*1156 PLAGER, Circuit Judge.

This is a patent infringement case involving an improved kaleidoscope 1 as described and claimed in the patent-in-suit, U.S. Patent No. 4,740,046 (the ’046 patent). A jury of the United States District Court for the Eastern District of California found in favor of the accused infringer, WildeWood Creative Products, Inc. (WildeWood), finding that the asserted claims of the ’046 patent were invalid and were not infringed by the accused ILLUSION kaleidoscope. The trial court, in accordance with the jury verdict, entered judgment for defendant on September 8, 1992, Civil Action No. CV-F-91-415 OWW. On January 15, 1993, the court denied plaintiffs’ motion for judgment as a matter of law (JMOL) or, in the alternative, a new trial. Plaintiffs appeal. We affirm-in-part and vacate-in-part.

BACKGROUND

Plaintiff Patrick MacCarthy is the inventor of the liquid kaleidoscope disclosed and claimed in the ’046 patent. Plaintiff Beachcombers is a Florida corporation and the exclusive licensee of MacCarthy under the patent. Kaleidoscopes are well known in the art, the invention of which is attributed to Sir David Brewster, a Scot, who patented his in 1816, and for whom the Brewster Society, an organization of designers, collectors, and lovers of kaleidoscopes is named. Defendant WildeWood is a California corporation that sells a kaleidoscope, known as the ILLUSION, which is the accused device.

MacCarthy’s invention, an improvement in a crowded field, involves changing the object cell of the typical kaleidoscope into one that is “comprised of one or more channels through which liquid can flow as the kaleidoscope is rotated.” (Spec., col. 4, lines 41-44.) The point is illustrated by Figure 1 of the ’046 patent, reproduced below:

[[Image here]]

The barrel of the kaleidoscope is identified by reference numeral 1. At the proximal end of the barrel is an opaque barrier 2 containing an eye-piece 3. Contained within the barrel and fixably attached to it are mirrors 4 and 5. These mirrors are aligned along the longitudinal axis 9 of the barrel, and are displaced at an angle 7 relative to one another.

At the distal end of the barrel is rotatable container 14. Within that container is a chamber which is bounded by the container 14 in conjunction with transparent window 8 *1157 and translucent/frosted window 22. Contained within that chamber is closed loop receptacle 15 which is positioned along axis 9 so that it lies in plane 10. The closed loop receptacle is partially filled with a. liquid, and is configured with a reservoir bulb 20 at corner 19. It is this closed loop receptacle that is the “object cell” in the context of MacCarthy’s invention.

The accused ILLUSION device is the commercial embodiment of U.S. Patent No. 5,029,954 2 , Figure 10 of which is reproduced below:

[[Image here]]

The major difference between this device and the device which is illustrated in Figure 1 of the ’046 patent is the configuration of the liquid-filled tube 46 which corresponds to the closed loop receptacle of Figure 1. As shown, that tube extends through a chamber formed at the distal end of the barrel 11. The closed loop receptacle of Figure 1, by contrast, is wholly contained within the chamber located at the distal end of the barrel.

Plaintiffs brought a patent infringement action against defendant on July 29, 1991 alleging infringement of the ’046 patent by defendant’s making and selling the ILLUSION kaleidoscope. Defendant denied infringement and sought a declaratory judgment that the patent was invalid.

At the close of discovery, defendant Wilde-Wood filed a motion for summary judgment of noninfringement and invalidity of the ’046 patent. The trial court denied the motion on July 17, 1992. The ease then went to trial before a jury. The jury returned 22 special verdicts, finding that ’046 claims 1, 6, 8, 9,11, and 13, the claims at issue, were not infringed, and were invalid as anticipated, obvious, and indefinite. 3 Judgment was entered accordingly.

MacCarthy’s post-trial motion for a new trial and a renewed motion for judgment as a matter of law were both denied by the trial court in an opinion and order dated January 25,1993. It is from that order that MacCar-thy appeals to this court.

*1158 DISCUSSION

Claim 1 of the ’046 patent — the only independent claim at issue; claims 6, 8, 9,11, and 13 depend from claim 1 — reads:

I. In a kaleidoscope comprising plural reflecting surfaces joined along one edge, a viewing means proximal to one end of said surfaces and an object cell proximal to the other end of said surfaces, the improvement comprising an object cell comprising at least one channel containing liquid.

The remaining dependent claims read:

6. A kaleidoscope as in [claim] 1 ... wherein the design changes as a function of the angle at which the kaleidoscope is held with respect to the horizontal.

8. A kaleidoscope as in [claim] 1 ... wherein said at least one channel is of such a shape that it allows the flowing liquid to enter and leave the field of view several times during a single revolution of the kaleidoscope.

9. A kaleidoscope as in [claim] 1 ... wherein:

(a) the kaleidoscope further comprises an eye-piece and

(b) said at least one channel enables said liquid to flow toward and away from said eye-piece as the kaleidoscope is rotated.

II. A kaleidoscope as in [claim] 1 ... wherein said liquid contains a solid.

13. A kaleidoscope as in [claim] 1 ... and further comprising a pump for impelling said liquid within said at least one channel.

We first consider plaintiffs’ assertion that the trial court erred in denying the motion for JMOL in relation to the jury’s conclusion of indefiniteness, keeping in mind our standard of review: whether the jury’s express or implied findings of fact are supported by substantial evidence, and whether those findings support the conclusion of indefiniteness. See Read Corp. v. Portec Inc., 970 F.2d 816, 821, 23 USPQ2d 1426, 1431 (Fed.Cir.1992). Plaintiffs’ point is that one skilled in the art would understand that the phrase “object cell” recited in the body of the ’046 claims is broad enough to encompass a liquid-filled tube extending through the end of the barrel, such as that of the accused device, and would not be confused so as to surmise, as defendants assert, that the phrase is limited to the chamber situated at the end of the barrel through which the liquid-filled tube extends.

Plaintiffs are correct. The relevant statute, 35 U.S.C. § 112 ¶ 2 (1988), requires that the claims “particularly [point] out and distinctly [claim] the subject matter which the applicant regards as his invention.” The operative standard for determining whether this requirement has been met is “whether those skilled in the art would understand what is claimed when the claim is read in light of the specification.” Orthokinetics Inc. v. Safety Travel Chairs Inc., 806 F.2d 1565, 1576, 1 USPQ2d 1081, 1088 (Fed.Cir.1986).

The background section of the ’046 specification begins with the conventional definition of the phrase “object cell”: the chamber at the end of the kaleidoscope barrel opposite to the end containing the eyepiece containing movable objects such as pieces of colored glass or the like. 4 However, the remainder of the specification clarifies that MaeCarthy intended the phrase to mean something different in the context of his invention. That is perfectly acceptable. As we have repeatedly said, a patentee can be his own lexicographer provided the patentee’s definition, to the extent it differs from the conventional definition, is clearly set forth in the specification. E.g., Hoganas AB v. Dresser Indus. Inc., 9 F.3d 948, 951, 28 USPQ2d 1936, 1938 (Fed.Cir.1993).

In Figs. 1, 3, 4, 5, and 6, closed-loop, liquid-filled channels are illustrated which extend through or are situated within the chamber conventionally known as the “object cell”. The specification expressly identifies these channels as object cells. 5 Then, at Figs. 7-9, reproduced below, liquid-filled tubes, both linear and nonlinear, are illustrated which extend through what is conventionally known as the “object cell”.

*1159 [[Image here]]

Again, the specification expressly identifies these tubes as object cells. 6 Thus, it is clear from the specification that a meaning other than the conventional one and broad enough to encompass the liquid-filled channels or tubes illustrated in the figures was intended. 7 Accordingly, the district court erred in denying the motion for JMOL in relation to the indefiniteness question.

We next consider plaintiffs’ assertion that the district court erred in denying the motion for JMOL in relation to the conclusion that the ODYLIC was in public use prior to May 27, 1985 8 , and thus constitutes prior art under 35 U.S.C. § 102(b) (1988).

The jury found that the ODYLIC was in public use by virtue of its display to guests at a party hosted by Carolyn Bennett, the designer and developer of the device, on April 12,1985. The jury’s factual findings in support of its conclusion are entitled to a deferential standard of review. E.g., Manville Sales Corp. v. Paramount Sys., Inc., 917 F.2d 544, 549, 16 USPQ2d 1587, 1591 (Fed.Cir.1990); Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 1266, 229 USPQ 805, 808 (Fed.Cir.1986), cert. denied, 479 U.S. 1030, 107 S.Ct. 875, 93 L.Ed.2d 829 (1987). Our task is to determine whether substantial evidence supports these findings and conclusion.

Bennett testified as to the specifics of the party in question. She explained that the *1160 April 12, 1985 date was corroborated by an entry in her date book and discussions with friends; that between 20-30 guests were present; that she personally demonstrated the device to some of the guests for the purpose of getting feedback on the device; and that she made no efforts to conceal the device or keep anything about it secret.

Mary Ann Regan, Bennett’s friend and a guest at the party, also testified at trial. She said that she observed the ODYLIC at Bennett’s party, and remembered noticing its distinctive characteristic — the liquid-filled tube extending through the end of the barrel. She testified that a lot of other guests were present at the time she observed the device; that other guests were looking at the device and picking it up; and that Bennett never asked her to maintain in secrecy any information about the device.

MacCarthy asks us to discount this evidence on the basis it is insufficiently corroborated by documentary evidence, that Bennett was not a disinterested witness, that it is inconsistent with Bennett’s testimony that she did not begin actively commercializing the device until 1986, that it is inconsistent with her deposition testimony, and that Bennett has no independent recollection of the date of the party.

This we may not do. Bennett’s oral testimony was adequately corroborated by physical evidence, i.e., the entry in her date book and the ODYLIC itself, and, if accepted at face value, meets the clear and convincing standard required. See Trans-World Mfg. Corp. v. Al Nyman & Sons, 750 F.2d 1552, 1559, 224 USPQ 259, 262 (Fed.Cir.1984). The jury chose to believe the evidence in spite of these purported deficiencies, which it was entitled to do as these factors all relate to Bennett’s and Regan’s credibility. We have no warrant to override the jury’s decision on the basis of these factors. That would usurp the prescribed function of the jury.

MacCarthy also questions the substantiality of the evidence in light of Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 229 USPQ 805 (Fed.Cir.1986), cert. denied, 479 U.S. 1030, 107 S.Ct. 875, 93 L.Ed.2d 829 (1987), a case in which the court found that a display of a device to friends and colleagues of the inventor was subject to an implied restriction of confidentiality, and thus did not constitute a ‘public use’. But in Moleculon the court found that the inventor at all times retained control over the use of the device as well as over the distribution of information concerning it. Id. at 1266, 229 USPQ at 808. Here, there was evidence upon which the jury could have reasonably concluded that Bennett did not retain control over the use of the device and the future dissemination of information about it — Bennett’s testimony that her purpose in demonstrating the device at her party was to generate discussion and garner feedback, that she never imposed any secrecy or confidentiality obligations, and Regan’s testimony to the effect that she did not believe she was subject to any secrecy or confidentiality restrictions — notwithstanding the closeness and ongoing nature of Bennett’s relationship with her guests. We have no warrant or reason to disturb that finding.

Thus, there is sufficient evidence to sustain the conclusion that the ODYLIC was in public use prior to May 27,1985. The trial court correctly denied the motion for JMOL on the issue.

MacCarthy argues that the trial court erred in denying its motion for JMOL on the anticipation and obviousness questions. We begin with the language of the claims. The first step involves the proper interpretation of the claims. The second step involves determining whether the limitations of the claims as properly interpreted are met by the prior art, and in particular by Bennett’s ODYLIC device.

a. Claim 1

The critical limitation of claim 1 is the requirement that the object cell be configured as one or more liquid-containing channels. Plaintiffs argue that the liquid-filled tube of the ODYLIC is not a “channel” as defined by claim 1 because the kaleidoscope barrel of the ODYLIC is attached to a fixed stand, and the liquid filled tube with the objects in it is in a horizontal orientation and *1161 is rotated around its own axis, the result of which is that liquid does not traverse the length of the tube.

We do not agree. Fig. 7 and the accompanying passage in the specification illustrate and describe an embodiment of the invention identical in all material respects to the ODYLIC. 9 As so understood, claim 1, being the broadest claim in the patent, covers this embodiment, and thus does not support the distinction being urged by plaintiffs. See SmithKline Diagnostics Inc. v. Helena Lab. Corp., 859 F.2d 878, 883, 8 USPQ2d 1468, 1473 (Fed.Cir.1988). The trial court was correct in denying plaintiffs’ motion for JMOL on the question of the novelty of this claim in relation to the ODYLIC.

b. Claim 6

The additional limitation introduced by claim 6 imposes the requirement that the design changes as a function of the angle at which the kaleidoscope is held with respect to the horizontal. That limitation is plainly met by the ODYLIC. The object cell of that device contains various solid objects, such as marbles and pieces of colored glass, which are movable and immersed in a viscous liquid. As the angle at which the kaleidoscope barrel is held with respect to the horizontal is changed, the solid objects in the object cell shift position, causing the pattern exhibited by the device to change.

The fact that the barrel of the ODYLIC is fixed to a stand while the device is in operation is of no moment. The limitation does not impose any requirement that the kaleidoscope barrel be tiltable relative to the horizontal. It relates solely to a characteristic of the object cell, i.e., that it be configured such that the design exhibited by the device changes if the barrel is' tilted relative to the horizontal. The object cell of the ODYLIC is so configured.

Thus, the jury could have reasonably concluded that claim 6, as well, is anticipated by the ODYLIC. The trial court did not err in denying Beachcombers’ motion for JMOL on this question.

c. Claim 8

Claim 8 adds the additional limitation that the channel or channels comprising the object cell be configured such that the liquid enters and leaves the field of view several times dining a single revolution of the kaleidoscope.

The only plausible interpretation of this claim language consistent with the present record is that the limitation imposes the requirement that the channel be only partially filled with liquid. That interpretation is directly supported by the body of the patent, a recognized tool of claim construction. See ZMI Corp. v. Cardiac Resuscitator Corp., 844 F.2d 1576, 1580, 6 USPQ2d 1557, 1560 (Fed.Cir.1988). Figs. 3 and 4, reproduced below, illustrate embodiments of the invention in which the object cell consists of one or more channels.

*1162 [[Image here]]

The channels axe shown as being only about half filled with liquid. These are the only embodiments identified in the specification in which liquid is characterized as entering and exiting the field of view several times in a single revolution. 10

There is no evidence that the ODYLIC, or any of the other prior art references, discloses or suggests partially filling the channel with liquid. The most pertinent of these is Bush, Reissue No. 5,649. However, the object cell disclosed in that reference is a conventional one and thus entirely within the field of view of the eyepiece. Accordingly, the reference does not remotely suggest configuring the object cell to cause liquid to enter and exit the field of view several times during a single rotation. We conclude that the trial court erred in denying plaintiffs’ motion for JMOL on the anticipation and obviousness questions in relation to this claim.

d. Claim 9

The additional limitation introduced by claim 9 imposes the requirement that the object cell be configured such that the liquid is allowed to flow toward and away from the eyepiece as the kaleidoscope is rotated. The jury, in concluding that this claim is anticipated by the ODYLIC, must have interpreted this limitation as embracing a channel which only extends widthwise towards the eyepiece, and does not longitudinally do so. However, since any channel extends to some degree widthwise or lengthwise towards the eyepiece, this interpretation suffers from an infirmity — it would result in claim 9 having the same scope as claim 1, and thus would render claim 9 superfluous. Accordingly, it is presumptively unreasonable. See United States v. Telectronics, Inc., 857 F.2d 778, 783-84, 8 USPQ2d 1217, 1221 (Fed.Cir.1988), cert. denied, 490 U.S. 1046, 109 S.Ct. 1954, 104 L.Ed.2d 423 (1989) (citing Tandon Corp. v. United States Int’l Trade Comm’n, 831 F.2d 1017, 1023, 4 USPQ2d 1283, 1288 (Fed.Cir.1987)).

A more reasonable interpretation, and one that gives significance to the distinction between claims 1 and 9, is that it imposes the requirement that the channel extend longitudinally in some degree towards the eyepiece. Since there is no evidence that the prior art, including the ODYLIC, discloses or suggests that interpretation, we conclude the trial court erred in denying plaintiffs’ motion for JMOL on the anticipation and obviousness questions in relation to this claim.

e. Claim 11

Claim 11 imposes the additional requirement that the liquid in the channel contain a solid. That limitation is plainly met by the object cell of the ODYLIC, which contains a liquid in which solid objects such as marbles and pieces of glass are immersed. Thus, the trial court properly denied plaintiffs’ motion for JMOL on the question of the *1163 novelty of this claim in relation to the ODYLIC.

f. Claim IS

Claim 13 imposes the additional requirement that the kaleidoscope be configured with a pump for impelling liquid within the channel. There is no evidence that the prior art discloses or suggests this limitation. Thus, the court erred in denying the motion for JMOL on the anticipation and obviousness questions in relation to this claim.

We next consider the question of infringement of the patent. We need consider this question only in relation to claims 8, 9, and 13, since we agree with the trial court that the other claims (claims 1, 6, and 11) are invalid for lack of novelty.

As with the validity question, a determination of patent infringement is a two-step analysis. E.g., Lemelson v. General Mills Inc., 968 F.2d 1202, 1206, 23 USPQ2d 1284, 1287 (Fed.Cir.1992), cert. denied, — U.S. -, 113 S.Ct. 976, 122 L.Ed.2d 131 (1993). The first step involves interpreting the claims to determine their proper scope and meaning. The second step involves determining whether an accused device or process is within the scope of the properly interpreted claims. Id.

We have already interpreted the claims for purposes of assessing their validity. The same interpretation of course applies to the infringement analysis. See Semned, Inc. v. Richard-Allen Medical, Indus., 888 F.2d 815, 828 n. 7, 12 USPQ2d 1508, 1511 n. 7 (Fed.Cir.1989) (citing W.L. Gore & Assoc. v. Garlock, Inc., 842 F.2d 1275, 1279, 6 USPQ2d 1277, 1280 (Fed.Cir.1988)).

The object cell of the accused ILLUSION device is completely filled with a volume of liquid in which solid objects are immersed. Moreover, the object cell does not extend longitudinally at all towards the eyepiece. Nor is the device configured with a pump. Consequently, the device does not infringe claims 8, 9, and 13. The trial court did not err in denying plaintiffs motion for JMOL on the infringement question in relation to these claims.

SUMMARY AND CONCLUSION

1. The trial court erred in denying plaintiffs’ motion for JMOL on the indefiniteness question.

2. The trial court was correct in denying plaintiffs’ motion for JMOL on the question of the status of the ODYLIC as prior art.

3. The trial court was correct in denying the motion for JMOL on the question of the novelty of claims 1, 6, and 11 in relation to the ODYLIC.

4. The trial court erred in denying plaintiffs’ motion for JMOL on the anticipation and obviousness questions in relation to claims 8, 9, and 13, but the court was correct in denying plaintiffs’ motion for JMOL on the infringement question in relation to these claims.

In view of our disposition of these issues, we do not reach the other issues raised by the parties. Nor do we find it necessary to remand the case for a new trial. Accordingly, the trial judge is

AFFIRMED IN PART and VACATED IN PART

COSTS

Each party to bear its own costs.

. Webster’s Third New International Dictionary (1966) defines a kaleidoscope as “an instrument that contains loose fragments of colored glass confined between two flat plates and two plane mirrors placed at an angle of 60° so that changes of position exhibit its contents in an endless variety of symmetrical varicolored forms."

. That patent issued on July 9, 1991 to Mark Eilrich and Cozy Baker. Cozy Baker is the designer of the ILLUSION.

. In those verdicts, the jury found that the ILLUSION avoided infringement,'either literally or by equivalence, of the claims at issue, that a prototype known as the ODYLIC was in public use prior to May 27, 1985, that the claims at issue lack novelty in view of the ODYLIC, that claims 1 and .11 lack novelty in view of Bush, Reissue No. 5,649, that Bush does not anticipate the remaining claims at issue, that all the claims at issue are obvious in view of the prior art, and that the claims are indefinite.

. Col. 1, 11. 46-51.

. Col. 8, 11. 67-68, and col. 9, line 1.

. Col. 9, 11. 10-12, 35-41.

. The district court is incorrect in its conclusion that "MacCarthy's own testimony was such that a reasonable jury could find that the patent was unclear, imprecise, and indefinite.” Slip op. at 2. The thrust of MacCarthy’s testimony is that one of skill in the art would have concluded, from a reading of the '046 specification, that a meaning different from the conventional one was intended.

.The significance of that date is that it is one year prior to the filing date of the patent-in-suit, and thus is the operative date for purposes of § 102(b).

. The specification states that the embodiment of Fig. 7 can be operated by rotating the object cell relative to the kaleidoscope barrel. Col. 9, 11. 28-30.

. Col. 7, 11. 5-9, 11. 53-58, and 11. 64-68.