Bio-rad Laboratories, Inc., a Delaware Corporation,plaintiff-appellee, Cross- Appellant, v. Nicolet Instrument Corporation, a Wisconsin Corporation,defendant-appellant, Cross-appellee, 739 F.2d 604 (Fed. Cir. 1984)

This is an appeal from an unpublished decision of the United States District Court for the Northern District of California holding claims 1-7 of U.S. Patent No. 3,319,515 (hereinafter '515 patent or Flournoy patent) for an "Interferometric Optical Phase Discrimination Apparatus" valid and infringed. Judgment was entered on a jury verdict rendered after fourty-four days of trial. We vacate the District Court's denial of prejudgment interest and remand for an award of prejudgment interest or a determination that some reason exists for denying prejudgment interest. In all other respects, the judgment is affirmed.

The Parties

Bio-Rad Laboratories, Inc. (Bio-Rad) is a Delaware corporation with its principal office in Richmond, California. Between 1978 and 1979, Bio-Rad acquired two divisions, Digilab, Inc. (Digilab) and Block Engineering, Inc. (Block Engineering). Digilab was formed as a subsidiary of Block Engineering in 1969 to produce and sell instruments that rapidly measure the crystalline surface layer on semiconductors, known as the epitaxial layer. Since 1970, Digilab has manufactured and sold an epitaxial measuring instrument known as the FTG-12 under an exclusive license from E.I. du Pont de Nemours and Company (du Pont), owner of the '515 patent. When Bio-Rad acquired Block Engineering and Digilab, it also acquired the exclusive license under the '515 patent which included the right to enforce the patent against infringers. The name Bio-Rad will be used throughout this opinion to refer to the conduct of Digilab, and Block Engineering, as well as the conduct of Bio-Rad.

Nicolet Instrument Corporation (Nicolet) is a Wisconsin corporation engaged in the production and sale of electronic instruments. In 1979, Nicolet began selling an epitaxial thickness measuring instrument known as the MX-ECO.

Procedural Background

In 1981, Bio-Rad sued Nicolet for willful infringement of the '515 patent by Nicolet's production and sale of the MX-ECO measuring instrument. Nicolet counterclaimed for a declaratory judgment that the '515 patent is invalid and not infringed, and for alleged violations of the antitrust laws.

Both Bio-Rad and Nicolet demanded a jury trial and agreed to submit the case to the jury on general verdicts without special interrogatories. The jury declared all seven claims of the '515 patent valid and infringed, but not willfully infringed. The jury also concluded that Bio-Rad did not violate sections 1 or 2 of the Sherman Antitrust Act. The jury awarded Bio-Rad $3,078,000 in money damages for the infringement.

Nicolet moved for judgment notwithstanding the verdict (JNOV) and, in the alternative, for a new trial. Nicolet also moved to deny enforcement of the '515 patent for misuse. The district court denied all three motions and denied Bio-Rad's separate request for prejudgment interest. The trial judge entered judgment on the jury verdicts without comment and permanently enjoined Nicolet from infringing the '515 patent in any way, including by the production, use or sale of the MX-ECO epitaxial thickness gauge. Interest was awarded from the date judgment was entered.

Nicolet appeals from the judgment entered by the District Court, and from the court's refusal to declare the '515 patent unenforceable for patent misuse. Bio-Rad cross-appeals from the lower court's denial of prejudgment interest. Although this appeal is from the judgment entered on the jury verdict, the District Court's denial of Nicolet's motions guides our analysis. Accordingly, we consider the issues raised by the parties in the following order:

I. Grounds For JNOV.

II. Grounds For New Trial.

III. Patent Misuse.

IV. Prejudgment Interest.

Standard of Review

Several cases have issued from this court describing the analysis trial judges should follow when they entertain motions for JNOV or new trial. Connell v. Sears, Roebuck & Co., 722 F.2d 1542, 1546, 220 USPQ 193, 197 (Fed. Cir. 1983); Railroad Dynamics, Inc. v. A. Stucki Co., 727 F.2d 1506, 1512-13, 220 USPQ 929, 936 (Fed. Cir. 1984); Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 221 USPQ 669, 672-673 (Fed. Cir. 1984). That analysis need not be repeated here, but description of our appellate review process is useful.

Nicolet has urged that we review the record de novo, in part because no interrogatories were submitted to the jury and because the trial judge did not produce a written opinion discussing reasons for denying Nicolet's motions or describing the facts supporting the jury verdict. Nicolet mistakes our role. The presence of special interrogatories or a written opinion by the trial court certainly makes our task easier but the absence of these aids does not alter our standard of review. In the absence of special interrogatories we presume the existence of factual findings and legal conclusions necessary to support the verdict reached by the jury. The findings and conclusions are controlled by the court's instructions to the jury. Perkin-Elmer, 732 F.2d at 893, 221 USPQ at 673. On appeal after denial of a motion for JNOV, appellant must show either: 1) that the jury's presumed or express findings are not supported by substantial evidence or, 2) if the jury's findings were supported by substantial evidence, that the legal conclusions implied from the verdict cannot be supported by those findings. Perkin-Elmer, 732 F.2d at 893, 221 USPQ at 673. In sum, appellant must show either that reasonable jurors could not have made the presumed findings or that the conclusions necessary to the general verdict were not supported. The District Court's denial of Nicolet's Motion for a new trial is reviewed on an abuse of discretion standard. Railroad Dynamics, Inc. v. A. Stucki Co., 727 F.2d at 1512, 220 USPQ at 935. As this description of our approach demonstrates, our role as a reviewing court is circumscribed by standards that place a difficult burden on appellant.

Analysis

To overturn the District Court's denial of the Motion For JNOV, Nicolet presents several grounds for declaring the '515 patent invalid. The various grounds for invalidity will be discussed under the following lettered sections:

A. Anticipation and Obviousness.

B. Fraud.

C. Compliance With The Requirements Of 35 U.S.C. § 112.

D. Infringement.A. Anticipation and Obviousness

The '515 patent claims a method and apparatus for determining the thickness of a film or a surface layer on a substrate. The apparatus, known as an interferometer, may be used to measure the thin crystalline surface layer of semiconductors, known as the epitaxial layer. The claimed method is represented by claim 1 of the '515 patent, reproduced below:

1. A method for determining a physical property of a specimen on the basis of interferometic optical phase discrimination comprising, in sequence, directing analytical radiation upon said specimen along a course such that there is developed an optical phase difference between radiation emanating from a first radiation-reflective surface of said specimen and from a second radiation-reflective surface of said specimen, cyclically varying at a predetermined time rate the length of the path of travel of said radiation emanating from a given one of the pair consisting of said first radiation-reflective surface of said specimen and said second radiation-reflective surface of said specimen, with respect to the length of the path of travel of the radiation emanating from the other of said pair over a span inclusive of a zero-order peak at which there exists constructive interference between said radiation emanating from said first radiation-reflective surface of said specimen and from said second radiation-reflective surface of said specimen, measuring the elapsed time existing between the scan traverse of said zero-order peak at which there exists said constructive interference and a preselected reference point synchronized in time with the cycle of said scan traverse, said preselected reference point being originally established in location by scanning said specimen through a span inclusive of a central zero-order peak, and determining said physical property of said specimen as a function of said elapsed time.

The apparatus claims correspond to the method claims as illustrated by claim 5, reproduced below:

5. Apparatus for determining a physical property of a specimen on the basis of interferometric optical phase discrimination comprising, in combination, a source directing radiation upon said specimen of a wavelength adapted to develop an optical phase difference between radiation emanating from a first radiation-reflective surface of said specimen and from a second radiation-reflective surface of said specimen, means cyclically varying at a predetermined time rate the length of the path of travel of said radiation emanating from a given one of the pair consisting of said first radiation-reflective surface of said specimen and said second radiation-reflective surface of said specimen with respect to the length of the path of travel of the radiation emanating from the other of said pair over a span inclusive of a zero-order peak at which there exists constructive interference between said radiation emanating from said first radiation-reflective surface of said specimen and from said second radiation-reflective surface of said specimen, means sensing the occurrence of said zero-order peak, means measuring the elapsed time existing between the scan traverse of said zero-order peak and a preselected reference point synchronized in time with the cycle of said scan traverse, said preselected reference point being originally established in location by scanning said specimen throughout a span inclusive of a central zero-order peak, and means indicating said physical property of said specimen as a function of said elapsed time.

Operation of a preferred embodiment of the claimed invention is illustrated with reference to Figure 1 from the patent in suit.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

Film or specimen 10 moves over guide rolls 11 and 12 as it is produced. A beam of light emanates from a light source 16 mounted within housing 17. Light 15 travels through convex lens 18 and is projected as a single beam onto the film 10 by a silvered prism 19.

Enlarged inset A shows light 15 impinging upon a front refractive surface of film 10 at Q. A first component of the light is reflected away from the film along R1, while a second component of the light is transmitted through the film to a back radiation-reflective surface at S. The transmitted component is reflected back along path R2 . Both components of light moving along paths R1 and R2 are identified in Figure 1 as 23. The reflected light components traveling along paths R1 and R2 are out of phase because of the difference in distance traveled by the first and second components after refraction and reflection. The phase difference results in interference that can be measured by an interferometer.

An interferometer used in appellee's claimed invention, known as a Michelson interferometer, is illustrated in Figure 2 of the '515 patent:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

As in conventional interferometers, the light 23 strikes an inclined beam-splitter 32 which delivers light to mirrors 30 and 31. The mirrors reflect the light back to beam splitter 32 which in turn directs the beams towards a photoelectric detector 33. The detector measures the interference between the beams of light. During operation, mirror 30 is reciprocated cyclically over a known distance. The reciprocal motion, or oscillation, is effected by a saw-tooth oscillator 37 and associated electronic power amplifier 38. The mirror 30 oscillates at about 50 cycles/second.

As explained above, the light making up beam 15 is in phase when it reaches the surface of film 10. However, after reflection and refraction, the light traveling along paths R1 and R2 are out of phase because of the difference in distance traveled. The differences in distance traveled is related to the thickness of film 10. In the interferometer, a beam splitter divides each of the two components of the light 23 in half, yielding four beams. Two beams are directed to a stationary mirror and the other two to the reciprocating mirror. The mirrors reflect the beam components back to the beam splitter which directs the light to the detector. As the beams emerge from the interferometer together, they interfere with each other. This interference is measured and an interferogram may be produced. The interferogram may look like the interferogram illustrated in Figure 3 of the '515 patent, reproduced below:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

Central peak P is generated when all four beam components interfere constructively. This occurs when the mirrors 30 and 31 are the same distance from the beam splitter 32. Side peaks g1 and g2 occur when the difference in the distances between the beam splitter and the two mirrors corresponds to the optical retardation caused by the film. The side peaks are of lesser intensity than the central peak because only two of the beam components are constructively interfering.

The invention claimed in the '515 patent uses the distance between peaks to determine the thickness of the film. Figure 3 shows the "mirror separation in microns." This is the distance the mirror has traveled from the position where the four beams interfere constructively, known as the zero position. The time between the occurrence of two reference peaks may be measured directly from the interferogram to determine the thickness of the film.

Nicolet argues that the claims, except claim 6, read on a publication by Lowenstein and are therefore invalid under 35 U.S.C. § 102. Lowenstein used infrared radiation and a lamellar grating interferometer which does not use mirrors that oscillate like those of a Michelson interferometer. Lamellar grating interferometers were used when light in the far infrared was employed because, at the time, there were no efficient beam splitters for light in the far infrared. One measurement or run of the interferometer used by Lowenstein could take six to eight hours. Nicolet's argument is that the substitution of a Michelson interferometer would have been obvious to one of ordinary skill in the art when not working with far infrared radiation. There was also evidence that the Lowenstein article differs from the claimed invention because Lowenstein does not teach the measurement of thin films from peaks of an interferogram. Because of these substantial differences between the claimed invention and the Lowenstein reference, the appropriate analysis is under 35 U.S.C. § 103.

The parties agreed that a person of ordinary skill in the art would be an instrument designer with a college degree in physics or engineering and some practical experience in the design of optical instruments. Each party proffered expert testimony on the factual question of the scope and content of the prior art. It is notable that Bio-Rad's expert, Gebbie, was the only expert capable of testifying on the basis of first-hand experience with the level of skill in the relevant time period. One of Nicolet's experts, Lowenstein, whose article was relied on to assert invalidity, said:

As far as the designers of commercial or industrial interferometers were concerned, I think I don't know anybody or, at least not well enough to be able to form an opinion about his level of skill.

This excerpt from the testimony is produced to illustrate the importance of the jury's role in ascertaining the content of the prior art. The jurors must discern how a person of ordinary skill in the art would have interpreted available art at the time the invention was made. The factfinder's interpretation of prior art references is necessarily guided by the credibility given the testimony of the experts. Here, the jury presumably credited the testimony of Gebbie rather than the testimony of Lowenstein in deciding what the Lowenstein article would have meant to one of ordinary skill in the art. Given the different familiarity of the experts with those skilled in the art at the relevant time, that evaluation by the jury was reasonable and cannot be successfully challenged on appeal. The only question for this court, therefore, is whether, on the facts testified to by Gebbie and presented by Bio-Rad, reasonable jurors could conclude that the invention would not have been obvious to one of ordinary skill in the art at the time it was made.

Nicolet argues the claimed invention would have been obvious in view of the Lowenstein publication. The Lowenstein publication used a lamellar grating interferometer to measure refractive properties of sapphire. The interferometer used by Lowenstein had a set of stationary plates and a set of moveable plates. To analyze the sample being studied, the moveable plates were run in one direction by a motor until the motor was shut off. One run, or scan, typically took 6 to 8 hours. After one scan was completed a motor drove the interferometer back to begin a new scan. The claimed invention requires cyclical mirror movement. Lowenstein did not disclose or suggest the mirror oscillation required by the claimed invention.

Dr. Gebbie testified that those of ordinary skill in the art were skeptical of Michelson rapid scan interferometers, which provide cyclical mirror movement, and would not have interpreted the Lowenstein article to suggest using a Michelson interferometer as required by the claimed invention. It appears from Gebbie's testimony that a person of ordinary skill in the relevant art would not have arrived at the claimed invention after reading the Lowenstein article. Nicolet has not shown that reasonable jurors could not agree with Gebbie that one of ordinary skill in the art would not have derived the claimed invention after reading the Lowenstein publication.

Further, there was evidence that the claimed invention enjoyed commercial success in the measurement of electrically conductive film and epitaxial layers of semiconductors. The claimed invention permitted simple accurate measurement of substrates moving at high speeds. It reduced labor costs and increased productivity. The savings were in some instances so great that one witness said the $65,000 instrument had paid for itself in about two weeks. The jury had to consider this success in reaching its presumed conclusion that the claimed invention would not have been obvious.

We have no doubt that conflicting evidence was presented at trial by Nicolet on the appropriate interpretation of the Lowenstein publication and commercial success. That, however, does not carry Nicolet's burden on appeal.

Nicolet argues the '515 patent is invalid because Flournoy, the named inventor, did not actually invent the claimed subject matter. Nicolet contends that Myron Block, not Flourney, was the inventor. According to Nicolet, Myron Block permitted Flournoy, a DuPont engineer, to claim inventorship to induce DuPont to invest in Block Associates (a predecessor of Block Engineering founded by Myron Block). There is conflicting testimony in the record with regard to how the invention was developed. Nicolet's case on fraud principally hinges on the credibility of Myron Block. However, the development of the claimed invention described by Bio-Rad during the trial adequately supports the finding that Flourney invented the claimed subject matter.

Dr. Philip Flournoy joined DuPont in 1960 as a research engineer in the Engineering Physics Laboratory at the DuPont Experimental Station. A group of scientists at DuPont, including Dr. Flournoy, were engaged in developing instruments for monitoring and controlling a wide variety of industrial manufacturing processes. Dr. Flournoy wanted to develop an instrument that could provide rapid information on the thickness of epitaxial layers on semiconductors of electrically conductive film, known as Mylar, during the production process. Previous techniques were undesirably slow because the spectrometers used required a relatively long period of time to produce one spectrum. Flournoy hoped that a rapid scan interferometer could provide desired spectral information rapidly in a short period of time.

In the course of investigations on measuring Mylar film and epitaxial layers, Dr. Flournoy observed the presence of small peaks ("side bursts") on either side of the main peak of an interferogram. To determine whether these small peaks were associated with the equipment or with the sample, Dr. Flournoy took interferograms of samples having different thicknesses and observed that the spacing between these peaks varied directly with the thickness of the sample. This determination led him to conclude that, with an adequate interferometer, he could construct an instrument for measuring film thickness, utilizing a rapid scan interferometer.

In March, 1961, Flournoy and his supervisor visited Block Associates to evaluate Block Associates as a vendor of rapid scan interferometers for DuPont's process control equipment needs. That trip satisfied Flournoy that Block Associates could produce sufficiently high quality equipment to satisfy DuPont's requirements.

By December 1961, Flournoy had obtained a suitable interferometer. With that instrument, he performed a series of experiments whereby he took interferograms from samples of Mylar, detected distinguishable side bursts in the interferogram, and utilized the distance between those side bursts to calculate the thickness of the Mylar samples. The accuracy of these experimentally obtained thickness readings was verified by comparing them to the actual thickness of the film, measured with a micrometer. By July 1962, Flournoy further refined the specifications for the various operational parameters for a thickness measuring instrument.

In July 1962, there were discussions between DuPont and Block Associates concerning the possibility of Block's manufacturing the thickness measuring instrument for DuPont. In the course of these discussions, the parties recognized that Block Associates and DuPont had possibly conflicting claims to inventorship of the process and apparatus embodied in the thickness gauge. It was agreed that the parties would collect and exchange documentation of their inventions. On the basis of the exchanged documentation, the patent attorneys would determine which party was entitled to any of the invention(s) that might reside in the thickness gauge under development.

After the documents were exchanged, Block Associate's patent attorney and DuPont's patent attorney determined that Flournoy was the inventor. Specifically, they found that Flournoy could prove he had conceived the invention of measuring film thickness with reference to the side burst spacing on the interferogram at least as early as March 23, 1961 and that he could prove a reduction to practice of that invention in December 1961. Block, on the other hand, could not prove that he or his employees had conceived the invention any earlier than May 4, 1961, nor could Block prove a reduction to practice of the invention any earlier than June 29, 1962. A memorandum was produced, and signed by both attorneys, wherein Myron Block recognized Flournoy as the inventor of the subject matter ultimately claimed in the '515 patent.

These facts adequately support the jury's presumed finding that Dr. Flournoy invented the claimed subject matter. Whether a different conclusion could have been reached on the basis of conflicting testimony presented by Nicolet at trial is not the issue before this court.

C. Compliance With The Requirements Of 35 U.S.C. § 112

Nicolet argues that the '515 patent is invalid for failure to disclose the best mode and because the disclosed invention is inoperative. Both arguments are premised on the omission of Automatic Gain Control circuitry (AGC) from the specification. On appeal, Bio-Rad identifies specific testimony establishing that the disclosed invention is fully operative and that at the time the patent application was filed, the AGC was not considered a necessary part of the invention or of the best mode for practicing the claimed invention. Nicolet's arguments on these issues amount to little more than bald assertions by counsel. As appellant, Nicolet has not satisfied this court that reasonable jurors could not find that the best mode requirement was satisfied or that the claimed invention was operative.

Nicolet also asserts that the claims are vague and indefinite. Here again the argument amounts to little more than a statement by counsel supported by some evidence in the record that the language "scanning said specimen" does not have a definite, understandable meaning. Bio-Rad presented testimony showing that a person of ordinary skill in the art would understand the phrase "scanning said specimen" to mean that a movable mirror of the interferometer must reciprocate when the phrase is read in light of the specification. Nicolet has not presented a basis for overturning the jury's presumed resolution of this issue in Bio-Rad's favor.

Nicolet contends its MX-ECO machines do not infringe any claims of the '515 patent for two reasons. According to Nicolet, the method claims require light to be directed from a source onto a sample, then reflected from the sample to an interferometer and then onto a detector. Both parties agree that the MX-ECO does not follow this sequence. During operation of the MX-ECO, light moves from a source into an interferometer, then onto the sample, and from the sample to a detector. The issue here is whether reasonable jurors, after reviewing all the evidence, could have interpreted the claims to include the sequence of events followed by the MX-ECO. Envirotech Corp. v. Al George, Inc., 730 F.2d 753, 758, 221 USPQ 473, 477 (Fed. Cir. 1984). The contested language of claim 1 is reproduced below:

1. A method for determining a physical property of a specimen on the basis of interferometric optical phase discrimination comprising,

in sequence,

directing analytical radiation upon said specimen,along a course such that there is developed an optical phase difference between radiation emanating from a first radiation-reflective surface of said specimen, second radiation reflective surface of said specimen cyclically varying at a predetermined time rate the length of the path of travel of said radiation emanating from a given one of the pair consisting of said first radiation-reflective surface of said specimen, and said second radiation-reflective surface of said specimen with respect to the length of the path of travel of the radiation emanating from the other of said pair, * * *.

At trial, Bio-Rad presented testimony of Mr. Johnston, a patent attorney, on the subject of claim interpretation. The witness testified that to clearly require performance of one step after another in the order listed, he would add the words "and thereafter" or similar language between the listed steps. Certainly such language in claim 1 would remove any ambiguity. Because the claim could have been more clearly written to require a specific sequence, but was not, there is room for reasonable people to interpret the claim as urged by Bio-Rad. We emphasize that our task is not to interpret the claims as though no trial had occurred. Both parties submitted testimony in support of their interpretation before the jury. Bio-Rad's interpretation prevailed and was not overturned by the trial judge. On appeal, we consider only whether reasonable jurors could have interpreted the claim in the manner presumed. Envirotech Corp. v. Al George, Inc., 730 F.2d at 758, 221 USPQ at 477. We conclude that reasonable jurors could conclude the claims simply require light to travel from a source to a detector in a manner that develops an optical phase difference.

Nicolet's second theory for overturning the judgment of infringement is that the '515 patent requires determining a sample's thickness by measuring time directly from an interferogram. In contrast, the MX-ECO measures thickness of a sample by measuring the distance traveled by the reciprocating mirror between occurrences of constructive interference.

The jury presumably concluded that one of ordinary skill in the art would understand that the MX-ECO's mirror time of travel is functionally equivalent to the mirror distance traveled. The patent specification uses "time" to encompass alternative approaches for determining the distance traveled by the mirror between the occurrence of a sideburst and a reference point. When the patent claims speak of "measuring the elapsed time" they also refer to the measurement of the distance the mirror traveled from a reference point to the position where a sideburst occurs. For instance, Figure 3 of the '515 patent shows both distance of mirror travel as well as time. Both distance and time may be used to determine the distance traveled by the mirror. The jury could also reasonably have drawn the presumed conclusion from other evidence produced by Bio-Rad such as Nicolet's sales brochure which described a Michelson interferometer used in the MX-ECO and stated that the mirror velocity must be constant during the scan. The evidence produced by Bio-Rad shows that although time and distance are different, they are related in this technological setting. Because of the close relationship between time and distance in this case, we cannot conclude that the jury's finding of infringement must be reversed.

Nicolet's argument that the MX-ECO applies mathematical formulas known as fourier transforms to the data of an interferogram does not avoid infringement. Steps or features added to a claimed invention do not necessarily avoid infringement. Radio Steel Mfg. Co. v. MTD Products, Inc., 731 F.2d 840, 221 USPQ 657 (Fed. Cir. 1984).

Nicolet argues that a new trial is required because the form of verdict used in this case was improper. Nicolet contends that the general verdict form permitted the jury to decide questions of law. This court has already held that there is no error inherent in submitting questions of law to a jury so long as the jury is adequately instructed. Railroad Dynamics, Inc. v. A. Stucki Co., 727 F.2d at 1514-15, 220 USPQ at 937-38. Nicolet's argument for a new trial therefore has no merit absent some harmful error in the instructions.

The instructions in this case were jointly submitted by the parties to the court. When the instructions were given to the jury, Nicolet raised no objection to any of them. Ordinarily, a party must object to an instruction in time to give the trial judge an opportunity to correct the alleged error before the jury returns its verdict. If no objection is raised, the instructions cannot be challenged on appeal unless great injustice would result. See American Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1364, 220 USPQ 763, 774 (Fed. Cir. 1984).

Nicolet also argues that the trial judge failed to properly instruct the jury on Nicolet's burden for proving invalidity. The jury was instructed that the patent must be presumed valid and that the burden of proving invalidity rests upon Nicolet. The trial judge also told the jury that " [t]he presumption of validity is rebuttable only by clear and convincing evidence." Nicolet contends that the jury should have been told that the presumption of validity may be weakened or destroyed when prior art more pertinent than that considered by the examiner is presented at trial. Because prior art presented at trial was more pertinent than art before the examiner, Nicolet says the jury should have been instructed that Nicolet only had to carry its burden of proof by a preponderance of the evidence.

Nicolet cannot possibly show that great injustice was caused by the challenged instruction since this court has repeatedly held that the party asserting invalidity must prove facts capable of overcoming the presumption by clear and convincing evidence. The introduction of prior art not considered by the PTO does not change the burden of proof or the requirement that evidence establish the presumption-defeating facts clearly and convincingly. Connell v. Sears, Roebuck & Co., 722 F.2d at 1549, 220 USPQ at 199. Nicolet's challenge to the burden of proof instruction is therefore without merit.

Here, Nicolet says that Bio-Rad abused the discovery process to such an extent that a new trial is required. The argument is not premised on failure to disclose documents but on the production of certain documents during the trial. The documents involved opinions from attorneys on validity questions.

Questions regarding abuse of the discovery process are within the trial judge's province and denial of a motion for new trial based on such abuse will be overturned only if the denial of a new trial is shown to be an abuse of discretion. Edgar v. Finley, 312 F.2d 533 (8th Cir. 1963). The disputed evidence was discovered during trial and the documents were entered as evidence in the case. The record indicates Nicolet had an adequate opportunity to examine witnesses necessary to explain the significance of these documents and to present arguments regarding their importance on the validity issues. Under these circumstances, we cannot say that an abuse of discretion occurred when Nicolet's request for new trial was denied.

Finally, to obtain a new trial, Nicolet argues the damages award is excessive. In attacking the jury award of $3,078,000, Nicolet assumes that lost profits were awarded on 91 MX-ECO instruments sold by Nicolet and that a reasonable royalty of $18,000 per instrument was assessed on each sale of 19 other MX-ECO instruments. Bio-Rad does not challenge this presumed analysis of the jury's award, as it appears to follow Bio-Rad's arguments to the jury on the damages issue. Although we are not required to follow the analysis presumed by the parties, we do so in this case and affirm the jury award.

To obtain lost profits, a patent owner must prove that he would have made the sales but for the infringing activity. The patent owner need not prove causation as a certainty, evidence showing a reasonable probability that Bio-Rad would have made the infringing sales Nicolet made will suffice. Gyromat Corp. v. Champion Spark Plug Co., 735 F.2d 549 at 553 (Fed. Cir. 1984); Lam, Inc. v. Johns-Manville Corp., 718 F.2d 1056, 1065, 219 USPQ 670, 675 (Fed. Cir. 1983). The jury was instructed that Bio-Rad is entitled to lost profits on sales Bio-Rad would have made but for Nicolet's infringement and that several factors should be considered, including:

whether plaintiff had the existing or at least potential manufacturing and marketing capabilities to obtain the infringing sales, whether plaintiff's FTG-12C and defendant's MX-ECO compete with one another, whether plaintiff's and defendant's marketing efforts were aimed at the same geographical markets and categories of customers, and whether there were any particularly close substitutes for the FTG-12C and the MX-ECO.

The instruction on lost profits is not challenged on appeal. We conclude that sufficient evidence was proffered by Bio-Rad to establish a reasonable probability that in the absence of Nicolet's infringement, Bio-Rad would have sold an additional 91 thickness measuring instruments. For instance, Bio-Rad showed demand for the claimed invention by proving Nicolet's sales of infringing instruments and licenses for the production and use of instruments embodying the claimed invention by two other corporations. Bio-Rad is the only company licensed to sell instruments embodying the invention of the '515 patent, with one inconsequential exception. Texas Instruments was licensed to sell five instruments embodying the claimed invention after Texas Instruments had used them for one year. This fact alone does not preclude the award of lost profits by the jury. In addition, Bio-Rad's witnesses testified that Bio-Rad's facilities were adequate or could have been made adequate to meet the additional demand for the claimed invention.

Nicolet argues that Bio-Rad is not entitled to lost profits because Bio-Rad did not establish the absence of acceptable noninfringing substitutes, one of the factors required by Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152, 1156 (6th Cir. 1978), to justify an award of lost profits.^*  We conclude there is sufficient evidence of record for a reasonable jury to find the absence of acceptable substitutes. Bio-Rad's market survey expert testified that in surveying purchasers of the Nicolet device, 70-90% of the interviewees would have purchased Bio-Rad's FTG-12 device had Nicolet not been present in the market. Bio-Rad also points to admissions by Nicolet that Bio-Rad was Nicolet's sole competitor. We are also satisfied that the jury could reasonably conclude from the evidence that there were no other devices which could be deemed acceptable substitutes, especially in view of the fact that Nicolet did not offer any evidence to the contrary. Indeed, it is Nicolet's burden on appeal to show that reasonable jurors could not have concluded that but for Nicolet's infringement, Bio-Rad would have sold 91 additional instruments. Nicolet has failed to carry that burden here.

Nicolet also contends that the amount of lost profits awarded is so large that the jury verdict must be overturned or a new trial ordered. The amount of lost profits awarded cannot be speculative but the amount need not be proven with unerring precision. See Lam, Inc. v. Johns -Manville Corp., 718 F.2d at 1065, 219 USPQ at 675; Livesay Window v. Livesay Industries, Inc., 251 F.2d 469, 472, 116 USPQ 167, 169 (5th Cir. 1958).

To support its contention, Nicolet describes the testimony of Bio-Rad's accounting expert as speculative and without any reasonable basis. Bio-Rad's expert relied on an incremental cost accounting method that resulted in a profit of about $39,650 for each domestic sale and $34,313 for each foreign sale. This calculation included all of Bio-Rad's costs, except for direct selling expenses, labor and materials. We cannot conclude that the size of the award for lost profits was too speculative simply because the jury was guided by an incremental cost accounting method.

Nicolet also argues that the reasonable royalty, presumed to be part of the damages award, is too high. They point out that the industry royalty rate runs from three to ten percent of sales while the royalty awarded by the jury approaches one third of the selling price of the MX-ECO. Though established royalty rates are normally applicable, Hanson v. Alpine Valley Ski Area, Inc., 718 F.2d 1075, 1078, 219 USPQ 679, 682 (Fed. Cir. 1983), they do not necessarily establish a ceiling for the royalty that may be assessed after an infringement trial. Cf. Tektronix, Inc. v. United States, 552 F.2d 343 at 347 n. 5, 213 Ct. Cl. 257, 193 USPQ 385 at 390 n. 5 (1977), cert. den., 439 U.S. 1048, 99 S. Ct. 724, 58 L. Ed. 2d 707 (1978). Appellant has not met its burden of showing that use of a higher rate was inappropriate under the circumstances of this case.

Finally, Nicolet argues it is entitled to a new trial on its antitrust counterclaims because the jury's incorrect analysis of validity and infringement necessarily resulted in their rejection of Nicolet's antitrust claims. In view of our affirmance of the decisions on validity and infringement, Nicolet's argument is without merit.

Nicolet's answer to Bio-Rad's complaint alleged patent misuse because of bad faith threats of patent infringement suits. Nicolet's answer said: "Plaintiff is estopped from enforcing said patent No. 3,319,515 by reason of misuse thereof in the form of bad faith threats of patent infringement suits based thereon." This was the only misuse theory presented to the jury. After the jury verdict was returned, Nicolet argued two different misuse theories to the trial judge in a motion to deny enforcement of the '515 patent. The trial judge denied the motion without comment. On appeal, Nicolet relies on the same two misuse theories that were presented only to the trial judge after the jury rendered its verdict. Nicolet would have this court declare the patent in suit unenforceable.

First, Nicolet argues that Bio-Rad misused its patent in 1969-1970 when Digilab refused to sell unpatented interferometers to Texas Instruments unless Texas Instruments took a license under the '515 patent. Second, Nicolet contends that misuse occurred when Bio-Rad and DuPont agreed to jointly enforce the '515 patent against Nicolet and created a joint litigation fund for that purpose. We do not address the merits of these contentions because Nicolet has waived the right to raise these misuse issues.

Misuse is an affirmative defense that must be pleaded and proved. 35 U.S.C. § 282 (1976); Trio Process Corp. v. L. Goldstein's Sons, Inc., 461 F.2d 66, 74, 174 USPQ 129, 135 (3d Cir. 1972), cert. denied, 409 U.S. 997, 93 S. Ct. 319, 34 L. Ed. 2d 262 (1972). Nicolet and Bio-Rad both asked for a jury trial on all the issues. No issues were reserved for trial to the court only. In fact, in its Joint Pretrial Statement, Nicolet said "Bifurcation or separate trials of any issues is neither desired nor feasible." Because the affirmative defenses argued on appeal were not pleaded in Nicolet's answer or presented to the jury, we hold these defenses were waived and we need not consider them further. See id. at 461 F.2d at 74, 174 USPQ at 135.

Nicolet relies on In re Yarn Processing Patent Validity Litigation, 472 F. Supp. 170, 205 USPQ 758 (S.D. Fla. 1979) (Yarn Processing) for the proposition that misuse must be decided by the court and not the jury. Nicolet's reliance on this case is unfounded. The plaintiff in Yarn Processing sued for infringement occurring after misuse had been purged and demanded a jury trial. The District Court granted a motion for separate trial to the court of the purged issues because they were separable from the infringement issues and their separate adjudication would expedite the litigation. The court concluded that trying the equitable misuse defense before a judge did not violate plaintiff's right to a jury trial. The decision in Yarn Processing, therefore, has no bearing on this case where all issues were to be tried to the jury.

The trial court denied Bio-Rad's request for prejudgment interest before the United States Supreme Court decision in Devex Corp. v. General Motors Corp., 461 U.S. 648, 103 S. Ct. 2058, 76 L. Ed. 2d 211 (1983) was available. In that case, the Supreme Court said that prejudgment interest is compensatory, not punitive, and should be awarded under 35 U.S.C. § 284 absent some justification for withholding such award. Here the district court denied prejudgment interest but did not indicate that there was any reason for denying prejudgment interest. The determination of whether such a reason is present is for the trial court to ascertain in the first instance. We therefore vacate the trial court's denial of prejudgment interest and remand for an award of prejudgment interest or a determination that some specific justification exists for denying prejudgment interest. In all other respects, the judgment is affirmed.

AFFIRMED-IN-PART, VACATED-IN-PART, AND REMANDED.