Opinion
No. 3769.
July 16, 1928.
Appeal from the District Court of the United States for the District of New Jersey; Joseph L. Bodine, Judge.
Patent infringement suit by the General Electric Company against the Atlas Specialty Company and others. Decree granting plaintiff a preliminary injunction, and defendants appeal. Affirmed.
Charles J. Holland, of New York City, for appellants.
Howson Howson, of New York City (Hubert Howson, of New York City, Albert G. Davis, of Schenectady, N.Y., and Frederick P. Fish, of Boston, Mass., of counsel), for appellee.
BUFFINGTON, WOOLLEY, and DAVIS, Circuit Judges.
This suit against the Atlas Specialty Company and sundry personal defendants is the second in a group of four suits instituted by the General Electric Company for infringement of several of its patents in relation to incandescent lamps.
The first case was against the Desmond Incandescent Lamp Co. et al. (C.C.A.) 27 F.2d 590, where, as here, the plaintiff moved for and was awarded a preliminary injunction. Here, as there, the defendants appealed, raising closely related questions of fact which we have considered and shall briefly discuss under the law which, as we stated in that case, is applicable to trial and appellate courts in cases of preliminary injunctions.
As the alleged infringing lamps in this case are all vacuum lamps, the Langmuir patent, No. 1,180,159, covering an incandescent lamp with a tungsten filament in an atmosphere of inert gas, does not figure. It is only the Just and Hanaman patent, No. 1,018,502, covering a substantially pure tungsten filament for incandescent lamps that is charged to be infringed and against further infringement of which the preliminary injunction runs. Similarly the question involved is whether or not the plaintiff's evidence gathered from its supporting affidavits is of a character sufficient to sway the discretion of the trial judge in concluding that the defendants infringed the previously adjudicated patent by the manufacture and sale of lamps with filaments made of substantially pure tungsten.
In this case the defendants, contrary to the action of the defendants in the Desmond Case, while denying that their filaments are made of the tungsten of the patent, openly told of what they are composed. Admitting that tungsten is present, they said it is not substantially pure tungsten or that it is not tungsten "in a coherent metallic state and homogeneous throughout" but it is an alloy of which tungsten is only one element and then an element functioning differently (because of the other elements) from tungsten in its pure state. To prove this they give two analyses; one made by Doctor Fink and not sworn to by him, and the other by Leslie G. Jenness, sworn to. Though one lacks proper proof, we have considered and shall quote both.
Doctor Fink, according to the statement of another, found that the filament which the defendants submitted to him for analysis contained 96.07 per cent. of tungsten, 2.1 of thorium, 0.4 of silica, 0.1 of yttrium, 0.4 of vanadium and less than 1 per cent. of sodium oxide and potassium oxide. Mr. Jenness found in the wire submitted to him 95.3 per cent. of tungsten, 3.4 of thorium, 0.6 of vanadium and 0.3 of silica. He was not instructed to look for other constituents. A composite of these analyses shows a content of substantially pure tungsten equal to 96 per cent. and that the remaining 4 per cent. was made up of a half dozen more or less inefficient constituents. The plaintiff on the other hand caused four analyses to be made of the defendants' filaments. One group of two (Brophy) showed 1.21 and 1.09 per cent. of thoria or other equivalent nonvolatile metal, no silica, no vanadium; explaining that if yttrium were present it was in the form of yttria and was not separable from thoria in the analyses. All the rest (about 98.79 per cent.) was tungsten. Another analysis (Schwartz) disclosed 1.20 per cent. of thoria, no vanadium and no silica; the balance, tungsten. Because of the difficulty of chemically separating yttria from thoria, a spectroscopic analysis for yttria was made and failed to disclose its presence. Still another analysis (Jeffries) yielded 1.07 per cent. of thoria; all else, tungsten. With these analyses before them both parties proceeded to compare them at length with the constituents of filaments involved in prior litigation. General Electric Co. v. Laco-Philips Co. (C.C.A.) 233 F. 96; General Electric Co. v. Alexander and Alpha Laboratories (D.C.) 277 F. 290; Id. (C.C.A.) 280 F. 852; Id., 260 U.S. 727, 43 S. Ct. 89, 67 L. Ed. 484; General Electric Co. v. P.R. Mallory Co. (D.C.) 286 F. 175; General Electric Co. v. Save Electric Corporation (D.C.) 286 F. 180. We discern danger and have found little help in these litigious comparisons. Nor have the analyses produced by both parties, regarding all as correct analyses of the samples severally submitted, greatly aided us in determining whether the evidence shows infringement for purposes of a preliminary injunction because from a rather wide experience in cases involving chemical processes we know that a small change in chemical formula sometimes makes a great difference in results and sometimes it makes no difference at all. Therefore we have kept constantly in view the capital question — how do the defendants' mixed or "alloyed" filaments act in comparison with the plaintiff's unmixed and substantially pure tungsten filaments? In this regard we looked first to their action in the sense of re-action to tests which science and the art have adopted and next to their action or function when put to the uses for which they were made.
In this case, as in the Desmond Case, the defendants' filaments were submitted to the melting point test, to the hot-to-cold resistance test and to qualitative chemical analysis, the reaction to each was, according to the plaintiff's affiants, that of substantially pure tungsten. To these tests the defendants in this case made the same challenge that the defendants in the Desmond Case made with, we think, no more success. True, thorium was present in the defendants' filaments in some percentage, varying by the analyses from 1.07 to 3.40, but we have not been convinced that the addition of this small quantity of thorium, long before employed as a means to prevent lateral sliding in a tungsten filament known as off-setting, or even the addition of silica, yttrium and vanadium in the percentages claimed by the defendants, as against the preponderating percentage of tungsten made an alloy or a new metal which destroyed or altered the essential properties of tungsten when functioning as a lighting means in a vacuum bulb. The evidence discloses no dissimilarity; indeed it tends to identity.
We have limited our consideration of this case to the Just and Hanaman patent, the one patent declared on in the complaint and mentioned in the decree awarding a preliminary injunction. We have purposely avoided considering the Coolidge patent, No. 1,082,933, involved in General Electric Co. v. DeForest Radio Co. and Robelen Piano Company, still pending in this court, and also the Laise patent, No. 1,569,095. It should therefore be understood that nothing said in this opinion, nor in any of the opinions delivered in this group of four cases, nor indeed the decisions themselves, have any bearing on that case or on the validity of the Laise patent. Restricting ourselves to the Just and Hanaman patent, the one patent here in suit, we are satisfied that the plaintiff's supporting affidavits, read in the light of the defendants' controverting affidavits, show such a case of infringement as justified the learned trial judge in entering the decree for a preliminary injunction. Grushlaw v. Phœnix Knitting Works (C.C.A.) 183 F. 222.
The decree is affirmed.