eoncurring-in-part and dissenting-in-part.
I join the court’s opinion as it pertains to the validity and infringement of the ’308 patent, and agree that the jury’s verdict on fraud cannot stand. I join Judge Bryson’s opinion sustaining the jury verdict on M3’s antitrust counterclaim and remanding. My views on the validity of the ’056 patent follow.
By special interrogatory, a jury found each of the disputed claims of the ’056 patent invalid because the claimed invention was on sale in the United States more than one year before July 30, 1986, the filing date of the ’056 patent’s parent application. M3 Systems presented the jury with two reasons why the invention may be invalid for violation of the on sale bar: a transfer from Radiplast to Pharmaseal of 250 needles in June 1985 and an offer from Radiplast to Dr. Ronald Phelps in November 1984. We may affirm the invalidity verdict on either basis. See, e.g., Ethicon Endo-Surgery, Inc. v. United States Surgical Corp., 93 F.3d 1572, 1582, 40 USPQ2d 1019, 1027 (Fed.Cir.1996). Because I believe that the jury had substantial evidence that Radiplast placed the invention claimed in the ’056 patent on sale in November 1984,1 would sustain the jury’s verdict of invalidity.
Discussion
An inventor who places his invention “in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States” loses his right to patent the invention. 35 U.S.C. § 102(b) (1994). A determination that a product was placed on sale under section 102(b) is a question of law, based on underlying facts. See, e.g., KeyStone Retaining Wall Sys. Inc. v. Westrock, Inc., 997 F.2d 1444, 1451, 27 USPQ2d 1297, 1303 (Fed.Cir.1993). While we review the trial court’s ultimate determination of a section 102(b) bar de novo, see, e.g., Ferag AG v. Quipp, Inc., 45 F.3d 1562, 1566, 33 USPQ2d 1512, 1515 (Fed.Cir.1995); U.S. Environmental Products Inc. v. Westall, 911 F.2d 713, 715, 15 USPQ2d 1898, 1900 (Fed.Cir.1990), in considering its denial of Bard’s motion for judgment as a matter of law, we review the jury’s verdict, as did the trial court, for substantial evidence. See, e.g., Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 619, 225 USPQ 634, 636 (Fed.Cir.1985); Railroad Dynamics, Inc. v. A. Stucki Co., 727 F.2d 1506, 1513, 220 USPQ 929, 936 (Fed.Cir.1984). “‘Substantial’ evidence is such relevant evidence from the record taken as a whole as might be accepted by a reasonable mind as adequate to support the finding under review.” Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 893, 221 USPQ 669, 673 (Fed.Cir.1984).
We are guided in our review of the legal conclusion by principles underlying the on sale bar: broad and prompt disclosure of inventions to the public; providing opportunity to experiment, improve, and determine the market value of inventions; discouraging inventors from withdrawing inventions that the public has already come to believe are freely available; and discouraging commercialization that expands the patent system’s grant of the right to exclude others. See, e.g. Envirotech Corp. v. Westech Eng’g, Inc., 904 F.2d 1571, 1574, 15 USPQ2d 1230, 1232 (Fed.Cir.1990); King Instrument Corp. v. Otari Corp., 767 F.2d 853, 860, 226 USPQ 402, 407 (Fed.Cir.1985); General Electric Co. v. United States, 228 Ct.Cl. 192, 654 F.2d 55, 61, 211 USPQ 867, 873 (Ct.Cl.1981). Because the ultimate determination of whether an on sale bar exists rests on the totality of the circumstances, that is, on consideration of the unique facts of each transaction or event, no one factor necessarily controls. See, e.g., *1375Ferag, 45 F.3d at 1566, 33 USPQ2d at 1515. Nevertheless, we have held that “[fjoremost among these is the policy of preventing inventors from exploiting the commercial value of their inventions while deferring the beginning of the statutory term. To this end, the inventor is strictly held to the requirement that he file his patent application within one year of any attempt to commercialize the invention.” Ferag, 45 F.3d at 1566, 33 USPQ2d at 1515 (internal citation omitted). The inventor is entitled to the full benefit of the patent regime; the public is entitled to full, timely disclosure of the protected invention.
We are likewise guided in our review by the principle that we must presume facts necessary to support the jury verdict. See, e.g., Perkin-Elmer, 732 F.2d at 893, 221 USPQ at 673; Railroad Dynamics, 727 F.2d at 1516, 220 USPQ at 939. Given the on sale bar verdict, we assume the jury found that Radiplast made a definite offer to sell certain subject matter and that this subject matter “fully anticipated the claimed invention or would have rendered the claimed invention obvious by its addition to the prior art.” UMC Elec. Co. v. United States, 816 F.2d 647, 656, 2 USPQ2d 1465, 1472 (Fed.Cir.1987); see also LaBounty Mfg., Inc. v. United States Int’l Trade Comm’n, 958 F.2d 1066, 1071, 22 USPQ2d 1025, 1028 (Fed.Cir.1992). Thus, on review we must affirm the verdict of invalidity of the ’056 patent if these factual findings are supported by substantial evidence, and within the context of the various policies underlying the on sale bar, the totality of circumstances supports the ultimate legal conclusion.
I. Offer for Sale
On September 25, 1984, Ronald Phelps, an Alabama medical doctor, sent Radiplast AB a letter that stated: “I am interested in learning more about the new device for percutaneous needle biopsy pictured on the enclosed brochure. I would appreciate it if you would send me all the information you have pertaining to the instrument. Also, please include a price list. Thank you.” Phelps included with this letter a brochure entitled “Radi-biospy device, a new device for percutaneous needle biopsy.” This brochure described previously existing technology and then stated:
A new device has been constructed in order to improve this biopsy method. With the aid of this instrument the biopsy procedure can be carried out with one hand, and as the movements of the obturator and cannula are automatized, better tissue specimens are obtained. All biopsies can be performed by one examiner under dynamic ultrasonic control, or under fi[uo]roscopy.
The new device consists of a spring-trigger system for firing the two different parts of the needle — the cannula and the obturator.
It is constructed of alloyed brass and, like the pressure rod, can be autoclaved.
See special instructions before using.
Manufactured by ... RADIPLAST AB....
By way of its managing director, Thomas Engstrom, Radiplast, replied as follows to Phelps’ letter on November 12,1984:
We thank you for your letter of [S]ept. 25[,] 1984 and for your interest in our BIOPSY DEVICE. I am truly sorry for my late reply.
Our generation No. 2 of the device will we, together with our new biopsy needles suitable for the device, start marketing in USA beginning of — 85, at the moment we do not know through which company.
If you do not want to wait until we have our representation in USA arranged, you can always [sic] order the device directly from us.
Our price for the device is SEK 9.900,- and for the needles SEK 75/ea.
The device is reusable and can be auto-claved. Very little service has to be done on the device due to reliable design. The needles are disposable and are designed to suit the device.
I am enclosing leaflet and article.
*1376I am looking forward to hearing from you.
(Emphasis added).
The Radiplast brochure that Phelps sent to Radiplast describes a device that can be operated with one hand, by one operator, leaving the physician’s other hand free to operate the ultrasound or fluoroscopy equipment. The brochure, describes both parts of the needle as automatized by way of a spring-trigger system. It describes the construction materials used to manufacture the device as well as a procedure by which it can be cleaned. In short, the brochure can be understood to describe either a first generation prior art device or the second generation device described in the ’056 patent.
Despite this ambiguity, Engstrom’s reply to Phelps’ letter in November 1984 is far more telling both in what it said and when it said it. His letter explicitly refers to the second generation device and “new biopsy needles suitable for the device.” Since the second generation device requires a needle that moves both forward and rearward, unlike the prior art TruCut needle, Engstrom’s letter is a clear offer for sale of the second generation device and new biopsy needles. With the exception of a reference to marketing efforts being made in the United States and the possibility of sales through a United States distributor thereafter, this letter was written entirely in the present tense.
The letter was also written after a series of correspondence between Radiplast and Hart Enterprises, a United States medical device manufacturer, addressing tooling and manufacturing costs for these new biopsy needles. On September 4, 1984, Engstróm had written: “Enclosed please find ... a drawing on the biopsy needle. The stainless steel parts are not the final ones, there could be changes in length, diam. and the design of the point.” On September 28, 1984, Hart Enterprises responded: “[Ejnclosed are two drawings, one of the Stylet Hub and one of the Cannula Hub for your Radiplast Biopsy Needle. If you approve these concepts we will proceed to make a prototype, and then production of the molds.” Radiplast replied on October 18, 1984: “Biopsy needles: Enclosed please find our order for tooling and engineering. We approve your design of the plastic parts. The dimension from the top surface to center line of both cannula and stylet should be 4.2 mm. Regarding the needles we will probably start with 2.000 — 3.000 units bulk packed.” Less than one month later, Eng-stróm sent Phelps the November 12, 1984, letter.
These facts alone are sufficient support for the jury’s verdict that there was a definite offer for sale of something more than the TruCut prior art or first generation needles. However, to apply the on sale bar, the jury also had to decide whether this offer for sale of new biopsy needles was an offer of the invention claimed in the ’056 patent. We review this second presumed factual finding for substantial evidence, and like the district court on its denial of Bard’s motion for judgment as a matter of law, we also consider whether there may be policy considerations against imposing the on sale bar.
II. Offer of the Claimed Invention
Bard claims that Radiplast’s November 1984 offer to sell second generation devices and new biopsy needles cannot trigger the bar because at that time no operable device had been made, FDA approval had not been obtained, Radiplast had not conducted clinical testing, it had not found a United States distributor, and it had not developed a final needle design. Bard misapprehends the legal significance of each of these. Clinical testing is not required before a sale can bar patent rights. Nor can subsequent clinical testing excuse a prior sale, if what was offered for sale was the claimed invention. Clinical testing is merely one possible policy reason why a particular sale might be excused from the bar. Since Radiplast did not contemplate sales to Engstróm for testing purposes, the possibility of subsequent clinical testing is of no moment. Likewise, FDA approval is not required before a sale can bar patent rights. Even an illegal sale of the claimed invention before the critical date can bar patent rights. Nor is a domestic distrib*1377utor relevant to the on sale bar inquiry; a sale by a foreign distributor, from a foreign countiy to the United States can bar patent rights. See, e.g., In re Caveney, 761 F.2d 671, 676-77, 226 USPQ 1, 4 (Fed.Cir.1985).
The first of Bard’s two remaining arguments — that no operable device had been made — is a feint because manufacture of an operable device is not a prerequisite for application of the on sale bar. See, e.g., Barmag Barmer Maschinenfabrik AG v. Murar ta Machinery, Ltd., 731 F.2d 831, 837, 221 USPQ 561, 565 (Fed.Cir.1984). While operability may or may not be relevant, see, e.g., UMC, 816 F.2d at 656, 2 USPQ2d at 1472 (reduction to practice is not a requirement for application of the on sale bar), manufacture of an operable device alone is not, see, e.g., Continental Plastic Containers v. Owens Brockway Plastic Products, Inc., 141 F.3d 1073, 1078-79, 46 U.S.P.Q.2d 1277, 1281 (Fed.Cir.1998) (declining to extend exception from public use bar under section 102(b) in design patent case). Operability is relevant only to the extent it demonstrates that a claimed element of the invention had not yet been invented, or the inventors did not know they had a workable invention and thus had nothing to offer for sale. See, e.g., Petrolite Corp. v. Baker Hughes, Inc., 96 F.3d 1423, 1427, 40 USPQ2d 1201, 1204 (Fed.Cir.1996) (“[T]he thrust of the on-sale inquiry is whether the inventor thought he had a product which could be and was offered to customers, not whether he could prevail under the technicalities of reduction to practice ....”) (quoting Paragon Podiatry Lab., Inc. v. KLM Lab., Inc., 984 F.2d 1182, 1187 n. 5, 25 USPQ2d 1561, 1570 n. 5 (Fed.Cir.1993)). Bard has not asserted the second circumstance, and as explained below, the alterations made after the offer for sale to Phelps did not address inventive aspects of the ’056 patent’s new biopsy needle.
As support for its remaining contention— that it had not developed the final design of the biopsy needle — Bard points to Eng-strom’s testimony, as managing director of Radiplast, and correspondence between Ra-diplast, American Pharmaseal, (one of Radi-plast’s potential distributors in the United States), and Alan Taylor (president of Hart Enterprises). Each of these letters was sent after the November 1984 offer for sale to Phelps, and each evidences continued testing of and proposed modifications to the second generation device and the new biopsy needles.*
Engstrom testified that American Pharma-seal’s research and development laboratories conducted in-house testing. A technical report produced after this testing says that “testing [was] to insure functionality of the spring loaded activatior, the Biopty™ device, and the needle before releasing them to the field trial.” As a result of its testing, American Pharmaseal recommended: “increasing] the strength of the stylet handle design and adding] the buffing operation to cannula grinding process.” Engstrom testified that this advice was “to, how do you say, make some changes on the plastic parts and also the— what do you call that— well, the, for some plastic parts broke actually, so we put some, a stopper in the second generation device to prevent, if that happened, to prevent the stylet to go further on.” Engstrom testified that on American Pharmaseal’s advice, Radiplast added a “stop” to the second generation device, after the offer to Phelps.
Engstrom also testified that Radiplast conducted field trials in December 1985, from which it learned that “there was a potential risk for this one snapping back and hurt the doctor’s hand,” and “many patients thought the noise of the instrument was very disturbing.” As a result, Radiplast added “an automatic retraction, a spring, actually, which took this handle back,” and “some damping things, you know, to reduce the noise of the instrument.” After these field trials, Eng-strom sent a letter to Hart Enterprises on *1378January 15, 1985, which stated: “The needle should be changed according to our phone discussion, which means that the wings of the cannula hub should have the same length. Both should be as long as the shortest wing.” A letter from Hart Enterprises to Engstrom on January 25,1985, enclosed three drawings that show “[t]he cannula and stylet hub dimensions are identical to the drawings and prototype you had previously received, with the exception that the cannula hub wings are now [sym]metrical.”
This evidence suggests that Radiplast modified the second generation device by altering the strength of the stylet handle design, adding a buffing operation to the cannula grinding process, a stopper, automatic retraction via a spring, damping to reduce noise, and equal length symmetrical cannula hub wings as long as the shortest wing. However, Bard cannot avoid the on sale bar merely by showing improvements to the invention after its commercialization. See, e.g., Seal-Flex, Inc. v. Athletic Track and Court Constr., 98 F.3d 1318, 1324, 40 USPQ2d 1450, 1454-55 (Fed.Cir.1996). These changes must be something more than obvious mechanical adjustments; they have to be inventive redesigns that are claimed by the ’056 patent. While some of Radiplast’s changes resulted in different possible embodiments of, or additions to, the new biopsy needle that is claimed by the ’056 patent, none of the changes are claimed in the text of the ’056 patent. Moreover, contrary to Bard’s contentions, its evidence suggests at the very least that Radiplast had “reason to expect” in November 1984, that its needle “would work for its intended purpose upon completion,” Micro Chemical, Inc. v. Great Plains Chemical Co., Inc., 103 F.3d 1538, 1545, 41 USPQ2d 1238, 1244, and that Radiplast had more than a mere conception from which it was working towards development, see UMC, 816 F.2d at 657, 2 USPQ2d at 1472.
Because Bard’s evidence shows nothing beyond unclaimed mechanical adjustments to the needle design claimed in the ’056 patent after the November 1984 offer for sale of new biopsy needles, the jury had substantial evidence in support of its finding that the November 1984 offer for sale generated a statutory bar. See, e.g., Robotic Vision Sys., Inc. v. View Eng’g, Inc., 112 F.3d 1163, 1167, 42 USPQ2d 1619, 1623 (Fed.Cir.1997). A contrary view would attribute to the ’056 patent additional limitations taken from later developed commercial embodiments. ■ Because the claimed invention had been completed, Engstrom’s new biopsy needle design calls for an outcome different from Robotic Vision, 112 F.3d 1163, 42 USPQ2d 1619 (remanded for further fact finding on the completion date of a computer software program), Micro Chemical, 103 F.3d at 1544, 41 USPQ2d at 1243 (only a proposed configuration existed and the invention remained to be completed), and Shatterproof Glass, 758 F.2d at 623, 225 USPQ at 640 (a reasonable jury could have found that “apparatus and method of the claims were not functional”).
III. Policy Considerations
Other than the need for sufficient time to test the new biopsy needle design, which is not a policy consideration summoned by the November 1984 offer, Bard has not argued that there are policy considerations weighing against imposition of the on sale bar. Since the policies that underlie the bar focus on the inventors attempts to exploit the invention, not whether a potential purchaser was made aware of or understood it, discussion of Phelps’ actual knowledge of the details of the invention or the differences between generations of the biopsy gun is irrelevant. See, e.g., Ferag, 45 F.3d at 1568, 33 USPQ2d at 1516 (“We emphasize that this is an objective test, and that at its heart lies the inventor’s attempt to commercialize the' invention_ [T]he measure of the bar is what was offered, not the patentee’s intent.”) In light of the strong policy of preventing exploitation of the commercial value of an invention while deferring commencement of the statutory term, I would affirm the jury’s application of the on sale bar.
Reliance on Engstrom’s trial testimony is inherently less reliable than contemporaneous documentary evidence. Cf. TP Lab., Inc. v. Professional Positioners, Inc., 724 F.2d 965, 972, 220 USPQ 577, 583 (Fed.Cir.1984) (inventor's expressions of “subjective intent ... particularly after institution of litigation, is generally of minimal value").