This action was brought to restrain the infringement of United States letters patent No. 470,813, for improve*63ment in electric railway signal apparatus, issued March 15, 1892, to A. J. Wilson, inventor, and by him assigned to the complainant. A diagram representing block sections in series on a line of railway, in-strumentalities consisting of batteries, relays, rail and signal circuits showing the arrangement of the apparatus, accompanies and illustrates the patent, and as it is thought indispensable to a complete understanding of the invention it is herewith reproduced.
*64The specification says:
“The greater number of blocks naturally found on any given installation in practice would be equipped by duplicating indefinitely often the arrangement of the circuits and of instruments shown in connection with the intermediate blocks of the series represented, as, for instance, with blocks marked C, D, E and E. In the installation represented in the drawings the length of the track is divided into separate sections, which are insulated from each other in the well-known way, and are marked in the drawings A, B, O, D, E, E, and G. The trains are supposed to be moving upon the track in the direction indicated by the arrows. Each track section has a battery near the end of the same, marked, respectively, A', B', O', I>', E', P', and G', and in the same familiar way the rails of each section form a part of the circuit of the battery pertaining to that section.”
Then follows a description of the various operations of the train entering a track section. The claims of the patent, two in number, read:
“(1) In a block signaling apparatus, the combination, with a series of normally-closed rail circuits, of a series of normally open or broken signaling-circuits, each signaling-circuit including a normally-open circuit-closer constructed to be closed by the action of the train upon one of the rail-circuits, and also including a normally-closed circuit-breaker constructed to be opened by the action of the train upon the next succeeding rail-circuit of the line, substantially as and for the purpose set forth.
“(2) In a block signaling apparatus, the combination, with a series of normally-closed rail-circuits, of a series of normally open or broken signaling-circuits, each signaling-circuit including a normally-open circuit-closer constructed to be closed by the action of the train upon one of the rail-circuits, and also including normally-closed circuit-breakers constructed on each alternate signaling-circuit to be opened by the action of the train upon the two next succeeding rail-circuits of the line and on the intermediate signaling-circuits to be opened also by the action of the train on the third succeeding rail-circuit of the line, substantially as and for the purpose set forth.”
The first claim, which alone is involved, is broad in its scope and defines invention. The second claim is for the specific method of construction of the signal apparatus in controversy. The defenses are anticipation, noninvention in view of the prior art, and noninfringement.
The practicability of the signal system under consideration depends upon the signaling-circuits which are acted upon in block sections in such a way that the alternate signaling-circuit co-operates with two block sections, and those interposed between co-operate with three block sections. This feature of the apparatus is explained by defendant’s expert witness Waterman, who testifies:
“It is characteristic of the system shown that on each block or track section of the indefinitely duplicatable portion a train acts to bring to safety one signal in advance, at the entrance of the next adjacent block or track section, and to set to danger one signal standing at the rear of the train at the entrance to the block in which the train finds itself, and also to hold at danger further to the rear two signals on alternate blocks, and on intermediate blocks one signal in addition to that immediately in its rear.”
The first question that requires consideration is whether, in view of the prior art, the combination of the rail and signal circuits, batteries, and relays in series arranged to co-operate so as to protect every •block in the series of blocks involved invention. Before discussing the signal apparatuses of the prior art and the improvement in contro*65versy, it is desirable that the objective use of the home and distant signals, which were commonly known in the art, should be thoroughly understood, for both are used in the block-signal systems of complainant and defendant. The distant or cautionary signal consists of a notched blade or arm mounted on an upright post, and is located in the rear of the home signal. It is designed to transmit information to the engineer in charge of an approaching train of the condition of the block section before the home signal is visible to him. The home signal is a narrow blade or arm mounted on a post at the entrance of the block which it guards. At night colored lights are used in place of the pivoted blade. It is also worth noting that railway signals generally, in addition to telegraphic signals, consist of interlocking signals which are used principally in the equipment of a railroad yard to control switches and crossings, while the block signals are operated automatically by electricity to protect a specified distance of track by keeping trains traveling in the same direction a certain distance apart. There were two different methods of automatic signaling known in the art at the date of the, invention in suit, namely, the normal clear or safety and the normal danger systems. We are primarily concerned in this case with signaling-circuits and rail-circuits that distinctively operate upon the normal danger plan.
To prevent confusion it probably should be stated that a normal bias danger signal implies a movement of the signal blade through gravitational tendency to a horizontal position. The phrase “normal danger system” more specifically has reference to a signaling device which is supplied with normally-closed rail-circuits, normally-open signal-circuits including a normally-open circuit-closer, so arranged that the current which upsets the signal from a right-angle position to the post upon which it is mounted flows only when a train approaches or occupies the track-circuit section to its rear. In short, if when there is no train present on the section the blade is at right angles to the post, the signal is of the normal danger class, and, when the signal is in an inclined position in the absence of the train, it is of the normal safety class. Thus we find a normal danger bias signal in a normal danger apparatus, the signal blade being firmly held in the danger position when no trains are present, and dropping to safety when the wheels and axles of the train establish a path of low resistance, causing the current to energize the relay to break or close th?. circuits. The normal danger bias tendency is likewise present in the, normal safety system, for when the magnetization is withdrawn bj the presence of a train the signal blade recedes from its angular or safety position to a position at right angles to the post upon which it is mounted.
The normal position of the signal while the block was unoccupied by a train was early in the history of the automatic signaling art regarded as a problem of perplexing importance, and it was mooted whether the semaphore arm should point to safety, indicating protection to a train approaching the block, or whether it should normally point to danger, indicating obstruction ahead, thus stopping the movements of the train unless the signal dropped to safety upon its approaching the block section, In the normal clear system of signaling, *66energy or force was required to retain the signal in its normal safety hanging or inclined position until the presence of the train in the block actuated it, while in the normal danger system the utilization of energy or force to hold the signal blade in a horizontal position became essential only when its hanging to a safety or oblique position by an approaching train was demanded. This automatic method of signaling was a part of the art as it existed prior to the invention in suit, and by its adaptation signalmen or any manual efforts were not required It was customary, and still is, to protect separate track sections and switches and railroad crossings by manual application of the horizontal or danger signal, as, for instance, an operator in charge of a track section on the approach of a train, having previously learned by a method of signaling that the frack ahead was clear, would lower the danger signal from a horizontal to an oblique position, the commonly accepted position to indicate a clear track. This having been the commonly accepted method of manual signaling, the complainant argues that the normal danger adaptation in suit by which the signal is, in the absence of a train actuating it, in the normal position, was recognized by signaling engineers as a distinct advance over the normal clear system. The defendant, in answering, argues, inter alia, that at the date of the Wilson invention both systems were thoroughly understood by the art, and simply were different methods accomplishing identical results; but of this I am not convinced.
Considering more particularly the question of anticipation, four patents are relied upon to negative the validity of the patent in controversy. These patents and publications prove that it was not new at the date of the invention in suit to actuate signaling apparatus by the presence of a train in a block section; that it was a common expedient in automatic signaling to divide a railway trade into blocks of different lengths, and place home and distant signals alongside the track section which were operated by the presence of the train; that it was old in the art to control the signal semaphore blade or arm by pressure or counterpressure to enable its dropping to an oblique position, and, indeed, it was old to permit the pivoted arm when released from the current pressure to assume a horizontal position through gravity owing to the presence of the train in the block section; that the magnetization of the relay to overcome the controlling or opposing force holding the signal in position was also understood in the art. The proofs, however, establish that to clear the signal ahead by the action of the train before it reached the entrance to the block, and at the same time protect said train by a rear signal to danger in a normal danger system, was a new achievement. The problem for solu- • tion was to seasonably clear the forward signal as the train approached the block without such function retarding or interfering with the signal to danger in the rear of the train as it left the rear block. The prior patents did not protect the block or afford means for adequately clearing the signals ahead and to the rear of the train.
In the Robinson patent, No. 130,661, of August SO, 1872, which was a normal safety system, the current passed through a continuously-closed signal-circuit to maintain the semaphore blade in a hanging position. The flow of the current was diminished and diverted *67from the relay when the rail-circuits were connected by the wheels and axles of a train. The relay was demagnetized, but again became magnetized upon the instant the train left the track section. These functional results were in consequence of the attraction of the armature by the magnet as the current passed through the circuit, magnetizing the relay and later demagnetizing it. In the normal safety type of signals, in the absence of the train from the track section, and while the armatures are attracted toward the magnet, the signal circuits are closed, and the blade descends downward to safety from its horizontal or danger position; but when the train on the train-circuit demagnetizes the relay, the armature recedes from it, opening Ihe signal-circuit and sending the signal upward to danger, and upon the train leaving the track section the relay is again magnetized and the signal assumes its normal position. Defendant claims that the essential difference between this feature of the Robinson invention and the later invention of the Gassett patent, No. 351,867 (the normal danger type of signaling), is simply that in the earlier type the relays close the signal circuits when they are magnetized, and in the latter they are closed or demagnetized by a back contact of the armature; that the primal difference between the normal dear and normal danger methods of signaling, as claimed in the patent in suit, consisted in adding to the latter a circuit-closer permitting the train to act upon it before reaching the signaling station. The expert witnesses for the defendant are strongly of the opinion that the addition of a circuit-closer to the normal danger system was plainly obvious in View of the Robinson and Gassett patents. Importance is attached to Robinson’s reference in his specification that the signal-circuit may be either open or closed. In his drawing accompanying the patent both normal bias to danger and normal bias to safety signals are indicated. The specification does not show, however, a successfully operating normal danger signal. Indeed, the art was familiar with a signal system of the normal safety type operated with a normally open signaling-circuit. This appears by the testimony of complainant’s witnesses, Blodgett and Dane, and I am unable to perceive that Robinson in the then state of the art implied the employment of an open signaling-circuit in connection with a normal danger plan. This view also finds support in a paper read in April, 1886, 14 years after the patent to Robinson was issued, by Mr. Hardy, and published in the Electrician and Electrical Engineer, from which the following may be quoted:
“Most automatic signals are now arranged to stand normally all clear and show red when there is danger in the way. This may be some time improved so as to stand normally at danger, showing red, and open to let an advancing train enter, which would he a more logical system, but its adoption is just now attended hy some undesirable complications which hardly warrant such a change.”
Accordingly it may be safely decided that the addition of a circuit-closer as used in the patent in suit, and an arrangement of the signaling-circuits as therein described, to enable the trains to operate the signals ahead and to the rear before entering the block section, was not an obvious expedient. On the contrary, the proofs show that the skilled in the art groped in the dark and persistently endeavored to *68develop the art to secure a satisfactory normal danger system of signaling.
Eight different normal danger systems are shown in the patents issued to Spang, and in each the clearing sections are shown to be from 60 to 150 to 500 feet in length. In patent No. 164,227, the theretofore unprotected clearing sections for the advance signals were' protected, but owing to the magnetization of a magnet, which in this type of signaling was necessary in order to hold the signal in its danger position, insufficient protection was afforded. The evidence of the complainant also indicates that the signaling devices of Spang were impracticable because of the shortness of the clearing sections. The Spang type of signaling is distinguished from that in suit in that it required current to maintain the signal in its danger position, and was based upon an obviously different principle than that by which complainant’s signals operate. That an absence of complete protection to the train section made the system an impracticable one is practically conceded by the defendant. Such a system, testifies the witness Dodgson, is not only unsafe, but it tends to produce, instead of prevent, collisions between trains. None of the systems described in the Spang patents have ever been in practical use, and I am satisfied that they taught little to the art. The shortness of the clearing section was a vital defect of construction in the prior art, for in such systems there necessarily would have been a retardation if not a complete stopping of trains at the signal post before entering the block section. The impracticability of the prior signaling devices and the absolute failures of the prior art rightly are attributed by the expert witnesses for the complainant to insufficient clearing space and imperfect signaling facilities in the rear of the train in the clearing section; while in the Wilson patent the clearing space behind the signal is employed for effectuating the successful signaling requirments which result in giving protection from other trains approaching in the rear on the same track.
The Gassett patent, No. 251,867, of 1882, for a normal danger system with an overlapping feature, was the subject of the closest controversy between the expert witnesses on both sides. The specification shows among other things that two signals were controlled by one direct circuit section, and that it was the object of the patent to have such signals safeguard two or more track sections. The defendant claims that these patents indubitably establish that Wilson was not the first to discover means for controlling a danger signal by more than one track section, and that the method of duplicating his unit for a single or double track is so clearly described in such patents that the system of Wilson was obvious to signaling engineers. None of the Gassett systems went into practical use excepting the first, which, however, was soon discontinued. In the Gassett, as in the prior normal danger devices, the clearing sections were left unprotected, and therefore they were impracticable. In his third normal danger system he made provision for safeguarding the entire track, but as the clearing was by a system of track instruments (treadles alongside the track), ■with the danger signal to be affected by the rail-circuit, it cannot be *69considered to anticipate the patent in suit, which concededly was based upon another and different principle.
It is probably undeniable that when the prior art is considered in its entirety the achievement oí Wilson was not of an amazing character, but by his conception of the aim, the method, and the arrangement of the instrumentalities he solved what at that time was regarded in the art as a serious and difficult problem. If I am correct in my estimation of, the impracticableness of the prior normal danger devices, as shown by the exhibits herein and the oral evidence, and that they were incapable of achieving the same functional result, the differences between them and the invention under consideration are clearly marked and justify applying the principle announced in the following cases: Loom Co. v. Higgins, 105 U. S. 580, 26 L. Ed. 1177; The Barbed Wire Patent, 143 U. S. 283, 12 Sup. Ct. 413, 36 L. Ed. 154; Brunswick-Balke-Collender Co. v. Thum, 111 Fed. 904, 50 C. C. A. 61; Doig v. Morgan Machine Co., 122 Fed. 460, 59 C. C. A. 616; Brown Bag Filling Machine Co. v. Drohen (C. C.) 140 Fed. 97. It is to be considered that the adaptation of the normal danger systems by railroads, in some instances to the exclusion of the normal clear system, has materially assisted in recent years in developing such system, and would seem to be ample testimony of the usefulness of the improvement. Old elements found in the prior signaling devices unquestionably were brought together by the patentee, tie was not the creator of all the elements of the combination, but I think it is established that by his arrangement or rearrangement of the rail and signal circuits and use of circuit-closers the elements of claim 1 perform a new and different operation, and hence protection of the patent should not be withheld. Electric Signal Co. v. Hall Signal Co., 114 U. S. 87, 5 Sup. Ct. 1069, 29 L. Ed. 96. It is not intended to hold that the prior art was absolutely worthless. It taught considerable, but the normal danger products were evidently fragments, and of value only as indicating an endeavor to make them practicable on the high plane required by railroads using them. This, then, is a case where the inventor is shown to have taken the last step which accomplished the desired result, and in consequence he became a pioneer in the art of normal danger signaling. He solved the problem in a simple enough way, but where others tried and successively failed he succeeded, and his endeavors, therefore in the field of railway signaling most puissantly point to invention entitling his claim to a'broad construction. Cimiotti Unhairing Co. v. American Fur Refining Co., 198 U. S. 399, 25 Sup. Ct. 697, 49 L. Ed. 1100; Computing Scale Co. v. Automatic Scale Co., 204 U. S. 609, 27 Sup. Ct. 307, 51 L. Ed. 645.
Does the defendant inflinge the claim in .suit? The specification describing the combination of claim 1 states:
“It will thus be observed that in the broad invention the rail-circuits are normally closed and the signaling-circuit —the circuits directly operating each signal — are normally open or broken, being normally open or broken at a circuit-closer which is constructed to be closed by the action of the train upon one of the rail-circuits — the one at or near the end of which the signal operated by that signal-circuit stands — and that the signaling-circuit is also capable of being broken by one or more normally-closed circuit-breakers which are con*70structed to be opened by tbe action of a train on one or more ot tbe next succeeding rail-circuits of tbe line. It will also be observed that in tbe invention as embodied in its preferred form tbe rail-circuits are, as before, normally closed, and tbe signaling-circuits normally open and open as before at a circuit-closer wbieb is constructed to be closed by tbe action of a train upon one of tbe rail-circuits, and that the signaling-circuit is also capable of being broken by normally-closed circuit-breakers which are constructed to be opened by tbe action of a train on either of tbe next two succeeding rail-circuits of tbe line, the alternate signaling-circuits having circuit-breakers constructed to be opened by the action of a train on either of the next three succeeding rail-circuits of tbe line.”
And further:
“It is not essential to tbe main invention, however, that tbe intermediate signals should be employed as cautionary signals, or be affected by tbe action of tbe train on the second or third succeeding rail-circuits, and in such event tbe connecting circuits and circuit-breakers affecting that end would be omitted.”
It will be noted that the structure of the claim, besides having normally-closed rail-circuits and normally-open signal-circuits, embodies a normally-open circuit-closer made by the armature of the magnet and back contact together with a normally-closed circuit-breaker made by the armature and the magnet at its front contact. When a train is on each rail-circuit the signal immediately goes to clear or safety, and correspondingly a signal blade to the rear of the train goes to danger, preventing its clearance for following trains. It' will not be necessary to recount the many different instrumentalities by which the successive operations of the signaling are accomplished. In my judgment the defendant’s system, though constructed somewhat differently, is nevertheless readable upon the claim.
The defendant contends that the system employed by it is radically different from that of complainant, that the specific claim of the patent is the only conception of the patentee, and that his home and distant signals are entirely dependent upon a unitary structure with signals having separate circuits; that the defendant’s signals which are not mounted on the same post are in a series in the same circuit; that its signals are set to safety by the train upon the third or fourth section in advance, while in the patent in suit the blade drops to safety by the train when it is in the block preceding the signal post; that in defendant’s system the distant signal sets to safety after the home signal, while in that of complainant the home signals are at safety after the distant. A further distinction insisted upon is that two normally-open circuit-closers in the Wilson device co-operate in the same series, while in defendant’s the signals shift to safety by a normally-open circuit-closer in the signal-circuit, and that some of the signals are placed to safety owing to a circuit-closer in the shunt circuit which contracts the primary circuit; that defendant’s method of duplicating the signal is different; and, finally, that the patent is devoid of showing a duplicatable system. A complete description of defendant’s apparatus based on Wagner diagram No. 41, and taken from complainant’s brief, follows:
“Tbe borne signal and its distant signal are controlled by a signaling-circuit energized by the battery B, this signaling-circuit being held normally open at tbe circuit-closer, b. Tbe signals stand normally at danger, and remain in that position at all times except when actuated to give a clear indication to an approaching train. Tbe rail-circuits are all normally closed; and in tbe part of *71defendant’s! circuits shown in this diagram each block is divided into three interconnected rail-circuits. When the approaching train is on the rail-circuit 02, which extends for 2,800 feet in rear of the distant signal DSo, its wheels and axles shnnt the relay, 7c, of that rail-circuit and thereby close the normally-opon circuit-closer, b, thereby closing the normally open signaling-circuit, if the track ahead is clear of trains. The first result of the closing of tlio signaling-circuit is that the home signal HS5 goes to clear, but this movement to clear varies the resistance of the signaling-circuit by the closing of a short branch through the contact, d, so that the distant signal is moved to clear position. Thus the closing of the normally-open circuit-closer, b, causes a dear indication of the home signal, if the conditions of traffic permit a clear indication, and also causes a clear indication of the distant signal if the home signal has cleared. The one normally-open signaling-circuit effects both operations, and the proximate cause of ihe clear indications of the signals is the closing of the normally-open circuit-closer, b.
“Considering now the distant signal DS8, we find that when the train is on the rail-circuit D next succeeding the rail-circuit 02, its wheels and axles shunt the relay II of the rail-circuit D, and open the normally-closed circuit-breaker, f, with the result that the distaut signal is moved to and held at the danger position so that it cannot clear for a following train.
“The home signal is not moved to danger until the train reaches it, and the armature movement which opens the normally-closed circuit-breaker, f, to put the distant signal to danger, also closes a normally-open circuit-closer, a. to continue the clear indications of the home signal. This circuit-closer, a, closes a normally-open branch of the signaling-circuit, excluding the distant signal. As the train proceeds through the block between the distant signal and the home signal, these signaling conditions are maintained, the passage of the train successively over the interconnected rail-circuits of the block continuing the de-energized condition of the relay, 11, of the rail-circuit D through nor-aily-closed circuit-breakers, each located in a rail-circuit in rear and controlled by the relay of the rail-circuit in advance, so that the presence of the train upon the rail-circuit D2, which is immediately in rear of the home signal, shunts the relay of the rail-circuit D2, and thereby maintains the de-energized condition of the relay, 11, with consequent closed condition of the normally-open circuit-closer, a, and consequent clear indication of the home signal HS5. When the train is upon the rail-circuit E, which is immediately in advance of the home signal, and next succeeds the rail-circuit X)2, it shunts the relay, 14, of the rail-circuit E, and opens the normally-closed circuit-breaker, e, whereby the home signal is moved to and held at the danger position so that it cannot clear for a following train.”
The rail-circuits in the defendant’s device likewise operate to protect the rear of the train and to clear the ahead signal, as in the apparatus of complainant:
“When a train is on the rail-circuit C2, it causes the home signal IIS5 to give a clear indication and the distant signal DS,‘3 to give a clear indication, and it prevenís a clear indication of the home signal HSI and distant signal 3>S1. When a train is on the rail-circuit D, it causes the home signal HS5 to give a clear indication, and it prevents a clear indication of the distant signal 1)S3 and home signal HS3, and also of the distant signal DS1. When the train is on the rail-circuit D.i, its effect is exactly the same as when on the rail-circuit D, causing a clear indication of one signal ahead, and preventing a clear indication of throe signals in roar. When the train is on rail-circuit 1)2, it causes the home signal HS7, homo signal 1IS5, and distant signal DB5 to give clear indications, and prevents a clear indication of home signal llsSÍS, distant signal DS8, and distant signal ESI. When the train is on the rail-circuit E, it causes the home signal IIS7 to give a clear indication, and prevents a clear indication of the home signal HS5, distant signal J)Sñ. and distant signal 1>S3.”
The diagram in evidence of defendant’s structure indicating the combination of circuits and apparatus for operating home and distant signáis with numerical references is herewith reproduced.
*72
The elements of the claim in controversy are at present in defendant’s system, and in my judgment they are operated to produce the double function of clearing to safety ahead and to danger in the rear of the train. In the patent to Westinghouse, No. 360,638, of April 5, 1887, for a normal safety home and distant signal system, the same end is not achieved as in the patent in suit. True, it shows normally-closed rail-circuits, but the specification does not indicate that such circuits could be functionally used as normally-open signal-circuits of the normal danger type of signals. Nor do I think, for the reason already dwelt upon, that changes and alterations from the Westinghouse normal safety system to the system in suit was obvious to the skilled in the art. That the home and distant signals of the defendant are mounted on the same post as in the Westinghouse patent is inconsequential, since such variation from the patent in suit nevertheless enables the distant signal to be functionally operated by taking the substance of the Wilson invention.
The next important point is that the claim in terms is limited by the words “substantially as and for the purpose set forth” to the device illustrated in the drawing and specifically described in the. specification. Concededly the patent is for a block-signaling system, and it may be supposed, in view of the primary character of the patent, that it includes a stretch of track protected by a successive number of blocks. In fact, the broad claim, as allowed without objection or criticism b)'- the patent ofiice, plainly is for a combination of a series of a specified type of signaling-circuits and rail-circuits in a block-signaling device. It is shown, as has been indicated, that when the different elements were combined as set forth in the specification and claim they co-operated with one another and with the rail and and signal circuits to create a block system which presumably included protection for a multiplicity of track *73sections. Such co-operation of old elements imparted a new method of operation producing a new and highly meritorious result, and therefore the authorities cited to show that a mere duplication of parts is not invention have no application. The defendant has appropriated the invention owned by the complainant, and the installations complained of in the bill were not infringements of claim 1.
A decree for injunction and accounting may be entered. The defendant in its brief has objected to paying the entire expenses of taking the rebuttal testimony given on behalf of the complainant. Complainant’s record contains much repetitious, if not irrelevant, testimony, and the large niunber of drawings and exhibits were not necessary to a complete presentation or understanding of the questions involved. Therefore, it is thought proper that the complainant should pay a portion of the costs and disbursements, the amount which it shall pay to be determined upon settlement of the decree.