resistances, which, in this supposed case, allow only 40 to pass. As the motor speeds up, and the counter electromotive force develops, the net current will be reduced to 8 units, and the device will be inefficient. To avoid this result, as soon as the net current falls to 20, a section of resistance should be cut out, and the amount of current admitted from the line increased to 60, and, in the same way, the amount should be successively stepped up to 80 and 100. The device by which the resistance is thus cut out step by step, and the current controlled, is called a rheostat, and it consists of a switch maintaining at one pole a constant contact with the line and at the other end selective contact with the resistances. In its simplest form, this is manually operated, as in the familiar instances of the controller of the electric street car or automobile. For many uses manual control is not practicable, and automatic electric control becomes necessary.

Before Carrichoff's improvement, this automatic control had been accomplished with more or less success by three classes or types of apparatus. In the first or time limit class, some timing device was arranged to operate the successive resistance cut-outs at predetermined and fixed time intervals. This is so far from Carrichoff's system that it, together with the manual control type, needs no further consideration.

In the second type, the current generated by the motor and representing the counter electromotive force, is led through a series of switches, the solenoid magnets of which are so wound that the first will be operated when this current reaches 40 units, and will then cut out a section of resistance; the second, when it reaches 60, and will then cut out another section, etc. It seems to be clearly established that this system, under many and perhaps under usual conditions, is efficient, but that under other conditions which are not uncommon, and which, with certain installations, are to be expected, is not satisfactory. It has direct bearing upon Carrichoff's improvement only in one respect hereafter to be mentioned.

The third type touched Carrichoff more closely. Indeed, this system was exemplified in the patent to Sprague, issued when he and Carrichoff were both in the employ of the Sprague Company, and Carrichoff, in his specifications, expressly declares his invention to be an improvement upon the plan of this Sprague patent. The Sprague plan, as far as it pertained to this particular subject, involved two features, which may be called his primary and his secondary features, and the claims. of his patent (whether valid or not) seem to be partly generic, as resting solely on this primary feature, and partly specific, as resting upon this secondary feature, when employed as a means of carrying out his generic thought. The primary feature consisted in the use of an automatic throttle to control the action of the successive resistance cut-outs. This throttle consisted of a solenoid magnet switch, interposed in the main circuit. So long as the current exceeded (e. g.) 20, the magnet remained energized and its switch contacts were held up and open. When the current dropped below this point, the contacts dropped, an independent circuit from the master switch and leading through these throttle contacts was closed, and current was carried

to and operated a resistance controller device, whereby some resistance was cut out, the line current admitted to the motor increased above 20, and the throttle switch contacts were again opened. This operation would be automatically repeated as often as the current fell below the predetermined point and until the resistances were all cut out. For his secondary feature or specific form of controller mechanism, Sprague provided a revolving drum carrying contacts so arranged that, as it revolved, it would successively cut out the resistance sections. This part of the operation-the actual throwing of the switches which controlled the resistance sections-was as completely mechanical as if the drum had been revolved by hand; but he gave electrically automatic revolution to the drum by operating it with a small independent motor, called a pilot motor, and this motor was actuated by current through the throttle contacts, when they closed as above described. With each such closing of the throttle, the pilot motor would be operated until the drum had revolved enough to cut out one resistance, and so increased the line current in the motor and at the throttle, and so opened the throttle and stopped the pilot motor. As with reference to the system last described, this pilot motor drum system gave fairly good satisfaction, and was and is considerably used, but has certain comparative disadvantages.

We come, now, to the Carrichoff improvement. He adopted and employed what we have called the generic or primary part of Sprague's invention, viz., the solenoid magnet throttle, as the means of automatically sending out an electric messenger whenever the motor current fell to the point which called for an increased current from the line, which messenger should operate one step in the resistance-shunting process. For Sprague's pilot motor drum, Carrichoff substituted a series of solenoid magnet resistance shunting switches (which, for convenience, we will hereafter refer to merely as magnets 1, 2, 3, etc.). The current which is caused by the closing of the throttle contact, energizes magnet 1, whereby its contacts are raised and closed, one section of resistance is cut out, and the line current is shunted through this closed switch around that resistance section. The lifting of the core of magnet 1 at the same time lifts a plate, closing contacts in a circuit from the master switch through the magnet winding to the ground, whereby this magnet continues to be energized and so to cut out this section of resistance, regardless of the subsequent opening or closing of the throttle. The same lifting action also closes contacts between magnets 1 and 2, whereby there is completed a circuit from the master switch through magnets 2 and 1 to the throttle and ground, so that the closing of the throttle contacts thereafter will close this circuit and energize magnet 2. The same thing then happens as with magnet 1; another section of resistance is cut out, and a maintaining circuit for 2 is closed. The same steps occur in succession with the other magnets. The result is that, when the current in the motor circuit falls below 20 (continuing to use our arbitrary illustration), the throttle closes, the first magnet is energized, its section of resistance is cut out, it is released from control by the throttle, the next magnet is rendered subject to that control ready for the next step in the process, and so on.

With this description, the recited claims will be intelligible. It is evident that claim 5 is intended to be somewhat the broadest of the three, in that it uses the word "controller" in the place of "rheostat," and yet it is difficult to see how the structure necessarily required by its language can differ materially from the structure contemplated by claim 6; also it is to be observed that claim 6 calls for a contact "in the circuit through each magnet," while claim 7 specifies that there is a branch circuit from the master switch through the rheostat magnet, and calls for a "contact in this branch circuit." It is not readily apparent that there can be a circuit through each magnet, as called for by claim 6, which will not necessarily be the branch circuit called for by claim 7. The relative construction of the three claims in this respect has not been argued, and we do not see that it is material. If, in fact, the Patent Office has granted three claims which cannot be distinguished from each other, the defendant is not harmed by this duplication; and the government is not asking cancellation of the patent. We may fairly take the seventh claim as the one which should be considered, and its consideration may well be approached from the standpoint stated by this court in Dayton Co. v. Westinghouse Co., 118 Fed. 562, 566, 55 C. C. A. 390, 394, as follows:

"But it is still contended that, conceding the fact to be that no one had made these particular inventions, yet that, so much was known to men learned in the science or skilled in the art of the subject, it did not involve invention to devise these ways and means for accomplishing the desired result. As to this it must be said that the subject is one of the most abstruse and subtle of all the practical sciences, and its pursuit involves the exercise of the keenest intelligence and most patient research that gifted men can bestow upon it. We ought, therefore to be cautious, when a distinct and practical improvement is made in so useful an art, in denying to the author the reward which the law gives to meritorious inventions."

We find no sufficient reason in the record for denying to this claim the merit of invention. Its distinctive thought lies in the idea that a series of rheostat magnet switches for resistance shunting should be so related to a throttle that the first is controlled by the throttle, and that the operation of the first, under the influence of the throttle, brings the second of the series within the throttle control and frees the first therefrom; the operation of the second brings the third to a condition ready to be operated by the throttle and frees the second, etc. This idea was wholly novel in this art and has been extensively put into practical use. It is unquestionably an efficient and useful combination, and the patent monopoly granted therefor must be sustained.

We say this idea was wholly novel. Of course, this depends upon the breadth which the idea is thought to have, and we do not mean to question the existence of suggestions and analogies which in the light of the patent disclosure can be developed into a close resemblance. Indeed, the very wealth of material which defendant puts forward as "perfect anticipations" makes it difficult to choose those things which merit discussion.

[2] The opinion below indicates that the Sprague patent was defendant's chief reliance. We have already described this patent. It seems that complainant, owning also the Sprague patent, had brought a suit thereon against defendant, involving the same infringing device

here involved, and that, in that case, Mr. Bentley, the expert for complainant there and here, had testified that defendant's device, which responds to the description we have given of Carrichoff's structure, was the equivalent of Sprague. Whereupon defendant's counsel say that as soon as, in this suit, defendant's device is claimed to be the equivalent of Carrichoff, it follows that Carrichoff must be invalid, because he then, also, must be the equivalent of Sprague. This argument overlooks the fact that equivalency in the patent law is not necessarily mutual. Whether the device of defendant is the equivalent of that to which a plaintiff patentee has been granted a monopoly depends upon the scope of the claim in suit. The instant case well illustrates the fallacy which we noted in Curry v. Union Co., 230 Fed. 422, 429, 144 C. C. A. 629. Mr. Bentley's testimony in the Sprague Case, taken altogether, makes it clear that he was finding equivalency in defendant's device only from the point of view of the broad claims of the Sprague patent, which were not concerned with the details which distinguished defendant and Carrichoff alike from Sprague. It must be obvious, when our attention is drawn, that the defendant's device may be the full and complete equivalent of Sprague as to the thought or feature which was the subject of his generic invention above described, and yet that either Carrichoff or defendant, whichever was earlier, may involve additional and subordinate features of patentable merit, so that Carrichoff, if earlier, would dominate defendant. More concretely, when we consider merely the subject of a throttle which, by its successive opening and closing, cuts out successive sections of resistance, it may well be said that a pilot drum and a series of magnetic switches are equivalents; and yet the planning and arrangement of the magnets in their relation to the throttle and the way in which they shall be operated by it may be meritorious invention and may support a valid patent.

[3] In this same connection, it is to be observed that the defendant has a patent upon its form of device and insists upon the benefit of some presumptions from this patent. We do not need to repeat that the issue of the later patent raises no presumption of noninfringement, and usually does not even tend to establish that conclusion. The contrary claim confuses the presumption of patentable difference with the presumption of noninfringement.2 Herman v. Youngstown (C. C. A. 6) 191 Fed. 579, 584, 112 C. C. A. 185; Curry v. Union Co., supra.

The Sprague patent is further relied upon as an anticipation because, in his specification, he says:

"Single or multiple magnets may be provided which operate rheostats or rheostat sections, either by single or stepped movements. to establish the particular connections."

Here is a suggestion that Sprague's system of throttle control could be used in connection with the type of apparatus which Carrichoff later

Our attention is called to the language of this court in National Co. v. Ralston Co., 172 Fed. 393, 398, 97 C. C. A. 91, in which there is a casual reference to this presumption of noninfringement. There was in that case no occasion to observe the difference between the two presumptions, and the reference thereto cannot be taken as a deliberate judgment upon the point.

243 F.-13

devised. This situation is united with the alleged testimony of Mr. Bentley that the chief merit of Carrichoff's invention lay in the conception that the thing could be done to make the argument that he invented nothing, since Sprague had disclosed the conception. This, again, overlooks the true substance of Mr. Bentley's testimony. It is not to the effect that Carrichoff's patent must rest on the conception that this ultimate result could be accomplished, but rather that its real merit lay in the conception that the throttle and successive magnets could be so arranged that they would operate in the manner pointed out in claim 7, and that, after this idea occurred to him, only ordinary skill was necessary to provide the apparatus which would carry it into effect. There is nothing to show that Sprague had this more specific idea. His suggestion is vague. It probably could be carried out more or less efficiently in more than one way besides that which Carrichoff devised.

Defendants also point out several old patents relating to a system of regulating the current furnished by a generator to a lamp circuit, and it is said that in this there is a throttle controlling a series of magnet resistance switches substantially as in Carrichoff. Likewise, the counter electromotive force system is said to point the way to even if not to embody the same combination. We think these things are too far away. Each of them shows a series of electro-magnetic switches adapted to cut out separate sections of resistance; but the latter has no throttle at all, and has a different principle of operation, while the former, though it has a regulating magnet, analogous in some respects to the throttle of Sprague and Carrichoff, does not disclose the idea that each magnet of the series can be in succession free from, subject to, and free from the throttle influence. The patents to Shepard and the second Sprague patent are said to show that it was old to use the throttle of Carrichoff for the operation of magnet switches for various motor grouping, instead of for throwing out resistance sections, and it is urged that there was no invention in the change, especially in view of the use of magnets for the latter purpose in the counter electromotive force system. It is enough to say of these patents that only by considerable elasticity of definition can their control devices in the main circuit be considered to be the throttle of Sprague and Carrichoff, and that they do not disclose Carrichoff's characteristic idea of having a series of magnets, each, except the first, successively (1) free from, (2) under, and (3) free from throttle control. If they rightly operate to limit somewhat the breadth of construction to be given to claim 7, that limitation is immaterial in this case.

Without further discussion of the many other earlier patents, we conclude that none of them are any closer than those which have been mentioned, and that they serve neither to anticipate Carrichoff nor to limit claim 7 in any essential degree. If it be thought that the maintenance circuit, whereby each magnet, after operation, remains free from the throttle control, is essential to an operative combination, but is not implied by the terms of claim 7, and cannot be read in, the only practical effect is that one of the other claims sued on, instead of claim 7, would be the proper basis for the decree.