed the examiner’s opinion that Claims 1-43 set forth an obvious extension over prior art: the 1976 cheese.
Schreiber returned to the preliminary examiner and submitted another CIP application, in which it tried to distinguish the 1976 cheese and the two-step claims presented in the 1978 application while simultaneously arguing that the “edible mass” claims were not new matter. (Sa-puto Ex. G, Tab 29, Preliminary Amendment of Dec. 20, 1991, at 8.) Again, the examiner rejected the proposal. The process (proposed amendments that were then rejected) was repeated numerous times (the record contains 7 formal amendment proposals and four written rejection — opinion-like documents called “Office Actions” — plus several references to amendments proposed during telephone conferences). On February 10, 1998, more than twenty years after the process began, [945]*945Schreiber received Reissue Patent No. 35,-728 (“Patent 728”4).
B. Patent 728
Patent 728 discloses a “cheese-like product ... produced by the direct conversion of dry, particulate rennet casein to a substantially homogenous mass of cheese-like consistency” and the process for making that product. (Compl. Ex. A, Patent 728 at Abstract.) According to the patent specification, “[t]he important factors of ... the present process are the relative amounts of dry rennet casein, the solvation agent, and water.” (Patent 728, col. 2, ll. 28-31.) Specifically,
[t]he cheese-like product of this invention is based [on?] a substantially homogenous mass comprising at least about 25 percent by weight of previously dry but solvated edible rennet casein as the principal protein source, a solvation agent and water. A bland edible lipid material, i.e., an edible oil or fat, and a suitable flavoring agent (or agents) can be present in quantities sufficient to impart to the product the characteristic fat content and flavor of the desired cheese analog.
(Patent 728, col. 2, ll. 39-47.)
The claims in Patent 728 fall into two general categories: the “food product” claims and the “edible mass” claims, defined as any claim that uses the phrase “edible mass” regardless of the presence of the phrase “food product.” Of the claims asserted by Schreiber against Sapu-to products, Claims 30, 32-34, 38-40, 42, and 44 are in the “edible mass” category; Claims 4, 10, 12-13, 16-17, 31, 35-36, and 43 relate to a “food product.”5 The parties dispute whether the two terms have different meanings: Saputo maintains that “edible mass” excludes lipids and “food product” includes lipids; Schreiber eon-tends that “edible mass” and “food product” are interchangeable and that both may, but do not necessarily, include lipids.
The enumerated claims in Patent 728 contain marked similarities. For example, every claim teaches a cooking temperature for solvation of “about 190° F. to about 205° F.” for either “about 3 minutes” or “about 2 minutes to about 4 minutes.” Likewise, each claim states that rennet casein is the “principal protein source.” Almost all of the claims disclose a product or mass that is “about 25%” or “about 30%” by weight of dry rennet casein; the sole exception is Claim 32, which dictates “particulate rennet casein constituting at least 25 percent by weight of said edible mass.” (Patent 728, col. 14, 11. 18-20 (emphasis added).) Finally, the first 31 claims reveal “about 2 to about 12 percent” of solvation agent, while Claims 32-43 disclose “about 3 to about 6 percent” solvation agent.
C. Terms at Issue
Schreiber and Saputo have identified five terms in need of construction: “edible mass,” “about 25%,” “about 2 minutes to about 4 minutes,” “temperature of about 190° F. to about 205° F.,” and “principal protein source.” Schreiber argues that “edible mass” means what it says: essentially, a glob you can eat. Thus, according to Schreiber, the phrase does not exclude lipids. Saputo relies almost exclusively on the prosecution history to demonstrate that “edible mass” excludes lipids, but also maintains that the use of the phrase in the actual claims shows that “edible mass” cannot, as a logical matter, contain lipids.
As to the next three terms, the primary dispute revolves around how much “wiggle room” the term “about” allows the identified ranges. For example, Schreiber argues that “about 25%” includes 20%; [946]*946“about 2 minutes to about 4 minutes” includes 30 seconds through ten minutes; and “about 190° F. to about 205° F.” means 150° to 300°. Saputo disagrees that “about” allows such expansive readings of the disputed ranges.
Finally, although the parties agree on the meaning of “principal protein source,” they differ about when, during the imitation cheese-making process, to measure the main source of protein. Neither party clearly advocates a specific time, but we believe Schreiber wants us to measure the principal protein source once the solvation process as set out in the asserted claims is complete and Saputo wants the measurement taken when the entire manufacturing process is complete.
ANALYSIS
“[A] patent must describe the exact scope of an invention and its manufacture.” Markman, 517 U.S. at 373, 116 S.Ct. 1384. It does so via “claims,” as distinguished from the specification and examples contained in a patent. Id. (“The claim defines the scope of the patent.”) (quotation omitted). Specific, enumerated claims “provide the metes and bounds of the invention.” Jeneric/Pentron, Inc. v. Dillon Co., 1999 WL 66537, at *7 (D.Conn. Feb.3, 1999). Our task, therefore, is to determine the scope of Schreiber’s patent by construing, as a matter of law, the disputed claim terms. Markman, 517 U.S. at 388-89, 116 S.Ct. 1384.
To accomplish this goal, we consult evidence intrinsic to the patent: specifically, the patent itself,' including the specification and claims; and its prosecution history. Finnigan Corp. v. International Trade Comm’n, 180 F.3d 1354, 1363-64 (Fed.Cir.1999); Vitronics Corp., 90 F.3d at 1582. Intrinsic evidence “constitute[s] the public record of a patentee’s claim.” Jeneric/Pentron, Inc., 1999 WL 66537, at *7. When intrinsic evidence yields an unambiguous result, reliance on extrinsic evidence is improper; it “would destroy the rights of competitors to rely on the public record and design around the claimed invention.” Id.
To construe the disputed terms, we first consider the words of the claims, both asserted and non-asserted. Vitronics Corp., 90 F.3d at 1582. Generally, the words in a claim are given their ordinary and customary meaning. But “a patentee may choose to be his own lexicographer ... as long as the special definition of the term is clearly stated in the patent specification or [prosecution] history.” Id.
Second, we evaluate the specification. “The specification contains a written description of the invention which must be clear and complete enough to enable those of ordinary skill in the art to make and use [the invention].” Vitronics Corp., 90 F.3d at 1582; see also 35 U.S.C. § 112 (Specification). Additionally, the specification “shall set forth the best mode contemplated by the inventor of carrying out his invention.” 35 U.S.C. § 112. It is important, however, to distinguish the claims from the specification: “[I]t is the claims, not the written description, [that] define the scope of the patent right.” Laitram Corp. v. NEC Corp., 163 F.3d 1342, 1347 (Fed.Cir.1998) (emphasis in original). Thus, “a court may not import limitations from the written description into the claims.” Id.
Finally, we analyze the asserted claims in light of the patent’s prosecution history. Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1457 (Fed.Cir.1998); Vitronics Corp., 90 F.3d at 1582. “Prosecution history limits the interpretation of claim terms so as to exclude any interpretation that was disclaimed during prosecution.” CVI/Beta Ventures, Inc. v. Tura LP, 112 F.3d 1146, 1155 (Fed.Cir.1997). “The relevant inquiry is whether a competitor would reasonably believe that the applicant had surrendered the relevant subject matter.” Cybor Corp., 138 F.3d at 1457.
Having set forth the relevant standards, we now construe the disputed claim terms.
[947]*947A. “Edible Mass”
At issue is whether the term “edible mass,” as used in Claims 30, 32-34, 38-40, 42, and 44, can include a lipid. The claim language suggests that it cannot. First, several of the asserted claims do not recite the presence of a lipid. See, e.g., Claims 32-34, 38-40, and 44.
Furthermore, according to the “edible mass” claims that do include lipids, the manufacture of the edible mass is complete prior to the addition of any lipid material. Thus, for example, Claim 41 (upon which Claim 42 is dependent) discloses a two-step process for “preparing a simulated, non-cultured food product” by
admixing dry particulate rennet casein as the principal source of protein, water and an edible solvation agent [deleting specific quantities and types of solvation agent] ... and maintaining said admixture at said temperature for a time period of about 2 minutes to about 4 minutes to solvate said rennet casein to form an edible mass ...
admixing with said edible mass a bland, edible lipid material ...
heating said new admixture to a temperature of about 150° F. to about 300° F. with agitation, and maintaining said new admixture at said temperature with agitation to produce said food product.
(Patent 728, col. 16, 1. 63—col.17, 1. 27.) The recipe for “edible mass,” found in the first quoted paragraph, lists rennet casein, water and an edible solvation agent as the only ingredients. The second quoted paragraph explicitly demonstrates that the “edible mass” is formed prior to the addition of a lipid, otherwise the phrase (edible mass) could not be used as it is. Finally, the third paragraph describes the edible-mass-plus-lipid created pursuant to the second paragraph as a “new admixture” which, after additional cooking, produces “said food product.” Again, if the edible mass was not complete until the addition of a lipid, the phrase “new admixture” would be inappropriate.6
The prosecution history supports our conclusion that “edible mass” as used in Patent 728 is complete prior to the addition of any lipid material; that it is made from rennet casein, water and solvation agents. In an amendment submitted to the patent examiner on September 20, 1995, Schreiber relied on the lack of lipid in the edible mass to distinguish prior art:
Claims 1-3, 26-28, 32-34 and 38-40 all require that an edible mass without lipid material be formed. These claims differ from the method used to make the [1976 cheese] because it was not made by first solvating the rennet casein to form an edible mass and then mixing a lipid material with the edible mass.
(Saputo Ex. G, Tab 49, Amendment of Sept. 20, 1995, at 17.) The only amendment submitted after the 1995 Amendment contains purely formalistic changes, such as adding commas and underlining material not contained in Patent 800. (See, generally, Saputo Ex. G, Tab 50, Amendment of Sept. 3, 1996.) Schreiber cannot successfully repudiate an interpretation so explicitly argued to the Patent Office, particularly when the interpretation was advanced to avoid prior art. See Southwall Techs., Inc. v.Cardinal IG Co., 54 F.3d 1570, 1576 (Fed.Cir.1995) (“Claims may not be construed one way in order to obtain their allowance and in a different way against accused infringers.”).
Schreiber advances three arguments in support of its position that edible mass may include lipids. First, it points to various parts of the specification. But neither the narrative sections nor the examples cited by Schreiber ever use the phrase “edible mass”; instead, they refer to a “paste-like mass,” (Patent 728, col. 2,1. 26), or a “smooth, fluid mass,” (Patent 728, col. 7, 1. 57). Schreiber specially defined “edi[948]*948ble mass” to the patent examiner and may not escape that definition by resort to other terms.
Second, Schreiber argues that the removal of the phrase “without lipid material” from Claim 32' disproves our conclusion. In 1993 Schreiber proposed an amendment to Claim 32: “A substantially homogenous, aqueous, [non-lipid containing] edible mass without lipid material.” (Saputo Ex. G, Tab 33, Amendment of January 14,1993, at 5 (brackets, underline, and double underline in the original).) Schreiber maintains that the deletion of this language shows that the Patent Office did not rely on a lipidless edible mass to distinguish the 1976 cheese. But both “[non-lipid containing] and ‘without lipid material’ ” were deleted by September 1994, long before the 1995 Amendment, in which Schreiber persisted in distinguishing the “edible mass” claims from the 1976 cheese on this basis.1 (See Saputo Ex. G, Tab 44, Amendment of Sept. 26, 1994, at 9 (“32. (Four times amended) A substantially homogenous, aqueous, edible mass tuhich comprises ... ”).) This earlier change, therefore, does not undermine our reliance on Schreiber’s 1995 representations.
Finally, Schreiber insists that the transitional term “comprises” used in many of the “edible mass” claims leaves the ingredient list open-ended. However, as Schreiber argued to the Patent Office Board of Appeals, “[w]hile it is true that the claims use the phrase ‘comprising,’ other portions of the claim require the exclusion of the lipid material found in the May, 1976 product.” (Saputo Ex. G, Tab 22, Schreiber Reply Br. to the Bd. at 8; see also Tab 44, Amendment of Sept. 26, 1994, at 16-17 ( In the context of “comprises/consisting of’ transition, Schreiber states, “[a]s noted during the interview, col. 5 line 65 to col. 6 line 11 of the patent specification provides support for an edible mass that includes solvated rennet casein and other ingredients but excludes lipids.”).)
At base, Schreiber spent twenty years trying to convince the Patent Office that its edible mass claims were not barred by the 1976 cheese because the “edible mass” step, unlike the first step in manufacturing the 1976 cheese, does not include a lipid; rather, an edible mass is made from rennet casein, water, and a solvation agent. It cannot, at this late date, shift gears for the sole purpose of alleging a cause of action against Saputo. The phrase “edible mass” excludes the presence of lipid material.
B. “About” Claims
The word “about” is at the core of the debate over the next three disputed claim terms. Although the use of “about” in a patent claim obviously demonstrates a pat-entee’s intent to avoid strict numerical boundaries, the parties clash about the breadth imposed on the specified ranges by its inclusion in Patent 728. Schreiber believes “about” must be read expansively; Saputo disagrees.
The Federal Circuit has provided guidance on the impact of “about” on claim construction. See, e.g., Modine Mfg. Co. v. International Trade Comm’n, 75 F.3d 1545, 1554 (Fed.Cir.1996); Pall Corp. v. Micron Separations, Inc., 66 F.3d 1211, 1217 (Fed.Cir.1995); Conopco, Inc. v. May Dep’t Stores Co., 46 F.3d 1556, 1561 (Fed.Cir.1994). “About” means “with some approach to exactness in quantity, number, or time: Approximately,” Conopco, 46 F.3d at 1561 n. 2 (quotation omitted); “approximately” means “reasonably close to,” Quantum Corp. v. Rodime, PLC, 65 F.3d 1577, 1581 (Fed.Cir.1995). On the other hand, “ ‘about’ must be given reasonable scope .... [and must] be understood in light of the technology embodied in the invention.” Modine Mfg., 75 F.3d at 1554. In the end, the Federal Circuit appears to construe “about” much as it does other patent language — by resort to the patent itself and the prosecution history. See Pall Corp., 66 F.3d at 1217 (“We thus consider how the term ‘about 5:1 to about 7:1’ was used in the patent specification, [949]*949the prosecution history, and other claims.”); Conopco, 46 F.3d at 1561 (“The 179 specification and prosecution history do not support the argument that the inventors used this term [‘about’] differently from its ordinary meaning.”).
1. “About 25%”
Sehreiber concedes that the phrase “at least 25%,” present in asserted Claims 32 and. 44, constitutes an absolute floor of 25%. It insists, however, that “about 25%,” as used in asserted Claims 10, 12-13, 16-17, 30-31, 33, 36, 38-39, and 42-43, means “greater than about 20%.” Aside from the difficulty of including an ambiguous descriptor in a definition intended to provide a limit on an already ambiguous phrase, the public record prohibits such a broad interpretation of “about 25%.”
Sehreiber initially sought to patent an imitation cheese “containing, at least about 20% by weight of dry, particulate rennet casein.” (Saputo Ex. F, Serial No. 772, 567 Tab 1, Patent Application of Feb. 28, 1977, at 1 (Abstract of the Disclosure).) In its 1978 application, however, Sehreiber not only described a German patent covering a cheese product “containing] up to 20 weight percent of rennet casein,” but also abandoned its claim to “about 20%” in favor of “at least about 25%.” (Saputo Ex. F, Serial No. 877,071 Tab 1, Patent Application of Feb. 14, 1978, at 3-4.) The description of the German patent and use of “at least about 25%” survived the next 19 years to be included in Patent 728. (Patent 728, col. 1, 1. 66—col. 2, 1. 2 (German patent description); id. col. 2,1. 23 (“about 25%”).) Clearly Sehreiber changed its position to avoid prior art.
Furthermore, after the patent examiner rejected the 1978. Patent Application, Sehreiber answered the examiner’s other prior art concerns:
The Examiner’s attention is courteously invited to the teachings of U.S. Patents No. 4,055,555 and No. 4,096,586 [Bad-ertscher Patents] that expressly teach that ... about 21.25 weight percent represents the theoretical upper limit for solvation. Inasmuch as the present invention contemplates at least about 25 weight percent ... and prefers about 30 to about 45 percent by weight there can be no question whatever that the present invention is unobvious.
(Saputo Ex. F, Serial No. 877,071 Tab 3, Amendment of Apr. 17, 1979, at 4-5.) Again, Sehreiber' expressly disowned any claim to 21.25% and below for the purpose of making its invention “unobvious” in light of the Badertscher.
Sehreiber now maintains that Bad-ertscher’s “theoretical” 21.25% could not actually have constituted prior art because hypothetical limits do not enable one skilled in the art to reproduce the product. Schreiber’s argument, however, is to the wrong tribunal at the wrong time. If it was unhappy that the examiner considered the Badertscher patents prior art, it should have addressed the issue during the prosecution of Patents 800 and 728. Our task is not to correct alleged errors during the patent process; rather, we are interested in Schreiber’s understanding of the claim terms as demonstrated by the patent process. Here, instead of arguing that the Badertscher patents were not prior art, Sehreiber chose to distinguish them on the ground that “about 25%” does not mean “21.25%.” Sehreiber must live with that choice.
At this point Saputo argues only that Sehreiber cannot recapture concentrations below 21.25%. We agree based on the prosecution history, but decline to delineate any more specific boundaries for “about 25%.”
2. “About 2 to about 4 minutes” and “about 190° F. to about 205° F.”
Sehreiber argues the time and temperature limitations separately, but we agree with Saputo that these claim terms are interdependent. At base, we are construing a recipe. Logic dictates that higher temperatures allow shorter cooking [950]*950times while lower temperatures require longer in the oven; in other words, the two measurements are related. This, however, is as far as we follow Saputo’s argument; based on the public record, we must construe the time and temperature ranges as Schreiber advocates.
Starting with the claim language, every single claim in Patent 728 teaches a cooking temperature of “about 190° F. to about 205° F.” for either “about 3 minutes” or “about 2 minutes to about 4 minutes.” The specification, however, consistently discloses “a temperature of about 150° F. to about 300°F., preferably at about 190° F. to about 205° F.... [for] about 30 seconds to about 10 minutes, preferably about 2 minutes to about 4 minutes.” (Patent 728, col. 2,11. 54-59; see also id. at col. 6, 11. 5-11.) Example 4 illustrates a recipe cooked “at about 170° F. to about 180° F.,” but no example reveals a cooking time greater than “about three minutes.”
The prosecution history is entirely unenlightening on the issue. Saputo does not point to any prior art that would limit the ranges to those stated in the claims, as opposed to those taught in the specification. Although Schreiber apparently attempted initially to include the broad time range in its claims, the record does not reveal why the attempt was abandoned. Finally, Patent 728 does not identify the time and temperature as “important factors” of the invention. (Patent 728, col. 2, 11. 19-63 (Summary of the Invention).)
Saputo’s invocation of “common sense” is appealing at some level: when a recipe dictates two to four minutes, one does not expect to cook the batter for ten. Likewise, when the recipe states a cooking temperature of 190° F. to 205° F, setting the oven for 150° or 300° would be silly. But the recipe here is not strictly limited to the small ranges and, in fact, consistently teaches the broad ranges throughout the narrative description. Saputo refers to nothing in the public record supporting its contention that we should ignore the specification. Therefore, we construe “about 2 minutes to about 4 minute” to encompass a range of 30 seconds to ten minutes; “about 190° F. to about 205° F.” includes 150° F. to 300° F.7
C. “Principal Protein Source”
We decline the parties’ invitation to “construe” this phrase except so far as to note that Patent 728 teaches rennet casein as the majority source (i.e. more than 50%) of the product’s protein. We will not determine when the measurement is to be taken because we cannot. Even the parties’ arguments regarding the issue demonstrate the impossibility of deciding that issue in a vacuum. At the Markman hearing, Schreiber argued that “later added sources of protein do not count.” Later than what? Saputo’s argument is no clearer absent some context: “[A] product [that] possesses a greater percentage of protein derived from some source other than rennet casein does not fall within this claim limitation.” Saputo’s statement begs the question of when to measure the protein source; when does the batter (for lack of a better word) become “a product.” We invite the parties to revisit the issue in their summary judgment briefs.
CONCLUSION
The disputed claim terms are contraed as stated above. Accordingly, we vacate our stay of the summary judgment proceedings. Schreiber must file its response to Saputo’s motion on or before March 20, 2000. Saputo may file a reply, if any, on or before April 10, 2000. Finally, given our construction of the asserted claims and for the reasons stated in our Order of January 19, 2000, (R. 65), we deny Schreiber’s 56(f) motion for additional discovery. (R. 60-2.)