In the court below the appellant, hereafter called the Perkins Company, the owner of patent No. 626,-927, issued June 13, 1899, to Charles C. Perkins, for an incandescent lamp socket, filed a bill against the appellee, hereafter called the United Company, charging infringement of the fourth and sixth claims thereof. In an opinion reported in 187 Fed. 942, that court held there was no infringement. From a decree dismissing the bill, the Perkins-Company appealed to this court.

*367The patent has been the subject of several cases. Tn this circuit its validity was sustained in a case reported as Perkins Electric Switch Co. v. Buchanan & Co. (C. C.) 129 Fed. 134. That decree was affirmed by this court in an opinion. Buchanan v. Perkins Electric Switch Mfg. Co., 135 Fed. 90, 67 C. C. A. 564. The Circuit Court of Appeals of the Sixth circuit, in Perkins Electric Switch Mfg. Co. v. Yost Electric Mfg. Co., 179 Fed. 511, affirmed a Circuit Court •decree sustaining its validity. Its validity was also sustained by the Circuit Court in the First circuit. Perkins Electric Switch Mfg. Co. v. Knowles, 187 Fed. 635. As to anticipatory evidence, which is now Cor the first time called to the attention of any court considering this patent, viz., the stipulated date, January 7, 1899, of the publication oí 1luisman and Cover’s British patent No. 30,065, we find the proofs unquestionably establish that Perkins’ invention was of an earlier date. Reference to the foregoing cases will fully explain the devices of the patent, its relation to the art, the contentions bearing on its validity, and the construction of its claims, and renders needless a present restatement thereof.

We content ourselves with saying that in a general way the Perkins invention consisted of forming in the body of a metallic electric light socket two insulated chambers, in each of which were placed -one of the. two binding posts of opposite polarity, and in one of which the switch was located. A solid wall of insulating material between the two chambers prevented, at that point, the objectionable features, first, of short-circuiting from binding post to binding post through the bearded or loose ends of the wires attached thereto; and, secondly, arcing from the switch. Perkins was the first to accomplish effective insulation through the use of separated chambers in a metallic electric light socket. Mis device was compact, effective, and had marked commercial success. It conduced to safety of life and property. The chamber shown and described in his patent had a solid insulating wall between the chambers, and while the solidity of this wall was emphasized in some of the opinions referred to, yet such statements must be read in the light'of the alleged infringement then being considered. But a study of the patent shows that the inventive novelty of Perkins -device lay in providing chambers insulated with reference to the path of possible current travel, and not merely to the particular chamber construction used to secure that result. In his device the location of the two binding posts, as well as that of the switch — and the consequent course of the only electric path of short-circuiting and sparking — -necessitated the placing of a wall at a particular place; that is, where the circuit path crossed the line where the chambers abutted. In other words, tlie wall was solid there, because it was necessary to make it so in'order to close the path of current course. The respond■ent, however, has found another and different way of preventing-short-circuiting through bearded wires — and in that regard it does not infringe — namely, by making the connection between its wires with outwardly extending binding screws and placing these outwardly extending screw connections in such electrically remote relations to each *368other that short-circuiting was practically impossible. Or, as said by respondent’s expert:

“Whereas, in defendant’s socket there is little or no danger of stray strands of wire of opposite polarity Coming in contact with each other by reason of the communicating passageway S, both because the binding screws are widely separated from each other and because the plate D stands between the head of the binding screw IS and the passageway if.”

It follows, therefore, that closure or nonclosure of the abutting chambers in respondent’s device at the place where it is left open is a negligible, functionless matter. Accordingly the respondent can have, and actually has, no wall at what would in Perkins’ location of parts be, and in respondent’s is not, the path of inter-binding^post, short-circuiting. Indeed, to have placed even a closed wall at that point _would not establish infringement by the respondent, for such wall would not accomplish any preventive insulating function. On the other hand, when respondent locates its switch, it does so at a recessed portion or end of the chamber, from which the switch spark cannot travel through the circuit pathway of- the Perkins device. But at that part of the recess where the circuit can travel, and across the path of such travel, it places a solid wall for the purpose of insulating the recess of the chamber — which recess alone requires insulation — from the abutting chamber. Having, by virtue of this recessed .location of the switch and the solid wall across the current path, made the two chambers electrically insulate, does the respondent become a noninfringer by breaking down the wall at a point beyond, where whether it be perforate or imperforate in no way aifects the working of the device ?

Assuredly not, unless mere form and functional substance are the same thing. With as much justice could it be said that a ship that was staunch and tight below the water line was leaky because of its scupper holes. It is, of course, true that Perkins’ chambers are described in his specification as “two separated chambers,” and in his claims as “recesses arranged to form insulated chambers”; but it is not to be overlooked that the specification describes these.terms as embodying a structure such “that a spark caused by opening the switch cannot communicate with the other side of the circuit.” And this is precisely what, by the location of its wall with reference to the switch, the respondent does. In that respect — and therein was the gist of the Perkins device — the chamber of the respondent is closed, separated, and insulated because, in the words of the claim, it has “recesses arranged to form insulated chambers,” and its effect is such that, in the words of the specification, “a spark caused by opening the switch cannot communicate with the other side of the circuit.” It follows, therefore, that the chamber, recess is separated and insulated both in letter and spirit, as those terms are used in the patent to describe Perkins’ invention. And this view is in accord with the former opinion of this court, wherein the language of an expert witness was approvingly quoted, who, in describing Perkins’ invention said:

“The essential feature of the construction, in so far as claims 3, 4, and 9 are concerned, as I understand the patent, consists in the form of tin "n*369fsuiatitig Works and the disposition of the several necessary electrical contact portions in such maimer that they are effectively separated from one another and serve to hold together the insulating pieces to make the interior mechanism and its insulation substantially a unit, exceedingly simple and strong in its construction and successfully meeting the requirements of practical use. This is attained by the use of two opposed chambered insulating blocks united by plates, serving also as electrical connections to form mibutantially separate indo*tires. whereby conducting parts of opposite polarities are separated from one another in a substantially perfect manner.'’

It follows, therefore, that the decree of the Circuit Court must be reversed, and the case remanded, with directions to enter a decree against the respondent for infringing the claims in controversy.