clusion, and that the requirement should be regarded as directory, rather than absolutely essential. The full six months are allowed by the statute to the applicant, and it can hardly be supposed that Congress intended that this period could be cut short by the exigencies of the Patent Office. He has no power to coerce its proceedings. Delays in a public office are not generally allowed to prejudice the right of one who has performed all that the law requires of him to secure official action. The object which Congress had in view was to compel the applicant to follow up his application by the reasonable performance of the conditions precedent to the issue of the patent. If he entitles himself to have the patent issue within six months, that object is subserved. Said Mr. Justice Field, in delivering the opinion of the court in French v. Edwards, 13 Wall. 506, 20 L. Ed. 702:

"There are undoubtedly many statutory requisitions intended for the guide of officers in the conduct of business devolved upon them, which do not limit their power, or render its exercise in disregard of the requisitions ineffectual. Such, generally, are regulations designed to secure order, system, and dispatch in proceedings, and by a disregard of which the rights of parties interested cannot be injuriously affected. Provisions of this character are not usually regarded as mandatory, unless accompanied by negative words importing that the acts required shall not be done in any other manner or time than that designated."

And in 2 Sutherland, Stat. Construction, § 612 (2d Ed.), it said: "In other words, as the cases universally hold, a statute specifying a time within which a public officer is to perform an official act regarding the rights and duties of others is directory, unless the nature of the act to be performed or the phraseology of the statute is such that the designation of time must be considered as a limitation of the power of the officer."

Again, the statute declares that, if the final fee is not paid within the six months, the patent shall be withheld, but it does not declare that, if not dated as directed, the patent shall not issue, or, if issued, shall be void. Besides all this, the officers charged with the administration of the law have for many years construed the law as giving power to the office to exercise control over applications until the patent should finally issue, and that it was justified in resorting to a matter of form in order that the right of the citizen should not be sacrificed. So far as we know, this is the first time the practical construction of the statute now called in question has been challenged, and much disturbance of things supposed to rest on solid foundations would ensue if such an objection should be sustained.

Whether the practice, if promptly challenged by the government in a direct proceeding to recall a patent would be sanctioned, we have not now to consider. We think the patents ought not to be held void for irregularity, if it be such, at the instance of a private party in a collateral proceeding.

Pursuant to the leave granted, the defendants answered the bill. They denied that the inventions covered by the several patents were new, alleged numerous anticipations, denied in general the validity

of the patents, renewed their objection to the patents on account of their dates, and also denied infringement. The case went to hearing on pleadings and proofs. The bill was dismissed, but upon what grounds the record does not disclose.

The patent to Scribner, No. 357,538, was granted for an improvement in spring-jack switches such as were described in a former patent to the same patentee. He states that the object of his improvement is to "so construct the switches that they may be light, strong, and efficient, and easily accessible for adjustment or repairs, while occupying the smallest possible space upon the switch board." In telephone switch boards such switches are employed for the purpose of opening and closing the circuits between the patrons, and the switches are necessarily numerous. In order that as many as possible may be brought within the reach of an operator, it is important that they be compactly arranged in the switch board. The switch jacks shown in this and the others of complainant's patents occupy a space at the outer end of about one-half inch square and in length about two inches. They are arranged side by side on plates each of which contains say 20 switches, and these plates of switches are multiplied laterally, and are laid up one on top of the other to the height desired. The front edge of these plates is in the face of the switch board, and an opening in the front of the plate or frame extends back into the switch, through which opening the plug at the end of the connecting piece held by the operator is passed in to make contact with the springs of the jack, the springs being conductors in the circuit. Normally, the springs effect a closure of the circuit in the jack. But when the plug is inserted, and the springs are moved away from their normal contact, the circuit in the jack is opened and extended through the plug to the line connected therewith. Fig. 1 of the patent No. 357,538, here shown, illustrates the invention patented:

The plate, a, contains several slots for receiving the switches. Two switches are inserted in their places. The lower one of the two is in its normal condition. The upper one shows the plug inserted in the jack and lifting the spring, g, off the contact piece on the spring below. The spring, f, has a ground connection, and is insulated

from the test-strap, e, and the latter is in connection with the testtube, d, which makes contact with the tip of the plug, as the operator begins the insertion. The frame is of rubber or some nonconducting material, and so are the insulating strips shown in black. The spring, strap, and insulating strips compose the jack, and are made to completely fill the slot in the frame at the rear end thereof when all are pressed down edgewise to their place. The test-strap connection with the test-tube enables the operator to find out, through the test-wire with which it connects at the rear, whether the line is in use or not. The springs and strap are all made of thin conducting material, to economize space. This explanation is not of all the characteristics of the switch jack, but is sufficient to an understanding of the claims involved.

The claims of this patent which are said to be infringed are the third and fifth, as follows:

"(3) In a spring-jack switch, the combination, with the strap, e, connected with the test-piece at the front of the board of the ground-piece, f, and the lever, g, said strap, ground-piece, and lever being punched out of sheet metal and placed in substantially parallel planes and insulated, substantially as shown and described."

"(5) The combination, with the rubber frame or support provided with the holes and slots, of spring-jack switches inserted therein and completely filling the slots at the rear, whereby the teeth or sections of said frame are rigidly supported, substantially as shown and described."

A great number of patents had been issued in this country and abroad for switch jacks in telephone and other electrical apparatus before Scribner devised the switch jack above described. The defendant's counsel say that every element employed by Scribner is to be found in the prior art, and in structures already patented, and that there was no invention in reassembling them to perform their accustomed functions in another organization. And we have no doubt that the fact is thus correctly stated. But we do not think the conclusion follows; for, while all the things employed by Scribner and their capacities had been known to those familiar with the general subject and in some forms and relations had been applied to effect similar purposes, yet no combination of them operating in the same way as that of Scribner's is shown to have ever existed. An examination of the older devices would only show that the concession we have made is justified. And it would also show that the more recent of the prior patents for such devices have rested upon distinctions in the methods of combination. But the patent under consideration is for a new and improved combination and we are convinced it is one of much utility. Being only for an improvement, the patent must be so limited as not to cover any combination shown by the prior art. But, when so limited, it is manifest that the defendant company infringes it. As it does not build upon the older models, but copies that of the complainant's, it bears witness to the value of Scribner's improvement. We shall recur to this subject after the examination which we propose to make of the other patents described in the bill and claimed by the complainant.

Patent No. 488,033, issued to Scribner and Warner, was for a form of construction based in part on the Scribner invention, already considered, and consisted in the application of it to metallic circuits with certain necessary additions and changes. It was stated in the application that the object of the invention was to provide ready means of looping or connecting together such circuits. It included a switch jack proper and the plug to operate it.

Fig. 3 illustrates the switch jack, and a part of Fig. 7 the opening in the frame in front of the jack, and the loop-plug.

In Fig. 3 a plate is shown having spaces, 1, 2, 3, 4, 5, for five jacks laid side by side. Three of the spaces are filled with jacks. The invention consisted of the two springs, preferably of different lengths, adjusted to press normally toward each other and upon different contact points (the posts c and d), a guide-tube in front of the ends of the springs; and a loop-plug, g, is shown in Fig. 7, adapted to be inserted in the tube and to spread the springs, which are line terminals, apart so as to separate them from their normal contacts, c, d, and closing each spring with a different terminal in the loopplug. The loop-plug has a metallic sleeve larger than the tip of the plug, and when the plug is inserted one of the springs is pressed

off its contact and comes in contact with the sleeve through which the circuit is extended when the other end of the loop is connected with the line beyond it by a similar plug and switch. The loopplug also presses off the other spring from its contact, and, the tip of the plug being in contact with that spring, another circuit connection is formed and extended through the plug. Proper insulation is provided throughout the apparatus described.

The claims of the patent involved in the present controversy are the first and third, as follows:

"(1) A circuit-changing device consisting of two springs or line-terminals, each spring adjusted to normally press against a different contact-piece, and a tube or guide placed in front of said springs, in combination with a loopplug having a tip smaller than the metallic sleeve of the shank thereof, said loop-plug being adapted to be inserted in said guide to separate said springs from their contacts, respectively, while each spring is at the same time closed to a different terminal of the loop-plug."

"(3) In a circuit-changing device, the combination, with two insulated springs of different lengths, having their free ends in the same direction, of a guide in front of said springs and a plug provided with two terminals, one terminal upon the tip and the other upon the shank thereof, whereby on inserting the plug into the guide the springs are respectively closed to different terminals of said loop-plug, the shorter spring being closed upon the tip and the longer spring upon the sleeve, substantially as and for the purpose specified."

Anticipation of this invention is claimed by the defendants upon similar grounds as those raised against the former patent to Scribner, which has been already considered; and that patent is itself relied upon as an anticipation of Scribner and Warner's. But this patent seems much less open to the defense of anticipation than the Scribner patent; for, although the elements employed in this invention and some of the details of methods of arrangement in respect of some of those elements are to be found in many former structures, there is nothing in the proof that shows the same combination, or anything similar to those of the patent under consideration. The reference to the Scribner patent, No. 357,538, as an anticipation, is unavailing. It is true that the means employed in that patent are to some extent utilized. There are similar springs, a plug, and a similar method of opening the circuit and of extending it. But it has not the test-strap of the Scribner combinations, nor are the normal contact-points the same. Besides, this patent has special provisions not only in the switch jack proper, but in the loop-plug also, and relatively to each other, for changing and extending the circuit, while nothing of the kind is shown in the former patent to Scribner. But, even if it had adopted the Scribner invention, the improvement upon it would nevertheless be patentable.

The Scribner and Warner invention appears to us to be of very considerable merit, and a marked improvement upon anything before attained. We have no doubt whatever of the patentability of the claims relied upon. Nor can there be any doubt, as we should suppose, that infringement is proven. The grounds for this latter. conclusion will be stated when we have reached our conclusion in regard to the other patent involved in the controversy.