609 F.2d 763
204 U.S.P.Q. 785
CONTROL COMPONENTS, INC., and Richard E. Self, Plaintiffs-Appellees,
VALTEK, INC. and Alpha Engineering Company, Defendants-Appellants.
United States Court of Appeals,
Jan. 9, 1980.
Fox, Edwards & Plumb, H. Ross Workman, Alan Jensen, Salt Lake City, Utah, Pravel, Wilson & Gambrell, B. R. Pravel, Albert B. Kimball, Jr., Houston, Tex., for defendants-appellants.
Arnold, White & Durkee, Thomas O. Arnold, Houston, Tex., Bryan & Bollo, Roland Towler Bryan, Paul L. Bollo, Stamford, Conn., for Control Components and Self.
Appeal from the United States District Court for the Southern District of Texas.
Before FAY, RUBIN and HATCHETT, Circuit Judges.
HATCHETT, Circuit Judge.
Defendants, Valtek, Inc., (Valtek) and Alpha Engineering Co., (Alpha) appeal from a district court judgment entered pursuant to a jury verdict finding that they infringed various claims of a United States patent on fluid control valves owned by Control Components, Inc. (CCI). We affirm.
Valtek and CCI manufacture industrial control valves used to regulate the flow of high pressure fluid in severe service applications. Conventional valves are unable to avoid two major problems associated with high pressure fluid flows: internal damage when the fluid is a vaporizable liquid and intense noise when the fluid is a gas.
Internal valve damage results from cavitation and erosion. Cavitation is a two-stage phenomenon beginning with the formation of vapor bubbles from a sudden reduction of pressure in a liquid flowing at high speed. The second stage of the cavitation process is the collapse or implosion of the vapor bubbles caused by deceleration of the fluid and a corresponding increase in pressure above its vapor point. Unable to exist at the higher pressure, the bubbles collapse with explosive energy, tearing away the interior surface of the valve. Erosion is the physical wearing away of metal as a result of liquid flowing at high velocity.
Severe noise accompanies the sonic and supersonic velocity of the fluid stream in gas applications. The noise can be so intense as to pose health problems for industry workers.
In December, 1966, plaintiff, Richard Self, applied for a patent on an invention designed to minimize the internal damage associated with conventional control valves. After testing the structure with a gas in 1967 and discovering that the valve also moderated aerodynamic noise, Self filed a continuation-in-part of the prior application in May, 1968, now U.S. Patent 3,514,074 (THE PATENT), which was granted May 26, 1970. CCI is the exclusive licensee of the patent.
The major objects of the invention as recited in the patent are to "effect energy losses in high pressure flowing fluids . . ., thus avoiding damage and erosion" and to limit "fluid velocity" while "Quietly effecting energy losses."
The patented valve utilizes a stack of annular disks encircling a chamber in which a movable plug is lodged (see Figure 1). On the face of each disk are a large number of angular turn inducing grooves that produce resistance to fluid flow. A variety of possible configurations for the passageway grooves are illustrated in Figures 2, 3, 3A, 4, and 5. When the disks are stacked one upon the other, "individual passageway grooves" are enclosed between their abutting faces and impart frictional resistance losses to the fluid as it flows through each groove. The position of the center plug may be varied causing fluid flow through more or fewer of the individual grooves.
As described more particularly in the patent claims which are the focus of this litigation, the device is:
1. . . . a rigid structure comprising a stack of members having abutting faces enclosing therebetween a plurality of individual passageway grooves angular between inlet and outlet ends thereof to turn the fluid and provide a substantially longer fluid flow length than the distance between the inlet and the outlet ends thereof, and each passageway groove having an effective long length to diameter ratio cooperating with the angular turn-inducing configuration thereof to impart high frictional resistance losses to fluid flow therethrough; and means for compelling flow of the fluid through said passageways whereby potential energy of the fluid will be dissipated and velocity of the fluid will be controlled.
16. A device according to claim 1, having in combination, a valve housing having a fluid passage of substantial cross-sectional flow area therethrough, said device comprising an annular structure mounted in said housing and across said passage to compel all fluid flowing through said passage to travel therethrough, and a valve plug movable in controlling relation reciprocably within said annular structure.
17. A combination according to claim 16, in which said annular structure comprises a stack of annular disks having said passageways in their faces and extending between and having openings at the inner and outer perimeters of the annular structure and adapted to be selectively opened and closed by movement of said plug in the annular structure.
The jury found that a valve manufactured by Valtek and marketed by Alpha infringed claim 17 of the patent.
The Accused Device
The Valtek valve also consists of a stack of annular disks surrounding an adjustable plug (see Figure 6). On the face of the Valtek disks are concentric grooves (see Figure 7). The grooves are cut successively deeper as they approach the circumference of the disks (see Figure 8). When the disks are stacked together they form a tooth-like passage as depicted in the cross-section diagram of Figure 8. Fluid flows from the center chamber radially outward, in a wave-like motion across the face of the disks. Promotional material emphasizes that the valve solves the cavitation problem because "pressure is reduced gradually across the faces of the disks without sharp pressure drop typical of conventional (valves)."
Case and Issues
Defendants appeal that portion of the judgment finding claim 17 of the patent valid and infringed; they also appeal that portion finding the intentional infringement of CCI's trademark, DRAG.
Defendants argue that: (1) claim 17 of the patent is invalid; (2) the lower court erred in refusing to admit into evidence statements made under oath by Richard Self and patent office findings on obviousness; (3) if valid, claim 17 was not infringed; and (4) no substantial evidence supports the finding of trademark infringement.
DISCUSSION OF THE ISSUES
I. Validity of the Patent.
A. Factors to be Considered.
A patent is invalid if the subject matter sought to be patented would have been "obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains." 35 U.S.C. § 103. "While the ultimate question of patent validity is one of law" the determination of obviousness "lends itself to several basic factual inquiries." Graham v. John Deere Co., 383 U.S. 1, 17, 86 S.Ct. 684, 694, 15 L.Ed.2d 545 (1966); Swofford v. B & W, Inc., 395 F.2d 362 (5th Cir.), Cert. denied, 393 U.S. 935, 89 S.Ct. 296, 21 L.Ed.2d 272 (1968); Robbins Co. v. Dresser Industries, Inc., 554 F.2d 1289 (5th Cir. 1977). Factual questions properly resolved by the jury include "the scope and content of the prior art . . . differences between the prior art and the claims at issue . . . and the level of ordinary skill in the pertinent art . . . ." Graham, 383 U.S. at 17, 86 S.Ct. at 694. Skepticism of experts, commercial success, long felt but unsolved needs, and the failure of others are relevant secondary considerations. United States v. Adams, 383 U.S. 39, 86 S.Ct. 708, 15 L.Ed.2d 572 (1966).
The court instructed the jury on the "primary factors to consider" in supporting a "determination of the issue of obviousness,"1 and submitted to the jury the following question on the issue of validity:
Are the differences between the subject matter patented by the claims of the 074 patent and the prior art such that the subject matter as a whole would have been obvious to one of ordinary skill in the valve art as of December 5, 1966?
A general verdict was entered with respect to claim 17 finding that "the differences would not have been obvious to one of ordinary skill in the valve art." No separate special interrogatories were entered on the factual inquiries underlying the obviousness question.
The court did not err in submitting the above question to the jury. The interrogatory was not so broadly framed as to leave the ultimate determination of obviousness to the jury. As noted, in jury cases the question of patent validity is to be determined by the judge "on the results of factual inquiries" made by the jury. National Filters, Inc. v. Research Products Corp., 384 F.2d 516, 517 (5th Cir. 1967); Cathodic Protection Service v. American Smelting and Refining Company, 594 F.2d 499 (5th Cir. 1979). Under the court's instructions, the jury was to base its general verdict on the primary factors for consideration outlined in the charge. These considerations were appropriate questions of fact for the jury under Graham.
B. Standard of Review.
In light of the charge, we think jury findings on the factual underpinnings were implicit in the general verdict. White v. Mar-Bel, Inc.,509 F.2d 287 (5th Cir. 1975). Under these circumstances we will "presume that the disputed matters of fact have been resolved favorably to the prevailing party in accordance with the trial judge's instructions." Panther Pumps & Equipment Co., Inc. v. Hydrocraft, Inc., 468 F.2d 225, 228 (7th Cir. 1972); See, Mar-Bel at 290-91.2 We presume as to the Nature of the findings, not as to their Correctness. We are limited in the latter analysis by the general proposition that jury findings on disputed matters of fact will be upheld by the reviewing court if substantial evidence exists to support them. Kiva Corporation v. Baker Oil Tools, Inc., 412 F.2d 546 (5th Cir. 1969); Boeing Co. v. Shipman, 411 F.2d 365 (5th Cir. 1969).
As this court noted in Swofford, however, the validity issue involves mixed questions of fact and law. Preliminary factual determinations are made on the scope and content of the prior art and on the differences between the prior art and the claims at issue. The trial judge then determines whether the improvement would have been obvious at the time of the invention to a person having ordinary skill in the art. This conclusion of law requires the application of correct legal criteria to the factual determinations made by the jury. Armour & Co. v. Wilson & Co., 274 F.2d 143 (7th Cir. 1960). The legal conclusion is "fully reviewable by the appellate court." Monroe Auto Equipment Co. v. Heckethorn Mfg. & Supply Co., 332 F.2d 406, 411 (6th Cir. 1964); Swofford at 368. Any difficulty, however, in applying legal criteria to findings of fact, is considerably diminished in cases where the jury has made implicit findings on each underlying factual inquiry. See, Kiva Corporation. If findings of fact on the scope of the prior art and the uniqueness of the claim are supported by substantial evidence, a legal conclusion consistent with those findings is likely to follow. See, e. g., Kiva Corporation; Steelcase, Inc. v. Delwood Furniture Co., Inc., 578 F.2d 74 (5th Cir. 1978); Kaspar Wire Works, Inc. v. Leco Engineering & Machine, Inc., 575 F.2d 530 (5th Cir. 1978).
" Full review" in this context amounts to an inquiry whether the judge "correctly applies the law set out in Graham." Swofford at 368; Kaspar at 543. Like the trial court, we are aided in our inquiry by the jury's findings of fact. If supported by substantial evidence, these findings are apt to strengthen the trial court's legal conclusion. This case is no exception. Our independent review of the record discloses competent substantial evidence to support the jury's findings on the factual inquiries underlying the determination of validity.
C. Scope and Content of the Prior Art.
Evidence was presented to show that prior to the invention in 1966, no valve existed which solved both the problems of noise and cavitation, nor had any valve been able to use all fluids (gas, liquid, etc.) without damage. A number of manufacturers and users, including the world's largest control valve manufacturer, had failed to develop a valve which could solve both problems.
D. Differences Between the Prior Art and the Claims at Issue.
Valtek contends that claim 17 is readable on certain prior art references. These include the "Sempell R06" publication and the Binkley, Schlegel, and Willmann patents.
Sempell R06 is a German publication describing the Sempell valve as having a stack of "annular resistance disks," each having "radially ground-in recesses which, when stacked one upon the other, form high-resistance flow channels." The Binkley patent describes a valve for regulating high velocity discharge of water at dams. To control velocity through the valve, Binkley uses an annular structure that surrounds a valve plug which is actuated in a piston-like fashion to permit fluid to enter into numerous passageways created by the annular structure. The "passageways" are designed to decrease fluid velocity. According to Valtek, Binkley demonstrates that it has long been known that "angular turn-inducing configurations" would "impart high frictional resistance losses to fluid flow." (Claim 17 of the patent.) The Schlegel device utilizes a set of disks, each having concentric "projections" over which fluid flows. Fluid is repeatedly deflected at right angles by the projections causing considerable pressure loss. The Willmann device is a valve used for emulsifying milk or other fatty liquids. It includes a stack of annular disks through which the fluid is compelled to flow. Each disk has on its face a number of grooves which are said to correspond with the "passageway grooves of the Self patent."
CCI presented the following evidence to distinguish these prior art references from the claims of the Self patent. The Sempell R06 valve was not designed to control velocity. The purpose of separate disks in the Sempell valve was to facilitate replacement of disks damaged from the uncontrolled velocity, avoiding the necessity of replacing the entire valve. The valve plug was designed to withstand high velocity whereas Self's plug acts only to close off passageways from which low velocity fluid flows. The Sempell valve would easily clog with dirt, was subject to excessive wear, and was withdrawn from the market in 1975. In the Binkley device, fluid flows over and around the structure formed by the element, and not through passageways between abutting faces of annular disks. Binkley is a variable resistor addressed to water noise only. In Binkley, all passageways are simultaneously filled, whereas the Self plug regulates water flow through selected disks. The Schlegel device does not have a stack of disks with abutting faces within the meaning of claim 17. Because it has no control plug within the resistor, it is not a valve. It does not have a plurality of passageways but one inlet and one outlet per pair of disks. It is used only for gases. It is a variable velocity device, not a constant velocity device as the Self patent. The Willmann apparatus was designed to homogenize milk and was used only in liquid applications. It was one of the prior art references considered by the patent examiner prior to granting the Self patent.
E. Secondary Factors.
Evidence was also before the jury showing: that the Self valve was a tremendous commercial success; that competitors were distressed over the widespread success of the Self valve; that experts in the valve art were at first skeptical over the claims of the inventor; that prior to the invention, experts in the art urged against the use of abrupt turns which ultimately proved the solution to the problem; and, that competitors viewed the Self valve as representing an important advance in the art.
In reaching their general verdict, the jury was able to consider extensive and often conflicting evidence of a technical nature. Scientific testimony included that of the inventor, three fluid dynamics experts, three control valve experts, the designer of the Valtek valve, several user experts, and two patent experts. Exhibits included flow demonstrations in the courtroom, approximately 200 documentary exhibits, and approximately 60 physical exhibits. We are satisfied on this record that the jury's factual findings are supported by substantial evidence.
We also conclude that the trial court correctly applied the law as set out in Graham to these factual findings to determine that the improvement embodied in the patent would not have been obvious to one skilled in the art at the time of the invention.
II. Evidentiary Rulings.
Defendants contend that the trial court erred in refusing to admit statements made under oath to the patent office by Richard Self, and written findings entered by the patent examiner. We disagree. The trial court found that the statements of Mr. Self were not in conflict with his testimony and were therefore not available to impeach him. The ruling was within the trial court's discretion. United States v. Hale, 422 U.S. 171, 95 S.Ct. 2133, 45 L.Ed.2d 99 (1975). The examiner's report was a non-final agency action which included the examiner's opinion on the ultimate issue of validity. The trial court did not abuse its discretion in excluding the report on the basis of possible prejudice. Perel v. Vanderford, 547 F.2d 278 (5th Cir. 1977); Fed.R.Evid. 403.
III. Patent Infringement.
Defendants argue that as a matter of law claim 17 is not infringed. First, it is argued that claim 17 is a dependent claim which must be construed to include all the limitations of claim 1. 35 U.S.C. § 112. The language " said passageways" in claim 17 must, therefore, be limited to the meaning of "individual passageway grooves" as described in claim 1. Because the Valtek disks do not have "passageway grooves" but concentric teeth, there can be no infringement. Next, Valtek argues that the file history of the patent and the prior art require a narrow construction of the phrase "individual passageway grooves." Because to uphold its validity claim 17 had to be narrowly construed to avoid the prior art, it cannot now be expanded by the doctrine of equivalents to recapture what was previously disclaimed. If the "individual passageway grooves" are read so broadly as to include the Valtek disks, claim 17 is invalid as readable on the prior art Schlegel disks. See, e. g., Sterner Lighting, Inc. v. Allied Electric Supply, Inc., 431 F.2d 539 (5th Cir. 1970). Finally, it is argued that changes made in the patent file wrapper support the limited construction. The passageway grooves language replaced the phrase "streams of fluid" which was readable on the Wilson prior art reference. The new language was chosen to avoid conflict with prior art closely corresponding to Valtek's valve. It cannot now form the basis for a finding of infringement against Valtek. See, e. g., Nationwide Chemical Corp. v. Wright, 584 F.2d 714 (5th Cir. 1978).
These arguments do not address the ultimate factual question. Agreeing that claim 17 is a dependent claim and that it must be construed with all the limitations upon the phrase "individual passageway grooves" as argued above, the question remains as to the proper meaning and significance of the phrase. That is a factual issue to be determined by the jury with reference to the specifications in the patent, the disclosures of the prior art, and the testimony of experts. National Filters, Inc. v. Research Products Corp., 384 F.2d 516 (5th Cir. 1967); Cameron Iron Works v. Stekoll, 242 F.2d 17 (5th Cir. 1957); Graver Tank and Manufacturing Co. v. Linde Air Products Co., 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097 (1950).
After considering the extrinsic evidence, the jury resolved the factual dispute over the meaning of the phrase against the defendants. We believe the judge properly denied the defendants' motion for judgment n. o. v. on this issue. Extrinsic evidence in support of the jury's findings was "of such quality and weight that reasonable and fair-minded men" might reach different conclusions. Boeing Co. v. Shipman, 411 F.2d 365, 374 (5th Cir. 1969). The jury verdict was properly left undisturbed.
IV. Trademark Infringement.
A trademark is infringed if use of the allegedly infringing mark is likely to cause confusion or mistake, or to deceive purchasers or users as to the source, endorsement, affiliation or sponsorship of the product. 15 U.S.C. § 1114; Roto-Rooter Corp. v. O'Neal, 513 F.2d 44 (5th Cir. 1975). " Proof of actual confusion is not necessary likelihood of confusion is the appropriate inquiry." Id., at 45; Continental Motors Corp. v. Continental Aviation Corp., 375 F.2d 857 (5th Cir. 1967). Factors to be considered in determining likelihood of confusion include:
. . . the type of trademark at issue; similarity of design; similarity of product; identity of retail outlet and purchasers; identity of advertising media utilized; defendant's intent; and actual confusion. (Id., at 45.)
At the close of the evidence, the question of trademark infringement was propounded to the jury with appropriate instructions by the court.3 The jury responded with a written verdict stating, "Yes, defendant's use of 'DRAGON TOOTH' is likely to cause confusion, or mistake or to deceive." The jury also found that "the trademark infringement was intentional." The trial court concluded that "there was substantial evidence to support the jury verdict." We agree and find no basis for disturbing the jury verdict or the judge's order denying judgment n. o. v. on this point.
Accordingly, the final order and judgment of the trial court is
Appendix to follow.APPENDIX
NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLENOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLENOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLENOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE
ALVIN B. RUBIN, Circuit Judge, concurring in part and dissenting in part:
In discussing the validity of the patent, the majority opinion demonstrates complete understanding of the issues. Because I think the result my brethren reach is correct, I concur. However, while the approach they take pursues a line indicated by prior jurisprudence, I think it takes us a step further into the Serbonian bog that threatens to engulf patent litigation. A different method of trial appears to me to be the only firm ground for traversing the terrain of obviousness in a manner consistent with the Supreme Court's determination in Graham v. John Deere Co., 1966, 383 U.S. 1, 17, 86 S.Ct. 684, 694, 15 L.Ed.2d 545, 556, that "the ultimate question of patent validity is one of law."
Patent validity and nonobviousness are not separate questions; validity embraces nonobviousness for it can be established only by proving that indispensable ingredient. In Swofford v. B & W, Inc., 5 Cir. 1968, 395 F.2d 362, Cert. denied, 393 U.S. 935, 89 S.Ct. 296, 21 L.Ed.2d 272, Judge Wisdom, implementing Graham for a panel of this court, attempted to provide a path to follow in determining the respective roles of judge and jury in deciding the obviousness issue. Noting the inconsistent trails we had previously followed, Swofford determined that obviousness is itself a question of law for the judge to decide. The decision is reached in three steps. First, what was the prior art? a factual question. Second, what, if any, improvement has the patentee made over the prior art? a question of fact that will usually turn on expert testimony. Third, would the improvement have been obvious to one skilled in the art? a question of law, fully reviewable by the appellate court.
The Supreme Court apparently takes the same view of the obviousness issue, See Sakraida v. Ag Pro, Inc., 1976, 425 U.S. 273, 96 S.Ct. 1532, 47 L.Ed.2d 784; and we have since attempted consistently to adhere to the approach in Swofford. See, e. g., Robbins Co. v. Dresser Industries, Inc., 5 Cir. 1977, 554 F.2d 1289, 1290; Gaddis v. Calgon Corp., 5 Cir. 1975, 506 F.2d 880, 884; Garret Corp. v. American Safety Flight Systems, Inc., 5 Cir. 1974, 502 F.2d 9, 14.
In White v. Mar-Bel, Inc., 5 Cir. 1975, 509 F.2d 287, we took another step on the Swofford course, stating:
(I)f the ultimate issue of validity depends on subsidiary fact questions, it is the court's duty to instruct the jury that it should return one verdict if the facts are found one way and a different verdict if the facts are found otherwise. In such event, as in other cases tried to a jury, the reviewing court will Presume that the disputed matters of fact have been resolved favorably to the prevailing party in accordance with the trial judge's instructions. (Emphasis supplied.)
Id. at 290-91. We thus adopted the view of the Seventh Circuit in Panther Pumps & Equipment Co. v. Hydrocraft, Inc., 7 Cir. 1972, 468 F.2d 225, 228, Cert. denied, 411 U.S. 965, 93 S.Ct. 2143, 36 L.Ed.2d 685, an opinion by Judge (now Justice) Stevens.
I respectfully submit that the path laid out in Panther Pumps and followed in Mar-Bel does not and cannot satisfy the Graham mandate. Both opinions rely on the thesis that, because the judge may, in other civil jury cases, instruct the jury to render a general verdict, he may (or should) also do so in patent validity cases. Under that view, the problem in patent cases is merely to instruct the jury adequately about the factual determinations that would point one way or the other; an instruction perhaps more complex than in other jury cases but different only in slight degree.
The verdict reached by the jury on such a charge escapes appellate review save for analysis of the correctness of the jury charge. Graham, I respectfully submit, commands not only how and by whom issues are to be decided at trial but how they are to be reviewed on appeal. The basic issue before us is how we, as an appellate court, review a general jury verdict that concludes merely (and categorically) that the subject matter of the patent was not obvious. My brethren, following Mar-Bel, conclude that there are implicit factual findings in the general verdict that are subject to review based only on the substantial evidence test.
I think meaningful devotion to Graham requires a different course, unique to patent litigation. In this area where decisional responsibility is so clearly divided, the methods of reaching a decision must be more sharply defined; the path to this end is to require jury verdicts on special interrogatories, as permitted by Fed.R.Civ.P. 49(a). See generally Guidry v. Kem Manufacturing Co., 5 Cir. 1979, 598 F.2d 402. If we do not do so, the result will be that validity will
effectively become a question for the jury, not one of law for the judge. If every special verdict on validity leads to implicit findings on non-obviousness, then the trial court cannot review either the verdict or the underlying findings unless the standard for judgment notwithstanding the verdict is used.
Ropski, Constitutional and Procedural Aspects of the Use of Juries in Patent Litigation, 58 J.Pat.Off.Soc'y 609, 685 (1976). Moreover that course has pragmatic difficulties: a relatively minor error in the charge may require a lengthy new trial. Submission on special interrogatories can avert that. See Brown, Federal Verdicts: The Doubt Eliminator, 1968, 44 F.R.D. 338.
Even if I thought Mar-Bel terra firma, I do not think the judge followed its guidance here. The jury was explicitly instructed, "You are to determine the question of obviousness. . . ." This imperative was qualified merely by identifying the underlying factual inquiries of the Graham test as "factors to consider" when the jury was resolving the question. The jury, not the judge, determined obviousness; and not even the less than fully satisfactory Panther Pumps procedure was followed. That case would at least have required an instruction of the on-the-one-hand and on-the-other-hand variety.
Appellants, however, do not challenge the instructions to the jury. Their attack is levelled against the conclusion that the patented invention was nonobvious. Because, in addressing the motion for a judgment notwithstanding the verdict, the district judge necessarily considered and rejected appellants' contention that the invention was obvious as a matter of law, and because, as my brethren ably demonstrate, there is adequate evidence in the record to support the conclusion of nonobviousness under the Graham standard, I concur in the affirmance of the judgment that the patent was valid and infringed. Moreover, in ruling on the motion the district judge indicated that his conclusion on the obviousness of the invention would not differ from the jury verdict. In view of the district judge's correct statement of the Graham standard in his instructions to the jury, and the substantial evidence of facts establishing nonobviousness under that standard, I cannot conclude that the district judge's holding was erroneous as to either facts or law. See Fed.R.Civ.P. 52(a); United States v. United States Gypsum Co., 1948, 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746, 765; W.R.B. Corp. v. Geer, 5 Cir. 1963, 313 F.2d 750, 753, Cert. denied, 1964, 379 U.S. 841, 85 S.Ct. 78, 13 L.Ed.2d 47.
Turning to the trademark issue, I cannot conclude that the evidence shows any possibility of confusion by the customers for these devices. See Roto-Rooter Corp. v. O'Neal, 5 Cir. 1975, 513 F.2d 44. That statement may simply reflect my personal perversity, for the jury apparently found such a possibility, and the able trial judge and two of my colleagues consider the evidence sufficient to warrant its verdict. The jury found that the trademark was intentionally copied, and I not only consider this supported by substantial evidence; I agree. The product was, however, as the majority show, sold only to sophisticated purchasers.1 The valve controls are expensive advance-order items, generally specially engineered for a particular application. The marks themselves DRAG and DRAGONTOOTH are similar only in the use of the syllable "drag". I do not find substantial evidence that the industrial purchasing agents who were the real customers would have been misled even had the imitator chosen the mark Drag II in a deliberate effort to imitate. In this market, there was neither real confusion nor likelihood of confusion.
The evidence overwhelmingly suggests the opposite: past purchasers of these valves were aware and future purchasers likely would be equally aware of the existence of two valves, two manufacturers and two trademarks, distinguishing between them with accuracy. No doubt Valtek's use of the trademark DRAGONTOOTH aided in it alerting the marketplace to the existence of competition to the DRAG valve, as did its sales through the sales representative previously used by Control Components. But such ploys to advise purchasers of the nature and availability of one's product are not the palming off of one's goods as those of a competitor required for trademark infringement. See B. H. Bunn Co. v. AAA Replacement Parts Co., 5 Cir. 1971, 451 F.2d 1254, 1261. The evidence does not suffice to demonstrate that the purchasers of these valves would be led by Valtek's trademark DRAGONTOOTH to purchase the Valtek valve in the erroneous belief that it is actually produced by Control Components. Valtek trades not on the good will attached to its competitor's trademark, but on the market's desire for the particular type of product previously produced only by Control Components. The latter appropriation of the market is the essence of free competition. If it has any limits, they are provided by the patent law. In my opinion the facts in the record and all reasonable inferences from those facts can support only one conclusion: there was no evidence of confusion among the purchasers of these valves As to the source of the Valtek valve. See Boeing Co. v. Shipman, 5 Cir. 1969, 411 F.2d 365. I therefore respectfully dissent from that portion of the court's opinion upholding judgment of the trademark infringement.
The jury was charged as follows:
Your analysis for the determination of the issue of obviousness should be based on all of the prior art taken as a whole. You may consider the inferences or teachings which one of ordinary skill in the art would draw from the prior art before December 5, 1966.
In evaluating whether the subject matter as a whole would have been obvious as of December 1966 to those of ordinary skill, the primary factors to consider are the content of the prior art and the differences or similarities between the 074 patent and the prior art and the level of skill of a person of ordinary skill in the art. You may also, if you believe the evidence warrants it, take into consideration the following secondary factors to the extent that they may be found to cast some light on the circumstances:
(1) Whether the valve described by the 074 patent yielded a new function or result not expected by those of ordinary skill in the art as of December 5, 1966.
(2) Whether the nearest reference had disadvantages which would naturally discourage the search for a new solution to problems of cavitation and noise.
(3) The commercial success of the 074 patented device and whether the commercial success, if any you find, was largely independent of advertising or of government regulation, such that the success was related to the merit of the 074 patented device, or whether the commercial success was merely the result of advertising, sales promotion or government regulation.
(4) The period during which the problem, if any, solved by the 074 patented device, remained unsolved after it was recognized as a problem.
(5) Whether the 074 patent satisfied a long-felt need of those skilled in the art to which the subject matter of the 074 patent pertains.
You are to determine the question of obviousness based on the perception of a person of ordinary skill in the art as of December 5, 1966 not as of the level of knowledge at later dates. You must be cautious not to decide this question in the light of hindsight which is always better than foresight.
This approach is most compelling in cases of this kind, where the patent "is presumed valid (and) the burden of establishing its invalidity rests with the party challenging the patent." Cathodic Protection Service v. American Smelting and Refining Co., 594 F.2d 499 (5th Cir. 1979); 35 U.S.C. § 282
The jury was charged as follows:
Plaintiffs' trademark DRAG is infringed by defendants only if defendants' use of DRAGON TOOTH on their valve creates a likelihood of confusion, mistake or deception as to source, as to endorsement, approval, affiliation or as to sponsorship of the DRAG valves or DRAGON TOOTH valves.
Likelihood of confusion, mistake or deception is determined by evaluating a variety of factors including:
(1) similarity or dissimilarity of the two marks in their entirety, as to appearance, sound, connotation and commercial impression;
(2) similarity of product;
(3) whether the purchasers or users of the valves are same;
(4) the level of sophistication of the valve purchasers and users;
(5) the conditions under which sales of the valves are made including their sales channels;
(6) whether the advertising media utilized are the same;
(7) defendants' intent; and
(8) actual confusion, if any.
Trademark infringement must rest upon a finding that the allegedly infringing mark is likely to confuse the product's "typical buyer". Armstrong Cork Co. v. World Carpets, Inc., 5 Cir. 1979, 597 F.2d 496, 500 n. 5; Kentucky Fried Chicken Corp. v. Diversified Packaging Corp., 5 Cir. 1977, 549 F.2d 368, 389 n. 26. The evidence of the sophistication of the normal purchaser and the care involved in the decisions leading to the purchase of the product is the touchstone for determining the likelihood of confusion. 3 R. Callmann, The Law of Unfair Competition, Trademarks and Monopolies § 81.2 (3d ed. 1969)