board and the annunciators and answering jacks of about 200 of the subscribers. If there were 1,200 to be served by the entire board it might consist of six sections, upon each of which a line jack would be placed connected with the line of each subscriber, while each section would contain the annunciators and answering jacks of only about 200 members of the exchange. For instance, the first section might contain the annunciators and answering jacks of subscribers numbered from 1 to 200, inclusive; the second section those numbered from 201 to 400 inclusive; the third section those numbered from 401 to 600, inclusive. But each section would contain a line jack for every subscriber served by the entire board. The annunciators and answering jacks are divided between the sections in order to enable a single operator to attend to the calls of all the subscribers whose annunciators appear upon a section, and it is impracticable for a single operator to serve more than 200 subscribers. In the combination of Seely the annunciators on each section of the board are formed into a group, and the answering jacks of the subscribers represented by these annunciators are formed into another group by themselves and placed upon the same section. The members of the several groups of annunciators, and their corresponding answering jacks, are placed in the same relative positions to each other on each section, so that the method of finding and plugging into the answering jacks of the subscribers is uniform on all the sections of the board. When a shutter drops the operator plugs into the answering jack of the subscriber who calls, switches her telephone into the circuit, answers the call, learns the number of the subscriber sought, plugs into the line jack of the latter upon her section of the board, calls the subscriber wanted, and then switches her own telephone out of the circuit and leaves the two subscribers electrically connected, so that they may talk with each other. Prior to Seely's invention the line jacks on a multiple switchboard performed the function of Seely's answering jacks. When a shutter dropped the operator was required to find the line jack of the calling subscriber, to plug into that, to answer the call, then to find the line jack of the subscriber called, to plug into this line jack, to call him, and then by means of a cord with plugs at each end to connect the line jack of the caller with that of the called. Seely's invention consisted in the combination with the line jacks and groups of annunciators on the sections of a multiple switchboard of groups of answering jacks in such a way that each group of annunciators should have on the same section with it a group of the answering jacks of the subscribers represented by the annunciators thereon, so that the members of the corresponding groups of annunciators and answering jacks should be placed in uniform relative positions to each other on every section of the board. The essence of his invention was the convenient and uniform grouping of the annunciators and their corresponding answering jacks relatively to each other. Some of the obstacles removed and some of the advantages derived from this combination were these: In the absence of Seely's groups of answering jacks, when a shutter dropped and disclosed the number of a cailing subscriber, the operator was obliged to search through all the line jacks on her section, perhaps 1,200 in number,

to find the jack of the subscriber and to plug into that line jack in order to answer the call and to learn what was wanted. The desired line jack might be at some distance from the annunciator, and time and thought would be required to find it and to make the connection with it. Seely placed the groups of answering jacks near to their corresponding groups of annunciators. No group contained more than 200 jacks. When a subscriber called, the operator was relieved from a search for his line jack among a very large number, perhaps 1,200 similar jacks. She was only required to find and connect with his answering jack in order to learn his wants. As the number of these answering jacks was small, not more than 200, and they were uniformly placed and arranged relatively to their corresponding annunciators, a little experience soon enabled her to plug into the right answering jack automatically, without noticing the number described by the annunciator or thinking of or searching for it. This automatic and almost involuntary habit of the operators greatly increased the accuracy, speed, and efficiency of the service. It is sometimes necessary to transfer an operator from the section at which she has been employed to another section of the board. The line jacks corresponding to the group of annunciators on one section of a multiple board are in a different place upon the section from those which correspond to the group of annunciators on any other section of the board. In the absence of Seely's uniformly grouped answering jacks an operator would become accustomed to find the line jacks through which she must answer on a certain part of her section. When she was sent to another section she would be compelled to devest herself of the habit of plugging into that part of the section to answer the calls, and to learn to answer them through another set of jacks on another portion of the section. This condition of things necessarily resulted in delay, mistakes, and confusion. But when the groups of answering jacks were combined with their corresponding groups of annunciators in the way discovered by Seely, an operator who once learned the relative positions of the answering jacks and their corresponding annunciators on one section knew their relation upon all the sections, because their relative grouping was uniform throughout, and she could serve equally well on any section of the board. It is frequently convenient, and sometimes necessary, to change the service of a subscriber from one section of a board to another, because the subscribers represented on a certain section require more service, while those assigned to another require less service, than a single operator can render. The only convenient method to make this transfer without the uniformly grouped answering jacks of Seely was by changing the number of the subscriber, and many subscribers seriously objected to changes in their numbers, because these changes. involved a change of thought and habit, and because their numbers came to have value in trade. Seely's groups of answering jacks enabled the telephone companies to transfer the service of a subscriber from one section to another by a simple transfer of his annunciator and answering jack without any change in his number, because the uniform relation of the respective members of the corresponding groups of annunciators and answering jacks renders numbers upon

them unnecessary, and enables the operators to serve the subscribers with equal facility in any section to which they are assigned. Seely's groups of answering jacks diminished the congestion of lines and the matting which resulted from the connection of the line jacks with each other for purposes of communication. Without these answering jacks the line jack of the calling subscriber was first used for answering purposes, and was then connected with that of the subscriber called by a cord attached to a plug at each end. One of these plugs was inserted in the line jack of the calling subscriber, and the other in that of the subscriber called. Every connection made stretched a cord across a portion of a section of the board. In busy times these cords became numerous and matted. When Seely's groups of answering jacks were placed upon the board it became unnecessary to use the line jacks to answer calls, and the electrical connection of the caller with the called could be effected by connecting the answering jack of the former with the line jack of the latter, thus greatly relieving the congestion of the cords upon that portion of the section covered by the line jacks. These, and perhaps other, benefits conferred by this invention, commended it to the trade, and it immediately went into general, perhaps almost universal, use in all the large telephone exchanges of the country. Its utility is not denied. But it is earnestly contended that it was not patentable because any mechanic skilled in the art could have produced it without the exercise of any of the genius of the inventor. In support of this contention it is argued that letters patent No. 246,481, issued on August 30, 1881, to Eldred and Durant for improvements in telephone exchange systems and apparatus, discloses an annunciator, a line jack, and an answering jack for each subscriber, while letters patent No. 258,234, Issued on May 23, 1882, to M. G. Kellogg, shows a board containing line jacks and other boards provided with annunciators and their corresponding answering jacks, so that when Seely made his combination there was nothing left for him but the grouping of the answering jacks, and that there could have been no invention in the mere arranging of these jacks to correspond with the annunciators. A careful examination and analysis of the devices described in these patents discloses the fact, however, that neither of them either suggests the improvement of Seely or was designed to or capable of removing the evils which he sought to remedy. His invention was directed to the improvement of the service on a multiple switchboard, to an effort to enable one operator to render more speedily and efficiently all the service required by the subscribers intrusted to her care. Neither of the patents cited contemplates the use of a multiple switchboard with the improvements it describes, and neither of them in any way suggests that separate but uniform grouping relative to each other of corresponding annunciators and answering jacks which constitutes the essence of Seely's invention. The patent to Eldred and Durant shows a signal box and two spring jacks at the exchange for each subscriber. The operator is provided with a telephone connected by two wires with the metal sides of a wedge. The sides of this wedge are insulated from each other by a piece of rubber between them. By the insertion of this wedge under a spring jack

upon the side of the signal box the telephone of the operator is switched into the circuit, so that she may answer a call or notify a subscriber that some one has called him. When the wedge is withdrawn the circuit is completed through another spring jack, and by the insertion of the plugs of a cord circuit under two of these spring jacks the lines of two subscribers may be electrically connected. It is true that this patent shows an annunciator, a spring jack for answering calls, and a second spring jack for connecting with other subscribers for each of the subscribers to the exchange. But the device lacks Seely's uniform grouping, his combination with and improvement on the multiple switchboard, his means of accomplishing his purpose, and it is utterly incapable of performing any of the functions or accomplishing any of the objects of his invention.

Nor is the contrivance of Kellogg much nearer in function, means, or effect to the combination of Seely. It has no multiple switchboard. It consists of one board, on which the line jacks of all the subscribers are placed, and two or more other boards, among which the corresponding annunciators are distributed. In practice each annunciator board is served by one operator, while the board which contains the line jacks is worked by switchmen who connect the line jacks as directed by the operators at the annunciator boards. Each operator at an annunciator board is provided with a telephone which she may switch into the circuit of a subscriber by pressing the plug attached to it upon a connecting bolt near or on his annunciator. Conceding that these connecting bolts are the equivalents of Seely's answering jacks, we have here a single switchboard containing the line jacks of all the lines and several switchboards, each one of which is provided with annunciators and corresponding answering jacks of a part of the lines. But here is no multiple switchboard, no separate and uniform grouping of corresponding annunciators and answering jacks, but an answering jack on each annunciator; no purpose or effort to concentrate all the work of serving each line in the hands of a single operator, but a successful effort to divide it between two or more operators, thereby increasing delay, confusion, and chance of mistake. The combination of this patent does not teach the way to remove the evils which Seely remedied or to reach the desiderata which he sought. The patent describes no means which could perform the function of improving the service of the multiple switchboard which the combination with that board of corresponding annunciators and answering jacks separately but uniformly grouped upon each section of the board effected. Our conclusion is that the patents of Eldred and Durant and of Kellogg do not anticipate or suggest the combination of Seely. The question recurs, did it require any exercise of the inventive faculty to produce his combination, in view of the state of the art which the multiple switchboard of Firman and these patents disclose? No one has yet been able to formulate a test whereby a line of demarkation between the products of the inventor's intuition and the results of the skill of the mechanic may be surely drawn in all cases as they arise. That question is and always must be left for determination by a careful exercise of the judgment, enlightened by a knowledge of the state of the art and of the advance in it which the device in question marks, and guided by the

established rules and principles of the law. The two classes of cases led by Atlantic Works v. Brady, 107 U. S. 192, 200, 2 Sup. Ct. 225, 27 L. Ed. 438, and Loom Co. v. Higgins, 105 U. S. 580, 591, 26 L. Ed. 1177, have been again cited and reviewed for our guidance, and have been carefully considered in reaching our conclusion. A plausible and persuasive argument may be made that this combination falls under either class of cases, that it might have been and was produced by the skill of the trained mechanic or by the intuitive genius of the inventor. The patent which describes it, however, raises a presumption in favor of its novelty and its patentability. It was a new combination. No such separate yet uniform grouping of corresponding annunciators and answering jacks with a multiple switchboard had ever been made or used before Seely conceived and described it. That combination was not a pioneer; perhaps it was not a great invention. But it discharged the functions of the multiple switchboard, its annunciators, and switches more speedily and efficiently than they had ever been performed without it, and a new combination of old elements by which an old result is attained in a more facile, economical, and efficient way may be protected by a patent. National Hollow Brake-Beam Co. v. Interchangeable Brake-Beam Co., 106 Fed. 693, 707, 45 C. C. A. 544, 557; Seymour v. Osborne, 11 Wall. 516, 542, 20 L. Ed. 33; Gould v. Rees, 15 Wall. 187, 189, 21 L. Ed. 39; Thomson v. Bank, 53 Fed. 250, 252, 3 C. C. A. 518, 520, 10 U. S. App. 500, 509.

The combination had great utility. It went into immediate and general use. While this fact is insufficient in itself to sustain a patent where the machine or combination is clearly without novelty, yet where the question of novelty is fairly open under the law, the fact that a patented device or combination has displaced others which had previously been used to perform its function and has gone into immediate and general use is pregnant and persuasive evidence that it involved invention. National Hollow Brake-Beam Co. v. Interchangeable Brake-Beam Co., 106 Fed. 693, 707, 45 C. C. A. 544, 558; Smith v. Vulcanite Co., 93 U. S. 486, 495, 23 L. Ed. 952; Loom Co. v. Higgins, 105 U. S. 580, 591, 26 L. Ed. 1177; Manufacturing Co. v. Adams, 151 U. S. 139, 143, 14 Sup. Ct. 295, 38 L. Ed. 103; Magowan v. Packing Co., 141 U. S. 332, 342, 12 Sup. Ct. 71, 35 L. Ed. 781; Graphophone Co. v. Leeds (C. C.) 87 Fed. 873; Topliff v. Topliff, 145 U. S. 156, 164, 12 Sup. Ct. 825, 36 L. Ed. 658. It cannot be truthfully said that it is so clear that there was no invention in Seely's device that the question whether or not it was the product of the inventive genius is not open for consideration under the law.

Again, the court below has considered this question in the light of the state of the art, and of the conflicting testimony of the witnesses, and has decided that Seely's combination was an invention. This conclusion is presumptively correct, and ought not to be reversed unless an obvious error has intervened in the application of the law, or some serious mistake has been made in the consideration of the facts by the circuit court. National Hollow Brake-Beam Co. v. Interchangeable Brake-Beam Co., 106 Fed. 693, 716, 45 C. C. A. 544, 567; Mann v. Bank, 86 Fed. 51, 53, 29 C. C. A. 547, 549, 57 U. S. App. 634, 637; Tilghman v. Proctor, 125 U. S. 136, 8 Sup. Ct. 894, 31