Gamewell Fire Alarm Telegraph Co. v. Mayor of Bayonne

194 F. 147 | D.N.J. | 1912

CROSS, District Judge.

Two patents belonging to the complainant are alleged in the bill of complaint to have been infringed by the defendant: The answer denies infringement, and alleges the invalidity of both patents. They were issued February 4, 1896, one, No. 553,873, to John J. Ruddick, the other, No. 553,839, to Frederick W. Cole. The mayor and council of the city of Bayonne is only a nominal defendant; the signal boxes in its use, it is stipulated, having been made and sold to it by the Star Electric Company subsequent to the acquisition by the complainant of the patents in suit, but prior to the verification of its bill of complaint.

[1] Both patents are for a “noninterference signal apparatus”; in other words, for fire alarm signal boxes, such as are in use in almost every large city. In practical use they are more often operated by persons unskilled than skilled, and are designed when operated to transmit to a central station the number of the box at which the alarm is given. The signals are transmitted electrically over a single circuit to which there may be, and not infrequently are, a large number of boxes attached. When the lever or hook of a given box is “pulled,” the circuit is opened and closed a definite number of times, with definite time intervals between to represent the number of the box. The receiver at the central station may be so arranged as to indicate the number of such box in different ways, as by marks upon paper, striking a bell, or blowing a whistle. It is obvious that, when two or more boxes located upon a single circuit are pulled simultaneously, the signals, unless in some way prevented, would intermingle and be confused and unintelligible at the central station; thus causing what is known in the art as “interference.” Speaking very generally, noninterference is accomplished by the presence in the circuit of an electro-magnet, which, with added means, controls and withholds the transmission of the signal, if the line is not in a condition successfully to transmit it at the time when the box is pulled. The elimination of interference was a problem which seems to have engaged the attention and perplexed the minds of inventors for many years. One of the complainant’s experts has shown, and his position is justified by the record, that the development of the art might, with propriety, be separated into four divisions — the first, being that in which no noninterference devices were used; the second, that in which a noninterference magnet was used; the third, that in which a time element, as well as a noninterference magnet was used; and the fourth, that in which perfect noninterference was accomplished so that no signal once given would be lost or confused, lie also mentions the various patents, which in his judgment represented the first period, which in point of time extended to about 1870. He pursued the same course with reference to the second and third periods, the former of which extended to 1880 and the latter to the time of the patents in suit; while the fourth period was made to embrace the time since then. He also considers somewhat in detail the numerous patents representing the different stages of the art, as above mentioned. It is unnecessary, however, to separately consider them, if for no other reason than that they, for the most part, were not relied, upon by the defendant at the argument. It is sufficient to say that in the first period no apparent attempt was made to overcome interference. In the second, the best of what was accomplished was undoubtedly represented by the Gamewell and Crane signal box, a construction which combined in part the claims of two patents. A physical exhibit of this box appears in the case, of which the same expert says:

“With this box, while it was a great step in the art over what had previously been done, there were still serious defects, for it is evident that a second box could easily be started at any of the closures of the circuit of another box in sending its signal. It is also evident that, if the pulls of two boxes were operated before the circuit was opened by either, then there would also be confusion and loss of both signals. It is also evident that, if a box was pulled while the circuit was accidently or otherwise momentarily opened, then the signal would not be sent and would be lost. It was also evident (hat if a second box was pulled while the circuit was being used by another box in sending its signal one signal would always be lost, and that if.' the second box was trailed on the closures, as above before mentioned, both the signals would have become lost, unless one box had already sent one round of its signal before the second one was nulled, or unless the first one hod got finished before the second one had sent all its rounds.”

That most of the objections to the Gamewell and Crane boxes just outlined in fact existed is not seriously controverted by the defendant. At all events, it is certain that interference was not thereby entirely eliminated. It still existed both as a possibility and a probability. Interference was hound to occur if a box of that type were pulled during the closed-circuit period of a box in operation. The probability of interference was, however, greatly minimized by having the signal-wheel so constructed that the circuit was held closed for short intervals and open for long intervals during the operation of the box. The Gardiner patent of 1880 may, in turn, be said to disclose a typical box in the third stage of the development of the art. It came into general use, and supplanted the Gamewell and Crane box. This box was so constructed that, when one box on the circuit was pulled, oilier boxes on the same circuit were cut out by time mechanism during what is called a test period; that is, for a period longer than the closure due to any closed circuit interval in the transmission of the signal of any box on the circuit. It should be understood that the circuit is normally held closed, and that a box when in operation closes the circuits for short intervals; hence it is possible to determine when a box in a circuit is in operation by ascertaining' whether or not the circuit remains closed for a longer period of time than the longest closed circuit period during the transmission of the signal of any box. The Gardiner patent to some extent applied this principle. His box reduced the probability of interference, but did not succeed in eliminating it. It still remained true that, if two boxes were pulled simultaneously, their signals would be confused, and that, if one box were pulled while another was transmitting its signal, the second signal would be lost.

On June 4, 1889, patent No. 404,438 was granted to Ruddick, the patentee of the patent in suit, for a noninterference fire alarm signal box, and it is this box, more than any other, which the defendant strenuously insists discloses, in substance, the essential features of the-patent in suit, or which, if it does not fully'- anticipate that patent, nevertheless requires its claims to be strictly construed and limited to the precise mechanism thereby shown, in which case the defendant does not infringe it. Ruddick’s first patent and the defendant's in-sistment in relation thereto will be considered later. As to the Rud-dick patent in suit, it is what is known as a “successive noninterference box.” It has previously been intimated that a noninterference box is designed to prevent a confusion of signals where two or more boxes upon the same circuit line are pulled at the same time. By the term “successive” is meant the inclusion of means whereby, when a box is pulled, the signal is withheld from transmission during the period when the line circuit is not clear or in a condition to successfully transmit it, but which, as soon as the line is in such a condition will automatically start, continue and complete the transmission of such signal. In order to show the object and character of the invention, it seems both proper and necessary to include herein, notwithstanding its prolixity, the explanation thereof given by Ruddick in the specification .of his patent, and this mainly for the reason that it is set forth therein in reasonably clear and general terms and without specific reference to the complicated drawings of the patent. The extract referred to, is as follows:

“This invention has for its object to construct a successive noninterference signal box. yet by the elimination of some features a noninterference signal box is produced which is not successive, but which possesses many points of merit.

“The signal-transmitting mechanism which is commonly referred to as the ‘signal-box’ comprises a motor mechanism and a box-number circuit-wheel adapted to be driven by it to operate an electric circuit to transmit a signal.

“In signal boxes previously constructed many different forms of motor mechanisms have been employed, and,' so far as my present invention is concerned, I may employ any well-known or suitable form.

“A controlling-lever is provided which is adapted to control the transmission of the signal, and said controlling-lever is governed by a noninterference magnet which is operated by or under the control of the signaling-circuit, so-that the said controlling-lever is therefore under the control of the circuit.

“The box is so arranged that whenever started, or whenever its signal has been ‘set,’ the controlling-lever will be placed under the control of the signaling-circuit for a predetermined length of time before the said circuit is operated or broken for its first time, by the said set signal, and if during such predetermined length of time the normal condition of said signaling-circuit shall be changed, regardless of the cause, the said controlling-lever will at once assume its abnormal position or operate to delay or it may be to prevent tlie transmission of the signal, according as to whether the box is arranged as a successive noninterference box, or simply as a noninterference box. *

“When a signal is being transmitted from any box the circuit is opened and closed repeatedly, and when the line remains closed a longer time than the longest closure in any signal, it is evident that no signal is being transmitted. and hence the predetermined length of time during which 1 desire to place the controlling lever solely under the control of the circuit is longer than tile longest closure in any signal.

"In signal boxes of the so-called ‘open-wheel’ variety the closures are all of the same duration or nearly so, being quite short, while in signal boxes of the so-called ‘closed-wheel’ variety some closures are of longer duration than the others. For instance, if but a single round is transmitted the longest closure will be between the groups of impulses, or what are commonly termed the ‘number groups.’ lmt if several rounds are transmitted the longest closure is usually between the successive rounds.

“The controlling-lever which controls the transmission of the signal, together with the noninterference magnet and its armature, I denominate a ‘determining device' which del ermines whether or not tile signal once set shill! be transmitted promptly or delayed, as will be described.

“The means or mechanism employed for placing the controlling-lever solely under the control of the circuit for a longer time than the longest closure in any signal after the said signal has been set and before its first impulse has boon transmitted is heroin denominated a ‘retarded operative device' because the operation of the circuit is delayed after the signal has been set.

“The locking or releasing lever for the train is herein shown as one having two distinct: movements, one of which is controlled or operated manually, as by a pull or starting-lever, and the other of which is controlled or operated by the train in running, the first movement of said lever effecting the release of Hie train or permitting it to operate, and the second movement of said lever, as herein shown, governing the action of the control ling-level' which controls the transmission of the signal.

"Suitable means are provided for holding the controlling-lever whenever the circuit is opened at (he home box. and for releasing said lever one or more or all of the times that the circuit is closed at the home box during a round of the alarm, the home box being understood as the ’particular box in circuit with others, which is being operated and described, if, however, the controlling-lover is allowed to assume its abnormal position, as it will bo if the normal condition of the circuit is changed during any interval of time that the controlling-lever is released or unlocked, and is therefore held only by ihe attractive force of the magnet, the signal will not be transmitted, or if the box is designed particularly as a successive box the said signal will bo retained to be thereafter transmitted, as will be described. In either event, however, means must be provided for restoring the controlling-lever, or permitting it to be restored, only at a certain time, as at the completion of a round or of a succession of rounds, so that if a sot signal has been retained it, will be let off or released only at the beginning of a round, in order that a correct signal may be transmitted.

"The devices employed for holding the controlling-lever whenever the circuit is opened at the home box, and releasing it (the last three words are inserted by stipulation) at certain times or every time, if desired, that ihe circuit is closed at the home box, that it may be moved into or allowed to assume its abnormal position if the line is in use and which allows said lever, when once thrown out, to resume its normal position only at the beginning of a succeeding round of ihe box number, constitute a trap for the controlling-lever.

“When the controlling-lever is restored, which, as before stated, is only ai or before the beginning of a round of the box number, if the box Is arranged as a successive box, the said lever is again placed under the control of the circuit for a longer time than the longest closure in any signal by means of the retarded operatin' device before the first impulse of the round is transmitted, so that if said controlling-lever is once thrown out it will require a closure, of the circuit which is longer than the longest closure in any signal before said lever will permit or effect the transmission of the retained signal; but if the box is arranged simply as a noninterference signal box the said controlling-lever will simply be restored just as or before the box stops running.

“In the box herein shown, the trap is arranged to carry out the function ascribed to it during the first round of the signal, and if the signal which is being transmitted is not interfered with, the co-operation of said trap with the controlling-lever will cease; but if the signal is interfered -with the trap will continue its co-operation with the controlling-lever until such time as a correct and uninterrupted round is transmitted; but while such feature has certain special advantages, it is unnecessary to provide means especially designed for so discontinuing the co-operation of the trap and controlling-lever.”

No attempt wall be made to explain this complicated device in detail, as to do so intelligibly would necessarily involve a complete reproduction of all of the drawings and specification of the patent, together with no inconsiderable portion of the explanations thereof given by the several experts who have testified in the case. The patent contains 59 claims, of which Nos. 2, 3, 4, 5, 6, 7, 8, 9, 11, 13, 15, 16, 17, 18, 21, 23, 24, 27, 54, 55, 57, and 58 are in issue. The claims in issue have been properly classified by counsel into three groups as follows: In the first group Nos. 11, 13, 15, 16, 17, 54, 55, 57, and 58; in the second, Nos. 2, 3, 4, 5, 6, 7, 8, and 9; and the third Nos. 18, 21, 23, 24, and 27. Claim No. 11, it is admitted, represents the first group; No. 4, the second; and No. 18, the third. The three claims just referred to follow in the order named.

“11. In a noninterference signal box, a motor, means for operating it, a retarded operative device consisting of a signal-wheel moved by the train to operate the circuit for its first time after the train has run a longer time than the longest closure in any signal, and a determining device comprising a noninterference magnet, its armature, and a controlling-lever, governed by said magnet, which acts during said lapse of time, to prevent an interfering signal being sent substantially as described.”

“4. In a successive noninterference signal box, a signaling-train, a controlling-lever which controls the transmission of the signal, a noninterference magnet and its armature, combined with a trap coextensive with a round of the box number .which positively holds the controlling-lever at each and every time the circuit is opened at the home box and releases said controlling-lever at each and every time that the circuit is closed at the homé box, to enable said lever to assume its abnormal position if the circuit is open at another point, and which allows said lever to resume its normal position when once thrown out only at the beginning of a round of the box number, substantially as described.”

“18. In a signal box, a signaling-train, a circuit-controller, a noninterference magnet and its armature, combined with a controlling-lever governed by said armature which controls the transmission of the signal, and an arm which vibrates to correspond with the makes and breaks of the circuitwvheel which engages said controlling-lever on the breaks and releases the same on the makes, substantially as described.”

With the aid of physical exhibits, a very clear exposition .of the subject-matter of the patent, and of the claims in issue, was made at the argument by the respective counsel. Claim 11 includes among other elements in combination the “retarded operative device” and “determining device,” the purpose and character of which fully appear in the above extract from the .specification. The first of said devices is furthermore, in that claim, said to consist of a “signal-wheel moved by the train to operate the circuit for its first time after the train has run a longer time than the longest closure in any signal,” and the other of said devices is characterized as comprising a noninterference magnet, its armature, and a controlling-lever governed by said magnet, which acts during said lapse of time to prevent an interfering signal from being sent. This is the broadest claim of the first group. The other claims introduce additional elements. Claim 16, for instance, introduces means for producing succession whereby after four ineffective turns of the signal-wheel the train, instead of being stopped, will continue to operate, the retarded operative device again come into operation, and the signal-wheel make four additional rounds.

The group of which claim 4 is a type provides a “trap” which is fully described therein as follows:

It “positively hotels the controlling-lever at each and every time the circuit is oponed at the home box and releases said controlling-lever at each and every rime that the circuit is closed at the home box, to enable said lever to assume its abnormal position if the circuit is opened at another point and which allows said lever to resume its normal position when once thrown out only at the beginning of a round of the box number.”

Its function is to prevent interference between two boxes when pulled simultaneously. The trap provides, among other things, means for holding the controlling-lever whenever the circuit is opened at the box pulled, and for releasing the lever whenever, and at such times as the circuit is closed at that box during a round of the alarm.

The claims of the third group also relate to the trap, but call for a vibrating arm; that is to say, an arm which vibrates to correspond with the makes and breaks of the circuit-wheel which engages the controlling-lever on the breaks and releases it on the makes. There is testimony in the case which with reason styles the Ruddick box of the patent in suit as:

“The first practical successive and noninterference box of any kind, it was the first box which produced succession within itself; that is, without the use of independent circuits or the use of independent instruments at central, * * * and this box of the Ruddick patent in issue was the first that combined the essentials that would send signals without confusion or loss. Rud-dick in applying this invention used only the signaling-train itself as a timing or operating mechanism, and had no independent timing mechanism in the box or associated with it to be depended upon for the carrying out. of this function of noninterference, and the principles disclosed by Ruddick are the ones, and the only ones in use to-day in producing all these desired results.”

This testimony is not, however, permitted to pass unchallenged, otherwise, at least one of the alleged defenses to this suit could not have been interposed with any probability of success. It is not surprising, therefore, to find that, according to the claim of the defendant, everything that Ruddick did by the patent in suit had been done before or so disclosed as to make it common knowledge in the art. In the attempt to establish this position, the defense has introduced into the case, and its expert has analyzed and discussed numerous patents which however, need not be considered at length, since whatever of weight there is in the defendant’s contention resides for the most part in the disclosures of the previous patent to Ruddick. Unless anticipation is found there, it cannot be found in the case. That patent fundamentally comes closer to the one under consideration than any other.

Ruddick’s application for his earlier patent was filed December 14, 1887, and that of the later June 28, 1890. The file wrappers of both are in evidence. The specification of the later patent as originally filed stated that:

“Tlie present invention constitutes an improvement on prior mechanism of this general character and especially upon the apparatus set forth and claimed in my patent No. 404,438, June 14, 1889.”

The foregoing language does not, however, appear in the specification as finally issued, which was introduced in due course by way of amendment to the original specification. It is claimed in behalf of the defendant that the substituted specification is broader than the invention shown in the original; that, however, was not the view of the examiner, nor does it seem justified. The original specification, although much amplified by the substitute, contained the essential features of the patent as issued. Again, while the application was pending in the Patent Office, the examiner, speaking of certain proposed claims, said “that a trap broadly in noninterference boxes is old,” citing Ruddick’s earlier patent, and that “a device to positively hold the armature of a magnet whenever the circuit is broken at the home box, and to release said armature .whenever the circuit is made at the home box, is old,” citing Herzog, 1883. He nevertheless allowed the first 11 claims, holding that a combination of the two devices was patentable, and in doing so said:

“Not only is the combination clearly patentable, but the structure of the device described in the application is so exceedingly complicated aud ingenious that a claim for the combination should be given the broadest possible construction and the broadest field of equivalency. The first inventor of the combination should be allowed to cover both the hypnotic and positive form of trap.”

The claims of the patent in suit were all allowed with Ruddick’s earlier patent in open view, and the patent itself was only issued after a strenuous interference contest in which the various issues in interference were closely examined and considered, and the examiner at the conclusion of his opinion says:

“The fact that the patent has been subjected to the closest scrutiny without developing any direct and positive statements setting forth the various elements of the interfering issues conclusively proves that those issues were not disclosed in the patent. The present application is therefore allowed.”

Nor did the earlier patent purport to accomplish all that is claimed for the patent in suit. Indeed, it is admitted in the specification of the earlier patent that in repeating the signal four times conditions might exist under which the signal the first time it was given would be confused. Thus it says:

“Supposing that the two boxes break the circuit at the same instant (and not one before the other, as in the case just cited), both boxes will sound the gong, and, if one mechanism rotates a little faster than the other there may be momentary confusion in regard to the first numeral, but as soon as the second notch is reached one or the other of them must hold the circuit, and there can he no further trouble. In any case three correct signals will be sent, leaving no possibility of mistake in the reading of the alarm. Of course, the wheel in every box is notched to correspond to the number of that box, as represented by the number and arrangement of the pins on the number wheel. There being no two numbers alike in any fire alarm telegraph .system, it is evident that no two wheels will be notched alike, and therefore that no two signals can be confused for longer than while the first box number is being signaled once.”

Confusion in fire alarm signals of any character or to any extent is naturally productive of delay and increased hazard. It were better to have the signal correctly given but once than to have it given four times, one of which, and that seemingly the most important, incorrectly given. That Ruddick’s earlier patent in view of his admission was not perfect needs no argument. Nor is it conceivable that the improvements which he embodied in his second patent, and which seem to have made it perfect and commercially valuable, were, as claimed on behalf of the defendant, purely mechanical and obvious to one skilled in the art. Were this true, it would reasonably follow that they would have occurred to Ruddick and been embodied in his earlier patent.

The two patents have been examined with care, without, however, finding that the earlier discloses the later, certainly not with that degree of clarity and definiteness which the law demands of that which may properly be adjudged an anticipation. The defendant, it may be said in this connection, largely relies to show anticipation by the earlier patent upon the arrangement in one of the drawings of certain pins upon the “number wheel”; but, admitting that the pins are so arranged as to afford plausibility to the argument, still, in the absence of any support therefor, either in the specification or claims of that patent, it cannot be held to be such a disclosure as in law would constitute an anticipation of the patent in suit, and this was likewise the conclusion of the examiner.

In his earlier patent Ruddick sought means to prevent interference in the signals of two simultaneously operated boxes, but his effort, like that of all who had preceded him, admittedly failed.

The defects in his first patent were, however, undoubtedly remedied in the second, which discloses novelty and invention. It not only carries with it a presumption of validity, but that presumption is materially strengthened by the contest which it successfully withstood in the Patent Office, and, even if it were admitted that the enthusiasm of the examiner was not entirely justified, still his opinion as an expert in a complicated patent of this character is entitled to great weight, far greater indeed, than it would have been had the patent passed under his supervision in due course without argument, or discussion and without his attention having been particularly directed to the disclosures of the prior art and particularly to those of the earlier Ruddick patent. The conclusion of the examiner is confirmed, moreover, by the opinions of two expert witnesses with which the court is entirely in accord. The Ruddick patent in suit is therefore held to be valid.

Upon the question of infringement, it should be said at the outset that the defendant’s box is also a successive noninterference box. The defendant,' however, attempts to distinguish it from the complainant’s in various particulars, for instance, in the first group of claims in issue, the question of infringement is made to depend largely upon the meaning of the word “signal” as found therein. In the defendant’s box the retarded operative device performs its function not only at the first round of the signal, as does the complainant’s, but also at the second, third, and fourth rounds, so that at the beginning of each round there is a long closure of the circuit. This is caused by the fact that the signal-wheel has one tooth longer than any of its other teeth, which hold the contact lever in a raised position for closing the circuit at the beginning of each of the rounds.

This difference, however, is not important since the defendant plainly uses an equivalent of the retarded device of the patent in suit. Upon the question of the meaning of the word “signal” above alluded to, it is insisted on behalf of the defendant that it includes not only the signal as first given, but three repetitions of it, while the complainant contends that it is complete when once given. I think the complainant’s contention is right. There is only one signal and it is that which is repeated three times. These repetitions cannot be tacked to the signal so as to form, or be said to form, a part of it; they cannot be otherwise than what they are, repetitions. Moreover, the prior art taught that a signal was completed in one round, the examiner in the Patent Office apparently held the same view, and the patent itself also shows that one round of the signal-wheel gives the signal. I conclude, therefore, that the first group of claims are infringed. Infringement of the second group does not seem to be seriously questioned. At all events, they may be read directly upon the defendant’s device, and are infringed by it. The third and last group in issue have as an element what is called a vibrating arm which serves to hold the controlling lever each time the circuit is opened at the box and releases it each time the circuit is closed at the box. It releases the noninterference device on the closure of the circuit, and holds it upon the break of the circuit. It vibrates according to the makes and breaks in the circuit, and thereby locks and unlocks the controlling-lever. The defendant’s illustrative drawing of its box shows a pin which is driven in and fixedly attached to the controlling-lever which performs the same function as the vibrating arm of Ruddick. It serves to lock and unlock the controlling-lever upon the makes and breaks of the circuit. However called, it is the equivalent of the vibrating arm of the patent in suit. My conclusion, therefore, is that the claims calling for that element are also infringed.

[2] The complainant does not claim that the Cole patent, the second in suit, does more than disclose mechanical improvements upon Ruddick’s device; that patent contains 23 claims, of which Nos. 11, 21, 22, and 23 are in issue. The claims in controversy point to a mechanical construction for the performance of some of the functions of a signal box employing the retarded operative device and the determining device of Ruddick’s patent. Thus claim 11 calls, in addition to other elements, for means controlled by the train in running which holds the signal-key with its circuit contacts closed during the test period, during which time the signal controller is free to act to mechanically lock the signal-key with its circuit contacts closed. Claim 21, among other elements, calls for a noninterference magnet and its armature which whenever retracted mechanically locks the signaling-key with its circuit contacts closed. Claim 22 specifies additionally that the signal-key whenever operated to open the circuit mechanically locks the armature in its attracted position, and claim 23 provides, with other elements, for a signal-controller governed by the armature which is free to retract on closures of the circuit at the home box, and whenever it retracts locks the signaling-key with its circuit contacts closed. The improvements called for by the claims in issue of the Cole patent appear, because of their simplicity and effectiveness, to have been adopted and largely used in putting Ruddick’s invention on the market. Defendant's expert admits that the “Cole patent appears to be a practical embodiment of the theoretical and evidently impractical mechanism- of the Ruddick patent in suit. It provides a test period extending one complete round of the circuit wheel, and provides for locking the key contacts during such periods.”

I think.that the claims in issue are valid, and have been infringed by the defendant.

That claim 11 is infringed is clearly shown by the testimony of one of complainant’s experts, who, after mentioning certain specific elements common to the boxes of Cole and the defendant, says:

“There is also required means controlled by the train in miming which holds the signal-key with its circuit contacts closed i'or predetermined length of time just before the beginning of a round of the box-number. This element corresponds to the retarded operative device which is operated by the train in running, and in this Cole patent it is the holder e, which rotates when the train is operated, and is so related to the other parts in the signaling-train that it engages with the projection e of the three-armed lever after the box has been pulled for the preliminary run indicated. In defendants box the long tooth upon the signal-wheel when moved by the train in running, after the box is pulled, co-operating- with the projection from underneath the signal-key S. performs the same function. The claim further requires that during the preliminary run for the predetermined length of time the signal-controller shall be free to act to mechanically lock the signal-key with its contacts closed. In the Cole patent, the signal-controller, as shown in my sketch Fig. 6, is free to assume its abnormal position, under the influence of the retractile spring, at any time during the preliminary run, and if it should do so, the pin 1)5, falling behind the shouldered end d' of the three-armed lever, will lock the signal-key with its contacts closed mechanically. In defendant’s box, the controlling lever 9, or signal-controller, is in the position indicated in my Fig. 10 during this preliminary run, and it is free to assume its abnormal position as indicated in my Fig. 12, in which position it mechanically locks the signal-key with its contacts closed. I, therefore, find that defendant’s box embodies the combination recited in claim 11 of the Cole patent.”

Claim 21 requires that the armature whenever retracted mechanically shall lock the signal-key with its circuit contacts closed. This element appears in the defendant’s box. Defendant’s counsel claim, however, that its structure does not infringe because there are times when the armature can be retracted at will without influencing the circuit contacts and without mechanically locking them closed. They are nevertheless locked at the critical time, and by means equivalent to those specified in the claim in question.

Claim .22 requires additionally that, whenever the signal-key is operated to open the circuit, it shall mechanically lock the armature in its attracted position. Cole does this by means of a shoulder on one of the ends of a three-armed lever, which is rotated when the circuit is opened; the shoulder referred to thereby passing under a pin, and locking the armature in its attracted position. The defendant’s device does the same thing, although not by precisely the same mechanical means, but which, nevertheless, perform the same function in the same way, and come within the terms of the claim.

Claim 23 requires that the signal-controller shall be free to retract on closure of the circuit at the home box, and that, whenever it retracts, it shall lock the signal-key with its circuit contacts closed. When, therefore, in the Cole box the circuit is closed at the home box, the controlling-lever can assume its abnormal position, the armature being in a retracted position, and when it does this, tile pin above referred to locks the signal-key with the circuit contacts closed. The defendant’s box is also provided with means whereby when the circuit is closed at the home box the controlling lever is free to retract to an abnormal position and thereby lock the signal-key so that the contacts are closed.

A decree in favor of the complainant and against the defendant with costs will accordingly be entered upon the claims in controversy of both patents.

It is understood from complainant’s brief that the complainant has no desire to interfere with the operation of the fire alarm system of the defendant, consequently no injunction will be issued against it, at least without giving it full opportunity to be heard, and then only in case it shall be made clearly to appear that the city can be amply protected in the premises.