This appeal arises from a patent infringement action brought by Pitney Bowes, Inc., against Hewlett-Packard Company in the United States District Court for the District of Connecticut. Hewlett-Packard counterclaimed for a declaratory judgment that the patent-in-suit was invalid, unenforceable, and not infringed. Pitney Bowes appeals from the grant of summary judgment of non-infringement to Hewlett-Packard. See Pitney Bowes, Inc. v. Hewlett-Packard Co., No. 3:95CV01764 (AVC) (D. Conn. Mar. 23, 1998) (the "March Non-Infringement Ruling"). Hewlett-Packard cross-appeals from the denial of its two earlier summary judgment motions with respect to non-infringement, based, respectively, upon different claim language and invalidity. See Pitney Bowes, Inc. v. Hewlett-Packard Co., No. 3:95CV01764 (AVC) (D. Conn. Feb. 9, 1998) (the "February Non-Infringement Ruling"); Pitney Bowes, Inc. v. Hewlett-Packard Co., No. 3:95CV01764 (AVC) (D. Conn. Feb. 9, 1998) (the "February Invalidity Ruling"). Our review of the claims, written description and prosecution history of the patent-in-suit leads us to conclude that, in its March Non-Infringement Ruling, the district court incorrectly construed the claim term "spots" to refer to spots of light generated by a laser beam, rather than spots of electrically discharged area on a photoreceptor. Because the grant of summary judgment of non-infringement was derived from the erroneous claim construction, we vacate the March Non-Infringement Ruling and remand for further proceedings in light of the correct claim construction. Because we vacate the grant of summary judgment and remand the case, Hewlett-Packard's cross-appeal from the denials of its two other summary judgment motions, not appealable independently, no longer arises from appealable orders and, therefore, we must dismiss both parts of the cross-appeal without addressing their merits.

The patent-in-suit, United States Patent No. 4,386,272 (the "'272 patent"), was issued on May 31, 1983 to Frank T. Cheek, Jr. and Ronald P. Sansome and assigned to Pitney Bowes. The '272 patent is entitled "Apparatus and Method for Generating Images by Producing Light Spots of Different Sizes" and is directed towards the technology of laser printing.

Laser printers, including the accused Hewlett-Packard devices in the instant case, convert electronic information into hard copy representations of images and characters, such as words and numbers. In general, laser printers operate by directing laser light onto a photoreceptor. Specifically, the photoreceptor consists of a drum, the surface of which is evenly covered with an electrical charge. When a beam of laser light strikes the drum, it dissipates a small area of the charge on the drum surface. This discharged area attracts charged toner, which is then transferred from the drum to the paper (by melting the toner particles into the paper fibers), thereby creating the final, permanent image. The photoreceptor drum is then cleaned and recharged so that the process can begin again. Each image (such as a letter or number) is composed of hundreds or thousands of these small dots of toner particles.

As a result of using similarly sized toner dots, the corners and edges of certain characters generated by laser printers can have an uneven appearance. This problem is known in the printing industry as "jaggies." The '272 patent teaches an apparatus and method for combating the problem of jaggies by using toner dots of different sizes. Indeed, the specification of the '272 patent explains that it can be used "to avoid roughened edges and improve character formation." '272 pat., col. 6, ll. 4-5.

Figure 1 of the '272 patent is illustrative of the laser printer apparatus which can be used to create such different sized toner dots:

[Tabular or Graphical Material Omitted]

As explained by reference to the numbering system depicted in Figure 1 of the '272 patent, a light source 10, such as a laser, generates a beam of light 12, which is directed through a neutral density filter 14 to control the light intensity. After passing through the neutral density filter, the light beam passes through several lenses 20, 22 and 24, until it reaches the reflecting face or facet 26 of a rotating polygonal mirror 28. The rotating polygonal mirror reflects multiple light beams 12, which then scan across, rather like a raster, onto a photoreceptor 32 mounted on a spinning drum. Each of these beams of light strikes the photoreceptor at a different location, causing the formation of a small discharged area at that point. As explained above, toner is attracted to each of these discharged areas and the image is then transferred to paper.

The '272 patent teaches two methods for varying toner dot size. First, as illustrated in Figure 1 of the '272 patent, an intensity modulator 64 can be attached to the source of the light beam. The degree of exposure on the photoreceptor is determined by two factors: (1) the intensity of the beam of light; and (2) the length of time that the beam of light remains in contact with the surface of the photoreceptor (which factor is termed the "pulse width"). Thus, as the intensity modulator increases or decreases the intensity of the beam of light, the greater or fewer the number of electrons that will be displaced on the photoreceptor. Because the size of toner dot that is ultimately produced on the paper correlates with the size of exposed area generated on the photoreceptor, the intensity modulator can be used to control toner dot size.

The second method taught by the '272 patent for varying toner dot size is that "[t]he system of this invention can also employ two power sources using parallel laser beams with each of the beams being of a different diameter and corresponding spot size." '272 pat., col. 6, ll. 5-8. Again, because different sized exposed areas are generated on the photoreceptor, this alternative, multiple beam embodiment creates different sized dots of toner on the paper, thereby reducing the problem of jaggies.

The first three claims of the '272 patent are the only claims at issue in this appeal. They claim:

1. A method of producing on a photoreceptor an image of generated shapes made up of spots, comprising:

directing a plurality of beams of light towards a photoreceptor, each beam of light generating a spot on the photoreceptor and controlling a parameter of the light beams to produce spots of different sizes whereby the appearance of smoothed edges are given to the generated shapes.

2. The method of claim 1 wherein the parameter controlled is light beam intensity.

3. Apparatus for producing on a photoreceptor an image of generated shapes made up of spots, comprising:

means for directing a plurality of beams of light toward a photoreceptor to generating [sic] a plurality of spots on the photoreceptor and means for generating spots of different sizes whereby the appearance of smoothed edges are given to the generated shapes.

'272 pat., col. 6, ll. 21-41.

The disclosure contained in the specification of the '272 patent also formed the disclosure used in the specification of two additional patents. The '272 patent stems from a parent application filed in July 1978. United States Patent No. 4,214,157 (the "'157 patent") also issued from that application and claims an invention for correcting imperfections in the polygonal mirror during the scanning process. A divisional application, filed from the same parent application, issued as United States Patent No. 4,310,757 (the "'757 patent") and claims an invention for correcting the scanning speed during the scanning process. The '272 patent resulted from a continuation application filed in connection with the application which issued as the '757 patent. A continuation-in-part application was also filed in connection with the '757 patent, which subsequently issued as United States Patent No. 4,809,021 (the "'021 patent"). The '021 patent claims an apparatus and method for smoothing the edges of laser printed characters by using toner dots of three different sizes. The description of the preferred embodiment contained in the specification of the '272 patent is apparently identical, or virtually identical, to that contained in the specifications of the '157 and the '757 patents.

On August 23, 1995, Pitney Bowes filed a complaint in the District of Connecticut alleging infringement of the '272 patent by Hewlett-Packard. In particular, Pitney Bowes contended that claims 1, 2 and 3 of the '272 patent were infringed by laser printers "manufactured, used and/or sold" by Hewlett-Packard. Complaint, 5 (Aug. 23, 1995). The accused devices are laser printers, which all incorporate Hewlett-Packard's "Resolution Enhancement technology" or "REt". It is not disputed that the accused Hewlett-Packard devices use essentially the same light scanning system as that illustrated in Figure 1 of the '272 patent. However, instead of adjusting the light beam's intensity, the accused devices solve the jaggies problem by modifying the pulse width of the light beam. On May 29, 1996, Hewlett-Packard answered the complaint and counterclaimed for a declaratory judgment that the '272 patent was invalid, unenforceable, and not infringed.

On August 11, 1997, Hewlett-Packard moved for summary judgment of non-infringement on the grounds that its accused laser printers do not create a "plurality of beams" as that term is used in claims 1 and 3 of the '272 patent. On September 3, 1997, Pitney Bowes cross-moved for summary judgment with respect to claim construction of that same term. In its February Non-Infringement Ruling, the district court denied Hewlett-Packard's motion and granted Pitney Bowes's motion. The court reasoned that the term "plurality" did not mean that the beams of light must issue simultaneously, but rather that they issue sequentially from the same light source. See February Non-Infringement Ruling, slip op. at 19. The court found support for this construction in the claims, the specification, and the prosecution history. See id. at 8-15. Consequently, the district court granted summary judgment to Pitney Bowes and denied Hewlett-Packard summary judgment of non-infringement because "the phrase 'plurality of beams of light' as used in the '272 patent means multiple beams of light generated sequentially from one or more light sources. The accused single light source printers do employ a 'plurality of beams of light' as defined in the '272 patent." Id. at 19.

On November 7, 1997, while its first motion for summary judgment was still pending, Hewlett-Packard moved for summary judgment of invalidity of the '272 patent, arguing that the inventors' oath failed to comply with 35 U.S.C. 115. On the same day that it ruled on Hewlett-Packard's first motion, the district court denied this second motion in its February Invalidity Ruling. Hewlett-Packard had argued that the inventors should have filed a supplemental declaration or oath with respect to the continuation application that resulted in the '272 patent, instead of submitting the declaration that had been submitted in connection with the parent application. In its February Invalidity Ruling, the district court rejected this argument, reasoning that the parent application and the accompanying oath adequately disclosed the matter subsequently claimed by the '272 patent. The district court also accorded a presumption of administrative correctness to the actions of the Patent and Trademark Office (the "PTO").

Also on November 7, 1997, Hewlett-Packard brought a third motion for summary judgment. This motion sought summary judgment of non-infringement based upon the claim term "spots of different sizes". The district court granted this motion in its March Non-Infringement Ruling. The dispute centered on whether the term refers to the spots of light generated by the light beam on the photoreceptor, as Hewlett-Packard argued, or describes the spots of discharged area on the photoreceptor that result from contact with the light beam, as urged by Pitney Bowes.

The court reasoned that "the plain language of the claims does not unambiguously support either proposed definition of the phrase 'spots of different sizes.'" March Non-Infringement Ruling, slip op. at 9. With respect to the specification, the court noted that Pitney Bowes had conceded that the first forty-two of the forty-four occurrences of the term "spot" referred to the spot of the light beam and that Pitney Bowes was arguing that only the final two occurrences of the term in the specification referred to the spot of discharged area on the photoreceptor. See id. at 12. However, the court rejected Pitney Bowes's argument, explaining that even these last two uses of the term "spot" were consistent with "spot" meaning the spot of the light beam. See id. at 12-13. Turning to the prosecution history, the court took note of the fact that the examiner amended the title of the patent to modify the term "spots" with the word "light" and reasoned that "[t]his is reflective of his understanding that the word 'spots' means 'light spots' in the '272 patent." Id. at 15. The court thus concluded that "the meaning of the phrase 'spots of different sizes' can be understood from a careful reading of the claims, the specification, and the prosecution history" and that that meaning referred to the different sized spots of light generated by the light beam on the photoreceptor. Id. at 16. Consequently, the court agreed with Hewlett-Packard's proposed construction and thus granted Hewlett-Packard summary judgment of no literal infringement because "[i]t is undisputed that the accused devices do not create light spots of different sizes" and summary judgment of no equivalent infringement because, by generating only single-sized spots of lights, the accused devices were specifically excluded from the scope of the claims. Id. at 18, 20.

Summary judgment of non-infringement requires a two-step analytical approach. First, the claims of the patent must be construed to determine their scope. See Carroll Touch, Inc. v. Electro Mechanical Sys., Inc., 15 F.3d 1573, 1576, 27 USPQ2d 1836, 1839 (Fed. Cir. 1993). Second, a determination must be made as to whether the properly construed claims read on the accused device. See id.

The first step of this analysis -- claim construction -- is a question of law. See Markman v. Westview Instruments, Inc., 52 F.3d 967, 979, 34 USPQ2d 1321, 1329 (Fed. Cir. 1995) (in banc), aff'd, 517 U.S. 370 (1996). Accordingly, it falls upon the district court to discern the meaning of the claim language. See id. Because it is a legal question, our review of the district court's claim construction is de novo. See Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1456, 46 USPQ2d 1169, 1174 (Fed. Cir. 1998) (in banc).

The second step of this analysis -- the determination of whether the properly construed claims read on the accused device -- is a question of fact. See Mannesmann Demag Corp. v. Engineered Metal Prods. Co., 793 F.2d 1279, 1282, 230 USPQ2d 45, 46 (Fed. Cir. 1986). Thus, summary judgment of non-infringement can only be granted if, after viewing the alleged facts in the light most favorable to the non-movant, there is no genuine issue whether the accused device is encompassed by the claims. See Fed. R. Civ. P. 56(c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986). Consequently, when summary judgment of non-infringement has been granted based upon undisputed material facts, our review of the district court's application of this second analytical step is de novo, construing, once more, all reasonable factual inferences in the non-movant's favor. See National Recovery Techs., Inc. v. Magnetic Separation Sys., Inc., 166 F.3d 1190, 1194, 49 USPQ2d 1671, 1674 (Fed. Cir. 1999).

The principal question we address on appeal concerns the first of these two analytical steps. Specifically, we must decide, based upon a de novo review, whether the district court's construction of the claim term "spots of different sizes" was erroneous. Our review of the patent document as well as the pertinent aspects of the prosecution history leads us to conclude that the district court erroneously interpreted the term to refer to spots of light generated by the beam of light, rather than the spots of discharged area on the photoreceptor. Because the district court's summary judgment of non-infringement derived from its erroneous claim construction, we vacate its judgment and remand for further proceedings consistent with this decision.

The starting point for any claim construction must be the claims themselves. See Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582, 39 USPQ2d 1573, 1576 (Fed. Cir. 1996). Here, both independent claims 1 and 3 contain a preamble stating that they claim either a method of, or an apparatus for, "producing on a photoreceptor an image of generated shapes made up of spots . . . ." '272 pat., col. 6, ll. 22-24, 33-35. Based upon this language, the "spots", which are the subject of this dispute, "produc[e] . . . an image of generated shapes". The "generated shapes" are, of course, the letters, numbers or other characters formed with fewer jaggies than under the prior art methods. These characters, "made up of spots", are "produc[ed] on a photoreceptor". Thus, based upon this preamble language, the "spots" are the constituent parts that make up the image of the desired character on the photoreceptor. The spots of light generated by the light beam cannot be the "spots" referenced in the preamble because the spots of light generated by the light beam are, it is not disputed, transient and do not themselves form the image of the desired character. The spots of discharged area created by the beam of light, however, last long enough to combine to form images of characters and they produce those images "on the photoreceptor". Accordingly, the language in the preamble of the claims strongly militates towards construing the claim term "spots of different sizes" to refer to the spots of discharged area on the photoreceptor, not the light spots generated by the beam of light.

Although our initial discussion has focused on the preamble, as opposed to the remainder of the claim language, this does not undercut its significance. "[A] claim preamble has the import that the claim as a whole suggests for it." Bell Communications Research, Inc. v. Vitalink Communications Corp., 55 F.3d 615, 620, 34 USPQ2d 1816, 1820 (Fed. Cir. 1995). If the claim preamble, when read in the context of the entire claim, recites limitations of the claim, or, if the claim preamble is "necessary to give life, meaning, and vitality" to the claim, then the claim preamble should be construed as if in the balance of the claim. Kropa v. Robie, 187 F.2d 150, 152, 88 U.S.P.Q. (BNA) 478, 480-81 (CCPA 1951); see also Rowe v. Dror, 112 F.3d 473, 478, 42 USPQ2d 1550, 1553 (Fed. Cir. 1997); Corning Glass Works v. Sumitomo Elec. U.S.A., Inc., 868 F.2d 1251, 1257, 9 USPQ2d 1962, 1966 (Fed. Cir. 1989). Indeed, when discussing the "claim" in such a circumstance, there is no meaningful distinction to be drawn between the claim preamble and the rest of the claim, for only together do they comprise the "claim". If, however, the body of the claim fully and intrinsically sets forth the complete invention, including all of its limitations, and the preamble offers no distinct definition of any of the claimed invention's limitations, but rather merely states, for example, the purpose or intended use of the invention, then the preamble is of no significance to claim construction because it cannot be said to constitute or explain a claim limitation. See Rowe, 112 F.3d at 478, 42 USPQ2d at 1553; Corning Glass, 868 F.2d at 1257, 9 USPQ2d at 1966; Kropa, 187 F.2d at 152, 88 USPQ at 480-81.

Here, the preamble is "necessary to give life, meaning, and vitality" to the claim. Kropa, 187 F.2d at 152, 88 USPQ at 480-81. The preamble statement that the patent claims a method of or apparatus for "producing on a photoreceptor an image of generated shapes made up of spots" is not merely a statement describing the invention's intended field of use. Instead, that statement is intimately meshed with the ensuing language in the claim. For example, both independent claims conclude with the clause "whereby the appearance of smoothed edges are given to the generated shapes". Because this is the first appearance in the claim body of the term "generated shapes", the term can only be understood in the context of the preamble statement "producing on a photoreceptor an image of generated shapes made up of spots". Similarly, the term "spots" is initially used in the preamble to refer to the elements that make up the image of generated shapes that are produced on the photoreceptor. The term "spots" then appears twice in each of the independent claims.1 That the claim term "spots" refers to the components that together make up the images of generated shapes on the photoreceptor is only discernible from the claim preamble. In such a case, it is essential that the court charged with claim construction construe the preamble and the remainder of the claim, as we have done here, as one unified and internally consistent recitation of the claimed invention.

Having noted the support in the preamble of the claims for construing the term "spots" to refer to the spots of discharged area on the photoreceptor and having further noted the necessity of the term's construction being consistent with that preamble language, we find further support for our claim construction in the language of the body of the claims. The independent claims refer to beams of light "generating a spot on the photoreceptor" (claim 1) or "generating a plurality of spots on the photoreceptor" (claim 3). '272 pat., col. 6, ll. 26-27, 37-38. While not dispositive, the reference to "spots" as being generated on the photoreceptor certainly supports a construction of "spots" as being the spots of discharged area on the photoreceptor. Indeed, it would seem to be more accurate to describe the spots of discharged area as being "generat[ed] . . . on the photoreceptor", than to describe the spots created by the light beams in that manner.

The final clause of both of the independent claims also supports construing the term "spots" to mean the spots of discharged area on the photoreceptor. This clause, which is materially the same for both claims, refers to producing or generating "spots of different sizes whereby the appearance of smoothed edges are given to the generated shapes". '272 pat., col. 6, ll. 28-30, 39-41. To be consistent with the preamble, this clause must mean that the "spots of different sizes" make up "the generated shapes", which have "the appearance of smoothed edges". However, the spots of light created by the beam of light do not make up "the generated shapes". Rather, they cause the elimination of charge at certain locations on the photoreceptor, which spots of discharged area make up "the generated shapes".

The use of the term "spots of different sizes" in the final clause of the independent claims also supports construing "spots" to refer to the spots of discharged area on the photoreceptor. The diameter of the beam of light is not altered when the intensity or pulse width is changed. However, changing these parameters does change the size of the discharged area on the photoreceptor. Thus, assuming arguendo that the size of the light spot generated by the beam of light is the same as the diameter of the beam of light, the parameters of pulse width and intensity cannot be used to create "spots of different sizes" and, consequently, based upon this assumption, the term "spots" would have to refer to the spots of discharged area on the photoreceptor.

Although Pitney Bowes argued before the district court that the size of the light spot generated by the beam of light does not alter in size when these parameters are adjusted, the district court rejected that argument. Instead, the district court reasoned that:

One common convention in the digital printing field is to define the size of a light spot as the area of light where the intensity exceeds a fixed threshold. Under this definition, the size of a projected light spot would change when the intensity of a light beam is varied.

March Non-Infringement Ruling, slip op. at 13 n.2. Pitney Bowes argues on appeal that, by relying upon this "common convention" in its claim construction, the district court erroneously relied upon extrinsic evidence to contradict the meaning of the claim term "spots" otherwise apparent from the intrinsic evidence.

The district court's statement that there exists a "common convention" in the "digital printing field" to define spot size "as the area of light where the intensity exceeds a fixed threshold", mirrors the terminology used in a supplemental affidavit submitted by William J. Hanson that was attached to Hewlett-Packard's reply brief in support of its motion for summary judgment of non-infringement based upon the "spots of different sizes" limitation. Adopting the terminology later used by the district court, Mr. Hanson averred that there was a "common convention" in the "digital printing field" to define spot size "as the area of the light spot where the intensity exceeds a fixed threshold". Mr. Hanson's affidavit responded to an earlier affidavit submitted by Michael Bass in support of Pitney Bowes's opposition. Professor Bass had declared:

[Hewlett-Packard's] interpretation of Claims 1, 2 and 3 as referring only to projected laser beam spots rather than the exposed regions on the photoreceptor or developed spots is wrong. Anyone skilled in the art of optics recognizes that a projected light spot does not change by varying the intensity of the laser beam. In other words, if a laser was pointed at a wall and the intensity of a laser was increased, the size of the projected beam on the wall would not change. By varying the intensity of that laser, however, one can vary the size of the developed spot in the electro photographic process. Thus, it is clear that the spot referred to in Claims 1, 2 and 3 is not simply referring to the size of the projected spot.

Rebuttal Aff. of Michael Bass, 6 (Oct. 3, 1997). In response, Mr. Hanson, Hewlett-Packard's expert, declared:

Professor Bass'[s] reasoning . . . may apply in the field of optics where optical spot size is typically defined as the diameter where the intensity is at a specified fraction of the maximum intensity. However, in fields other than optics, such as the digital printing field of the '272 patent, various definitions of light spot size are employed depending on the circumstances. For example, in high contrast imaging processes (of which laser printing is one), a common convention is to define the spot size as the area of the light spot where the intensity exceeds a fixed threshold. Under this common convention, the size of the project light spot does change by varying the intensity of the laser beam.

Supplemental Decl. of William J. Hanson, 3 (Dec. 9, 1997). In earlier deposition testimony, Professor Bass had testified that he had "never heard of" spot size being measured in this manner. Dep. of Michael Bass, at 154, l. 23 (Jun. 17, 1997). Neither party cites, nor have we found, any other reference to this "common convention" in the record. Pitney Bowes argues that, while the district court might have considered the existence of this "common convention" to be an uncontested background fact, the affidavits demonstrate that there was conflicting expert testimony regarding whether there was a "common convention" to define spot size in this particular manner. Pitney Bowes contends that reliance upon this "common convention" was ultimately determinative of the March Non-Infringement Ruling because, unless this "common convention" is accepted, the claim language and the written description both suggest that the "spots" are the spots of discharged area on the photoreceptor.

In Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1584, 39 USPQ2d 1573, 1578 (Fed. Cir. 1996), we explained that "[e]xtrinsic evidence is that evidence which is external to the patent and file history, such as expert testimony, inventor testimony, dictionaries, and technical treatises and articles." Despite the district court's statements to the contrary, Vitronics does not prohibit courts from examining extrinsic evidence, even when the patent document is itself clear. See, e.g., March Non-Infringement Ruling, slip op. at 15 ("In this case, the meaning of the phrase 'spots of different sizes' can be understood from a careful reading of the claims, the specification, and the prosecution history. Accordingly, consideration of extrinsic evidence is improper."). Moreover, Vitronics does not set forth any rules regarding the admissibility of expert testimony into evidence.2 Certainly, there are no prohibitions in Vitronics on courts hearing evidence from experts. Rather, Vitronics merely warned courts not to rely on extrinsic evidence in claim construction to contradict the meaning of claims discernible from thoughtful examination of the claims, the written description, and the prosecution history -- the intrinsic evidence. See id., 90 F.3d at 1583, 39 USPQ2d at 1579 ("In most situations, an analysis of the intrinsic evidence alone will resolve any ambiguity in a disputed claim term. In such circumstances, it is improper to rely on extrinsic evidence. In those cases where the public record unambiguously describes the scope of the patented invention, reliance on any extrinsic evidence is improper." (citations omitted and emphasis added)); see also Bell & Howell Document Management Prods. Co. v. Altek Sys., 132 F.3d 701, 706, 45 USPQ2d 1034, 1038 (Fed. Cir. 1997) ("Use of expert testimony to explain an invention may be useful. But reliance on extrinsic evidence to interpret claims is proper only when the claim language remains genuinely ambiguous after consideration of the intrinsic evidence . . . ." (emphasis added)). As Vitronics states:

Had the district court relied on the expert testimony and other extrinsic evidence solely to help it understand the underlying technology, we could not say the district court was in error. But testimony on the technology is far different from other expert testimony, whether it be of an attorney, a technical expert, or the inventor, on the proper construction of a disputed claim term, relied on by the district court in this case. The latter kind of testimony may only be relied upon if the patent documents, taken as a whole, are insufficient to enable the court to construe disputed claim terms. Such instances will rarely, if ever, occur.

Vitronics, 90 F.3d at 1585, 39 USPQ2d at 1579. Thus, under Vitronics, it is entirely appropriate, perhaps even preferable, for a court to consult trustworthy extrinsic evidence to ensure that the claim construction it is tending to from the patent file is not inconsistent with clearly expressed, plainly apposite, and widely held understandings in the pertinent technical field. This is especially the case with respect to technical terms, as opposed to non-technical terms in general usage or terms of art in the claim-drafting art, such as "comprising". Indeed, a patent is both a technical and a legal document. While a judge is well-equipped to interpret the legal aspects of the document, he or she must also interpret the technical aspects of the document, and indeed its overall meaning, from the vantage point of one skilled in the art. See Smith-Kline Diagnostics, Inc. v. Helena Lab. Corp., 859 F.2d 878, 882, 8 USPQ2d 1468, 1471 (Fed. Cir. 1988). Although the patent file may often be sufficient to permit the judge to interpret the technical aspects of the patent properly, consultation of extrinsic evidence is particularly appropriate to ensure that his or her understanding of the technical aspects of the patent is not entirely at variance with the understanding of one skilled in the art. See Mantech Envtl. Corp. v. Hudson Envtl. Servs., Inc., 152 F.3d 1368, 1373, 47 USPQ2d 1732, 1737 (Fed. Cir. 1998) ("[A]lthough [expert testimony] always may be admitted by the trial court to educate itself about the patent and the relevant technology, the claims and the written description remain the primary and more authoritative sources of claim construction."); Fromson v. Anitec Printing Plates, Inc., 132 F.3d 1437, 1444, 45 USPQ2d 1269, 1274 (Fed. Cir. 1997), abrogated on other grounds by, Cybor, 138 F.3d at 1456, 46 USPQ2d at 1174 ("Extrinsic evidence may be particularly helpful to the court when a specific technical aspect that is potentially of dispositive weight was not discussed in the specification or explored during the patent prosecution.").

In the instant case, therefore, had the district court relied upon the extrinsic evidence to contradict the claim construction unambiguously apparent from the intrinsic evidence it would have been error. However, we do not doubt the district court's express statements that it did not rely on extrinsic evidence in its claim construction. Having concluded that the claim language "did not unambiguously support either proposed definition", March Non-Infringement Ruling, slip op. at 9, the district court found support for its claim construction in the written description and the examiner's amendment of the patent title, see id. at 11-15. The district court merely referred in footnote two of its opinion to there being "[o]ne common convention in the digital printing field" in order to address the collateral argument made by Pitney Bowes that Hewlett-Packard's proffered construction would exclude the preferred embodiment from being covered by the claims of the '272 patent. The extrinsic evidence of this "common convention", therefore, was not relied upon by the district court to contradict the meaning otherwise apparent from the intrinsic evidence. Rather, the district court relied upon the intrinsic evidence to reach its claim construction, albeit an erroneous construction, and referred only briefly to the extrinsic evidence, which it quite properly examined, in discussing a collateral argument made by Pitney Bowes. This did not constitute error.

The written description and the prosecution history

To ascertain the meaning of claims, we consider three sources: the claims, the written description, and the prosecution history. See Markman, 52 F.3d at 979, 34 USPQ2d at 1329 (quoting Unique Concepts, Inc. v. Brown, 939 F.2d 1558, 1561, 19 USPQ2d 1500, 1558 (Fed. Cir. 1991)). We thus turn next to the two remaining sources for claim construction -- the written description and the prosecution history.

The district court placed significant weight on Pitney Bowes's concession that in forty-two out of forty-four uses of the term "spot" in the written description that usage referred to the spots of light generated by the beam of light and not the spots of discharged area on the photoreceptor. See March Non-Infringement Ruling, slip op. at 12. Based essentially upon this, the district court reasoned that "since inventors 'must use words in the same way in the claims and in the specification,' it is logical to conclude that the word spot also means a light spot when used in the claims of the '272 patent."3 Id. at 12-13 (quoting Fonar Corp. v. Johnson & Johnson, 821 F.2d 627, 632, 3 USPQ2d 1109, 1113 (Fed. Cir. 1987)).

As a general principle, we agree with the district court's "logic". Certainly, "the same word appearing in the same claim should be interpreted consistently." Digital Biometrics, Inc. v. Identix, Inc., 149 F.3d 1335, 1345, 47 USPQ2d 1418, 1425 (Fed. Cir. 1998). Indeed, our conclusion that the district court's claim construction was erroneous because it would assign a different meaning to the term "spot" in the preamble from that in the rest of the claims is broadly based upon this very principle. Nevertheless, we have also recognized that a patent's written description can set forth more than one definition of a claim term. In Genentech, Inc. v. Wellcome Foundation, Ltd., 29 F.3d 1555, 31 USPQ2d 1161 (Fed. Cir. 1994), we noted that with respect to the claim term "human tissue plasminogen activator", "there are at least four possible definitions of the phrase set forth in the specification." Id. at 1563, 31 USPQ2d at 1167. We explained that, to select among these definitions, we would "avoid those definitions upon which the PTO could not reasonably have relied when it issued the patent." Id. at 1564, 31 USPQ2d at 1167.

The case at bar presents a somewhat analogous situation to Genentech. The portion of the written description which Pitney Bowes contends uses the term "spot" to mean a spot of discharged area on the photoreceptor occurs at the very end of the written description and reads, in relevant part:

The scanning system of this invention, however, can be implemented by introducing an intensity modulator 64 for applying an amplitude modulated correction signal for maintaining laser illumination at a constant level. The intensity modulator 64 could also be used for control of spot size by varying the intensity. The use of different spot sizes can effectively be employed as letters or numbers are created so as to avoid roughened edges and improve character formation.

'272 pat., col. 5, l. 64 to col. 6, l. 5 (emphasis added). That the PTO focused on this part of the written description in granting the '272 patent is demonstrated by the irrelevance of the earlier parts of the written description to the invention actually claimed in the '272 patent. Indeed, the earlier parts of the written description relate to the inventions claimed by the '157 and the '757 patents. For instance, the written description describes in detail an apparatus and method for correcting imperfections in the polygonal mirror during the scanning process, as claimed in the '157 patent. The written description further describes in detail apparatus for correcting the scanning speed during the scanning process, as claimed in the '757 patent. As may be recalled, the '272, '157, and '757 patents all stemmed from the same parent application, or a divisional or continuation application thereof. Thus, it is perhaps unsurprising that their written descriptions are apparently identical and, consequently, not relevant in toto to all three inventions.

The earlier parts of this apparently identical written description do, indeed, use the term "spot". However, the usage of the term "spot" is noticeably different in the earlier portions of the written description. These earlier uses of the term refer to a "spot" in the sense of a moving spot, which surely must be the spot of light produced by the beam. Thus, for example, the written description explains that:

The edge detector 62 will indicate when spot 34 has passed a fixed terminal point beyond the scan path. The time differential as detected between the first edge detector 58 and the second detector 62 can be interpreted through logic circuitry to indicate the flight time for spot 34 to cover a fixed length scan path.

'272 pat., col. 5, ll. 52-57. The spot of discharged area on the photoreceptor cannot, of course, have a "flight time" and thus it is apparent that this usage of the term "spot" must refer to a spot of light. The two uses of the term "spot" in the later part of the written description, however, do not refer to "spot" in the context of a moving spot, but rather in the context of the "spot size" to be employed "so as to avoid roughed edges and improve character formation." This usage of the term appears quite distinct from the dynamic "spot" referred to earlier in the written description. Indeed, read in context, this usage of the term "spot size" relates to the spots that are used as the constituent parts of the printed characters, not the "spot" that moves within the apparatus described in the written description. Moreover, the earlier uses of the term referred generally to "the spot 34", in reference to the identical Figure 1 of all three patents, whereas these later uses referred only to "spot size", again indicating a different usage.

In circumstances such as this, where the language of the written description is sufficient to put a reader on notice of the different uses of a term, and where those uses are further apparent from publicly-available documents referenced in the patent file, it is appropriate to depart from the normal rule of construing seemingly identical terms in the same manner. This entirely accords with the public notice function of claims. See Vitronics, 90 F.3d at 1583, 39 USPQ2d at 1577; Hoganas AB v. Dresser Indus., 9 F.3d 948, 951, 28 USPQ2d 1936, 1939 (Fed. Cir. 1993). The prosecution history indicates to a reviewing member of the public that the '272 patent was one of several patents to be issued based upon the same written description disclosure. Parsing the written description, in the context of the prosecution history, puts the reader on notice that the term "spot" has different meanings in the written description depending on its context. Like Genentech, therefore, the term must be read to correspond to the only plausible meaning in each context. In light of the prosecution history, the only plausible meaning of the term "spot size", as used in the disputed part of the written description, is the area of discharge on the photoreceptor. The district court therefore erred when it relied upon the frequency of occurrences of the term "spot", in the context which all parties agreed meant the spot of light from the laser beam, to draw a "logical" conclusion that the two disputed occurrences of the term in the written description and all the occurrences of the term in the claims must also have that meaning.

Under the general heading of "The Prosecution History", the district court considered the effect of the amendment of the title of the '272 patent to include the term "Light Spots". See March Non-Infringement Ruling, slip op. at 13-15. The district court reasoned that:

In the instant case, the prosecution history is consistent with the conclusion that the word "spots" means "light spots" in the '272 patent. According to Patent Office procedure, "[w]here the title is not descriptive of the invention claimed, the examiner should require the substitution of a new title that is clearly indicative of the invention." M.P.E.P. [The Manual of Patent Examining Procedure] 606.01 (1979). After evaluating the application, the examiner was in the best position to fully understand the nature of the invention and the meaning of the terms used in the patent. It is relevant that he chose to modify the noun "spots" with the adjective "light." This is reflective of his understanding that the word "spots" means "light spots" in the '272 patent.

Id. at 14-15.

We conclude that the district court attached too much weight in its claim construction to the patent title and to the amendment of that title. Certainly, the district court is correct that Section 606.01 of the M.P.E.P. provides that the examiner may require a change in the title if the proffered "title is not descriptive of the invention claimed". However, Section 606.01 goes on to explain that "[t]his may result in slightly longer titles, but the loss in brevity of the title will be more than offset by the gain in its informative value in indexing, classifying, searching, etc." Thus, as indicated by the M.P.E.P., the purpose of the title is not to demarcate the precise boundaries of the claimed invention but rather to provide a useful reference tool for future classification purposes. In any event, if we do not read limitations into the claims from the specification that are not found in the claims themselves, then we certainly will not read limitations into the claims from the patent title. See E.I. du Pont de Nemours & Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433, 7 USPQ2d 1129, 1132 (Fed. Cir. 1988). Consequently, an amendment of the patent title during prosecution should not be regarded as having the same or similar effect as an amendment of the claims themselves by the applicant.

The near irrelevancy of the patent title to claim construction is further demonstrated by the dearth of case law in which the patent title has been used as an aid to claim construction. We are only aware of one case from this court in which the patent title was accorded any significance whatsoever in a claim construction. In Exxon Chemical Patents, Inc. v. Lubrizol Corp., 64 F.3d 1553, 35 USPQ2d 1801 (Fed. Cir. 1995), the court reasoned that:

Exxon's claims are drawn to a specific product which has particularly defined ingredients. Nothing in the claims, the specification, or the prosecution history suggests that Exxon's claims are not drawn to a product that contains particular ingredients. Indeed, to the contrary, the title to the '890 patent reads . . . "Lubricating Oil Compositions Containing Ashless Dispersant, [ZDDP], Metal Detergent and a Copper Compound". See Titanium Metals Corp. of Am. v. Banner, 778 F.2d 775, 780, 227 U.S.P.Q. (BNA) 773, 777-78 (Fed. Cir. 1985) (referring to patent's title as interpretative aid).4

Id. at 1557, 35 USPQ2d at 1804. While the court in Exxon Chemical Patents mentioned the patent title in its claim construction, this one-sentence statement was made simply to illustrate the point that the patent-in-suit claimed a product containing certain ingredients. Indeed, the claim construction relied, not upon the patent title, but upon the standard sources of the claims, the specification, and the prosecution history. See id. ("In sum, a review of the claims, the specification, and the prosecution history all point to the conclusion that Exxon claims a product, not merely a recipe for making whatever product results from the use of the recipe ingredients."). Consequently, that the patent title has only been mentioned once by this court in the context of claim construction and, even then, merely to make an illustrative point in one sentence, makes a powerful statement as to the unimportance of a patent's title to claim construction. It was therefore error for the district court to impart as much weight in its claim construction as it did to the amendment of the title of the '272 patent.

The grant of summary judgment of no literal or equivalent infringement based upon the claim construction of the term "spots"

The district court's grant of summary judgment of no literal or equivalent infringement was entirely derivative of its claim construction of the term "spots". Because the accused Hewlett-Packard devices do not have "spots" of light produced by their laser beams that can be "of different sizes", the district court concluded that they could not infringe either literally or equivalently. See March Non-Infringement Ruling, slip op. at 16-20. However, because we hold that the "spots" that can be "of different sizes" are the spots of discharged area on the photoreceptor, not the spots of light produced by the laser beam, we vacate the district court's grant of summary judgment to Hewlett-Packard and remand for further proceedings in accordance with this decision.

Hewlett-Packard cross-appeals from the denial of two summary judgment motions. In the February Non-Infringement Ruling, the district court denied Hewlett-Packard summary judgment of non-infringement based upon the claim limitation "a plurality of beams of light". In the February Invalidity Ruling, the district court denied Hewlett-Packard summary judgment of invalidity for failure to file a proper oath of invention. The statutes that set forth our appellate jurisdiction, however, limit us, with certain narrow exceptions, to hearing appeals from final judgments. See 28 U.S.C. 1292(c), (d), 1295 (1994). This is known as the final judgment rule. "The final judgment rule prohibits a party from appealing a district court's denial of a motion for summary judgment." Lermer Germany GmbH v. Lermer Corp., 94 F.3d 1575, 1576, 39 USPQ2d 2014, 2015 (Fed. Cir. 1996); see also Pacific Union Conference of Seventh-Day Adventists v. Marshall, 434 U.S. 1305, 1306 (1977); Gerber Garment Tech., Inc. v. Lectra Sys., Inc., 916 F.2d 683, 686, 16 USPQ2d 1436, 1438-39 (Fed. Cir. 1990). Because we are vacating the district court's grant of summary judgment of non-infringement to Hewlett-Packard and remanding the case for further proceedings, the two denials of Hewlett-Packard's summary judgment motions are not, at present, final, appealable orders, for they are no longer part of a final judgment. Consequently, we must dismiss Hewlett-Packard's cross-appeal without addressing its merits.5

The district court erroneously construed the claim term "spots". Because the district court's grant of summary judgment of no literal or equivalent infringement was derived from this erroneous claim construction, we vacate its judgment and remand for further proceedings in accordance with this decision. Given our order of vacatur and remand, the cross-appeal is dismissed. The summary judgment is thus

VACATED and REMANDED.

Defendant-Cross Appellant to bear costs.

RADER, Circuit Judge, with whom PLAGER, Circuit Judge, joins, additional views.

Appellate courts can err by giving too much guidance. In assessing the reliability of evidence, for instance, rarely can distant appellate hindsight improve upon the immediate and informed judgment of the trial judge. Moreover, no strict, uniform rules can anticipate every variable in assessing complex technical evidence. "Too much depends upon the particular circumstances of the particular case at issue." Kumho Tire Co. v. Carmichael, 119 S. Ct. 1167, 1175, 50 USPQ2d 1177, 1184 (1999). Addressing the analogous issue of admissibility of expert testimony, the Supreme Court recently cautioned against rigid tests for the reliability of expert testimony: "[W]e conclude that the trial judge must have considerable leeway in deciding how to go about determining whether particular expert testimony is reliable." Id. at 1176.

In Vitronics Corp. v. Conceptronic, Inc., this court gave trial courts detailed guidance to avoid expert testimony in claim construction, ultimately condemning reliance on such testimony as "rarely, if ever," proper. 90 F.3d 1576, 1585, 39 USPQ2d 1573, 1579 (Fed. Cir. 1996). This appellate perspective discounted the relevance and helpfulness of testimony from experts skilled in the art to determine the meaning of claims. Compare SmithKline Diagnostics, Inc. v. Helena Lab. Corp., 859 F.2d 878, 882, 8 USPQ2d 1468, 1471 (Fed. Cir. 1988) ("To ascertain the meaning of the claims, we look to the claim language, the specification, and the prosecution history. Also relevant are the other claims and expert testimony." (citations omitted)); James B. Altman et al., The Law of Patent Claim Interpretation: The Revolution Isn't Finished, 8 Fed. Circuit B. J. 93, 102-03 (1998) ("Vitronics -- at least as a number of courts appear to have applied it -- is inconsistent with the established standard for construing patent claims. The courts traditionally have attempted to interpret them consistently with the understanding of persons skilled in the relevant field of technology. Yet Vitronics has been read to encourage judges to ignore that understanding so long as the court thinks that the patent documents are clear.").

Vitronics offers good counsel when it urges trial judges to focus on the patent document -- notably the claims themselves -- to ascertain the scope of patent coverage. This appellate court, however, should refrain from dictating a claim interpretation process that excludes reliable expert testimony. The process of claim construction at the trial court level will often benefit from expert testimony which may (1) supply a proper technological context to understand the claims (words often have meaning only in context), (2) explain the meaning of claim terms as understood by one of skill in the art (the ultimate standard for claim meaning, see Markman v. Westview Instruments Inc., 52 F.3d 967, 986, 34 USPQ2d 1321, 1335 (Fed. Cir. 1995) (en banc), aff'd, 517 U.S. 370, 38 USPQ2d 1461 (1996)), and (3) help the trial court understand the patent process itself (complex prosecution histories -- not to mention specifications -- are not familiar to most trial courts). Because trial judges are trained as "gatekeepers" for expert testimony, Kumho, 119 S. Ct. at 1174, they are likely to admit such evidence only when reliable and keep it in its proper role.

Today this court takes the opportunity to restate the role of expert testimony. I applaud this court's effort to express more trust in the "broad latitude" and "considerable leeway" afforded presiding trial judges in assessing the reliability of expert testimony. Id. at 1171, 1176. In this case, however, this court holds that the trial court did not err by improper reliance on expert testimony. Rather, the trial court erred in placing too much reliance upon the written description when the claim language admitted of a broader reading. The resulting construction of the claims thus contained limitations improperly imported from the written description.