(after stating the facts as above). [1] Whether it was invention to take the ajutage from a pipe leading to an exhauster, which is one kind of an air pump, and put it on a pipe leading from a blower, which is another kind of an air pump, or to take it from a pipe carrying gas from a gas holder and put it on a pipe carrying an element of the gas on its way into the gas holder, is, to say the least, a question of doubt; but this case may be more clearly, as we think, disposed of on another ground. We conclude that the relation of Rusby’s ajutage and gauge to the “set” is the relation between the items of an aggregate, and not the relation between the elements of a combination. This conclusion depends on the general proposition that a metering device, the only office of which is to' measure the raw material entering a machine so that the operator may stop when he has put in enough, does not so coact with the machine itself as to be properly characterized as being in combination therewith.
The authoritative cases are few which have considered the distinction between aggregations and combinations, except as it is complicated and confused by the effort to find the line between invention and mechanical skill. See Walker on Patents (4th Ed.) § 32, and cases cited. The rule of aggregation is stated — probably as well as possible — in Macomber’s Fixed Law of Patents (2d Ed.) p. 6 (with reference to sections 42-51, q. v.):,
“The distinction between an aggregation and a true combination is not always clear. The main test lies in an examination of the result — the function performed. If that result is the sum of the several actions of the elements, it is an aggregation; if it is the product of those actions — if the action of one element so modifies the action of another that the resultant action differs from the sum of the separate actions — it is a true combination.”
We must first disclaim doubting that there would be a combination if the changes in the ajutage or its pressure gauge directly actuated the regulating valve — as, for example, on the principle of the household thermostat. There would then be a direct inter-relation, just as there is between tire air regulating valve and the parts of the apparatus further along whose variable action depends on the amount of air passing through. Likewise, it must be remembered that the existing ajutage was not modified or adapted in any degree or applied in any physically new way,- to make it fit this situation. There is no claim of novelty in this respect.
*687We disclaim, also, the idea that there can be no true combination in any case where the action of cither the operator or material operated upon is necessary to cause one element to modify the action of any of the others. Such a case was Krell Co. v. Story Co. (C. C. A. 7) 207 Fed. 946, 952, 125 C. C. A. 394. The piano case, the controlling levers and the swinging fallboard all were parts of one structure when at rest, and when the operator was moving the levers the fallboard in another position assisted the operator’s hand. Not only would all the elements of the combination be in action at the same time (not a controlling consideration of itself), but each would be causing the whole device to act. To make that case parallel with this, we must assume — if the piano device was pneumatic, worked by foot power — that a gauge indicated the air pressure, so that the operator might not pump too much air in, and that it was sought to put this gauge into combination with the piano case. Of the same kind was the decision of the same court in Oshkosh Co. v. Waite Co., 207 Fed. 937, 941, 125 C. C. A. 385. _ This involved feeding mechanism and wrapping mechanism, but it was found that both acted upon the same material passing along, and one conveyed it to the other. It was thought that there was the necessary degree of co-operation to make it a combination; and the analogue of that combination is found in the water gas set in the relation between (e. g.) generator and carburetor.
Of the same type is the Automobile Case (Columbia Co. v. Duerr, 184 Fed. 893, 107 C. C. A. 215), decided by the Court of Appeals of the Second Circuit, and referred to in the Krell Case. The existence of a patentable combination was assumed as between the rear wheels and their driving devices and the front wheels and their steering devices, although only the mediation of the human driver causes them to work efficiently together. Here, too, when the operator’s action develops their inherent interrelation, .the front wheels, turned by the steering column, direct the course of the rear wheels, and the rear wheels, driven by their motive power, compel the revolution of the front wheels. We can find in an automobile the same question as in our present case by supposing that a combination was claimed between the speedometer, which tells the driver when he has reached the speed limit, and the gas throttle which the driver manipulates in obedience to the speedometer’s suggestion.
We should also1 discard the thought that Rusby discovered the necessity for having information about the quantity of air entering during the blow. If he had been the first to learn that the quantity of air should he adjusted to existing conditions, and that if conditions remained the same in successive blows, the amount which had been proper in one was the amount which should be admitted on the .next, and in connection with this discovery he had put it to use through being the first to provide any measuring or indicating apparatus, the defense of mere aggregation might have a different aspect — though it seems that this pertains rather to the question, “Invention or skill?” The discovery that the air should be measured would be the principal thing, and providing the means for measuring would be incidental. The record does not permit us to think that Rusby made any such dis*688covery. His patent itself seems to assume that it was already common to predetermine the amount of air to be used during a blow; but, if we misjudge his specification in this respect, it can make no difference. The whole purpose of the blow was to get the three members of the set hot enough for the ensuing run, and the operator must continue the blow long enough to get this result. Whether the air. blast is spoken of in terms of volume or of time makes no difference. Each pertains to the question, “How much air?” Each run was observed and tested in different ways, and, according to these conditions and observations, the operator concluded whether there had been just enough, or too much, or too little air on the previous blow, and then he regulated tire next blow accordingly. After the practice was established, or in so far as it was established, of .having a fixed period of time for tire blow, the quantity of air was regulated-by opening or closing the valve. It may be true that it had not been customary to “predetermine” the quantity of air in terms of cubic feet, nor, indeed, does that method ever appear to have been used by defendant; but with each' blow after the first it was “predetermined” by the operator that for the next blow he would have either the same amount of air or more or less. The operator was not dependent merely on his observation of the run. He was, provided with the water gauge, and he knew that the maintenance of certain indicated pressures would show the passage of the “predetermined” amount of air with fair accuracy.
[2] So we are led to the proposition first stated. If we apply the test of the quotation from Macomber, can we say that the resultant action differs from the sum of the separate actions? The resultant action — the effect upon the fire, the carburetor, and the superheater —has been modified by the turning of the air valve; but how is it modified by the pressure gauge? The gauge gave notice to the operator that he should do something, and thfen its function was finished until it again gave him similar notice. The final result — satisfactory .gas — is the sum of the information which the gauge gives the operator plus the complete gas-making action of the machine as managed and controlled by the operator. If the operator does not pay attention, nothing happens. The action of the gauge may continue, but it affects nothing. To draw upon the law of negligence for a figure, between the gauge and the air valve stands the free-willed operator as an intervening cause, and the chain of causation is broken. So a mechanical combination cannot exist when the product delivered by one device is, by the attendant workman, carried over and used on another machine, and it cannot be controlling whether he does this carrying over — as typically — in his hands or, as here, in his head.
Meters have a complete function in and by themselves. Any operator of a machine may guess at tire quantities and proportions of raw materials going in, and may gutess at the amount of product coming out. He frequently wants to know more accurately, and then he puts on a suitable meter, wherever it is needed; but the meter does not affect anything except the operator, and the operator is not a part of the combination. In a water gas set, four different raw materials are fed to the apparatus: Air, steam, coke, and oil. The operator must *689know that the proper quantity of each is being supplied. The general behavior of the apparatus gives the approximate information; but, if lie wants to be more exact, he can be. The coke can be, and doubtless is, measured by weight or by volume, and the operator controls the quantity; the oil may be, and in fact is, measured as to its rate of flow by a pressure gauge displayed before the operator, and he manipulates the valve in the oil feed pipe; the steam may be, and is, measured by a pressure gauge, and if the operator wants more or less, he changes the valve in the steam pipe; and, by the device now in question, the air is measured, the result displayed to the operator, and if he wants more or less, he changes the valve in the air pipe. As to each of these entering materials, the operator is merely supplied with data from which he exercises his judgment as to what he will do.
The operation of the ajutage is as obscure as its name is unfamiliar, but in spite of that it does nothing excepting to indicate, and in a sense measure, the entering air. It does not co-operate with the three main members of the set any more than does the oil measuring device or the steam measuring device, or, for that matter, the scales that weigh the coal. Indeed, it is just as important to the operator to know the period of elapsed time as it is to know the rate of flow of the air, and the clock that tells him when fully to close the valve is as essential to the gas-making operation as is the gauge which tells him when to shut it part way.
It is only another form of expressing the lack of real combination to say that the gas-making apparatus does not extend beyond the gates at which the materials enter, and it is only fortuitous that the air measuring device happens to be set close to the entrance valve. It might as well be in another room.
The fallacy involved in the theory that the ajutage is in combination with the set is materialized in the “whereby” clause of the claim. This says, “whereby the attendant is enabled to introduce a definite volume of air,” etc. This is a mistake. The air valve is the means whereby the operator is “enabled to introduce” the definite volume of air. The ajutage does not enable him to do this. It only enables him. to decide whether he will introduce more or less.
[3] Upon principle, it seems clear to us that a mere measuring device, which does nothing and can do nothing except to give the operator information as to how much' material is going into the machine, is no part of a true combination of the operative elements of the machine; and we can find no case where the contrary has been held.
The Supreme Court, in a long line of comparatively early cases which consider — or seem to — the precise «question, has found mere aggregation in: Bringing together parts of a coal stove;1 the eraser and the lead of a pencil;2 the former and the mould;3 the tack and 1he washer;4 the stove reservoir and the base pan;5 the mirror and the street *690car front hood ;6 two pairs of dies in series ;7 the fuel magazine and the heater;8 rollers under a frame and the machine carried by the frame;9 the reversing device and its machine;10 the grating and the means for operating a lock on the other side thereof j11 the tool and its finger rest;12 buildings and railway tracks.13 In later years, several patents have been sustained against the defense of aggregation, but in these cases the dominant question was one of fact whether more than mechanical skill had been employed, and did not involve the question of law whether there could be a true combination between tire elements considered. Of this class are the decisions of this court and the Supreme Court in the Rubber Tire Cases (Goodyear Co. v. Rubber Co., 116 Fed. 363, 53 C. C. A. 583; Diamond Co. v. Rubber Co., 220 U. S. 438, 31 Sup. Ct. 444, 55 L. Ed. 527), and the decisions of this court cited in the margin.14 These are not particularly helpful here.
In National Co. v. Powers Co. (C. C. A. 7) 160 Fed. 460, 463, 87 C. C. A. 444, 447, it was held that there could be no combination between the motor that would run any kind of machine and the machine that would run with any kind of motor. We can paraphrase Judge Baker’s language by saying:
“If in truth a burst of inspiration points to the measuring of A.’s air current with B.’s meter, nevertheless a monopoly cannot be based merely on bringing the two together.”
Perhaps the most apposite comment is found in Diamond Co. v. Ruby Co. (C. C.) 127 Fed. 341, where, in speaking of different parts of the mechanism, Judge Archbald says (page 348):
“No doubt they are all factors to make up a complete machine, and turn out good matches, and in that remote sense may be said to work together. But mchanically they are distinct parts, acting separately, there being no more connection between the carrier and the mechanism for raising the paraffine wheel under which it passes than between that and the steam boiler or the standards of the supporting frame. To make the raising mechanism effective, a human factor, the hand of the operator, guided by his judgment, must intervene, and without this it has no influence on the product. Unless he sees *691occasion to act, the match produced will be equally good or equally bad, whether the wheel-raising mechanism is there or not. It is not as though it were automatic, raising the matches out of the bath on the stoppage of the carrier, and lowering that again on the resumption of its course. That might be regarded as directly co-operating with the other parts named to effect the general result.”
The question we are deciding is in some measure similar to that which we recently passed upon in Autosales Co. v. Caille Co., 224 Fed. 473, - C. C. A. -. We there held there was only an aggregate relation between the automatic scales and the table of weights. In that case, the table gave information to the observer as a means of advising him to act by weighing himself. In this case, the meter gives information to the operator, advising him how to act. In each case, when the information is given, the knowledge-imparting element is functus officio.
The holding of the Circuit Court of Appeals for the Third Circuit in Standard Co. v. Burdett Co., 197 Fed. 743, 117 C. C. A. 206 (adopting the opinion in [C. C.] 196 Fed. 43), doubtless “lies within the twilight zone’’ (196 Fed. 46); but, assuming the patent there involved to have been rightly sustained, we think the conclusion rests on Judge McPherson’s statement (196 Fed. 46) that “the result produced by the patent is a method of operation”; and though the claims seem also in the twilight zone between claims for mechanical combinations and claims for method, the stated object of the patent was “to provide a signalling system” (196 Fed. 44). It is not longer to be doubted that successive mechanical steps, taken by an automatic machine or by the human operator, may he patentable as an art (Expanded Co. v. Bradford, 214 U. S. 366, 382, 29 Sup. Ct. 652, 53 L. Ed. 1034); and the “one point control” of the elevator in the Standard-Burdett Case manifestly did constitute a system or method' of operation. In the present case, even if we overlook the form of the claim, there is no novelty in the method, as distinguished from the means, the ajutage, by which it is carried out.
The decree is reversed, and the case remanded, with instructions to dismiss the bill.