Opinion for the court filed by Circuit Judge NEWMAN. Separate opinion, concurring in part and dissenting in part, filed by Circuit Judge CLEVENGER.
PAULINE NEWMAN, Circuit Judge.ATD Corporation appeals the final judgment 1 of the United States District Court for the Eastern District of Michigan, holding the claims in suit of United States Patents No. 5,011,743 (the ’743 patent) and No. 5,111,577 (the ’577 patent), both entitled “Pad Including Heat Sink and Thermal Insulation Areas,” invalid and not infringed. We affirm the rulings of non-infringement and reverse the rulings of invalidity. On Lydall’s cross-appeal we affirm that inequitable conduct was not established. The challenged eviden-tiary rulings are sustained.
BACKGROUND
The ’743 patent relates to a flexible insulating pad that includes heat sink and thermal insulation areas. It is described as particularly useful for providing “hot spot” insulation, and is used primarily in automotive underbodies. It achieved prompt commercial acceptance, as an economical and efficient structure for dissipating heat. Two of the patent drawings are reproduced below. Fig. 1 a top view and Fig. 2 a cross section of the patented pad:
*538[[Image here]]
The thermal insulation area 5 is made of layers of metal foil 2 separated by air gaps, and the heat sink area 4 is formed by compressing the edges of the foil layers. The depressions at 6 are called “embossments,” and the dashed lines at 11 represent optional layers of heat-resistant scrim.
Claims 1 and 3 of the ’743 patent are at issue, shown with emphases added to highlight the disputed subject matter with respect to infringement; all of the other claim elements and limitations are conceded to be present in the accused pads:
1. A pad including thermal insulation and heat sink areas comprising:
a plurality of layers of metal foil forming a stack wherein said layers are arranged one above another in a vertical direction,
said stack including at least one heat sink area and at least one thermal insulating area adjacent to said heat sink area,
said layers providing better heat condition in said vertical direction at said heat sink area than at said insulating area,
at least two of said layers including a plurality of embossments therein separating said layers in said insulating area so as to provide gaps therebetween,
one of said layers in said insulating area being adjacent to and not metallurgically bonded to another one of said layers, said heat sink area comprising a compressed portion of said stack.
3. The pad of claim 1, wherein said heat sink area at least partly surrounds said insulating area and said layers in said heat sink area are interengaged with each other by securing means.
*539The ’577 patent is a division of the ’743 patent, and relates to the manufacture of the pad. Claims 1, 11, and 19 of the ’577 patent are as follows, with emphases added to show the usages of “embossments,” the only point of dispute.
I. A method of making a heat insulating pad having insulating and heat sink areas, comprising:
a step of assembling a plurality of layers of metal foil in a stack wherein said layers are arranged one above another in a vertical direction, at least two of said layers being separated from each other by a plurality of embossments on at least one of said layers;
a step of compressing at least one portion of said stack such that heat sink and insulating areas are formed therein with said layers providing better heat conduction in said vertical direction at said heat sink area than at said insulating area,
said embossments in said insulating area separating said layers so as to provide a gap therebetween;
and a step of securing said layers together in said heat sink area, said securing step including interengaging said layers in said heat sink area with each other.
II. A method of making a heat insulating pad having insulating and heat sink areas, comprising:
a step of assembling a plurality of layers of metal foil in a stack wherein said layers are arranged one above another in a vertical direction, at least two of said layers being separated from each other by a plurality of embossments on at least one of said layers; and
a step of compressing at least one portion of said stack such that heat sink and insulating areas are formed therein with said layers providing better heat conduction in said vertical direction at said heat sink area than at said insulating area,
said embossments in said insulating area separating said layers so as to provide a gap therebetween.
19. The method of claim 11, wherein said assembling step comprises assembling a plurality of layers of metal foil which make said pad flexible.
The accused Lydall pads contain heat sink and insulating areas, in accordance with the claims, and are the same as the ATD pads except that the Lydall foil layers are separated by knitted or woven mesh instead of by depressions in the foil. ATD argues that due to compression forces applied during manufacture, the Lydall foil layers are “embossed” with the impressions of the intervening mesh. ATD states that these “emboss-ments,” along with the mesh reach to the adjacent layers of foil, and thus that the product and process claims are infringed, literally or in accordance with the doctrine of equivalents.
Upon ATD’s suit for patent infringement, Lydall raised the defenses of non-infringement, invalidity, and unenforceability. On cross-motions for summary judgment, the district court granted Lydall’s motion that there was not literal infringement and no willful infringement. The district court also granted ATD’s motion that there was not inequitable conduct in the prosecution of the patents. The court ruled that there were genuine issues of material fact on the issue of infringement under the doctrine of equivalents and the issue of validity; these issues were tried to a jury. The jury found that Lydall did not infringe, under the doctrine of equivalents, claim 3 of the ’743 patent, nor claims 1, 5-7, 11, or 19 of the ’577 patent. The jury deadlocked on the issue of infringement by equivalents of claim 1 of the ’743 patent. The jury also found that claims 1 and 3 of the ’743 patent and claims 1,11, and 19 of the ’577 were invalid based on prior art. The district court entered judgment accordingly, and denied all post-trial motions. Each side appeals the rulings adverse to it.
I
LITERAL INFRINGEMENT
Determination of the issue of literal infringement involves the steps of first construing the claims, a matter of law assigned to the judge whether or not a jury trial has been demanded, and then applying the construed claims to the accused device, a factual *540determination performed by the jury. See Markman v. Westview Instruments, Inc., 52 F.3d 967, 976, 34 U.S.P.Q.2d 1321, 1326 (Fed.Cir.1995) (in banc), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996). In this case, as often occurs, the question of literal infringement was resolved upon the court’s construction of the claims. See id. at 999, 34 U.S.P.Q.2d at 1346 (Newman, J., dissenting) (“Deciding the meaning of the words used in the patent is often dispositive of the question of infringement.”) Thus a court may grant summary judgment when, upon construction of the claims and with all reasonable factual inferences drawn in favor of the non-movant, it is apparent that only one conclusion as to infringement could be reached by a reasonable jury. See Fed.R.Civ.P. 56(c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986) (purpose of summary judgment is to avoid an unnecessary trial); Multiform Desiccants Inc. v. Medzam Ltd., 133 F.3d 1473, 1476, 45 U.S.P.Q.2d 1429, 1431 (Fed.Cir.1998) (affirming ruling that given correct claim construction no reasonable jury could find literal infringement). Claim construction is determined de novo on appeal, Cybor Corp. v. FAS Technologies, Inc., 138 F.3d 1448, 1456, 46 U.S.P.Q.2d 1169, 1174 (Fed.Cir.1998) (in banc); Markman, 52 F.3d at 979, 34 USPQ2d at 1329, as is the correctness of the grant of summary judgment, Cole v. Kimberly-Clark Corp., 102 F.3d 524, 532, 41 U.S.P.Q.2d 1001, 1007 (Fed.Cir.1996).
A. Claim Construction
The claims of both the ’577 and ’543 patents recite layers of metal foil having “a plurality of embossments” that separate the layers and establish gaps of insulating air between the layers. Lydall’s position is that its foil layers are not “embossed”; ATD argues that they are, and that the district court erred in its construction of this term, leading to error in the ensuing decisions of non-infringement.
The district court drew on the two patent specifications for the meaning of “em-bossments” as used in the claims. See Slimfold Mfg. Co. v. Kinkead Industries, Inc., 810 F.2d 1113, 1116, 1 U.S.P.Q.2d 1563, 1566 (Fed.Cir.1987) (claims are understood in light of the specification of which they are a part). When “the specification explains and defines a term used in the claims, without ambiguity or incompleteness, there is no need to search further for the meaning of the term.” Multiform, 133 F.3d at 1478, 45 U.S.P.Q.2d at 1433. However, when such definition is challenged it is often appropriate, despite facial clarity and sufficiency of the specification and the prosecution history, to receive evidence of the meaning and usage of terms of art from persons experienced in the field of the invention. See Fed.R.Evid. 702-706.
The specifications of both patents in suit define the embossments as “depressions” and “bumps or projections,” the numbers referring to Fig. 2 supra:
The embossments 6 form depressions on one side of a respective one of the layers 2 and bumps or projections on an opposite side of the respective layer.
’743 patent, col. 7, lines 37-39; ’577 patent, col. 7, lines 41-44. The district court observed that this definition is consistent with the dictionary definition of “emboss” as meaning to “raise in relief from a surface.” ATD argues that this definition supports its position, since Lydall’s foil layers have “bumps and indentations.” ATD argues that there is no reason to believe that the inventor used the term differently from its dictionary meaning, leaving no ambiguity and thus no need for “claim construction.” ATD states that the district court erred in law in construing “embossments” as also requiring that the bumps and indentations reach from layer to layer of the foil. Thus ATD argues that since the claims are not ambiguous, and since “embossments” has a clear meaning, it is incorrect to add to the definition of “em-bossments” the requirement that they also serve to separate the layers of metal foil by the depth of the embossments that contact the adjacent layer of foil.
The specifications teach that the emboss-ments make point contact between the adjacent layers of foil:
The pad 1 can include two layers 2 only one of which includes the embossments 6. In a preferred embodiment, however, at least two of the layers adjacent to each *541other include a pattern of the emboss-ments 6, the layers being offset with respect to each other such that at least some of the embossments are not aligned in the vertical direction. With this arrangement, the layers 2 can be provided in point contact to minimize heat transfer therebet-ween in the vertical direction A.
’743 patent, col. 7, lines 9-17; ’577 patent, col. 7, lines 13-21. The district court also reviewed the patent drawings, and correctly described them as showing “the emboss-ments as being raised reliefs on the various layers such that the embossments come in direct point contact with adjacent layers (Figures 2 and 5 of the ’743 and ’577 patents.)” The court concluded that “emboss-ments” as used in the claims require raised reliefs that separate adjacent layers of foil by point contact.
ATD states that the district court erroneously incorporated into the definition of “embossments” in the claims in suit the structural feature of “point contact” from non-asserted claim 11 of the ’743 patent, which reads:
11. The pad of claim 1, wherein said pad is flexible and at least some of said em-bossments form depressions on one side of a respective one of said layers and bumps on an opposite side of said respective layer, said embossments providing point contact between the layers.
(Emphasis added.) ATD argues that point contact is described in the specification as simply a preferred embodiment, and is not a limitation to the claims that do not include it. ATD states that a definition of embossments without point contact is also supported by the description in the ’743 and ’577 specifications of an embodiment wherein layers of a heat-resistant mesh or scrim are interposed between foil layers. ATD states that since claim 1 is generic to the various embodiments shown in the specification, it was an error of law to restrict the broadest claim to the embodiment illustrated in Fig. 2 wherein the embossments are in point contact with adjacent foil layers.
Lydall responds that the mere fact that non-asserted claim 11 specifically states that the embossments are in point contact does not preclude the construction of claim 1 as requiring that the embossments separate the layers. Lydall points out that for the embodiment in which ATD uses mesh between some foil layers, the specification shows that at least some of the layers are separated by point contact of the embossments. Non-asserted dependent claim 14 of the ’743 patent, which specifies one or more layers of scrim, also incorporates the claim 1 limitation of at least two “layers including a plurality of em-bossments therein separating said layers in said insulating area so as to provide gaps therebetween.”
The separation of layers and ensuing provision of insulating gaps arise from the point contact. Thus we agree that the correct interpretation of the claims in suit is that, whether or not mesh is used between some of the foil layers, the embossments serve to separate at least some of the layers. The doctrine of claim differentiation can not broaden claims beyond the scope that is supported by the specification. Multiform Desiccants, 133 F.3d at 1480, 45 U.S.P.Q.2d at 1434 (“claim differentiation can not broaden claims beyond their correct scope”); Tandon Corp. v. United States Int’l Trade Comm’n, 831 F.2d 1017, 1023, 4 U.S.P.Q.2d 1283, 1288 (Fed.Cir.1987) (“Whether or not claims differ from each other, one can not interpret a claim to be broader than what is contained in the specification and claims as filed.”) The presumption that separate claims have different scope “is a guide, not a rigid rule.” Autogiro Co. of America v. United States, 181 Ct.Cl. 55, 384 F.2d 391, 404, 155 U.S.P.Q. 697, 708 (1967).
ATD also argues that the district court improperly added a method limitation to the product claims when it interpreted the method claims as requiring the embossments to be made on the metal sheets before the assembly process. The district court held that “[t]he specifications make clear that the ‘embossments’ as contemplated under the patents are made on the metal sheets prior to any assembly process” (district court’s emphasis), and that “the ’577 patent is very clear on pre-assembly embossment.” The *542court cited the specification of the ’577 patent:
The method according to the invention can also include a step of embossing a plurality of the layers 2 such that a plurality of the embossments 6 are formed therein, the embossing step being performed by simultaneously embossing a plurality of overlapping sheets 2 of the metal foil, each of the sheets after the embossing step comprising a respective one of the layers. The embossments can be provided in a random or uniform repeated pattern. It is also within the scope of the invention to emboss each sheet separately. The embossments can be provided by passing a single sheet or stack of sheets between a pair of rollers having the desired embossment pattern thereon.
’577 patent, col. 8, lines 32^44 (district court’s emphasis).
This aspect is relevant to both the product and the method claims, for it was not disputed that in the Lydall pad a pattern of impressions is produced on the metal foil after the foil is assembled into layers, by compression of the foil against the intervening mesh. The specifications make clear that the “emboss-ments” of the ’743 and ’577 patents are “something more than ‘impressions’ on the surface of the foil layers which are merely ‘incidental,’” in the district court’s words. The claims require that the “embossments” be deep enough to separate the adjacent layers of foil; the product claims are not dependent on how the embossments are made, and were not construed otherwise.
Upon plenary review we confirm the district court’s claim construction. The ’743 and ’577 specifications show and the claims require that the insulating gap between at least some of the metal foil layers is formed by the embossments that space apart the foil layers. In the patent specifications the presence of embossments making contact with adjacent layers of foil to separate the layers is described as a material aspect of the invention. The use of a mesh or scrim between some of the layers, even without “emboss-ments” in those layers, does not defeat the requirement of even the broadest claims that the pad contain at least two layers of foil bearing embossments that separate the layers. Claim 1 of the ’743 patent, correctly interpreted, embraces generically the use of mesh between some of the foil layers, but also requires that embossments on some of the foil layers make contact with and separate adjacent foil layers. All of the claims of both patents require embossments that serve this function, and the specifications make clear that this is essential to the patented invention. We conclude, as did the district court, that “embossments” as used in the ’743 and ’577 patents means depressions or bumps that separate and form a gap between at least some of the foil layers by point contact of the embossments with adjacent layers. This definition applies to all the claims in suit.
B. Application to Accused Devices
We give plenary review to the district court’s grant of Lydall’s motion for summary judgment of non-infringement, reapplying the standard for summary adjudication as applied by the district court.
Lydall’s original product (discontinued after June 1, 1995) included an outer layer of metal foil with a “decorative pattern” in relief. The ensuing product did not include this outer pattern in relief. The court found that these indentations “projected outward, only — i.e., they did not provide point contact with adjacent interior solid metal layers, nor did they provide gaps between the layers.” The court held that although these outer layers were “embossed” in the ordinary usage of the word, they did not contain “em-bossments” within the meaning of that term in the patents. We agree that no reasonable jury could have found otherwise.
In the Lydall product the metal foil layers are each separated by an intervening layer of metal mesh. When the mesh is pressed between the foil layers during the assembly process, there is created a patterned impression of the mesh on the smooth metal foil sheets. Applying the claims as construed, the district court found that these impressions do not constitute “embossments” as that term is used in the ATD claims, and that the Lydall impressions are “merely inciden*543tal to the contact of the mesh with the surface of the metal layers.” The district court found that in the Lydall product it is the mesh that separates the layers and creates the insulating gap, and not the embossed depressions that are formed in the metal foil where it is pressed against the mesh. The court found that there was no point contact between the Lydall layers, as required by the court’s claim construction, for “[t]he metal mesh in fact prevents all contact between the adjacent metal sheet layers.” The court concluded that because there were no emboss-ments providing point contact, summary judgment of no literal infringement was appropriate because an essential limitation of the claims was absent from the accused structure and method.
ATD argues that the Lydall embossments literally meet the claims’ requirement of em-bossments, not only in the plain meaning of “emboss,” but also because the Lydall em-bossments contribute along with the mesh to the gaps between the foil layers. ATD stresses that point contact is not a limitation to the claims in suit, and that the presence of the wire mesh between the foil layers of the Lydall product does not avoid literal infringement because there are embossments on the Lydall foil layers. ATD also argues that the issue of whether the Lydall embossments contribute to the gap between the layers of foil presents a question of material fact that could not be summarily resolved against ATD, and requests trial of this question.
It is not disputed that in the Lydall product and method there is no contact between the impressed relief pattern on the foil layers, and any adjacent layer. There are no “embossments,” as we and the district court construe the term as used in the patents, to separate the foil layers. This separation function is an express limitation of the asserted claims. In view of its absence from the accused product and method, no reasonable jury could have found that the asserted claims are literally infringed by the Lydall product or method. See Cole, 102 F.3d at 532, 41 U.S.P.Q.2d at 1007 (affirming summary ruling of non-infringement when “accused products do not literally meet all of the claim limitations”). Therefore, summary judgment of no literal infringement was properly granted.
II
INFRINGEMENT BY EQUIVALENTS
The question of infringement in terms of the doctrine of equivalents was given to the jury. The district court construed the term “embossments,” in Jury Instruction No. 25, as follows:
25.... You are instructed that emboss-ments are purposefully created patterns directly formed on individual foil layers that produced depressions on one side of a respective layer and bumps or projections on the opposite side of the respective layer. The embossments create projections providing direct point contact with adjacent layers. The embossments must be large enough to separate the layers and provide gaps between the layers.
ATD argues that the district court incorrectly construed the claims, thereby distorting the jury’s finding of non-infringement and providing an incorrect basis for the jury verdict. See United States Surgical Corp. v. Ethicon, Inc., 103 F.3d 1554, 1568, 41 U.S.P.Q.2d 1225, 1236 (Fed.Cir.1997) (reviewing whether jury verdict was in accordance with correct claim construction). As we have discussed, the district court’s construction of the term “embossments” correctly included the requirement of point contact.
On the issue of equivalency the jury was instructed as follows:
27. ... Under the doctrine of equivalents, you must find that there are insubstantial differences between the patent claims and the alleged infringing product or method of making the product. In this regard, you may consider whether the defendant’s product or method performs (1) substantially the same function (2) in substantially the same way (3) to produce substantially the same result when compared to the plaintiffs patented product or method, even though they may differ in name, shape or form.
As accused product or method does not infringe under the doctrine of equivalents if it performs the function and achieves the *544result in a substantially different way than the claimed invention.
The doctrine of equivalents does not involve the application of a formula and is not an absolute to be considered in a vacuum. Rather, the question of whether one component of the allegedly infringing product or method is equivalent to an element in the patented claim is a factual matter. It requires you to consider the context of the entire claim. Your answer will depend upon the drawings and written description, the patent application history, the prior art and all the circumstances of this case.
* * * *
Other factors to be considered in determining infringement is the known interchangeability of the accused and claimed elements or other objective technological evidence demonstrating that the substitute nevertheless represents a change that one of ordinary skill in the art would have considered “insubstantial” at the time of infringement. Evidence of copying and evidence of “designing around” are also relevant to the question of infringement under the doctrine of equivalents.
The issue of infringement under the doctrine of equivalents was fully litigated. One of Lydall’s expert witnesses testified that the Lydall mesh is substantially different from the ATD embossments. He testified that although the Lydall mesh and the ATD em-bossments perform the function of separating the foil layers, they do so in substantially different ways, leading to substantially different results. According to Lydall’s expert, the use of the Lydall mesh preserves sheet reflectivity and results in greater crush resistance, providing advantages as well as differences. ATD disputed this evidence and its significance, and stressed that Lydall used the combination of a heat sink and insulator, the most important part of the ATD invention. Reviewing the record, we conclude that there is substantial evidence to support the jury verdict that there is not infringement under the doctrine of equivalents.2 That judgment is affirmed.
Ill
PATENT VALIDITY
The jury verdict was that claims 1 and 3 of the ’743 patent and claims 1,11, and 19 of the ’577 patent were invalid based on prior art. The district court entered judgment accordingly.3 The evidence is reviewed to ascertain whether the jury’s express or implicit factual findings were supported by substantial evidence, and whether the legal conclusion represented by the verdict was adequately based on supported findings, accompanied by correct application of the law to the facts. See Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 619, 225 U.S.P.Q. 634, 636 (Fed.Cir.1985) (“In reviewing a decision denying a motion for judgment notwithstanding the verdict, we do not approach the issues as if there had been no trial. We review the evidence as a whole, and ascertain whether the verdict is in accordance with law, and whether there was substantial evidence in support of the jury’s verdict.”) In so doing, we resolve any disputed facts and draw all reasonable factual inferences in favor of the jury verdict. Richardson v. Suzuki Motor Co., 868 F.2d 1226, 1235, 9 U.S.P.Q.2d 1913, 1919 (Fed.Cir.1989); DMI, Inc. v. Deere & Co., 802 F.2d 421, 425, 231 U.S.P.Q. 276, 278-79 (Fed.Cir.1986).
Although the verdict form did not state the ground on which the jury relied, Lydall had raised grounds of both anticipation and obviousness, and the parties have argued both of these grounds on appeal.
*545Anticipation
A patent is invalid for anticipation when the same device or method, having all of the elements and limitations contained in the claims, is described in a single prior art reference. Richardson, 868 F.2d at 1236, 9 U.S.P.Q.2d at 1920; Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 894, 221 U.S.P.Q. 669, 673 (Fed.Cir.1984). An anticipating reference must describe the patented subject matter with sufficient clarity and detail to establish that the subject matter existed and that its existence was recognized by persons of ordinary skill in the field of the invention. See In re Spada, 911 F.2d 705, 708, 15 U.S.P.Q.2d 1655, 1657 (Fed.Cir.1990); Diversitech Corp. v. Century Steps, Inc., 850 F.2d 675, 7 U.S.P.Q.2d 1315, 1317 (Fed.Cir.1988). Lydall relied on seven references, as follows:
The Collier U.K. Patent No. 126,780, entitled Improvements in or connected with the Construction of Boxes or Cases or like objects and of the Floors, Roofs, Walls, Partitions or other Parts of Railway or other Vehicles or like objects, was cited in the prosecution of the ’743 patent. Collier, discussing insulation of railway cars, shows the use of embossed projections to separate metal layers for insulation, but does not describe foil layers and heat sink areas, as required by each of the claims.
The Meekenstock German utility model shows a heat shield for gas-exhausting parts of motor vehicles. Meekenstock describes the use of two sheet metal plates, at least one of them including projections to separate the plates, but shows no foil layers as required by each of the claims.
The ASTM (American Society for Testing Materials) Pub. C 667-80, entitled Standard Recommended Practice for Prefabricated Reflective Insulation Systems for Equipment and Pipe Operating at Temperatures Above Ambient Air, describes a rigid inner and outer case protecting embossed liners, which may be foil. The reference does not show compressed foil to form heat sink areas, as required by each of the claims, and cautions against such deformation of the liners.
The ASTM Pub. C 740-82 entitled Standard Practice for Evacuated Reflective Insulation in Cryogenic Service, describes an insulator made from multiple layers of metal foil, the foil layers having embossments that separate the layers and form air gaps between the layers. This reference does not show compressing the layers to form heat sinks, and cautions against compression of the layers. Neither of the two ASTM references shows compressed layers of foil creating heat sink areas accompanying the insulating areas.
The article by F.E. Swalley et al., entitled Practical Problems in Design of High-Performance Multilayer Insulation System for Cryogenic Stages, Advances in Cryogenio Engineering, Vol. 10 at 208 (1965), describes a multilayer insulation system for cryogenic stages, again showing projections to space the layers apart, but does not show compressed layers of foil creating heat sink areas.
The article by L.D. Stimpson et al., entitled Predicting Spacecraft Multilayer Insulation Performance from Heat Transfer Measurements, Heat Transmission Measurements in Thermal Insulations, ASTM STP 544 (1974), is directed to multilayer insulation blankets in spacecraft. It describes multilayered insulation as “a series of radiation shields with low-conductivity spacers.” Lydall does not discuss the substance of the Stimpson article in its brief, and we agree with ATD that Stimpson does not disclose an insulating area together with a heat sink area.
Logan et al. U.S Patent No. 4,489,852 describes a cooking utensil, such as a cookie sheet, “of a double walled construction providing an insulating layer or volume of air therebetween,” but does not show a stack of foil layers as called for by the claims. The parties stipulated that Logan et al. “do not disclose every element of the claims of the ATD patents arranged as in the claims.”
None of these references meets the criteria of anticipation, for none show the combination of embossed insulating layers and compressed heat sink areas. There was not substantial evidence to support a verdict of invalidity based on anticipation.
*546Obviousness
Obviousness is a legal conclusion based on underlying facts of four general types, all of which must be considered by the trier of fact: (1) the scope and content of the prior art; (2) the level of ordinary skill in the art; (3) the differences between the claimed invention and the prior art; and (4) any objective indicia of nonobviousness. See Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Continental Can Co. USA, Inc. v. Monsanto Co., 948 F.2d 1264, 1270, 20 USPQ2d 1746, 1750 (Fed.Cir.1991); Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561, 1566-68, 1 U.S.P.Q.2d 1593, 1595-97 (Fed.Cir.1987).
Determination of obviousness can not be based on the hindsight combination of components selectively culled from the prior art to fit the parameters of the patented invention. There must be a teaching or suggestion within the prior art, or within the general knowledge of a person of ordinary skill in the field of the invention, to look to particular sources of information, to select particular elements, and to combine them in the way they were combined by the inventor. See Heidelberger Druckmasckinen AG v. Hantscho Commercial Prods., Inc., 21 F.3d 1068, 1072, 30 U.S.P.Q.2d 1377, 1379 (Fed.Cir.1994) (“When the patented invention is made by combining known components to achieve a new system, the prior art must provide a suggestion or motivation to make such a combination.”); Northern Telecom, Inc. v. Datapoint Corp., 908 F.2d 931, 935, 15 U.S.P.Q.2d 1321, 1324 (Fed.Cir.1990) (the prior art must suggest to one of ordinary skill in the art the desirability of the claimed composition); Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1143, 227 U.S.P.Q. 543, 551 (Fed.Cir.1985).
ATD argues that there was not substantial evidence to support a finding that the prior art contained a teaching or suggestion to combine selected portions of the prior art in order to create the patented structure or method. ATD argues that the jury must have improperly reasoned with the hindsight of ATD’s successful accomplishment. Lydall relied on the same group of references as for anticipation. However, Lydall points to no evidence supporting the obviousness determination, other than the conclusory opinion of its expert witness.
We observe that embossments, similar in general shape to those of the ATD patents, have been used to space insulating layers of various forms. However, some of the cited references cautioned against compressing the layers in a multilayer insulator, and none showed compressing the insulating layers to form a heat sink as in the patented device and method. Lydall does not direct us to any evidence of a teaching or suggestion to select the components that ATD’s inventors selected, from the crowded field of insulation technology, to produce the product and method of the ’743 and ’577 patents. Lydall’s witnesses themselves expressed the view that such compression would be undesirable, providing cogent evidence that one of ordinary skill would not have deemed it obvious to compress the layers of an insulating device for heat sink purposes. Absent substantial evidence of such teaching or suggestion in the prior art or in the general knowledge of persons of ordinary skill in the field, there was not substantial evidence to support the jury’s verdict of obviousness. It was undisputed that the product met an unsolved need and was quickly adopted by the automotive industry, this commercial success also weighing against obviousness.
Because there was not substantial evidence supporting the verdict on the ground of either anticipation or obviousness, the judgment of invalidity is reversed.
IV
INEQUITABLE CONDUCT
Determination that a patent applicant engaged in inequitable conduct before the Patent and Trademark Office requires, as threshold findings of fact, both that the applicant failed to disclose material information to the PTO, and that he intended thereby to mislead or deceive the patent examiner into granting the patent. Materiality of the non-disclosed information, and culpable intent, must be established by clear and convincing evidence. When these facts are established, *547the court will weigh the findings and their premises and decide, in the court’s exercise of discretion, whether to hold the patent unenforceable. Kingsdown Medical Consultants, Ltd. v. Hollister Inc., 863 F.2d 867, 872, 9 U.S.P.Q.2d 1384, 1389 (Fed.Cir.1988) (in banc). We review the district court’s ruling on the ultimate issue of inequitable conduct on the standard of abuse of discretion, Kingsdown, 863 F.2d at 876, 9 U.S.P.Q.2d at 1392, while review of the underlying facts is on the clearly erroneous standard, with due consideration of the burden to establish both materiality and intent to deceive by clear and convincing evidence.
The district court ruled on summary judgment that the ’743 and ’577 patents were not unenforceable for inequitable conduct. Lydall appeals this ruling, stating that there were disputed facts as to whether certain information was material to patentability, and therefore that the issue was not amenable to summary disposition. Although the premises of inequitable conduct require findings based on all the evidence, a procedure that may preclude summary determination, see KangaROOS U.S.A., Inc. v. Caldor, Inc., 778 F.2d 1571, 1577, 228 U.S.P.Q. 32, 35 (Fed.Cir.1985), a motion for summary judgment may be granted when, drawing all reasonable factual inferences in favor of the non-movant, the evidence is such that the non-movant can not prevail.
Lydall argued that during prosecution of the ’577 divisional application ATD withheld U.S. Patent No. 2,212,481 to Sendzimir, as well as a PCT search report and prosecution records applying Sendzimir to the corresponding PCT application. Lydall argues that this inequitable conduct in prosecuting the ’577 patent “infects” the ’743 parent patent, and that both patents should be held unenforceable.
The Sendzimir patent was cited in the prosecution of the ’743 patent. Thus the district court found that ATD’s failure to provide the Sendzimir information during the prosecution of the ’577 divisional patent was not clear and convincing evidence of intent to deceive, observing that in accordance with M.P.E.P. § 609 (Rev.14, Nov. 1992)4 a reference is not required to be resubmitted in prosecuting a divisional application:
§ 609 .... the patent examiner will consider information cited or submitted to the Office in a parent application when examining a continuing application, and a list of the information need not be submitted in the continuing application unless applicant desires the information to be printed on the patent.
In view of § 609 it can not be inequitable conduct for an applicant not to resubmit, in the divisional application, the information that was cited or submitted in the parent application. See Transmatic, Inc. v. Gulton Industries, Inc., 849 F.Supp. 526, 31 U.S.P.Q.2d 1225 (E.D.Mich.1994), aff'd in pertinent part, rev’d in part, 53 F.3d 1270, 35 U.S.P.Q.2d 1035 (Fed.Cir.1995) (a material reference that is already of record in parent application need not be resubmitted by the applicant in a continuing application). Nor was ATD required to submit the documents relating to Sendzimir in the record of the PCT application, when Sendzimir was already of record in the United States parent application. Although international search reports may contain information material to patentability if they contain closer prior art than that which was before the United States examiner, it is the reference itself, not the information generated in prosecuting foreign counterparts, that is material to prosecution in the United States. The details of foreign prosecution are not an additional category of material information. See Molins PLC v. Textron, Inc., 48 F.3d 1172, 1180, 33 U.S.P.Q.2d 1823, 1828 (Fed.Cir.1995) (duty to cite material references arising in related foreign applications).
We discern no clear error in the district court’s ruling that because the Sendzimir reference was of record in the parent ’743 application, and because M.P.E.P. § 609 states that the information need not be resubmitted, there was not clear and convincing evidence of material withholding with intent to deceive. The summary judgment of no inequitable conduct is affirmed.
*548V
EVIDENTIARY RULINGS
Evidentiary rulings, when appeal-able, are reviewed under the criteria of the regional circuit, unless the evidentiary issue is unique to patent law or litigation and thus would benefit from national uniformity. See National Presto v. West Bend Co., 76 F.3d 1185, 1188 n. 2, 37 U.S.P.Q.2d 1685, 1686 n. 2 (Fed.Cir.1996) (“On procedural matters not unique to the areas that are exclusively assigned to the Federal Circuit, the law of the regional circuit shall be applied.”)
The abuse of discretion standard is applied to review of evidentiary rulings in the Sixth Circuit, as it is generally in the federal system. E.g., Schrand v. Federal Pacific Elec. Co., 851 F.2d 152, 156-57 (6th Cir.1988) (“This court applies an abuse of discretion standard in reviewing decisions of a trial court on the admission of evidence.”) In addition, Fed.R.Civ.P. 61 provides, “No error in either the admission or the exclusion of evidence ... is ground for granting a new trial or for setting aside a verdict ... unless refusal to take such action appears to the court inconsistent with substantial justice.” See Schrand, 851 F.2d at 156-57. Thus evi-dentiary decisions are reviewed for abuse of discretion, and may be reversed only when that abuse has led to harmful error or the denial of substantial justice. Cooley v. Carmike Cinemas, Inc., 25 F.3d 1325 (6th Cir.1994).
A
ATD asserts that the district court abused its discretion in allowing Lydall to present Lydall’s U.S. Patent No. 5,424,139 (“the Lydall patent”) as evidence that its products were substantially different from and superior to that covered by ATD’s patents, and that substantial justice requires a new trial. ATD states that Lydall, by its tactics, denied ATD adequate time to respond to this evidence.
During discovery, for each of the 19 claims of the Lydall patent ATD submitted a request for admission, asking Lydall to admit that “each accused Lydall shield is made in accordance with” that claim of the Lydall patent. Lydall’s response, for each claim, was “Denied.” Lydall retained that position for five months after the close of discovery. Twenty-nine days before the start of trial, Lydall changed its responses to “Denied, except as to shields made with twisted expanded metal mesh.” Based on this tardy change in position ATD moved in limine to exclude evidence of the Lydall patent. The motion was denied on the first day of trial, the district court stating that Fed.R.Civ.P. 36 binds a party to an admission, but not to a denial.
Lydall presented its patent as showing that its products were deemed patentable by the Patent Office, despite citation of the ATD patents, and for the comparative data in the Lydall patent which showed superiority of the accused products over the ATD product. Lydall presented five witnesses on these issues, giving substance to ATD’s complaint about the tactics of the deliberately tardy identification of this issue. ATD states that by this tactic it was denied reasonable time to investigate the patent and its data, depose Lydall’s witnesses, and present contrary evidence. ATD states that it was surprised, ambushed, and severely prejudiced. See Erskine v. Consolidated Rail Corp., 814 F.2d 266, 272 (6th Cir.1987) (“One of the primary objectives of the discovery provisions embodied in the Federal Rules of Civil Procedure is elimination of surprise in civil trials.”)
Lydall responds that the fact of separate patentability is admissible evidence, and thus that the district court acted within its discretion in admitting it. As for the tardiness of the discovery response, Lydall states that it properly denied ATD’s requests for admission five months earlier because “ ‘each’ accused product was not so made.” Lydall states that it amended its answers consistent with Fed.R.Civ.P. 26(e).
ATD states that all of the accused Lydall products are within the Lydall patent claims, but for a few prototypes. Lydall does not dispute this point. Nor does Lydall cite a change in circumstance justifying the change in its interrogatory responses. Fed.R.Civ.P. 36 requires that “[a] denial shall fairly meet the substance of the requested admission, *549and when good faith requires that a party qualify an answer or deny only part of the matter ..., the party shall specify so much of it as is true and qualify or deny the remainder.” At the trial Lydall presented five witnesses on various aspects of its patent, its prosecution, and its comparisons with the ATD product. It is not reasonable to assume that Lydall planned this trial presentation within the twenty-nine days after it gave notice to ATD that it would rely on the patent. By no stretch of the Rules can Ly-dall or its counsel be deemed to have met reasonable standards of fairness, good faith, or professionalism.
The district court, in its opinion denying ATD’s request for a new trial, stated that even if there were error in admitting the evidence, it was harmless because there was substantial evidence supporting a verdict of non-infringement even without the evidence of the Lydall patent:
Even if Lydall’s patent were erroneously admitted, such an error was harmless. An error in the admission of evidence is not a ground for granting a new trial where the admission of the evidence was harmless error. Fed.R.Civ.P. 61. In this case, Ly-dall presented substantial evidence at trial of its effort to design around ATD’s patents and also presented substantial evidence which showed that Lydall’s metal mesh separated layers in a different manner than ATD’s embossments. This evidence would support a non-infringement verdict even without evidence of Ly-dall’s ’139 patent.
That is not the correct standard. The question is not whether there was “substantial evidence” of non-infringement without the tainted evidence. See 11 Charles Alan Weight, Arthur R. Miller & Mary Kay Kane, Federal Practioe & Procedure § 2806 at 65 (1984 & Supp.1998) (“[0]n a motion for a new trial ... the judge may set aside the verdict even though there is substantial evidence to support it.”) The question is whether the admission of the Lydall patent affected the substantial rights of ATD, see Fed.R.Civ.P. 61; Fed.R.Evid. 103, or was indeed harmless error. See Schrand, 851 F.2d at 157. Rule 61 provides:
Harmless Error. No error in either the admission or the exclusion of evidence ... is ground for granting a new trial or for setting aside a verdict ... unless refusal to take such action appears to the court inconsistent with substantial justice. The court .,. must disregard any error or defect in the proceeding which does not affect the substantial rights of parties.
See also Fed.R.Civ.P. 59; 11 Wright, Miller & Kane, supra § 2805 at 60 (“The importance of Rule 61 in its application to motions for a new trial cannot be overlooked.... [I]t is only those errors that have caused substantial harm to the losing party that justify a new trial. Those errors that are not prejudicial do not call for relief under Rule 59.”) Therefore, we review whether the admission of this evidence, under the circumstances of the tardy discovery response, reasonably affected the outcome of the case. See Schrand, 851 F.2d at 157 (citing Fed.R.Civ.P. 61 and quoting Jordan v. Medley, 711 F.2d 211 (D.C.Cir.1983) (Scalia, J.) (“assessment of the likelihood that the error affected the outcome of the case”)); 1 Jack B. Weinstein & Margaret A. Berger, Federal Evidenoe § 103.41[2] (1998) (“In general terms, the test of whether a substantial right of a party has been affected is whether the error in question affected the outcome of the case.”)
Circuit Courts in civil cases not involving constitutional error have articulated the standard in different ways, e.g., whether it is “highly probable” that the erroneous admission did not affect the jury verdict, 1 Wein-stein & Berger, supra § 103.41 [2] at 103-53 n. 32 (citing First, Third, and Eleventh Circuit cases); and whether it is “more probable than not” that the erroneous admission did not affect the jury verdict, 1 Weinstein & Berger, supra § 103.41[2] at 103-53 n. 33 (citing Seventh, Eighth, Ninth, and Tenth Circuit cases). See also 3 James Wm. Moore, Federal Practice § 61.02[3] (1998) (standard of Rule 61 articulated in different ways). A number of factors have guided the courts in their determinations of whether error is harmless, including (1) whether erroneously admitted evidence was the primary evidence relied upon, (2) whether the aggrieved party was nonetheless able to *550present the substance of its claim, (3) the existence and usefulness of curative jury instructions, (4) the extent of jury argument based on tainted evidence, (5) whether erroneously admitted evidence was merely cumulative, and (6) whether other evidence was overwhelming. See 1 Weinstein & BERGER, supra § 103.41 [5] [a]-[h]; see also 11 Wright, Miller & Kane, supra § 2885 [Rule 61] at 464 (“erroneous admission of evidence may be found not to have been prejudicial if the fact already had been shown by admissible evidence, or prejudice may be avoided by a curative instruction”).
The district court noted that it gave a limiting instruction to the jury concerning the Lydall patent:
23. Where there is an issued patent, the later issuance of a patent for a device or method raises no presumption of nonin-fringement of the previously issued patent. You may consider the later issued Lydall patent in your decision and give it the appropriate weight, but you must keep in mind that even where improvements and modifications are separately patentable, the improved device or method may still infringe the previously issued ATD patents.
A later patented device or method may include additional elements or steps beyond those claimed in the earlier issued patent. But if the later patented device or method contains each and every element of a claim of the earlier issued patent, or an equivalent of any element not literally included, then that claim of the earlier issued patent is infringed.
This instruction, however, did not relate to ATD’s asserted inability to present a defense on the substance of the Lydall patent.
Following a careful review of the trial record, we do not grant ATD’s request for a new trial. Considering ATD’s emphasis on the use of embossed foil layers to separate the layers, and the use of metal scrim in both the prior art and between one or two of the ATD outer layers in addition to the emboss-ments, we are convinced that in view of the correctly construed claims, a reasonable jury could not have found either literal infringement or infringement under the doctrine of equivalents. On this basis the admission of the Lydall patent did not affect the outcome of this case.
B
Lydall cross-appeals the district court’s refusal to allow Lydall to present U.S. Patent No. 2,037,813 (the Munters patent) as evidence of invalidity of ATD’s patents. The court excluded the Munters patent because Lydall did not produce the reference during the designated discovery period.
According to the record, on December 20, 1994 ATD asked Lydall to identify each document Lydall considered to relate to validity and invalidity of the ATD patents. Lydall did not list the Munters patent in its reply of January 20, 1995. Subsequently, ATD served additional interrogatories and document requests, asking Lydall to identify prior art of which it was aware. Lydall did not list the Munters patent in any response. Following an extension, discovery closed on August 31, 1995. On December 1, 1995, Ly-dall served ATD with a “Notice of Prior Art Pursuant to 35 U.S.C. § 282” listing the Munters patent. This was one month before a first rescheduled pretrial conference. Rejecting the submission, the court explained:
I want the record to be clear as to the basis that I am denying Lydall the opportunity to use that prior art at trial. The basis is not relevance. The basis is not any evidentiary ruling as to the value of this prior art. The basis is simply I’m exercising my discretion under the rule of the Federal Rules of Civil Procedure to preclude a party from relying on theories not made available or not disclosed to the opposing side.
Lydall states that although the Munters patent was not disclosed in response to any of the discovery requests or during the discovery period, it should not have been excluded because it was disclosed in accordance with 35 U.S.C. § 282:
§ 282.... In actions involving the validity or infringement of a patent, the party asserting invalidity or noninfringement shall give notice in the pleadings or otherwise in writing to the adverse party at *551least thirty days before the trial, of the country, number, date, and name of the patentee of any patent, the title, date, and page numbers of any publication to be relied upon as anticipation of the patent in suit.... In the absence of such notice proof of said matters may not be made at the trial except on such terms as the court requires....
Lydall argues that § 282 overrides any discovery schedule set under the Federal Rules of Civil Procedure, because these Rules were instituted in 1938 whereas § 282 was reenacted as part of the 1952 Patent Act. P.J. Federico, commenting on the 1952 Patent Act, reported the relationship in broad, and ambiguous terms:
The last paragraph of section 282 relating to the giving of notice of various details relating to certain defenses is based on part of the last paragraph of old R.S. 4920, with modifications. The old provision was in fact superseded by the Federal Rules of Civil Procedure [28 U.S.C.A.] but, as modified, has been reinstated.
P.J. Federico, Commentary on the New Patent Act (1954), reprinted, in 75 J. Pat. TRADEMARK Off. Soc’y 161, 216 (1993). Since 1952, the matter has not been squarely resolved. In Thermo King Corp. v. White’s Trucking Serv., Inc., 292 F.2d 668, 674, 130 U.S.P.Q. 90, 94 (5th Cir.1961), that court referred to § 282 as “coexisting” with the Federal Rules. We agree that they coexist. However, when the court has set and the parties have agreed to a discovery period, that procedure necessarily governs that trial. Thus although § 282 sets a minimum period for the identification of prior art to be introduced as evidence of anticipation, a specific judicial directive for the timing of discovery establishes the procedures to which the parties are bound.
In Eaton Corp. v. Appliance Valves Corp., 790 F.2d 874, 229 U.S.P.Q. 668 (Fed.Cir.1986) this court held that the trial court could, in its discretion, allow into evidence a reference that was not disclosed at least thirty days in advance despite § 282, for in that case it was clear that the patentee was not surprised and was not prejudiced:
The objective of section 282’s provision for advance notice is to prevent unfair and prejudicial surprise by the production of unexpected and unprepared-for prior art references at trial. To this end, section 282 is to be read with the Federal Rules of Civil Procedure.
790 F.2d at 879,229 U.S.P.Q. at 672 (citations omitted). The purpose of § 282, like that of the Federal Rules, is to prevent unfair and prejudicial surprise, not to facilitate last-minute production of evidence. The district court in the instant case was well within its discretion in excluding the Munters patent, for the record shows that Lydall offered no reason to justify its submission long after the close of discovery. The solid entrenchment of the Federal Rules and the principles of orderly discovery weigh heavily against Ly-dall’s argument that § 282 governs the requirement of notice of prior art despite the elaborate discovery procedures, interrogatories, and explicit directives by which the trial was managed.
Costs
Costs are taxed against Lydall. See Fed. R.App. P. 39; Fed. Cir. R. 39.
AFFIRMED-IN-PART and REVERSED-IN-PART.
. ATD Corp. v. Lydall, Inc., 43 U.S.P.Q.2d 1170, 1997 WL 111783 (E.D.Mich.1997).
. The concurring/dissenting opinion draws analogy to a case styled Vehicular 'Technologies. However, the cases are not the same. In the case at bar the function of the embossments separating the layers is stated in the claims, whereas in Vehicular the function of back-up is staled only in the descriptive text of the specification.
. Although we have affirmed the judgment of non-infringement we review the merits of the judgment of invalidity, in accordance with Cardinal Chem. Co. v. Morton Int'l, Inc., 508 U.S. 83, 98, 113 S.Ct. 1967, 124 L.Ed.2d 1 (1993).
. Further elaboration of the rule was subsequently made. See M.P.E.P. § 609 (Rev.3, July 1997).