Before L. HAND, AUGUSTUS N. HAND, and CLARK, Circuit Judges.
This appeal involves the validity of Patent No. 2,118,468 for a "Method of Casting Articles of Intricate Design and A Product Thereof," issued to the plaintiff on May 24, 1938. Though the answers of both the original and the intervening defendants to the plaintiff's claim of infringement denied infringement as well as the validity of the patent, the district court found infringement by all the defendants as to at least some of their operations; and that is conceded on this appeal. The court, however, held all the claims of the patent invalid for want of invention and because anticipated, and ordered the action dismissed with prejudice. Judge Rifkind wrote a carefully considered opinion in which he described in detail the state of the prior art and discussed the contentions of the parties. D.C.S.D.N.Y., 69 F.Supp. 922. We agree with the views he expressed, and in our discussion shall assume the reader's familiarity with the extensive statement he has made.
The patent in suit consists of five claims for the method of casting he sets forth and a sixth claim for a product or article produced by this method. The process described is, with certain modifications, similar to the ancient "cire perdue" or the "lost wax" process. In brief, Jungersen's method is as follows: First a model of the article to be reproduced is made and a flexible mould-forming material such as rubber is built around it. The model is then removed from this mould, and wax or other material of a low fusing point is centrifugally cast into the mould. Next the wax pattern so obtained is removed and invested in plaster of Paris or other suitable material in order to form a secondary mould in which the metal may be cast. The wax is then melted out and all traces of it are removed from the mould. And finally, molten metal is centrifugally cast into this secondary mould. The resulting article is an exact replica of the original model. By using centrifugal casting, the wax and the metal are forced into the finest recesses of the moulds so as to assure accuracy of reproduction, no matter how intricate the design. Of the claims the first four are based directly on the process described, referring to the forcing "by centrifugal action" of both the molten wax into the primary mould and the molten metal into the secondary mould. See the second claim quoted at page 922 of 69 F.Supp. The fifth claim requires only the introducing of the material "into the mould by force sufficient to deposit" it in the depressions of the mould. And the sixth or product claim is similarly general.
Admittedly much of this is in the prior art. Indeed, plaintiff in effect so acknowledges. Thus he states in his British Patent No. 449,062, for "Improvements in Method and Means of Casting Metals," June 19, 1936, covering this same process: "In the casting of metal it is not broadly new to make an elastic mould from an original sample or model, to produce a wax or fusible pattern with the aid of the elastic mould and to remove the pattern from the latter and invest it in a final mould from which the wax or fusible pattern is subsequently melted out to permit of the pouring of metal into the mould cavity thus vacated." The inventor's appraisal of his own invention is of course of importance. Timken-Detroit Axle Co. v. Cleveland Steel Products Corp., 6 Cir., 148 F.2d 267, 270, certiorari denied 326 U.S. 725, 66 S. Ct. 30, 90 L. Ed. 430. It is clear from this admission that all but the use of centrifugal casting to force the wax and the metal into their respective moulds was deemed old. The validity of this statement in Jungersen's British patent is amply demonstrated by the prior art.
Cellini, the great Florentine artist of the sixteenth century, described with painstaking detail the use of the multipart mould and the lost wax process. In fact, of course, it is much older. 4 Encyc. Brit., 14th Ed., 240, 20 Id. 229; and authorities cited in the opinion below, page 927 of 69 F.Supp. La Gravure, in the French publication "Engraving, Carving, Moulding," 1924, received in this country in 1926, explicitly applied the process to small pieces of jewelry. So, too, Haseltine's British Patent, No. 2467, issued 1875, and Spencer's U.S. Patent, No. 748,996, issued 1904, taught the use of the primary and secondary moulds to reproduce an original model in all its detail. Moreover, both these latter patents also revealed the use of a flexible mould into which the wax is poured.
Thus plaintiff's patent can be sustained only if the use of force to cast the wax and the metal is such a change of a known method as to amount to invention. We think it is not. Indeed, the use of applied force to cast metal was shown in several patents prior to the plaintiff's. Haseltine taught the use of pressure - about 20 pounds to the square inch - on the fluid metal where necessary to make the metal lie closely to the mould and produce a perfect casting. And McManus's U.S. Patent, No. 1,457,040, issued 1923, described a centrifugal casting machine. Although the machine seemed to have been intended primarily for use in making dental castings, yet notwithstanding this, he expressly stated that it might, with slight modifications, "be equally as well applied to the casting of jewelry," while Kralund's U.S. Patent, No. 1,238,789, issued 1917, taught the use of force for introducing wax into intimate connection with the first die as well as upon the molten metal of the final casting. So we see that all the details embodied in the plaintiff's patent - save perhaps the centrifugal casting of the wax into the primary mould - were present in the art. It was known that if force were applied when metal was cast an accurate reproduction of the configurations and details of a model was assured. And McManus taught that the application of such force might be obtained by means of centrifugal casting. Applying these principles to a different material does not constitute invention; moreover, as we have seen, Kralund even applied force in introducing the wax to the die.
Plaintiff's most vigorous contention is, however, that the jewelry art has special requirements such as the complexity and minuteness of the mould and hence is nonanalogous to the prior art. But plaintiff himself indicates the contrary in his patent, for he says: "The principal object of this invention is to facilitate the casting of small metal articles, particularly articles of intricate detail such as jewelry." Thus the patent refers generally to the art of casting with a reference to jewelry as an example of the use to which the method may be put. Nowhere does the distinction that plaintiff now attempts to draw appear in his patent. And his British patent sheds some light as to what he claimed to introduce by way of novelty. As we have seen, that patent contained an express acknowledgment that many of the steps he proposed were old in the art. The reference was not to the art of jewelry, for the use of the primary and secondary moulds had not been used in the jewelry trade prior to that time. Obviously it was to the art of casting. So in that patent, when plaintiff sought to distinguish what he contributed from what previously existed in the field, he did not assert or even indicate that his contribution was nonanalogous to the prior art. Rather, he treated as invention the introduction of what he considered to be a new step in a familiar method of casting, namely, the use of centrifugal force for casting the wax and the metal into the moulds. Yet, as we have seen, the use of such force in casting metal for jewelry had been indicated by McManus. Moreover, it was well known as a general method; thus Stern's "Die-Casting Practice," published by the McGraw-Hill Book Company in 1930, compares and explains centrifugal casting and die casting, and shows how it is but a method of forcing the material into the mould. This is decisively shown also by the plaintiff himself in describing his course of invention as recounted in the opinion below. Thus he finally "fell on the idea of filling by centrifugal force," making "a swing out of a wired clothes-hanger that I could swing around in my hand * * * and after that * * * I went to a mechanical dentist" to make the final casting through the use of the dentist's centrifugal machine.
Hence the "flash of genius" must be found, if anywhere, in the plaintiff's recollection of the use of this force in dental casting and decision to apply it to this old process. Such, indeed, is the basis for the not inconsiderable success which the plaintiff appears to have had in prior litigation. In Jungersen v. Jenkins, D.C. Md., 30 F.Supp. 615, and in an unreported case before an English judge - both substantially undefended by other manufacturers - and in Jungersen v. Morris Kaysen Co., D.C.E.D.Pa., 31 F.Supp. 703, the use of centrifugal force, as well as the commercial success of the process, is stressed to uphold the invention. In Ostby & Barton Co. v. Jungersen, D.C.N.J., 65 F.Supp. 652, the same argument is developed extensively to reach the conclusion that claims 1 to 4 of the patent were valid, and claims 5 and 6 were invalid. Since there was no contention that the contesting parties had infringed the first four claims, the victory for the claimed invention had something of a pyrrhic character. At any rate both parties appealed, the decision was affirmed per curiam, 3 Cir., 163 F.2d 312, on the reasons assigned in the opinion below, and petitions for writs of certiorari by both parties were denied, 68 S. Ct. 356, Mr. Justice Black disagreeing as to the contestors' petition. The effect of the decision must necessarily be to limit plaintiff to the one method of use of centrifugal force. In view of the history of the process and the close equivalence of other methods, this would seem to be a substantial limitation on the patent.
The judge in the Ostby & Barton case referred to the use of centrifugal force as an "additional step" in the art which was both "new" and "novel." We are forced to disagree. It could hardly be an additional step; at most it can be only a modification or a different execution of an existing step. Nor, in view of its past employment, can we consider it new or novel. True, the process in its entirety had not been assembled for use in the manufacture of jewelry; but the addition made by plaintiff had been plainly indicated. The mere fact that the earlier references seemed to have made but little impression on the art of casting is in itself unimportant. Western States Mach. Co. v. S. S. Hepworth Co., 2 Cir., 147 F.2d 345, 350, certiorari denied 325 U.S. 873, 65 S. Ct. 1414, 89 L. Ed. 1991; Condenser Corp. of America v. Micamold Radio Corp., 2 Cir., 145 F.2d 878, certiorari denied 324 U.S, 861, 65 S. Ct. 869, 89 L. Ed. 1418. The combination of such known elements is not invention. Toledo Pressed Steel Co. v. Standard Parts, 307 U.S. 350, 355, 356, 59 S. Ct. 897, 83 L. Ed. 1334; Textile Mach. Works v. Louis Hirsch Textile Machines, 302 U.S. 490, 497, 58 S. Ct. 291, 82 L. Ed. 382; Paramount Publix Corp. v. American Tri-Ergon Corp., 294 U.S. 464, 474, 55 S. Ct. 449, 79 L. Ed. 997; Powers-Kennedy Contracting Corp. v. Concrete Mixing & Conveying Co., 282 U.S. 175, 186, 51 S. Ct. 95, 75 L. Ed. 278. Nor is the application of an old process to a new and analogous use. Dow Chemical Co. v. Halliburton Oil Well Cementing Co., 324 U.S. 320, 327, 65 S. Ct. 647, 89 L. Ed. 973. In the past the commercial success enjoyed by plaintiff might well have sufficed to tip the balance in favor of validity. Recent Supreme Court pronouncements indicate clearly that commercial success cannot raise a combination of known elements to the exalted level of invention. Funk Bros. Seed Co. v. Kalo Inoculant Co., 68 S. Ct. 440; Sinclair & Carroll Co. v. Interchemical Corp., 325 U.S. 327, 65 S. Ct. 1143, 89 L. Ed. 1644; Paramount Publix Corp. v. American Tri-Ergon Corp., supra. See Myerson v. Dentists' Supply Co., 2 Cir., 159 F.2d 681. The elements here stressed by plaintiff were present in even stronger form in some of these cases, as witness our view in Interchemical Corp. v. Sinclair & Carroll Co., 2 Cir., 144 F.2d 842, and that of the Seventh Circuit in Kalo Inoculant Co. v. Funk Bros. Seed Co., 7 Cir., 161 F.2d 981, each reversed in the recent Supreme Court decisions just cited. We should heed the admonition of these reversals.
What we have said of course disposes of claim 6, the product claim. There is nothing new in the product. Risdon Iron & Locomotive Works v. Medart, 158 U.S. 68, 81, 15 S. Ct. 745, 39 L. Ed. 899; Pfanstiehl Chemical Co. v. American Platinum Works, 3 Cir., 135 F.2d 171, 173.
L. HAND, Circuit Judge (dissenting).
In Jungersen's British patent, as my brothers truly say, he based his invention solely upon forcing the wax and the metal into completely intimate contact with every crevice of the mould, and for this he disclosed a centrifuge as the means. Moreover, it had already been known by other moulders of fine patterns that the metal might not fill all the spaces necessary for perfect reproduction. For example, in 1873 Haseltine disclosed a device which set up a pressure of twenty pounds to the square inch; and this too in a "lost wax" process. True, he did not disclose using similar pressure for the wax, and he did not use a centrifuge; but McManus used a centrifuge to force fusible metal into all the crevices of the mould, and that too in a "lost wax" process, the knowledge of which he appears to have assumed, for he does not disclose how to make the wax model. Kralund also showed a pressure die-casting process, as applied to ...