The bill in this case, which was filed by the appellee, complains of the infringement by the appellant of rights secured by letters patent No. 626,927, granted to Charles G. Perkins, as assignor of the appellee, June 13, 1899, for improvements in incandescent lamp sockets. “This invention,” the patentee says in his specification, “relates to a receptacle for incandescent lamps having a switch mechanism inclosed in a chamber that is insulated from the shell, and from the chamber containing the parts forming the other side of the circuit through the socket.” The answer denies that Perkins was the first inventor of any material part of the thing patented, and cites 17 former American and English patents on which reliance is made to prove anticipation. And the answer also denies infringement. Eleven claims are included in the patent in suit. The Circuit Court found and held that the claims numbered 3, 4, 6, and 9 were valid and were infringed by the manufacture and sale by appellant of certain forms of such sockets which are identified *512by numbers given them in the testimony and in the decree. Upon this finding the court entered the usual decreé in favor of the complainant.
The specifications disclose a socket having .an outer shell, lined with an insulating sheet, inclosing two blocks of insulating material placed one above the other and of a diameter equal to the space inside the lining of the shell. The meeting faces of the blocks are made to closely fit each other, except that on the two opposite sides of the blocks at their juncture a portion of each is taken out so as to form chambers on each side completely insulated from each other by the material of which the blocks are made. These chambers are much alike, except that one is larger than the other. The larger one contains not only a binding post, a connecting plate, and other parts mak.ing a connection of the positive circuit wire with the lamp, but also a switch mechanism for closing the circuit. The smaller chamber contains like parts for the transmission of the return current, but no switching apparatus. The wires bearing the current are each brought into these chambers through grooves or openings in the periphery of the upper block on the same opposite sides as the chambers. Fig. 3 of the drawings, here inserted, gives a general idea of the structure:
23 and 14 show the larger and the' smaller chambers, 24 and 17 show the connecting plates, and 40 is the switch-block. This goes far enough into detail for the present purpose.
The claims of the patent on which the decree was founded are as follows:
“3. In combination in a lamp-socket, a cap, a shell, two blocks of insulating material with recesses arranged to form two insulating-chambers, a plate with a binding-screw located in one of the chambers and having its ends secured to the respective blocks, a plate with a binding-screw located in the other of the chambers and having its ends secured to the respective blocks, and grooves in the edges of the upper block for the passage of the circuit-wires of the respective binding-screws, substantially as specified.
“4. In combination in a lamp-socket, a shell, two blocks of insulating material with recesses arranged to form insulated chambers, a plate with a binding-screw located in one of the chambers, a plate with a binding-screw located in the other of the chambers, and a switch-block located in one of the chambers and adapted to make contact With the end of the plate in the same chamber, substantially as specified.”
“6. In combination in a lamp-socket, a cap, a shell, an insulating-lining fitting the shell, two blocks of insulating material with recesses arranged to form two insulated chambers, a plate with a binding-screw located in one of the chambers, a plate with a binding-screw located in the other of the chambers, and a switch-block located in one of the chambers and adapted to make contact with the end of the plate in the same chamber, Substantially as specified.”
“9. In combination in a lamp-socket, a cap, a shell, two blocks of insulating material located within the shell, insulated chambers formed by recesses *513in the insulation, and plates bearing outwardly-extending binding-screws located in the recesses and having their ends secured by screws to the respective insulating-blocks, substantially as specified.”
From the preceding statement of the inventor’s announcement of the object he had in mind and the terms of each of these claims, it is perceived that the dominating feature of his invention was the provision of means for the insulation of the two chambers from the shell and from each other. The other elements associated in the combinations were incidental and appropriate to an organization which should effect his main purpose. The other features were found id earlier disclosures of the same art. There could have been no invention in bringing these old elements into a new combination, where each would perform the same service as they did in the earlier art. Burt v. Evory, 133 U. S. 349, 10 Sup. Ct. 394, 33 L. Ed. 647; Florsheim v. Schilling, 137 U. S. 77, 11 Sup. Ct. 20, 34 L. Ed. 574. We have on several occasions recognized and applied this rule. Burnham v. Union Mfg. Co., 110 Fed. 765, 49 C. C. A. 163; Overweight Counterbalance El. Co. v. Henry Vogt Mach. Co., 102 Fed. 957, 43 C. C. A. 80; Campbell Printing Press Co. v. Duplex Printing Press Co., 101 Fed. 282, 41 C. C. A. 351. The novel idea was to have two separate chambers, one for the apparatus connected with the wire carrying the positive current, and the switch to close the circuit, and the other for the apparatus connected with the negative wire, and to completely insulate both of the chambers. There was nothing new in lining the shell of the socket with an insulating sheet, nor in the use of connecting plates, nor in bringing the wires into the socket at a comparatively safe distance from each other. The objection to former structures which the inventor proposed to avoid were dangers arising from exposing the apparatus connecting the wire to the filament of the lamp to the close proximity of those connecting the other wire. It was known that in the operation of former devices sparks were sent off by the sudden opening of the switch, tha1> stray strands of the wire at its terminus, resulting from careless binding, would get too near the other wire or its connections and establish short circuits, and that arcs were liable to be formed in the socket from exposed surfaces of the wires at near distances. And the conception of the inventor was to use two chambers and build an insulating wall to separate them and which would absolutely prevent the danger of such occurrences. And the proof satisfactorily shows that he was the first person who conceived this idea of complete insulation, or rather, he was the first to disclose such an idea in a practical way. But others had come so near it that it was but a step from them to his own conception. And there is a temptation to say that the step was in the same direction and was indicated by what had been done before, and to recall what has been said of such conditions by able judges expert in patent law. In Smith v. Nichols, 21 Wall. 112, 22 L. Ed. 566, the law was formulated, and it has been followed ever since. In Smith v. Nichols, 21 Wall., at page 119 (22 L. Ed. 566), it was held that a mere carrying forward of an original conception patented, a new and more extended application of it, involving only change of form, proportion, or degree, the substitution of *514equivalents doin’g'the same thing as did the original 'invention by substantially the same means with better effects, is not such invention as will sustain a patent. Some of the decisions of the Supreme Court are: Roberts v. Ryer, 91 U. S. 150, 23 L. Ed. 267; Phillips v. Detroit, 111 U. S. 604, 4 Sup. Ct. 580, 28 L. Ed. 532; Burt v. Evory, 133 U. S. 349, 10 Sup. Ct. 394, 33 L. Ed. 647; Ansonia Co. v. Electrical Supply Co., 144 U. S. 19, 12 Sup. Ct. 601, 36 L. Ed. 327; Lovell Mfg. Co. v. Cary, 147 U. S. 623, 13 Sup. Ct. 472, 37 L. Ed. 307.
In several earlier patents the objections above referred to were recognized; and devices to mitigate them were shown. We shall refer to only a few of them which ere sufficient to show the state of the art at the date of the Perkins invention. A patent to Hubbell, No. 565,541, dated August 11, 1896, showed all the elements of the Perkins combination except that, instead of two chambers, there was but one which filled the whole interior of the socket. The respective congeries of apparatus connecting the respective wires with the lamp were located as far apart as might be. But there was only air insulation between them. In a patent to Wirt, No. 560,667, dated May 26,1896, there was one entire insulating block instead of two, and the respective parts of the connecting' apparatus were lodged in peripheral recesses on opposite sides of the block. But Wirt did not sufficiently guard the metallic parts em_ oyed by him from getting in contact, one side with the other. Then, there was a patent to Pass and Seymour, No. 568,919, dated October 6, 1896, for a “Socket for Incandescent Electric Ramps,” which shows an upper and lower block of insulating material and the other features of the Perkins patent, except that the two chambers were not completely insulated from each other. In their specification the patentees' dwell at length upon the dangers attending the use of former devices in the particulars above recited, and stated how they proposed to obviate them. They say that they arrange the wires as far apart as possible. And of their construction they further say:
“The wires and other metallic parts are all positively separated from each other, so that accidental contact of the wires or of detached strands is prevented by walls or projections formed in the porcelain”
—which is the insulating material. And, further, they locate their contact piece on the negative side in a depression in the upper face of the lower block and guard it from the other side by a “semi-circular wall.” This wall between the chambers does not go up to the bottom of' the upper block and is not long enough to go quite around to the wall of the chamber. The purpose of this depression and of this wall is evident. It was to shield the parts in one chamber from those in the other. Apparently they thought they had made the wall high and long enough to practically effect their purpose. What Perkins did was to carry the wall quite up to the bottom of the upper block and .quite around to the inside of the wall. By this construction there would be no- possibility of the apparatus in one chamher being disturbed by what was happening in the other.
The Third Circuit Court of Appeals in a suit for- infringement brought by_ one Buchanan against the appellee, 135 Fed. 90,1 held the Perkins patent to be valid,, upon the ground that Perkins had devised *515a more complete and effective separation and insulation of the chambers than had been done before him. Out of respect for that court and. the comity due to it, and notwithstanding the doubt implied from what we said on a former page of this opinion, we are disposed to follow the decision of the Third Circuit Court of Appeals. We are not certain in regard to the structure which was there held to be an infringement, but we infer from the opinion written by Judge Gray that it possessed the feature of complete separation ,and insulation of the chamber. This feature of complete insulation is present in all the claims of the Perkins patent here in question, and we need not refer to them in detail.
We come, then, to the question of infringement. Several styles of sockets, numbered from 1 to 4, are exhibited as made and sold at different times by the appellant. No. 1 of these, about which we will speak later, is clearly an infringement. It contains the two chambers separated and insulated from each other by a solid wall of insulating material, and both are insulated from the shell. The other elements are substantially equivalents. The other types have two chambers, but they are not completely separated from each other. There is a wall, but it is not entire. It is open in one of them for a space large enough to see through one chamber into the other and enable the operator to blow through it, and in others there is scarcely any division of the chambers. As to the first of these, it is contended that the breaking away of the wall is a mere subterfuge to differentiate the sockets from the Perkins patent. If this were so, if the opening in the wall were so small as to be negligible and to serve no useful purpose, we should think the contention ought to be sustained. But it is to be said that the opening is substantial. It differs from that of Pass and Seymour only in degree; and we have distinguished the Pass and Seymour patent from that of Perkins only by the fact that the wall of the former did not completely separate and insulate the chambers one from the other, while the latter does. Then, it appears, and in this we are confirmed by expert testimony, there is a substantial utility in being able to look into the socket without taking it apart and especially in being able to blow out the dust and debris that gathers in the socket by use. In the other types the difference is still more manifest. We are therefore constrained to hold that the appellant does not employ the feature which distinguished the Perkins patent, and on account of which that patent was granted. It seems impossible to doubt that but for it, and in view of the earlier patents, these claims of the patent could not have been granted.
As to the appellant’s socket No. 1, before mentioned, these facts appear : The appellant for two or three months at a period of about one year before the bill was filed, made and sold some of these sockets; that they then discontinued the use of that form and adopted the later numbered styles. It does not appear that the appellee was notified of the discontinuance of the infringing style or had any knowledge that it had been discontinued. We cannot say that it had no reason for apprehending that its use would be discontinued, or would not be renewed. In these circumstances, the appellee was entitled to file its bill for an injunction.
*516The decree will be reversed in all other respects, with costs of this court to the appellant, but affirmed in respect to appellant’s socket No. 1. The controversy over the socket No. 1 is of so small account that the general rule in regard to costs should not be varied.
67 C. C. A. 564.