Andrew Corporation (Andrew) and Gabriel Electronics, Inc. (Gabriel) appeal and cross-appeal two final judgments of the United States District Court for the District of Maine. The court held1 Andrew’s U.S. Patent No. 4,410,892 (the Knop patent) invalid for indefiniteness, but if valid infringed by Gabriel. The court’s second judgment2 held Gabriel’s U.S. Patent No. 3,550,142 (the Dawson patent) not infringed by Andrew. We reverse the judgment of invalidity of the Knop patent and affirm the other aspects of both judgments.
Background
Both the Knop and the Dawson patents relate to improvements in horn reflector microwave antennas used in long distance telephone and data communication networks. As described by the district court, a horn reflector antenna generally is constructed of an inverted vertical “feed” cone and a horizontal cylinder, which intersect at right angles. The microwave beam enters the feed cone vertically from the apex of the cone; an angled paraboloidal reflector catches the unfocused beam as it rises, focuses it into a coherent beam, and reflects it out the horizontal cylinder and on to the next antenna in the network.
The performance of such antennas is evaluated using the standard criteria of “gain” and “pattern”. “Gain” refers to the strength of the focused beam relative to the original unfocused beam; higher gain allows transmission over longer distances. “Pattern”, or “Radiation Pattern Envelope” (“RPE”), refers to the distribution of microwave energy outside the main beam in what are referred to as “side-lobes”.
These patterns are measured in both the horizontal (“E-plane”) and the vertical (“H-plane”) directions. Normally, the E-plane has more energy distributed outside the main beam. This is referred to as having “higher sidelobes” or “a wider pattern”, and results in unwanted interference with nearby antennas. The overall performance of an antenna is measurable by the E-plane and H-plane patterns.
THE KNOP PATENT
The Knop patent specification describes a horn reflector antenna that reduces inter*821ference with other antennas without significant loss of gain. This improved result is obtained by placing absorber material deep inside the conical feed horn. The use of absorber material had been shown in the prior art, placed in the first few inches of the conical feed horn to dampen stray radiation. According to the prior art, placing absorber material deeper into the cone causes unsatisfactory loss of gain. Knop teaches that this deep absorber acts by reducing the width of the E-plane RPE without significantly affecting the quality of the H-plane RPE, thereby improving overall performance of the antenna.
Indefiniteness — 35 U.S.C. § 112 A
The district court held the Knop patent claims invalid, stating that terms in the claims such as “approach each other”, “close to”, “substantially equal”, and “closely approximate”, with reference to the E-plane and H-plane RPEs, were too vague to satisfy the requirement of definiteness stated in 35 U.S.C. § 112.3 One or more of these terms appears in each of the claims, as illustrated in the following independent and dependent claims:
1. A conical horn-reflector antenna comprising the combination of:
a paraboloidal reflector forming a par-aboloidal reflecting surface for transmitting and receiving microwave energy,
a smooth-walled conical feed horn for guiding microwave energy from the focus of said paraboloidal reflecting surface to said reflector, and
a lining of absorber material on the inside wall of the horn for reducing the width of the RPE in the E plane of the antenna without significantly increasing the width of the RPE in the H plane, said absorber increasing the Eigen value E and the spherical hybridicity factor Rs sufficiently to cause the E plane and H plane RPEs to approach each other.
3. A conical horn-reflector antenna as set forth in claim 2 which produces substantially equal E and H plane illumination patterns.
6. A method as set forth in claim 5 wherein said lining of absorber material increases the taper of the field distribution along the radii of said horn in the E plane to closely approximate the taper of the field distribution along the radii of said horn in the H plane, [emphases added]
The criticized words are ubiquitous in patent claims. Such usages, when serving reasonably to describe the claimed subject matter to those of skill in the field of the invention, and to distinguish the claimed subject matter from the prior art, have been accepted in patent examination and upheld by the courts. As this court put it in Rosemount, Inc. v. Beckman Instruments, Inc., 727 F.2d 1540, 1546-47, 221 USPQ 1, 7 (Fed.Cir.1984):
Beckman attacks the claims as indefinite, primarily because “close proximity” is not specifically or precisely defined. As stated in the district court’s Memorandum Decision, “to accept Beckman’s contention would turn the construction of a patent into a mere semantic quibble that serves no useful purpose.”
In Rosemount the district court found that “ ‘close proximity’ is as precise as the subject matter permits”. Id. In Seattle Box Co. v. Industrial Crating & Packing, 731 F.2d 818, 826, 221 USPQ 568, 573-74 (Fed.Cir.1984) (citing W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1557, 220 USPQ 303, 316 (Fed.Cir.1983), cert. denied, 469 U.S. 851, 105 S.Ct. 172, 83 L.Ed.2d 107 (1984)), the court remarked that “substantially equal” is a term of degree, and that its acceptability depends on “whether one of ordinary skill in the art would understand what is claimed ... in light of the specification”, even if experimentation may be needed.
In W.L. Gore & Associates, Inc. v. Garlock, Inc., No. 87-1296, 842 F.2d 1275, 1280, 6 USPQ2d 1277, 1282 (Fed.Cir.1988), this court stated that an “imprecise claim *822limitation, such as the phrase ‘about 100% per second’ ” does not impart invalidity to the claims, but is to be considered in determination of infringement. See also Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1385, 231 USPQ 81, 95 (Fed.Cir.1986) (“the claims, read in light of the specification, reasonably apprise those skilled in the art and are as precise as the subject matter permits. As a matter of law, no court can demand more”), cert. denied, — U.S. -, 107 S.Ct. 1606, 94 L.Ed.2d 792 (1987); Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 624, 225 USPQ 634, 641 (Fed.Cir.), cert. dismissed, 474 U.S. 976, 106 S.Ct. 340, 88 L.Ed.2d 326 (1985):
“If the claims, read in the light of the specification, reasonably apprise those skilled in the art both of the utilization and scope of the invention, and if the language is as precise as the subject matter permits, the courts can demand no more.”
(quoting Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124, 136, 118 USPQ 122, 132. (2d Cir.), cert. denied, 358 U.S. 884, 79 S.Ct. 124, 3 L.Ed.2d 112 (1958)).
The Manual of Patent Examining Procedure instructs examiners in a similar vein. See MPEP § 706.03(d):
[An examiner] should allow claims which define the patentable novelty with a reasonable degree of particularity and distinctness. Some latitude in the manner of expression and the aptness of terms should be permitted even though the claim language is not as precise as the examiner might desire, [emphasis in original]
Andrew asserted that the Knop claims could not reasonably be expressed more precisely; and indeed the court found that it “became very clear during trial ... that curves showing RPEs for horn antennas will never be identical”. Words similar to those used in the Knop claims appear in prior art patents that were of record in the district court, dealing with similar technology. For example, claim 1 of U.S. Patent No. 3,305,870 to Webb describes a “radiation pattern whose magnitudes in both the E and H planes are substantially equal”.
Neither the record nor the law supports Gabriel’s position that one of ordinary skill in the art would not know when the RPEs were “substantially equal” or “closely approximate”. The district court’s ruling is contrary to authority, and the holding of invalidity on this ground is reversed.
B
The district court observed that the RPEs change gradually with increased depth of absorber in the cone. The court held that the claims must specifically delineate the point at which infringement starts, and that if such point does not correspond to an unexpected change in properties, the claims are fatally flawed under 35 U.S.C. § 112. The court said:
There is no unexpected change in the E-plane at any of the levels of absorber which gives a clue to what is intended by the patent claims. The choice of a point beyond which there is infringement, therefore, is arbitrary and the point unforeseeable. [footnote omitted]
The district court held that the outer limit of the claim scope must coincide precisely with the point at which the claimed invention comes into scientific being. The court referred to Brown-Bridge Mills, Inc. v. Eastern Fine Paper, Inc., 700 F.2d 759, 763, 217 USPQ 651, 655 (1st Cir.1983), which quoted with approval the holding originating in Kwik-Set, Inc. v. Welch Grape Juice Co., 86 F.2d 945, 947, 32 USPQ 104, 106 (2d Cir.1936) that “It is only where the selected [i.e., claimed] point corresponds with the physical phenomenon and the patentee has discovered the point at which that physical phenomenon occurs that the maintenance of a patent monopoly is admissible.”
This erroneous theory would prevent a patentee from obtaining claims that do not coincide with, or claim less than, the exact point at which a change in the physical phenomenon occurs; it would require the patentee always to discover that point, no matter how prolonged or expensive the additional research; and would bar patent *823protection when the change is by nature gradual or incremental at its transition, whether or not it is unobvious in view of the prior art.
Patentability is not measured against the closest point on the road to invention. Much technological change that meets the criterion of unobviousness, when viewed in light of the prior art, has a fuzzy boundary at its point of origin. Technological differences from prior art usually become more pronounced with distance from the boundary, but the changes may become manifest gradually. Indeed, the location of the boundary may well change with the available precision of measurement.
It is the prior public knowledge— the “prior art” — by which patentability is tested. A patentee may set the metes and bounds of that which is sought to be patented, and it is not material whether the phenomena just outside these claim limits are qualitatively different from that which is claimed. The patentee is not required to show that some technological discontinuity exists between the claimed invention and the subject matter just outside the claims, but only that the claimed subject matter would have been nonobvious in view of the prior art. 35 U.S.C. § 103.
The law imposes no obligation on a patent applicant to determine what is going on in the technological gap between the claimed invention and the prior art, or to set the claim limits at the precise technological edge of the invention. A claim is not fatally indefinite for failing specifically to delineate the point at which the change in physical phenomenon occurs. See, for example, the extensive body of case law holding that a patentee may claim less than the entire invention:
Nothing is better settled in the law of patents than that the patentee may claim the whole or only a part of his invention, and that if he only describe and claim a part, he is presumed to have abandoned the residue to the public.
McClain v. Ortmayer, 141 U.S. 419, 423-24, 12 S.Ct. 76, 77, 35 L.Ed. 800 (1891).
To the extent that Brovm-Bridge may be read as holding, as the district court appears to have believed, that a claim is invalid unless it sets as a limitation the exact technological border of the invention, we expressly reject that holding.4 The district court’s judgment of invalidity on this ground is reversed.
Obviousness and Enablement
Gabriel appeals the district court’s judgment that the Knop patent had not been proved invalid for obviousness, 35 U.S.C. § 103, or for lack of enablement, § 112. We have considered all of the arguments on both sides, and conclude that the court did not err in holding that the presumption of validity concerning these issues had not been overcome by clear and convincing evidence. See Hybritech Inc., 802 F.2d at 1375, 231 USPQ at 87 (“the presumption of validity goes to validity of the patent in relation to the patent statute as a whole, not just to non-obviousness”) (emphasis in original).
Inequitable Conduct
? asserted that Knop withheld material references from the patent examiner, thereby rendering the patent unenforceable for inequitable conduct. The district court determined that a European patent application which was disclosed to the patent examiner in a Rule 312 amendment, 37 C.F.R. § 1.312, was reasonably timely disclosed and was not material to the prosecution of the Knop patent. The court also found that a horn antenna that was in the *824applicant’s possession was “disclosed generically in the patent application” and that there was no intent to withhold this information. None of these findings has been shown to be clearly erroneous. Both materiality and intent to withhold must be shown, before these elements can be balanced in determining whether the applicant was guilty of inequitable conduct in patent prosecution. J.P. Stevens & Co. v. Lex-Tex Ltd., 747 F.2d 1553, 1560, 223 USPQ 1089, 1092 (Fed.Cir.1984), cert. denied, 474 U.S. 822, 106 S.Ct. 73, 88 L.Ed.2d 60 (1985).
We affirm the district court’s conclusion that inequitable conduct has not been proved by clear and convincing evidence.
Infringement of the Knop Patent
Andrew asserted infringement of Knop claims 1, 2, 3, 5, and 6 by the Gabriel antenna, either literally or in terms of the doctrine of equivalents. The district court first considered Andrew’s assertion of literal infringement.
Construing the claims in light of the prosecution history, the court concluded that the final phrase of all the independent claims
said absorber increasing the Eigen value E and the spherical hybridicity factor Rs sufficiently to cause the E plane and H plane RPEs to approach each other
was a material limitation because it was added during prosecution. The court heard testimony on whether the absorber in the accused Gabriel antenna achieved this result. Two experts testified, one called by Gabriel and one by Andrew, taking contrary positions on this issue. The district court found both experts to be “competent” and "credible”. The court stated that it could not decide between the opposing positions of the experts concerning the Eigen value and the spherical hybridicity, and that “the court has no basis for accepting either proposition over the other”. The court thus concluded that the evidence was “in equipoise”, and therefore that the plaintiff had not met its burden of proving infringement by preponderant evidence.
A true equipoise of evidence may indeed defeat the party with the burden of proof, see Aero Spacelines, Inc. v. United States, 530 F.2d 324, 332, 208 Ct.Cl. 704 (1976); see also Wilson v. Omaha Indian Tribe, 442 U.S. 653, 669, 99 S.Ct. 2529, 2538, 61 L.Ed.2d 153 (1979), but there is no authority for holding evidence to be in equipoise for the sole reason that the court could not decide between conflicting experts. We agree with the statement in United States v. General Motors Corp., 561 F.2d 923, 933 (D.C.Cir.1977), cert. denied, 434 U.S. 1033, 98 S.Ct. 765, 54 L.Ed.2d 780 (1978):
The mere fact that experts disagree does not mean that the party with the burden of proof loses. The finder of fact has to make the effort to decide which side has the stronger case. This can be based on the demeanor of the witnesses (if so, the trial judge should say so) or the intellectual strength of the evidence and arguments based thereon.
Given the complexity of modem technology, it may well happen that qualified experts will appear on both sides, that their testimony will conflict, and that the testimony or the technology or both of them will be difficult to understand. However, to decline to decide the issue when conflicting evidence appears to be counterbalancing solely because the subject matter is technically complex, will defeat the party with the burden of proof without fair hearing.
Such treatment would remove complex technological issues from the purview of justice; this can not reflect the correct judicial response to a world increasingly bound to technology. Thus we do not endorse in principle the district court’s treatment of literal infringement. In this case, however, the issue of infringement was adequately resolved by the court’s analysis in terms of the doctrine of equivalents.
The district court found that Gabriel’s antenna infringed the Knop patent under the doctrine of equivalents, applying the classical analysis of Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097, 85 USPQ 328 (1950). The court stated, “From exhibits produced at trial, and without necessarily understanding the mathematical expía-*825nation, the Court can see that both the [accused device and] Andrew’s embodiment of the patent function in the same way.” The court thus found that although it did not know the precise changes in the Eigen value and spherical hybridicity factors of the final clause of claim 1, equivalency had been established based on the similarities in structure and RPEs of the accused antenna and the patented antenna, which showed that the two antennas achieved the same results by the same means in the same way. Thus, although the court had found that the evidence as to the subject matter of the final phrase of the claim was inconclusive, the court stated:
At trial, the Court was thoroughly persuaded that for all practical purposes the [accused device] produced virtually the same patterns as those produced by the [patented device].
The court concluded that
by use of the same means, the [accused device] produces the same results in the same way as the invention described by the Knop patent.
The court, applying the doctrine of equivalents to the accused device in light of the claimed invention viewed as a whole, held that the accused device was within the ambit of the claim, despite the absence of preponderant evidence as to the final phrase when analyzed in isolation from the rest of the claim.
Gabriel argues that the doctrine of prosecution history estoppel requires that the limitation of the final claim phrase be affirmatively found in the accused device, and that it can not be inferred based on similarities in structure, way, and result. Gabriel argues that since the district court held itself incapable of finding this fact, infringement can not be found, even under the doctrine of equivalents.
The district court had held, responding to this argument by referring to the prosecution history, that “it seems more appropriate to characterize the [addition during examination of the final claim phrase] as a clarification of the function of the absorber in the cone”. The court concluded that “where the amendments were made for the purposes of explication and clarity, the Court does not think it appropriate to invoke estoppel against the application of the doctrine of equivalents.”
The purpose of an amendment must be taken into account when considering prosecution history estoppel. Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1363, 219 USPQ 473, 481 (Fed.Cir.1983) (“Depending on the nature and purpose of an amendment, it may have a limiting effect within a spectrum ranging from great to small to zero”). See also Moeller v. Ionetics, Inc., 794 F.2d 653, 659-60, 229 USPQ 992, 996-97 (Fed.Cir.1986) (amendment was not intended to limit, but “to ‘particularly point out’ the invention”, quoting the examiner).
We agree with the district court’s determination that this amendment to the Knop claims did not bar application of the doctrine of equivalents. Relying on the totality of similarities between the accused device and the claimed structure, including the similarities in the RPEs that are referred to in the final phrase of the claims, the district court found that the accused antennas perform the same function in substantially the same way to achieve substantially the same result as the claimed device. No clear error having been shown, the district court’s finding of infringement is affirmed.
THE DAWSON PATENT
The court held that Gabriel’s Dawson patent for a horn reflector microwave antenna was not literally infringed by the Andrew antenna since the claimed isolation of the reflector assembly was not literally present. Andrew Corp. v. Gabriel Electronics, Inc., 2 USPQ2d 1792, 1793-94 (D.Me.1987) [available on WESTLAW, 1987 WL 14966]. The court also held that the patent was not infringed under the doctrine of equivalents.
It was not contested that the “isolation of the reflector assembly” is a critical limitation in the Dawson patent claims. The accused Andrew device was tested to determine if isolation existed. The court re*826viewed these tests and the accompanying testimony, and held that neither isolation nor an equivalent was present. We discern no clear error in the court’s analysis and reasoning, and affirm its decision of non-infringement of the Dawson patent.
Summary
The court’s judgment of invalidity of the Knop patent due to claim indefiniteness is reversed. The holding of validity on all other grounds is affirmed.
The judgment that the Gabriel antenna infringes the Knop patent is affirmed; the matter is remanded for determination of damages.
The judgment of non-infringement of the Dawson patent is affirmed.
Costs
Costs in appeal and cross-appeal in favor of Andrew Corporation, the prevailing party-
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.
. Andrew Corp. v. Gabriel Electronics, Inc., No. 83-0372-P (D.Me. August 1, 1986).
. Andrew Corp. v. Gabriel Electronics, Inc., 2 USPQ2d 1792 (D.Me.1987) [available on WEST-LAW, 1987 WL 14966].
. 35 U.S.C. § 112 ¶ 2: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
. The case authorities relied on in Brown-Bridge v. Kwik-Set and Helene Curtis Indus. v. Sales Affiliates, 233 F.2d 148, 154, 109 USPQ 159, 163-64 (2d Cir.1956), do not support the application that the district court gave them. In both Kwik-Set and Helene Curtis the claim limitation fell in the midst of the prior art; for instance the claim in Helene Curtis was to an optimum range of a chemical property where the chemical was already known to have that property. See also Dow Co. v. Halliburton Co., 324 U.S. 320, 65 S.Ct. 647, 89 L.Ed. 973 (1945), which held on similar facts that the claimed range must be proved critical in order to show "invention”. These cases were all concerned with showing "invention” in light of prior art, rather than the § 112 issues here involved.