Opinion
Nos. 116, 117, Dockets 33486, 33535.
Argued October 30, 1969.
Decided January 14, 1970.
James C. Wood, Chicago, Ill. (Richard Whiting, New York City, and William E. Recktenwald and Hofgren, Wegner, Allen, Stellman McCord, Chicago, Ill., on the brief), for plaintiff-appellant.
Harvey E. Bumgardner, Jr., New York City (Ira J. Krakower, Martin J. Cohen, and Arthur, Dry, Kalish, Taylor Wood, New York City, on the brief), for defendant-appellee.
Before MEDINA, MOORE and FEINBERG, Circuit Judges.
The issue involved in this case is the validity of United States Patent No. 3,036,009, issued to Georg Zerbes for the invention of a five-oxide component soft ferrite manifesting a high Q-factor at high frequencies and assigned by him to Indiana General Corporation, plaintiff-appellant. The court below, Judge Tenney sitting without a jury, held the patent invalid. Opinion reported at 297 F. Supp. 427 (S.D.N.Y. 1969). Judge Tenney concluded that the patent was invalid under 35 U.S.C. § 112 in that it included overly broad statements and misrepresented the scope of the inventive advance; under 35 U.S.C. § 102 because of commercial use for more than one year before August 11, 1958, the filing date of the continuation-in-part application which eventually became the patent in suit; under 35 U.S.C. § 112 for claiming a utility not supported by the specification and known by Zerbes at the time the first application for a patent was filed in the United States on November 9, 1954; and under 35 U.S.C. § 103 for obviousness in view of the state of the art as it existed prior to November 27, 1953, the date on which the original application for a patent was filed in West Germany. Judge Tenney also held that Krystinel Corporation's accused products K-401-1, K-411, and K-502 were non-infringing. However, he concluded that Krystinel's K-401, discontinued on November 1, 1963, would have infringed the patent if the patent were held to be valid. He also concluded that Indiana General, its corporate predecessor General Ceramics Corporation, and Zerbes did not engage in fraud before the Patent Office although their concealment of certain facts rendered their conduct less than candid, and that the case was not "exceptional" within the meaning of 35 U.S.C. § 285 to warrant the awarding of attorneys' fees to Krystinel. Although, in affirming the judgment below we conclude that Judge Tenney's findings of fact are not "clearly erroneous" and that his conclusions of law are correct, we find it necessary to discuss only the issue of obviousness.
I. Discussion of Relevant Technical Terms
The invention to which the patent in suit relates is a soft ferrite which allegedly has a high Q-factor in high frequency communication applications. In order to understand the interaction of the patent in suit with the prior art, it is essential that the characteristics of soft ferrites and their relevant magnetic properties be explained.
The patent made only one claim, which reads as follows:
I claim:
A magnetic core body having a high Q-factor in the high frequency range of up to 300 mcs. comprising a sintered nickel-zinc ferrite prepared by firing at about 1100° C. to about 1350° C. a composition consisting essentially of the NiO, ZnO and ferric oxide components set forth in weight percent within the area A — B — C — D of the drawing, said composition comprising in addition 1 to 5 weight percent of manganese oxide based on the total weight percent, and 1/60% by weight of cobalt oxide based on the proportion of nickel oxide in the said composition, the amount of cobalt oxide being within 0.1 to 3.0% by weight of the total composition, the proportion of iron oxide in said batch being at least 50 mol percent of the entire composition.
The triaxial diagram referred to in the claim is set forth below:
A ferrite is a chemical composition containing iron (ferric) oxide and other metallic oxides which is magnetic although not metallic. "Soft" ferrites are temporary magnets; they become magnetized when an electric current or radio waves pass through them and lose most of their magnetism when such passage ceases. "Hard" ferrites are permanent magnets.
The ferrites here involved have an "inverse spinel" structure. Their formula is MO Fe2O3, where M represents any divalent metallic ion. A divalent metallic ion is one in which two electrons have been given up to form a neutral compound. Examples of such ions are cobalt (Co²+), manganese (Mn²+), zinc (Zn²+), and nickel (Ni²+). Each of these ions displays different magnetic properties.
The inverse spinel structure is cubic and crystalline, and the geometrical arrangement of the iron and divalent metallic ions and the oxygen ions within the crystal is responsible for the various magnetic properties of the substance. Generally some of the iron ions form a cube within which other iron ions, the ions of the divalent metals, and oxygen ions are arranged. Within the cubic structure, oxygen ions are closely packed together; and the iron ions, obtained from ferric oxide, and the divalent metallic ions are situated at interstices of lattices formed by the oxygen ions. There are two types of interstices, called A sites and B sites. An inverse spinel ferrite contains twice the number of B sites as A sites. In such a ferrite the iron ions occupy all the A sites and one-half the B sites. Consequently, the number of iron ions occupying the A sites is equal to the number occupying the B sites.
The interactions between the metallic ions of the A and B sites are very numerous, so much so that all other ionic interactions are relatively unimportant. Because the iron ions are arranged in opposite alignments on the A sites and the B sites, they cancel each other out. As a result, the magnetic properties of an inverse spinel ferrite are largely determined by the magnetic properties of the divalent metallic ions — i.e., cobalt, manganese, zinc, nickel, magnesium, vanadium, and copper ions — occupying the remainder of the B sites.
These data are the A B Cs of the art. Those skilled in the art are physicists, chemists and chemical engineers such as most of those who testified at the trial. In due course we shall discuss the investigations of the practitioners of this art and the nature of the progress made in the development of the art that constitute the background against which the issue of obviousness must be adjudicated.
The importance of soft ferrites in high frequency communication applications is derived from their high resistivity. The reason is that this property causes the material to be more efficient by having low eddy-current losses. The result is that less power is wasted, and the time lag in the operation of a given electrical device is decreased. Also important is the existence of high permeability and high quality factor. Permeability, symbolized by the Greek letter Mu (), is a measure of the ability of a given material to concentrate magnetic lines of force. It is the ratio of the flux (lines-of-force) density of the material in a given magnetic field compared with the flux density of air in the same field. Because radio and other electromagnetic waves have a magnetic component as well as an electric component, permeability can also measure the ability of a substance to concentrate such waves. Quality factor (the "Q-factor") is the inverse of loss factor, which is the amount of energy dissipated when magnetic lines of force or electro-magnetic waves pass through it. Thus, a high quality factor indicates a material which loses relatively small amounts of energy. A ferrite with high permeability and high quality factor at high frequencies is efficient and well suited to use as a radio antenna.
See J. Thompson, The Magnetic Properties of Materials 55 (1968); A. Morrish, The Physical Principles of Magnetism 489 (1965); R. Bozorth, Ferromagnetism 244 (1951).
II. The Development of the Patent in Suit
As the ensuing discussion will make all too apparent, the patent in suit had a rough voyage through the Patent Office and was finally issued only after the Commissioner had overruled the Examiner, who had protested again and again that the combination of elements whose properties were already well known and described in the prior art should not override the claim of obviousness. Accordingly, we think it helpful to discuss the course of Indiana General's experimentation with various types of ferrites and the progress of this particular alleged invention over a period of many years in the Patent Office.
The events which led to the development of United States Patent No. 3,036,009 were part of an effort by physicists, chemists and chemical engineers in the United States and Europe to discover the properties of ferrites and make them more efficient. Prior to the development of the material which formed the basis for the patent in suit, General Ceramics Corporation, one of Indiana General's corporate predecessors, marketed ferrites under the trade names of Ferramic J and Ferramic N. The former was put on the market in 1951 and the latter in 1952. Neither was commerically successful, although Ferramic N was better suited to use in radio antennas than Ferramic J. Among the major shortcomings of Ferramic N were its tendencies to become unstable because of changes in temperature, lapse of time (disaccommodation), and shock, as from the bumping of a delivery truck on an uneven road. It also had a relatively low Q-factor at 1 mc., which was then considered a high frequency, although it had a relatively high permeability. Various attempts were made to improve the stability of Ferramic N, such as by artificially aging the antenna rods which contained it. The results of these efforts were not satisfactory.
Appellant Indiana General claims that by the middle of 1953 none of its competitors had developed a ferrite which solved the problems with which Ferramic N was afflicted. Working on improvements in ferrites were such persons as Jacob L. Snoek of the Netherlands, working for Philips Research Laboratories; Godshalk Berge; Robert L. Harvey, working for Radio Corporation of America; and Dr. Ernst Albers-Schoenberg, Director of Research for General Ceramics.
General Ceramics had an agreement for the exchange of technical data with a West German corporation, Steatit-Magnesia Aktiengesellschaft ("Stemag"). The evidence indicates that during the summer of 1953 Dr. Albers-Schoenberg made one of a series of annual visits to Stemag. At that time he spoke to Georg Zerbes and found that the latter had come across a composition which added small quantities of cobalt oxide and manganese oxide to a nickel-zinc ferrite, allegedly resulting in a dramatic improvement in Q-factor at 1 mc. Dr. Albers-Schoenberg reported this development to Salvatore J. Ortepio, then a project engineer for General Ceramics, by letter of August 5, 1953, which stated in pertinent part:
The Stemag people have found, that a very small addition of Cobalt oxide (below 1%) improves greatly the Q-factor and temp. coefficient of antenna body. * * * (This is almost 874-body.) [General Ceramics' code name for Ferramic N].
Keenly interested in this development, Ortepio assigned work on this product to a Dr. Wetzel. Laboratory samples were prepared, and their Q-factors were measured. Finally, in October 1953 the decision was made to produce a commercial-sized batch of 300 kg. Parts of the batch were sent to certain radio manufacturers with a specification sheet dated November 18, 1953, identifying the product as 2017-C with the trade name of Ferramic Q. Its Q-factor was reported to be 470 at 1 mc.
The comparative chemical compositions of Ferramic N and Ferramic Q by weight per cents are as follows:
Ferramic N Ferramic Q
Fe2O3 75.50% 75.14% NiO 9.20% 9.06% ZnO 13.50% 14.05% MnO 1.80% 1.40% CoO ______ 0.88%
While the details will be considered later on in this opinion, it is important to note at this point, and to bear in mind throughout that this combination is critical to the issue of obviousness. It is critical because Ferramic N contains a small amount of manganese oxide (MnO) and to this combination is now added a small amount of cobalt oxide (CoO) which has the effect of greatly improving the Q-factor. This is precisely the teaching of the Crowley Patent (2,736,708, filed June 8, 1951). Whether or not Zerbes knew of the Crowley application when he is said to have found the new composition does not appear, nor is it material on the issue of obviousness, for the question is not what was or was not obvious to a particular person possessed of more or less limited knowledge of the prior art, but what would have been obvious to a hypothetical person who did have knowledge of the prior art.
On November 27, 1953, Stemag and Steatite Research Corp., a subsidiary of General Ceramics, filed an application in West Germany for patent Serial No. St 7388 VIb/80b (the "grandparent application"). The application disclosed a system of five-oxide component ferrites containing Fe2O3, ZnO, NiO, CoO, and MnO. A range was given for these components in terms of a three-component iron-zinc-nickel oxide ferrite with the statement that Q-factor deteriorates when "iron oxide contents [fall] below 66% by weight the named combination of additives."
We do not find it necessary to pursue the controversy between the parties concerning the meaning of this ambiguous statement.
Ferramic Q apparently met with commercial acceptance. For example, Zenith Radio Corp., on December 2, 1953, approved its purchase for use in radio antennas.
On November 9, 1954, United States Patent Application Serial No. 467,828 (the "parent application") was filed by Zerbes with the Patent Office. It disclosed that the addition of cobalt oxide made it possible to obtain ferrites with "high quality values or Q-factors in the high frequency region, 200-300 mcs, for example, which have not hitherto been attained." A triaxial diagram showed the components of the three basic oxides (Fe2O3, ZnO, and NiO), with Fe2O3 ranging from 58% by weight to 80% by weight. The application also spoke of a "preferred range" which encompassed still a smaller area of the triaxial diagram, with the weight percentages of Fe2O3 ranging from 62% to 80%. Moreover, the application stated that the products of the invention, without manganese oxide additions, have a proportion of Fe2O3 which is slightly less than 58% by weight. The application used imprecise terms to describe the quantities of the product under discussion, e.g., batch.
Table 2 of the parent application set forth the Q-factor at 1 mc. and illustrated dramatic increases in Q as a result of adding CoO and MnO to the basic three-oxide component ferrite. Claim 2 was for a nickel-zinc ferrite containing MnO of 1-5% by weight of the total composition and 1-60% by weight of CoO based on the proportion of NiO in the composition but not to exceed 3.0% of the entire composition.
U.S. Serial Application No. 467,828. Table 2 2 3
On April 25, 1955 the Examiner rejected all claims in the parent application on the ground that it was unpatentable over U.S. Patent No. 2,656,319, issued to Godshalk Berge on October 20, 1953. The Berge patent was said to disclose cobaltnickel-zinc ferrites. The Examiner stated that although the quantity of CoO claimed was less than that claimed by Berge, the mere reduction in quantity was insufficient reason for granting a patent in the absence of unexpected results. Claim 2 was also rejected because it was indefinite, merely indicating the physical mixing of oxides.
Zerbes amended the application on October 12, 1955 in what purported to be an attempt to remedy or rectify the deficiency of indefiniteness and he asked for reconsideration of the effect of Berge Patent No. 2,656,319, filed January 3, 1949. The further attempted amendments, the abandonment of the original application and the substitution of the two continuation-in-part applications No. 754,371 and No. 754,372 and the further proceedings in the Patent Office are sufficiently described in Judge Tenney's comprehensive opinion below, 297 F. Supp at 434-437. The net result was that the Examiner stuck to his ground and relied heavily upon the prior art, including Crowley Patent No. 2,736,708, filed June 8, 1951, Harvey Patent No. 2,723,239, filed September 29, 1952, Berge Patent No. 2,656,319, filed January 3, 1949, Albers-Schoenberg Patent No. 2,700,023, filed March 10, 1951, and a text book, College Chemistry, by A. Smith, at 643 (6th ed. W. Ehret 1946). In view of these references the Examiner could not see why the mere mixing of different metallic-oxide additives whose properties were already well understood was not "clearly within the teaching of the art." But the Examiner was overruled by the Commissioner on August 25, 1961. Zerbes amended his application again on January 22, 1962 by changing the limitation in the frequency range in the claim from "200-300 mcs." to "in the high frequency range of up to 300 mcs.," and the Patent No. 3,036,009 issued on May 22, 1962.
III. Application of the Law of Obviousness
We follow the rubric of Graham v. John Deere Co., 383 U.S. 1, 17, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). The scope and content of the prior art have already been delineated and the "level of ordinary skill in the pertinent art" has been disclosed by the foregoing discussion.
What put the whole series of maneuvers by the patentee, and the application for the patent itself, completely out of focus in the Patent Office was the failure of the patentee to disclose the existence of Ferramic N and Ferramic Q and the relationship between them. Appellant's predecessor manufactured and promoted both of these compositions with the trade before the filing of the grand-parent application in West Germany. We have already set forth the chemical compositions of these two ferrites, and it is quite apparent that Ferramic Q is substantially the same as Ferramic N with the addition of a small amount of cobalt oxide (CoO). As Ferramic N contained a small amount of manganese oxide (MnO), the new product, Ferramic Q, represented no more than the addition of the small amount of cobalt oxide. Ferramic N was, of course, part of the prior art and so were the Crowley and the Harvey patents above referred to, each of which taught that the Q-factor was considerably increased by the addition of the cobalt oxide. The Stemag people may well have thought that they discovered the effect of the addition of a small amount of cobalt oxide to a substance containing manganese oxide, which turned out to be the same as Ferramic N. But this does not negate the conclusion arrived at by Judge Tenney that a hypothetical person with knowledge of all this prior art and possessed of ordinary skill in the art of producing ferrites could be expected to add the small amount of cobalt oxide required to increase the Q-factor. We agree with Judge Tenney that this is sufficient to dispose of the issue of obviousness.
The close study of the subject matter that we have been required to make has convinced us, however, that this case presents an opportunity to develop, or at least to illuminate some fundamental principles.
Much is said by Indiana General about the search for a substance which would solve the problem caused by the deficiencies of Ferramic N. Undoubtedly, Indiana General was seeking some way to improve its product. But this record does not indicate any "long felt but unsolved needs" and the "failure of others to find a solution," such as has been the subject of comment in other cases where emphasis is placed upon commercial success.
See, e.g., Goodyear Tire Rubber Co., Inc. v. Ray-O-Vac Co., 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721 (1944); Carnegie Steel Co. v. Cambria Iron Co., 185 U.S. 403, 22 S.Ct. 698, 46 L.Ed. 968 (1902); The Barbed Wire Patent, 143 U.S. 275, 12 S.Ct. 443, 36 L.Ed. 154 (1892); Loom Co. v. Higgins, 105 U.S. 580, 26 L.Ed. 1177 (1881); Technical Tape Corp. v. Minnesota Mining Mfg. Co., 247 F.2d 343 (2d Cir. 1957), cert. denied, 355 U.S. 952, 78 S.Ct. 537, 2 L.Ed.2d 529 (1958).
Here we have an art that was practically non-existent prior to World War II. What the physicists, chemists, chemical engineers and others skilled in this particular art, including many of those whose writings and applications for patents were referred to in the File Wrapper of the patent at issue, were endeavoring to accomplish by their experiments and researches was of a more general character and of greater utility to the public in the long run. Thus these pioneers, both in the United States and abroad, were seeking to discover the properties of ferrites and to make them more efficient, not to discover some means of making Indiana General's Ferramic N more merchantable or to remedy its defects.
The result of these experiments and researches was, in part, to establish that the addition of a small quantity of manganese oxide had the effect of increasing permeability and that the addition of a small quantity of cobalt oxide improved the Q-factor. The desideratum at all times, of course, was the ascertainment of the combination of the two that would produce the optimum joining together of a sufficient permeability with a high Q-factor. The quantities of the divalent metallic oxides to be added to an Fe2O3 base could vary in proportion to the investigations and experiments of an unknown but surely large number of those who were skilled in the art. If each such combination, alleged to improve upon its predecessors, is to be patentable, and regarded as not the workaday product of those skilled in the art, the statute as interpreted by the Supreme Court in Graham v. John Deere Co. would, we think, be subverted, the definition of what constitutes invention would be greatly enlarged, and the public interest would suffer. That such a view would grievously prejudice the development of this art seems plain.
In view of the above it seems but natural that many of the observations relative to the properties of various additives appear as glosses or marginalia to the disclosures in such patents as Crowley, which have to do with the methods to be followed in the process of reducing the composition to its final form.
Thus the only difference between the prior art and Zerbes's alleged invention is the ascertainment of the proportions of manganese oxide and cobalt oxide that will produce a high quality factor and a sufficient permeability. Just what will be the exact proportions to supply the needs of particular products may require a large amount of experimentation. But this very experimentation is of the essence of the art involved in this lawsuit; it is on the level of ordinary skill in the art. This is all the more true because the practitioners of this art are, and of necessity must be highly educated, sophisticated persons who generally have at their disposal laboratory facilities and staffs of competent assistants.
Thus we hold that when the properties of the metallic-oxide additives that control the permeability and the Q-factor of a five-component nickel-zinc ferrite are well known to those skilled in the art, the product of individual experimentation with respect to the proportions of such additives required to produce such a ferrite with high permeability and high quality factor is not patentable. The reason for this ruling is that, as we have just said, such experimentation is on the level of ordinary skill in the art.
We limit this ruling to the subject matter before us, as we must, but see no reason to doubt that the principles we have formulated may have a wider application, especially in the field of chemistry.
The cases cited in the brief of Indiana General are all distinguishable.
White v. Fafnir Bearing Company, 389 F.2d 750 (2d Cir. 1968), affirming 263 F. Supp. 788 (D.Conn. 1966) dealt with the use of a new material in low friction bearings. The inventor used Teflon cloth to eliminate the need for special lubricants, resulting in a dry bearing surface in which the cloth was bonded to the resin surface of the bearing. Nobody before the inventor had used Teflon cloth as a lubricating material; other patents made claims for combinations of lubricants, none of which contained Teflon. The patent in suit provided a totally new solution to what had been for years a recurring problem.
Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263 (2d Cir. 1967) was a case in which the patented product was conceptually different from the prior art, which failed to solve a recurring and increasingly significant problem. The basic patent in the prior art, described in the opinion as a "paper patent," used a stylus and opaque armature in an aircraft-missle tracking apparatus. The patent in suit solved the problem of the stylus's obliterating the trace on the viewing screen by eliminating the armature altogether and mounting the stylus on a transparent glass or plastic plate so positioned that it did not obstruct the beam of light in the projector when it was inscribing. Other patents relied on to show that the prior art contemplated the invention either were irrelevant or did not focus on the specific problem.
Rich Products Corp. v. Mitchell Foods, Inc., 357 F.2d 176 (2d Cir.), cert. denied, 385 U.S. 821, 87 S.Ct. 46, 17 L.Ed.2d 58 (1966) involved the patent for a whipped emulsion liquid suitable for salad or dessert toppings. The patent was upheld against an attack of obviousness because it used an emulsifier different from that employed by a prior patent. The party claiming invalidity did not prove that there was no chemical difference between the two emulsifiers. Moreover, the prior patent relegated the emulsifier to a subsidiary role as an alternate fat-soluble vehicle for the coloring constituent. In the patent in suit, the emulsifier played a role of central importance to the whippability of the product.
Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124 (2d Cir.), cert. denied, 358 U.S. 884, 79 S.Ct. 124, 3 L.Ed.2d 112 (1958) involved the solution by the patent in suit of an old and serious industrial problem. Products made under prior patents were unsatisfactory and resulted in a considerable number of complaints from users. In the instant case the Crowley Patent made claims not for a specific product or range of products but for a method of producing ferrites.
Technical Tape Corp. v. Minnesota Mining Manufacturing Company, 247 F.2d 343 (2d Cir. 1957), cert. denied, 355 U.S. 952, 78 S.Ct. 537, 2 L.Ed.2d 529 (1958) dealt with a patent combining substances, the properties of which were well known. The validity of the patent was upheld against a claim of obviousness because expert chemists had tried unsuccessfully for years to produce just such a product as was effected by the invention disclosed in the patent. The Court said (247 F.2d at 347):
That Drew's patent involved invention is established in several ways. It had been sought by expert chemists of the DuPont Co. and of Johnson Johnson, for a number of years, unsuccessfully. Expanded Metal Co. v. Bradford, 214 U.S. 366, 381, 29 S.Ct. 652, 53 L.Ed. 1034; Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L.Ed. 1177. In his own experiments Drew stumbled upon it. Of course, after Drew's discovery it all seemed simple * * * [T]he demand had long existed; skilled men had sought to meet the demand without success.
The last point that Indiana General makes is that the patent in suit is not obvious because the prior art did not contemplate the specific range of components which Zerbes discovered. This argument has no merit because Zerbes claimed a very broad range of possible combinations, and in view of the foregoing discussion someone of ordinary skill in the art of producing soft ferrites could have produced a composition within the range. Furthermore, the patent in suit admittedly is one dealing with the combination of materials. The Supreme Court said in Great Atlantic Pacific Tea Co. v. Supermarket Equipment Co., 340 U.S. 147, 152-153, 71 S.Ct. 127, 130, 95 L.Ed. 162 (1950) that a stringent test of obviousness for combination patents is to be applied:
Courts should scrutinize combination patent claims with a care proportioned to the difficulty and improbability of finding invention in an assembly of old elements. The function of a patent is to add to the sum of useful knowledge. Patents cannot be sustained when, on the contrary, their effect is to subtract from former resources freely available to skilled artisans. A patent for a combination which only unites old elements with no change in their respective functions, * * * obviously withdraws what already is known into the field of its monopoly and diminishes the resources available to skillful men.
Although the patent involved in that case was for a mechanical device and the Court said that "[e]lements may, of course, especially in chemistry or electronics, take on some new quality or function from being brought into concert * * *," id. at 152, 71 S.Ct. at 130, the test should also apply to chemical combinations. Cf. Ruth v. Blue River Constructors, 224 F. Supp. 717, 725 (D.Colo. 1963); compare Application of Henderson, 52 C.C.P.A. 1656, 348 F.2d 550 (1965).
Applying this test, we conclude that Indiana General's argument is irrelevant. Ferramic N displayed higher permeability and Q-factor than Ferramic J. Ferramic Q, through the addition of cobalt oxide, improved the Q-factor of what essentially was Ferramic N.
IV. The Alleged Fraud Practised Upon the Patent Office by Indiana General
Krystinel Corporation has made a vigorous and earnest effort to induce us to reverse Judge Tenney's holding on the subject of attorneys' fees. The charge is that Indiana General and its predecessors deliberately and willfully entered upon a scheme to befuddle the Patent Office by various expedients. Judge Tenney recognized, as do we, that the conduct of Indiana General was "less than candid," but in the exercise of his discretion, viewing the situation as a whole, he balked at a finding of deliberate fraud. We have carefully studied this phase of the case, which adds further illumination to the conclusions at which we have arrived on the main issue, and we cannot find any abuse of discretion. The refusal to award counsel fees to Krystinel must stand.
What then are these alleged fraudulent practices? Probably the most important of all is the failure to disclose Ferramic N and its relationship to Ferramic Q. In view of the effect of such disclosure to Judge Tenney, and to us, it may well be that in the light of Crowley, the proceedings might have ended abruptly in final rejection of the application. But these patent cases have a way of developing in various directions; and if we approach the patent from what is stated on its face, including the triaxial diagram, the charge of willful fraud becomes more tenuous.
It is insisted, however, that the terms of the patent were so ingeniously devised as to carve out a large area of monopoly without disclosing precisely how the patentee's "magnetic core body having a high Q-factor in the high frequency range of up to 300 mcs." was to be obtained. This is a serious charge, and, except for the willful intent to deceive, the charge seems to be sustained by the evidence in the record. Judge Tenney found (297 F. Supp. at 440):
It is the opinion of this Court, however, that neither the 1954 parent application nor the patent in suit described the invention in such a way as to permit a person skilled in the art to reproduce the preferred composition or achieve the advancement claimed by the patent without himself having to conduct extensive experimentation. If, on the one hand, all the compositions located within the specified ranges produced improved results, the specification contained in both the parent application and the patent in suit would have been an adequate teaching of the claim. However, inasmuch as only certain of the compositions, none of which was specifically set forth, falling within the described ranges produced the desired result, any person skilled in the art was left with no alternative but to experiment with a multitude of possible combinations in order to search out and discover the actual invention.
Of course it is essential that the patent disclose the alleged invention in such clear terms that any person skilled in the art can put the invention into effect. A clear showing of Indiana General's Ferramic N and Ferramic Q would have shed considerable light on the subject. But Indiana General seems to have been more interested in the acquisition of an extensive, vaguely defined monopoly than it was in demonstrating just what the invention was.
There were other strange features. The specification includes the phrase "so that high quality values or Q-factors are obtained in the high frequency range of 200-300 mcs." The similar phrase in the claim itself was amended, just prior to the issuance of the patent, to "up to 300 mcs." As Judge Tenney remarked in substance, anyone familiar with the art could suppose that the patentee was representing that the alleged invention produced positive effects "falling within the specified high frequency range." When it was pointed out that Indiana General, on its own admission, could not support any such claim, and that "the useful frequency range had an upper limitation of approximately 30 mcs.," the answer made to us on the oral argument was that "up to 300 mcs." included 1 mc. This may indicate a certain furtiveness on the part of someone but it scarcely supports a charge of deliberate fraud.
The accumulation of these omissions fully supports Judge Tenney's finding that there was a lack of candor in the dealings of Indiana General with the Patent Office. It also evokes a suspicion that, if all the basic facts had been disclosed in the beginning the patent would have been summarily denied on the ground of obviousness. On the whole, however, and bearing in mind the fact that Judge Tenney saw many of the witnesses and weighed the testimony by deposition of others in the light of the witnesses he had listened to, we cannot say that the denial of the motion for attorneys' fees was based upon any abuse of discretion. We also sustain the rulings on whether or not this is an "exceptional case" within the meaning of 35 U.S.C. § 285, and to the effect that the showing made by Indiana General, while not measuring up to the standards of patentability, was "sufficient to support a bona fide belief in the patent's validity."
In view of the foregoing discussion, we do not reach the question of whether the four accused Krystinel products infringed the patent.
The motions submitted at the time of the oral argument are denied.
Judgment affirmed.
_________________________________________________________________________________________________ | Composition in Weight | | Q at 1 Megacycle | | | | percent | | | | | |_________________________| |_____________________________| | | | | | | | | Without cobalt | Q Improvement, | CoO: | | ZnO | NiO | FeO | o | | manganese | percent | 100 NiO | | | | | | Without | additions, CoO | | | | | | | | addition | =0.7% of batch, | | | | | | | | | MnO=2% of | | | | | | | | | batch | | | |______|________|_________|___________|___________|_________________|________________|____________| | 12.0 | 16.0 | 72.0 | 60 | 88 | 153 | 74 | 4.4 | | 19.2 | 4.2 | 76.6 | 60 | 56 | 124 | 121 | 16.7 | | 22.0 | 11.1 | 66.9 | 250 | 20 | 63 | 214 | 6.3 | |_________________________________________________________________________________________________|See, e.g., Goodyear Tire Rubber Co., Inc. v. Ray-O-Vac Co., 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721 (1944); Carnegie Steel Co. v. Cambria Iron Co., 185 U.S. 403, 22 S.Ct. 698, 46 L.Ed. 968 (1902); The Barbed Wire Patent, 143 U.S. 275, 12 S.Ct. 443, 36 L.Ed. 154 (1892); Loom Co. v. Higgins, 105 U.S. 580, 26 L.Ed. 1177 (1881); Technical Tape Corp. v. Minnesota Mining Mfg. Co., 247 F.2d 343 (2d Cir. 1957), cert. denied, 355 U.S. 952, 78 S.Ct. 537, 2 L.Ed.2d 529 (1958).