L. HAND, Circuit Judge.

The defendant appeals from the usual judgment of injunction and for an accounting, in an action for the infringement of patent No. 1,982,406, issued on November 27, 1934 to Hans F. Tonnies. Infringement being conceded, the sole question that we shall consider is the validity of the claims. The patent is for “an exposure meter which directly indicates the time of exposure necessary for taking photographs” (page one, lines 5, 6). It operates by means of a “photo-electric cell,” which, when struck by rays of light, .generates a current of electricity whose strength depends upon the brightness of the light. A galvanometer set in the circuit measures the strength of this'current and thus of the brightness of the entering light; the deflection of the galvanometer *779is translated upon a scale from which the photographer can read how much to open his lens and how long to expose his film to get the best photograph of the scene. So far there was concededly nothing new about the invention. The “photo-electric cell” went back to 1885, and it had been used as early as 1917 to determine the opening of the lens and the proper time for exposure of the plate or film. (Adsit, Pat. No. 1,244,159.) Other instances of the same use were Kohnen (German Pat. No. 312,805, 1917) and Kinsey (Pat. No. 1,864,442, 1932). These — except Figures 3, 4 and 5 of Kinsey — had, however, operated with a current generated by a battery, the “photo-electric cell” being introduced into the circuit as a variable resistance, and the current variations being measured by a galvanometer in the usual way. The presence of the battery made such meters too large to put into one’s pocket and carry about in one’s' hand, although Adsit did disclose a meter designed to be attached to the camera — Figure 3. The disadvantages of the use of.the “photo-electric cell” as a resistance were commented upon in Wein’s patent (No. 1,779,574) applied for in November, 1929, in the specifications of which he said: “such a cell does not of itself supply current and the combination not only requires a battery, but is subject to other well known difficulties inherent where selenium cells are used” (page one, lines 60-64). On the other hand Kinsey (whose application was filed in June 1929), in Figures 3, 4 and 5 of his patent, assumed that the current generated by the cell alone would be sufficient, and so did Tonnies in the patent-in-suit, whose German application was filed in August, 1930. So far as appears these are the first commercial applications of the self-generating cell.

Independently of this evidence it appears from the testimony that such a cell had not been developed earlier which could generate a “current sufficiently large to actuate an indicating instrument.” Fritts, the original inventor of the cell, had indeed in 1885 disclosed one that would generate “a very little current,” but its measurement “required a sensitive galvanometer” which “would not be portable,” but “would be a laboratory type of instrument”; indeed, “it was purely a laboratory device.” The plaintiff is an old and skilled manufacturer of electrical instruments; “competitive with many of the best makers of Europe.” It believes itself to “lead the trade in this respect and” to “have contributed most of the advances that have come in this trade, and the fundamental theories involved therein.” It “is founded on research and was one of the pioneers in industrial research in the entire electrical industry.” Nevertheless, although Goodwin, its engineer, had for a number of years been interested in getting an accurate meter for sunlight to control the exposure of a photographic plate, it was not until 1931 that the path was cleared, “when our company brought out, developed a photo-electrical cell that was the means that I was looking for. And we proceeded immediately to develop an exposure meter.” A compact, portable exposure meter had been awaiting a “photo-electric cell” of the self-generating type, which did not operate as a variable resistance in a battery circuit; and such a cell did not appear commercially much before 1929, else the plaintiff would have known of it.

The claims in suit were not however directed to that part of the disclosure; there was another difficulty which had to be overcome before such a meter could be made, and the patent depends for its validity upon the way that Tonnies overcame it. The “acceptance angle” of a camera means the greatest angle between any of the rays which enter the lens and fall upon the plate. Obviously, no rays can be photographed which do not reach the plate, so that the brightness of the scene for the photographic purposes means the light which is within the “acceptance angle.” A proper exposure meter should register only such rays as fall upon the plate, and the most obvious way to make one would be to duplicate the camera, merely substituting a “photo-electric cell” for the photographic plate. That is exactly what Adsit disclosed in Figures 1 and 2, though, apparently realizing that this was a clumsy device, he also disclosed in Figures 3 and 4 the much smaller boxlike motor we have mentioned, whose “acceptance angle” was to be the same as that of the camera. That meter would have anticipated Tonnies if the cell had been self-generating and there had been “a plurality” of admission chambers instead of one.

The patent in suit did “pluralize” the admission chambers, making a kind of honeycomb structure at the open end of the meter; each chamber being so proportioned in width and depth that its “acceptance angle” should correspond with that. *780of the camera with which the meter was to be used. Thus each chamber admitted to the surface of the cell only the same rays that were to be photographed, and there resulted a perfect correlation between the brightness of the scene metered and that of the scene photographed. It was necessary so to multiply the admission chambers in a “self-generating” cell, as it had not been in a “resistance” cell like Adsit’s, for the following reasons. Even in the fully developed “self-generating” cell the current developed is small, and it is proportional to the area of the cell upon which the rays strike. Thus it is desirable, if not imperative, that the whole surface of the cell should be exposed to the light of the scene. On the other hand, the greater the area of the exposed surface, the longer must be the distance between the surface of the cell and the aperture of the angle, if the “acceptance angle” of the meter is to be the same as that of the camera. Hence the proper amount of current demands a long tube, if the “acceptance angle” is to be maintained. It was to avoid this that Tonnies divided the surface of the cel! into five equal square compartments, the depth of which could — because of its small area — be made very small.

We do not know by what steps he came by this contrivance; but we do know how Goodwin independently did so in February, 1932, before the plaintiff had yet learned of the Tonnies patent. It had perfected the cell in 1931, as we have.said, but its engineers did not think of cutting up its surface and of using shallow chambers. Instead they set a tube at the open end, which, as its inner opening included the whole surface, had to be much too long for a pocket meter. They succeeded in producing only a bulky, awkward machine, even after they had made the tube telescope, so that it could be pushed in when not in use. The meter so designed they showed to a convention of motion-picture engineers in October of 1931, among whom was one, White, then in charge of the photographic laboratories of the Du Pont Company in Parlin, New Jersey. He saw the meter and “was playing with” it “a great deal of the time as well as were a number of other engineers.” But nobody suggested any substitute for the tube, and the plaintiff’s engineers were balked until one of them suggested taking the meter to White and getting his advice. This they did in January, 1932, and White— apparently at once — “suggested that we use a honeycomb baffle,” which completed the meter in the form in which it has been put upon the market.

In the light of this history the plaintiff makes the customary argument that what had halted such experts as the plaintiff’s engineers, judges ought not dismiss as obvious. To this we think however that the defendant’s answer is a good one: the plaintiff’s engineers did not know the optical art, and as soon as the problem was put to someone who did, the answer was immediate. The art had already several 'times provided tubes to shut out all but direct rays from impinging upon a “photoelectric cell.” Adsit (supra); Weiss, No. 321,069 (1885); McQuillen, No. 1,351,182 (1920); Bergstein, No. 1,840,014 (application filed 1928). And indeed, although the plaintiff’s engineers did not know of these disclosures, they had got so far independently. But the optical art had gone farther; it had used a “honeycomb” screen or “baffle” several times. As a means of centering the rays of an automobile headlight it was disclosed in Harrison, No. 1,136,006 (1915); Martin, No. 1,141,718 (1915); Bone, No. 1,280,953 (1918); Fuligni & Muhsgnug, No. 1,336,951 (1920). (Walker, No. 1,246,509 (1917) applied the same device to light shades.) All these were indeed to control emitted rays, but Otte, No. 1,266,525 (1918) applied such a “baffle” to restrict the admission of entrant light rays; and so did Bucky, No. 1,164,987 (1915) to entrant Rontgen rays. With this knowledge it does not seem to us that it was difficult to see that to reduce the length of the admission chamber, one needed only to multiply the chambers; and, this is confirmed by the fact that apparently as soon as someone familiar with the optical art was seriously consulted, he at once adopted that method.

As we have very often said, we put more reliance in deciding the issue of invention upon the history of the art than upon any tests which can _ be stated in general terms; indeed, there are, we think, none such which perceptibly help to a solution. H. C. White Co. v. Morton E. Converse & Son Co., 2 Cir., 20 F.2d 311, 313; R. Hoe & Co. v. Goss Printing Press Co., 2 Cir., 30 F.2d 271, 274; E. I. DuPont de Nemours & Co. v. Glidden Co., 2 Cir., 67 F.2d 392, 395; Picard v. United Aircraft Corp., 2 Cir., 128 F.2d 632, 636. In every case the issue is whether a per*781son acquainted with what had already been done, and endowed with no more than average imagination would have conceived the combination; and, while so to state the problem is not much of a step towards its answer, at least it clears the path of false directions. As we have seen, appearance of a portable exposure meter had been held up until the art had developed a self-generating “photo-electric cell” with power enough to move the indicating mechanism; and there had been no need of multiplying the admission chambers, while a battery furnished the power. Adsit had disclosed such a meter — satisfactory at least on paper and upon this record, standing unimpeached as to its operability. Even so, we might have been more impressed as to the talent needed to apply the existing means .to the new end, if the interval had been longer between the time when a compact portable meter became possible at all and when the patented combination was made. But that interval was only two or three years at most, and during that period two persons, independently and almost simultaneously, thought of the expedient. In the face of such a speedy answer as soon as an answer became serviceable, we see no reason not to yield our naive, lay judgment that to substitute a number of small chambers for one big one was an accomplishment within the compass of very mediocre abilities.

Judgment reversed; complaint dismissed.