These cross appeals stem from a patent infringement action, involving five patents relating to electrical resistors, their structure, method of construction, and design.1 The proceedings also included a claim of unfair competition. The district court, after an extensive trial, made a number of rulings. Since these are cross appeals, we shall refer to plaintiff below in the major action, Dale Electronics, Inc., simply as Dale, the owner of the patents, and defendant, the original defendant and counterclaimant, R.C.L. Electronics, Inc., as R.C.L. Since we do not reach many of the rulings, we shall not list them here but shall indicate our holding under each heading and shall indicate the rulings in the court below which are not reached. At the conclusion of the opinion we shall summarize our holdings.

The electrical resistors involved in these patents consist of an initial resistor wire wound around a ceramic core. Their purpose is to conduct electricity in an electrical circuit while converting part of the electricity to heat. The quality of "resistance" is the ability of the wire to use up electricity. The function of the core, whether a tube or cylinder, is to support the coiled wire, without taking electricity from the wire. The core, in other words, must be an insulator. But, since much heat is associated with resistance, a resistor core also must be able to absorb and dispose of heat through terminals at its ends. In short, a core should not only be a good insulator; it should be a good conductor. Generally, a good conductor of heat is also a good conductor of electricity; and a poor conductor of electricity, i.e., a good insulator, is a poor conductor of heat. The ceramic known as beryllium oxide (BeO), has the distinction of being both a good electrical insulator and a good thermal conductor. Herein lies much of our story.

The district court, 356 F. Supp. 1117, characterized the essence of this patent as "the use of beryllium oxide of at least ninety percent purity as the core of the resistor." This is consistent with Dale's description in its brief: "The single claim of this patent sets forth the BeO core in combination with the conventional resistor components, and requires further that the BeO comprise at least 90 percent by weight of the core to insure that the requisites of certain load life test be met." The court fully acknowledged the fact that the use of BeO as a core in electrical resistors, first devised by Dale's assignor and employee, Hay, was eminently successful, enabling resistors, avoiding concentrated build-up of heat by BeO's high conductivity, to achieve longer life at higher temperatures than traditional ceramic cores and to be produced in smaller size, fitting "the miniaturization revolution". Nevertheless, the court held the patent invalid for obviousness.2

In so doing the court relied, in varying degrees, on several sources of prior art. The first, and the only prior art referred to in the file wrapper, consisted of two patents: the Von Wedel Patent (No. 2,075,876) granted April 6, 1937, which observed that BeO, when mixed with materials such as fluorides, made an admirable insulating coating for twisted wires operating at high temperatures; and the Lindenblad Patent (No. 2,734,344), granted February 14, 1956, which used BeO as rings or sleeves around banks of thermo-couples, because they were both thermally conductive but electrically insulating, thus giving excellent heat transfer from one bank of thermo-couples to another. The court cited these patents as bearing "directly on the issue", but did not rely heavily on them.

The court then referred, without further exegesis, to a number of scientific articles published in the 1950's explaining the use of BeO as a combination insulator and conductor.3 It then went on to what it termed more "crucial" aspects of the prior art which it said lay not in the patents or scientific articles but in the "manner" in which the inventor, Hay, came to use BeO. The court noted that Hay's acquaintance with BeO resulted from a conversation with a salesman of a company supplying ceramic cores to Dale, leading to an order for BeO which was never fulfilled. Next came Hay's visit to a trade show in August of 1961 where Hay ordered sample cores of BeO from National Beryllia Corporation. National Beryllia, the court observed, had, in 1959, advertised in a trade journal the availability in quantity of "Berlox" (pure BeO), "extremely high in thermal conductivity and electrical resistivity", in various shapes and forms for insulators and other uses.4 The court found that in 1961 Hay himself received a graph, prepared by National Beryllia, illustrating a dramatic exponential increase in thermal conductivity for materials comprising a BeO component of at least 90 per cent in weight. The court concluded that "the adaptation by Hay of beryllium oxide was obvious to him and would have been to anyone skilled in the art."

We agree. We begin with observing that the file wrapper history leaves us somewhat puzzled. The thrust of Dale's effort before us, and before the district court has been the uniqueness of using 90 per cent BeO as the core of a resistor. Yet this application required almost five years of siege before the Patent Examiner yielded. He had firmly, up to almost four years after filing, resisted a positive recommendation, saying knowledge of the properties of 90 per cent BeO was old in the art. Finally, the examiner, while rejecting all of the structural claims, stated that a new claim, describing the performance of the core at maximum temperature for a fixed period of time, specifying a maximum temperature and a minimum deviation in resistance value for the resistor, would be acceptable. Whereupon without explanation for the about-face, all the previously rejected structural descriptions, conjoined with the new performance assurance, gained a favorable decision.

We therefore face a Patent Office ruling that approaches the point where "[w]e are at a loss to explain the Examiner's allowance on the basis of such a distinction." Graham v. John Deere Co., 383 U.S. 1, 34, 86 S. Ct. 684, 702, 15 L. Ed. 2d 545 (1966). While the grant of a patent carries with it a presumption of validity, we tend to view the strength of the presumption in this case as minimal. See Hawley Products Co. v. United States Trunk Co., 259 F.2d 69, 72, 74 (1st Cir. 1958). On the other hand, we deal with a decision made by a district court in a field which lends itself to several "factual inquiries": "the scope and content of the prior art", the "difference between the prior art and the claims at issue", and "the level of ordinary skill in the pertinent art". 383 U.S. at 17, 86 S.Ct. at 694. These factual determinations are entitled to weight in our deliberations. See discussion infra.

Dale does not challenge reference to the VonWedel and Lindenblad patents and the scientific articles we have referred to in n. 3. As we have implied, these are more than insubstantial bases for a finding of obviousness, particularly when coupled with the 1959 National Beryllia advertisement.5 Dale concentrates its fire on the propriety of the court's heavy reliance on Hay's conversation with the salesman, his attendance at the trade show, and the National Beryllia graph. Preliminarily, Dale argues that none of these can constitute prior art because they are not or have not been proven to be "publications".6 Dale seems to be seizing on one category of prior art as exclusive. In Graham, supra, 383 U.S. at 15, 86 S. Ct. at 692, the Court refers to the Congressional reports in relation to Sec. 103 defining the condition of obviousness as referring to "the difference between the subject matter sought to be patented and the prior art, meaning what was known before as described in section 102." Section 102 refers to the conditions which foreclose invention. Among them are that the invention was known . . . by others", Sec. 102(a), and that the supposed inventor "did not himself invent the subject matter", Sec. 102(f). Since Sec. 102 is the referent for Sec. 103, we draw the conclusion that if the facts that the whole of an invention was known to others or that none of the invention was created by the patent applicant bar entitlement under Sec. 102, the condition of knowledge by others or the borrowing by the applicant of a sufficient body of lore to make the invention obvious bars entitlement under Sec. 103. General Instruments Corp. v. Hughes Aircraft Co., 399 F.2d 373, 384 (1st Cir. 1968) (notebooks recording experimental work were accepted as evidence of prior art); see Colourpicture Publishers v. Mike Roberts Color Productions, 394 F.2d 431, 434-435 (1st Cir. 1968) (common objects were contemplated as eligible prior art).

Coming to the facts, the whole significance of Hay's conversation with the salesman is muted somewhat by the fact that Dale could not find a copy of Hay's letter to the salesman following the conversation. But enough remains. Hay testified that, had it not been for his conversation with the salesman, in which the latter suggested BeO as a core material with high thermal conductivity, the idea of making a resistor core from BeO "wouldn't come up in my mind". He also testified that the salesman, whose company was already selling Dale a material for a capacitor product, did not mention BeO for that use and that the only product Dale was making to which the salesman's suggestion could be relevant was a resistor.

Hay's subsequent visit to a trade show, where he inquired of National Beryllia about the availability of BeO, does not seem significant. But flowing from this contact, was Hay's receipt of samples of BeO rods in November, 1961, from National Beryllia. Before that time he had received data sheets. Asked if the graph showing the remarkable conductivity of BeO exceeding 90 per cent purity had come with the data sheets, Hay replied, "I don't know if it came in that particular group of mail."7 In any event, it was this graph, which Hay referred to as a "publication" of National Beryllia, which told him not only that the thermal conductivity of BeO was related to its purity but that at around 90 per cent purity there was a sharp increase in such conductivity.8

*****

* * *

*****

* * *

On receipt of the BeO rods, Hay wound resistor wires around them, thus making his first resistors. This took place in mid-November. In December temperature rise tests were run, showing excellent conductivity. In January of 1962 load life tests showed that the resistors proved more stable, with smaller resistance deviations, than Dale's standard resistors made with aluminum cores. On January 26, 1962, Hay signed a patent disclosure which essentially described his concept as that of using BeO as a core. No reference to the required relative purity of BeO was made. On February 15, 1962 Hay signed the patent application which was filed on February 26. It summarized the temperature and load life tests, described his method of preparing his core material "by mixing substantially 95 per cent by weight of beryllium oxide with substantially 5% by weight of earth clay." One of the claims specified that BeO must be approximately 90 per cent by weight of the core.

The record reveals, despite some initial testimony to the contrary, that no tests showing the critical necessity for BeO of approximately 90 per cent purity were conducted prior to the filing of the patent application. Hay testified that the National Beryllia graph would tell one looking for a core material of high thermal conductivity to pick BeO of more than 90 per cent. Three years later, in buttressing efforts in the Patent Office, Hay included, without attribution to National Beryllia, a replica of its graph as an exhibit illustrating BeO's conducting properties, calling particular attention to the sharp changes in the slope of the curve at the point where the concentration of BeO is at least 90 per cent.9

It seems abundantly clear to us that, given the latitude possessed by the district court in resolving the "factual inquiries" required in determining the issue of obviousness, the record supports the court's conclusion. Graham v. John Deere Co., 383 U.S. 1, 5, 17, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966); Nashua Corp. v. RCA Corp., 431 F.2d 220, 222 (1st Cir. 1970); Columbia Broadcasting Sys. v. Sylvania Electronic Prod. Inc., 415 F.2d 719, 728 (1st Cir. 1969); General Instrument Corp. v. Hughes Aircraft Co., 399 F.2d 373, 384 (1st Cir. 1968); Koppers Co. v. Foster Grant Co., 396 F.2d 370, 372 (1st Cir. 1968). The allowance of the patent, after a long and arduous process, on the basis of a description of performance10 of the structure which had steadily been rejected, renders more significant in our eyes any idea sources which were not revealed to the examiner. The existence of widespread literature in the 1950's and early 1960's including advertisements, concerning the increasing feasibility for many uses of the highly conductive and insulative BeO would have suggested to a wider reader than Hay what in fact he learned from the salesman-that BeO had arrived at the point where it might be excellent material for a resistor core. The National Beryllia graph, described constantly by Hay as a publication, provided Hay with the precise information as to purity required in a high performance resistor. Hay struck a rich lode only after all of the technology had led him to the marked spot. The knowledge of BeO's qualities and the new processes that made it more readily available combined in drawing the map. Hay needed only the knowledge of one skilled in the art to come upon the discovery. His advantage was only one of time. That is not enough.

So holding, we do not have occasion to review the court's conclusion that Hay had met the novelty requirement of 35 U.S.C. Sec. 102(f); that '090 contained sufficient instruction to satisfy the requirements of Sec. 112; or that, were the patent valid, R.C.L.'s BeO core resistors infringe '090.11

These patents concern the method ('855) of producing a resistor inbedded in a metal housing ('704). Both relate to an insulative fluid injected under pressure between the resistor core and the housing. Both patents refer to the recommended hardenable insulative material as "materials in the classification of epoxys, phenolics or silicones." To find a particular material which would work, according to Hay, the inventor, one would have to experiment. He also testified that at the time he filed his applications, he knew a specific material that worked very well, Rogers RX 600.12 Indeed, he had failed to make work many other materials in the classes described.

The district court held that "the plaintiff's desire to bring all types of insulating material within the scope of its patent claim without disclosing the one material that worked effectively has resulted in a description too vague for an accurate and effective disclosure of the invention" and ruled the patent invalid under the "best mode" language of Sec. 112.13 We agree.

Dale's major argument is that the nondisclosure of Rogers RX 600 was not intentional. We do not accept the proposition that where an inventor, as here, clearly knows a specific material that will make possible the successful reproduction of the effects claimed by his patent, but does not disclose it, speaking instead in terms of broad categories, he may nevertheless be considered as having described the best mode contemplated by him. The statutory language seems to contemplate precisely this situation. In Indiana General Corp. v. Krystinel Corp., 297 F. Supp. 427 (S.D. N.Y.1969), aff'd, 421 F.2d 1023 (2d Cir. 1970), the court upheld a finding of failure to disclose best mode, but at the same time refused to order attorney's fees, finding no deliberate fraud. Unintentional obtuseness or obfuscation might be a reason not to penalize someone; we do not see it as a reason for granting a seventeen year monopoly.

This holding makes unnecessary any review of the court's holding that these two patents met the nonobviousness requirement of Sec. 103; that '704 met the no-public-sale-for-a-year requirement of Sec. 102(b); that, if '855 and '704 are valid, R.C.L.'s in-house molding process and resulting product infringe; or that R. C.L.'s resistors made by the "stuffing" method do not infringe '704.

Dale challenges the district court denial of its Rule 59(e) motion to amend the judgment to read: "Nothing in this Opinion or Judgment shall be prejudicial to plaintiff's right to obtain and bring suit for infringement of reissue patents emanating from Patent 3,201,855 and U.S. Patent 3,206,704."

This, Dale suggests, was "not typical of the usual motion to amend judgment". This seems to us an understatement, the motion not being aimed at the judgment but at implications which might be drawn from it. Upon the district court's denial, Dale insisted on digging itself in deeper by asking if the denial was "on the merits", to which the court gave an affirmative answer.

It seems to us that the parties are trying to litigate something that as yet does not exist. We do not know whether or not Dale may be able to obtain a reissue patent, or, if it may, on what grounds. The cases cited by R.C.L. to support the district court ruling are appeals from denials of the reapplications or claims of infringement with counterclaims of invalidity of reissue patents. Application of Hafner, 410 F.2d 1403 (Cust. & Pat.App.1969); Application of Nelson, 280 F.2d 172, 47 C.C.Pa. 1031 (1960); Bendix Corp. v. Balax, Inc., 421 F.2d 809 (7th Cir. 1970); Maxant Button & Supply Co. v. Sears Roebuck & Co., 388 F.2d 912 (7th Cir. 1968). They are distinguishable from this case where there is no challenge to a reissue patent nor a claim that a reissue has been improperly denied.

We think it inappropriate that the court's decision be construed as wider than the issues which were before it. We accordingly hold that the court's denial of the motion was within its discretion, Walker v. Bank of America Nat'l Trust & Saving Ass'n, 268 F.2d 16 (9th Cir.), cert. denied, 361 U.S. 903, 80 S. Ct. 211, 4 L. Ed. 2d 158 (1959); Boaz v. Mutual Life Ins. Co. of New York, 146 F.2d 321 (8th Cir. 1944), but has no broader substantive effect.R.C.L.'s Counterclaim

R.C.L. challenges the court's dismissal of its counterclaim seeking an affirmative ruling that the '356 patent be declared invalid. This counterclaim was interposed to Dale's claim of infringement. Before trial, Dale moved to withdraw its claim of infringement of the '356 patent and for dismissal of R.C.L.'s counterclaim alleging invalidity. The court reserved judgment on this motion until after trial at which time it granted the plaintiff's motion and dismissed the counterclaim on the theory that there was no longer a justiciable controversy. R.C.L. cites several cases for the proposition that a counterclaim in a patent suit which seeks a declaration of invalidity of a patent can remain for independent adjudication, notwithstanding a stipulation of noninfringement.

In response, Dale claims that R. C.L. waived the right to proceed on the basis of a letter dated almost a year before trial began and well before Dale moved to drop its claim of infringement. The letter, on another point entirely, does not affect R.C.L.'s rights. Legal authorities support R.C.L.'s position that it was entitled to a decision on the validity of the '356 patent. Hawley Products Co. v. United States Trunk Co., 259 F.2d 69, 75-76 (1st Cir. 1958); Trico Products Corp. v. Anderson Co., 147 F.2d 721, 722 (7th Cir. 1945); Marshall v. RCA Corp., F.Supp., 177 U.S.P.Q. 426, 430 (S.D.Ind.1973); International Min. & Chem. Corp. v. Golding-Keene Co. (W.D.N.Y.1958), 164 F. Supp. 101.

Accordingly, this issue must be remanded, and, if R.C.L. wishes to pursue it, the district court has discretion to consider it on the record made to date or may receive further evidence if it sees fit.

This patent described an ornamental design for a housed resistor of the kind shown in the '855 and '704 patents. Housed resistors do not have much opportunity for originality of basic design. They must cover the core and wire. They require external ribbing to help dissipate heat. They also need a flat surface on which a manufacturer's label can be imprinted. Prior patents, one of which was not disclosed to the Patent Office, had the ribs running along the top of the housing. '884 had the ribs on the sides and the flat surface for the label on top. The court held that "It was not only the prior art, but the dictates of the manufacturing process that made the design obvious." It also held that '884 did not describe "any new, original and ornamental design", as required by Sec. 171. Despite the Patent Office's approval and the testimony of a hostile witness that somehow the virtue of '884 lay in a design consistent with other computer equipment, we cannot bring ourselves to say that the district court overlooked "some exceptional talent beyond the skill of the ordinary designer". G. B. Lewis Co., et al. v. Gould Products, Inc., 436 F.2d 1176, 1178 (2d Cir. 1971). "It seems to us at the most to call only for such routine skills as might be expected of any designer in the field . . .." The design patent shows "some minor variation of the prior art well within the competence 'of an ordinary practitioner of the craft exercising merely the skill of his calling"'. Hawley Products Co. v. United States Trunk Co., 259 F.2d 69, 74 (1st Cir. 1958).

Dale had merchandized its panel housed resistors, using the letters PH. Dale's Vice President disavowed any trademark claim. Dale's products had been used for years with panel housed (or mounted) resistors. One of Dale's customers asked R.C.L. to copy Dale's units. R.C.L. supplied its own resistor, noting thereon "RCL PH-10-150". The customer returned some of these units to Dale. The court held that only the buyer was confused, that any prospective buyer could ascertain the source by looking at the marking, that "PH" was a generic term, meaning "panel housed", that R.C.L. attempted to sell to no one but the customer who asked R.C.L. to manufacture to Dale's specifications. It concluded that Dale had not proven a case of unfair competition. We affirm.

Dale seeks fees and costs in connection with the class action phase of this litigation. The complaint, filed in March of 1971, naming R.C.L. as representative of a class including twelve other alleged infringers. Expensive, courtapproved notices were prepared and published. Finally, the court dismissed the action as to the class. Dale claims that R.C.L. sat back in bad faith and that the court abused its discretion in refusing to approve Dale's request. R.C.L. denies any bad faith alleging that it could care "less if it was led to the courthouse as a lone sheep or as part of a herd". R.C.L. for its part claims an abuse of discretion in the court's refusal to award reasonable attorney's fees and taxing certain costs.

As to Dale's claims for class action related fees and costs and R.C.L.'s claim for fees, we see no basis for concluding that the court abused its discretion. In such a trade-off situation it would be particularly difficult to try to second guess the district court. The taxing of the prevailing party's deposition costs in his favor is also a matter committed to the sound discretion of the trial court. The trial court's exercise of discretion will normally not be interfered with by an appellate court. Emerson v. National Cylinder Gas Co., 251 F.2d 152 (1st Cir. 1958). We readily recognize that daily copy in such a long and complex trial as this may be deemed essential by the parties. Their agreement to share the costs witnesses their similar belief. The court apparently felt that each side had realized benefit from the agreement and decided to allow the burdens to rest where they lay. We cannot say the court abused its discretion in denying reimbursement of some of R.C.L.'s costs.

We therefore summarize our holdings as follows:

-We affirm the judgment of invalidity of the '090 patent for obviousness;

-We affirm the judgment of invalidity of the '855 and '704 patents for failure to describe the best mode of implementation, in violation of 35 U.S.C. Sec. 112;

-We affirm the judgment of invalidity of the '884 patent for obviousness and failure to meet the standard of originality of 35 U.S.C. Sec. 171;

-We affirm the judgment for defendant in plaintiff's claim of unfair competition;

-We affirm the denial of plaintiff's Rule 59 motion;

-We reverse the judgment dismissing defendant's counterclaim and remand for further proceedings;

-We affirm the court's denial of fees and costs requested by plaintiff in connection with the class action; we also affirm the court's denial of counsel fees requested by defendant and its denial of costs of defendant's share of court reporters' fees and deposition.