(after stating the facts as above). It is not necessary to go very far into the state of the art. It is clear and is conceded that, with one exception, everything about Porter’s invention was entirely common. Switches and switch points were old. Operating them by rods or cables running to a distant point was old. The use of a spring or weight to hold the switch point normally in one of its two alternative positions, so that it would always be in that position, excepting while temporary force held it in another position, was old. It was old to operate a switch of this particular kind from a distant point. This kind of switch, properly called semiautomatic or automatic return, had been used in connection with'a main track and a side track for the purpose of making certain that a switch, opened to let a train from the main track onto the side track, should not accidentally remain open, but should automatically return, so as to keep the main track normally always with a continuous rail with the switch closed and the track open for traffic. On this record, Porter is entitled to say that he was the first who reversed this use and employed this mechanism in order to keep the switch normally open, so as not to permit traffic on the main line to get by except while the switch was manually held in abnormal position. Porter’s idea had merit, as applied to electric car traffic; it was a distinct contribution to the safety of travel; and his patent is entitled to consideration as favorable as the law permits.
[ 1 ] In approaching a comparison with other devices, we must first observe that this is a “derailing switch.” If there might be any distinction between a switch and a mere derailer, placed in one rail only, which would do nothing except to force the wheels onto the ground, Porter can take nothing by such distinction. He illustrated and described a connection between a main track and a side track and a device which would turn the wheels from the former to the latter. Neither his specification nor his claims are concerned with what becomes of the car or where it goes after it is diverted; and a switch which sends the car onto a side track or a switch track is just as much an *672infringement or just as much an anticipation as if the car ■ was sent into the ditch. Indeed, Porter expressly says that the track onto which the car is diverted may be a siding.
It is complainant’s theory that changing the adjustment and relation of the parts so as to hold the switch point normally against the main' rail and away from the switch rail, instead of normally away from the main rail and against the switch rail, was a sufficient change upon which to'predicate invention, when taken in connection with the new operating result which followed. This theory is not easy of acceptance, particularly in view of one of the earlier patents (Woodville, No. 140,-230, June-24, 1873), which shows a switch-point operated by a vertical lever, having a horizontal cross-arm extending in both directions, and a weight which could be hung upon one or the other arm as it was desired to set the switch point normally in one or the other position. Whatever might be the ultimate conclusion on this theory, we think the case is clearer and disposed of more satisfactorily by'approaching it from another standpoint.
For a better understanding of the comparison to be made,.we reproduce a drawing .of Porter’s device alongside the drawing of Martel, No. 243,933, July 5, 1881.
J. Y- PORTER.
DEBAILING SWITCH.
*673Martel’s general description and second claim are as follows:
‘■My invention reíales to railway switches; and it consists in a point or tongue pivoted at the end of the inner siding rail and adapted to he swung against the inner main rail, and a lever connected with said point or tongue, and having two arms, to one oí which a weight is applied to hold the point or tongue normally away from the main rail, and to automatically open the switch after the passage of a train to or from the siding, while the other arm is furnished with t wire rope, or its equivalent, by which the switch may be closed by the attendant in the switchhouse or at a. distance.”
“2. in combination with the main and siding rails, arranged substantially as shown and described, the pivoted tongue A, and the weighted lever M, provided with an actuating rope, or its equivalent, and connected with the tongue A, as set forth.”
Martel shows a cable running to a distant point, for the manual operation, while Porter shows a rod; but defendant here uses a cable, and for the purposes of this case the two are equivalents. While Martel uses, for his normalizing agent, a weight applied to a shifting lever at the switch, Porter uses a weight on a shifting lever at the distant point; but defendant employs a normalizing agent (a spring) operating directly at the switch point; and again, for the purposes of this case, Martel’s weight (normalizing agent) and Porter’s weight are equivalents. It follows that there is no material, mechanical difference between Martel and Porter, save in the selected, normal position of the movable switch point. If we might, at first thought, regard the two structures as mechanically different because of this different setting of the parts, we would then observe that the two cannot be distinguished from each other, excepting by the names of the tracks. A main track and a side track are structurally identical; they can be distinguished only by the use to which each is put. The fact that one track curves away while another continues the tangent does not differentiate. If these two devices, Martel and Porter, were put upon the ground side by side, with their above-mentioned, immaterial differences reconciled, and without extensions and continuations not covered by the patents, the skilled observer could not tell which was which — and for the very sufficient reason that there would be no difference whatever. The tracks must be labeled “main track” and “side track,” before the observer could distinguish.
“Main track” and “switch track” are arbitrary names of things; they are mechanically identical, and are intended for, and capable of, the same primary use, viz., to run cars over; they are substantially interchangeable; what is switch to-day may become main line to-morrow because of a blockade on the main line or because the superintendent changes liis mind. If the superintendent operating the road including Martel’s device should, for some temporary reason, send" his regular trains on the ¿rack O i?, opening the switch for them each time, and should use the track P S for his siding, running cars thereon without touching the switch, he would have Porter’s use and structure and would infringe the Porter patent. If, then, the next day, he went back to the old plan of running his trains, he would not infringe. Whether the apparatus infringed or did not infringe would not depend upon any rearrangement, construction, adaptation, or hair’s breadth of physical change, but only upon the shifting manner of its use.
*674In this view of the facts in the present case, we can see only a typical instance of that double use which will not support a patent. A review of a considerable number of decisions (cited in the margin)1 where this double use has been found to exist, confirms our conclusion. Remembering that Porter’s idea was that a new rule of conduct in using an existing structure would be highly desirable, we find special pertinence in Judge Grosscup’s decision in the Voightmann Case (margin). A fusible link, which had been used in other places, was applied to an external shutter. Judge Grosscup said (138 Fed. 57, 70 C. C. A. 483):
“It is' possible that Voightmann was the first to conceive that windows thus constructed would be a valuable adjunct to fireproof buildings. If so, it is the previousness of his conception that constitutes the merit of his co-ealled invention; for the mechanical embodiment of that conception is old. But it does not follow that a conception is patentable merely because it is first in time. Concept, alone, is not patentable. Concept must be accompanied by mechanical embodiment; and, as the law now stands, the mechanical embodiment, to make the invention patentable, must itself be unanticipated. * * * Voightmann possibly has pointed out to the world a wider use of the pre-existing art than was before known. But the discovery of an enlarged use is not, of itself, patentable invention.”
So, also, in the Bullock Case (margin), Judge Gray said (162 Fed. 28, 36, 89 C. C. A. 68, 76);
“But this function was dormant in the [device of the prior art]. Surely invention cannot be claimed in the appropriation of an old device, by reason of the- unthought of and undisclosed function in question.”
The cases relied upon by appellee (margin),2 to the effect that a mere *675change of location of elements may involve invention, all refer- to a rearrangement causing a new mutual interrelationship of the old elements. They do not involve the use of the old elements, in the old position and in the old relationship to each other, to get a “new result.” For example, in the Keystone-Adams Case, it seems true that the revolving shaft with wings was like an older one, save that it revolved in an opposite direction; but to permit this reverse revolution required mechanical adaption in other parts of the device, and results could have been made certain only by experiments. In the Kellogg-Dean Case, the condenser was for the first time associated with the combination which was patented. In the Hobbs-Beach Case as well as in the Potts-Creager Case, we have the typical transfer from one art to another, and a decision that, while there was analogy between the arts, it could not be said that there was identily of function, and hence that the comparatively slight mechanical changes and adaptations which were made were sufficient to support invention. We have observed no more extreme instance, where a slight physical change accompanied by new use and result, has been held sufficient to sustain a patent than is furnished by our own decision in Dunn Co. v. Standard Co., 163 Fed. 521, 90 C. C. A. 331, and on second appeal, 204 Fed. 617, decided March 13, 1913 ; but here, also, there was physical change which, (hough slight, signified transformation of one thing into another.
It often is not easy to determine whether a case discloses that “new result” which is one of the criteria of invention. The phrase is not of absolute meaning nor always applied with discrimination. In one sense every new use gives a new result — freezing fish with a particular form of device (Brown v. Piper), dredging with a propeller screw (Atlantic v. Brady), operating jail doors from a distance (Fond du Lac v. May), a safety gate for street car platforms (Aron v. Railway), a soft metal switch-plate holder (Johnson v. Toledo)— each was a thing accomplished which had not been done before. But when it was seen that the means employed were old, even in substantially the same form and substantially the same association, each was held to be not a new result, but only a new use of an old device — a newly observed or discovered function of a well-known thing. To the same point are all the double-use cases.
On the contrary, in cases where a new result has been adjudged to support patentability, the inquiry whether there was a “new result” has been for the purpose of testing out whether the observed novelty in the means employed could be considered patentable novelty. If there is a new structure or combination of means, then the fact that a “new result” follows tends to indicate the presence of the inventive faculty in the rearrangement of mechanical, electrical, or chemical elements or method steps which constitute the new combination; and to be possibly effective for this purpose it is immaterial whether the result be ultimate or intermediate — in the final product or in the manner of its making. Examples are: A more rapid production (Loom Co. v. Higgins, 105 U. S. 580, 26 L. Ed. 1177); a different product (Potts v. Creager); special qualities in the product (Diamond v. Consolidated, *676220 U. S. 428, 31 Sup. Ct. 444, 55 L. Ed. 527, and Ferro Concrete Const. Co. v. Concrete Steel Co., 206 Fed. 666, opinion of this court filed May 6, 1913); more perfect operation of the device (Kellogg v. Dean); etc. But when the means employed are wholly old, inherently and as well in their combination, apparently similar “new results” are, of necessity, only new or double uses, -and no amount of them will persuade that invention exists.
What then is Porter’s “new result,” considered either mechanically or functionally ? Is it anything more than maintaining one track normally closed, capable of being held open by force and while the force is applied, and then automatically closing again? Is not his safety plan for handling electric cars approaching a railroad crossing merely a new utility of an old result? We think it should be so classified, and that, as said in the Voightmann Case, supra, “the mechanical embodiment to make the function patentable must itself be unanticipated”; or, as said in the Platform Gate Case (Aron v. Manhattan Ry. Co.); 132 U. S. 84, 10 Sup. Ct. 24, 33 L. Ed. 272, “his right to a patent” must depend.upon the novelty of the means employed'“and not the end accomplished.”
There is another consideration (common, indeed, to all double-use cases) which further confirms our conclusion. An inventor and pat-entee is entitled to all the uses of his structure, even if he did not know or foresee all those uses. Goshen Co. v. Bissell Co. (C. C. A. 6) 72 Fed. 67, 74, 19 C. C. A. 13. So, Martel would be entitled to use his device for the safety purpose of Porter, since to do so required no mechanical change or adaption; and, in this connection, it should be* noted that Martel’s second claim above quoted reads on and covers Porter’s device, and Porter’s device, within the life of the Martel patent, would have been an infringement thereof. Such a comparison with the claim of an old patent is not always important; but, when the old patent and the new stand side by side as alternatives rather than as merely original and improvement, the comparison is at least significant.
There is still another consideration. The steam railroad that Porter is about to cross and which is the foundation of the operating merit in his plan is not named as an element in his combination; but if it were named, or if it were to be read in, the language of Mr. Justice Blatchford in Fond du Lac County v. May, 137 U. S. 407, 11 Sup. Ct. 102, 34 L. Ed. 714, would seem applicable:
“As tire mechanical operation and effect of the patented devices are the same, whether there he a grating or other barrier [steam railroad] or not, there is no patentable combination between the devices and the grating [steam railroad]. The grating performs no mechanical function and has no mechanical effect.”
[2] Quite obviously, we think, it cannot be important whether the cause of stopping before deciding which track to take is another railroad crossing with its inherent danger, or is another train ahead which it may be best to wait for and may be best to go around, or is any' other common, operating situation — or whether the man sent ahead to *677turn the switch is a freight brakeman who sets it as he has been directed or a conductor who sets it as he may decide.
The decree must be reversed, with costs, and the record remanded, with instructions to dismiss the bill.
In the Supreme Court: Phillips v. Page, 65 U. S. (24 How.) 164, 16 L. Ed. 639; Brown v. Piper, 91 U. S. 37, 23 L. Ed. 200; Roberts v. Ryer, 91 U. S. 150, 23 L. Ed. 267; Planing Machine Co. v. Keith, 101 U. S. 479, 25 L. Ed. 939; Heald v. Rice, 104 U. S. 737, 754, 26 L. Ed. 910; Atlantic Works v. Brady, 107 U. S. 192, 2 Sup. Ct. 225, 27 L. Ed. 438; Pennsylvania Co. v. Locomotive Co., 110 U. S. 490, 4 Sup. Ct. 220, 28 L. Ed. 222; Miller v. Foree, 116 U. S. 22, 6 Sup. Ct. 204, 29 L. Ed. 552; Fond du Lac County v. May, 137 U. S. 395, 11 Sup. Ct. 98, 34 L. Ed. 714; Lovell Co. v. Cary, 147 U. S. 623, 13 Sup. Ct. 472, 37 L. Ed. 307; National Co. v. Boston Co., 156 U. S. 502, 515, 516, 15 Sup. Ct. 434, 39 L. Ed. 511, reviewing the platform gate case (Aron v. Manhattan Ry. Co., 132 U. S. 84, 10 Sup. Ct. 24, 33 L. Ed. 272), the transom case (Wollensak v. Sargent, 151 U. S. 221, 14 Sup. Ct. 291, 38 L. Ed. 137), the engine valve case (Blake v. San Francisco, 113 U. S. 679, 5 Sup. Ct. 692, 28 L. Ed. 1070), and the billiard cue rack case (St. Germain v. Brunswick, 135 U. S. 227, 10 Sup. Ct. 822, 34 L. Ed. 122); Mast F. & Co. v. Stover Mfg. Co., 177 U. S. 485, 20 Sup. Ct. 708, 44 L. Ed. 856.
In this court: Steiner Co. v. Adrian, 59 Fed. 132, 8 C. C. A. 44; Griswold v. Wagner, 68 Fed. 494, 15 C. C. A. 525; Stearns v. Bussel, 85 Fed. 218, 29 C. C. A. 121; Fry v. Rockwood Pottery, 101 Fed. 723, 41 C. C. A. 634; Johnson v. Toledo Co., 119 Fed. 885, 56 C. C. A. 415.
In other Circuit Courts of Appeals: Bettendorf v. Little, 123 Fed. 433, 59 C. C. A. 473; Jones v. Cyphers, 126 Fed. 753, 62 C. C. A. 21; Voightmann v. Perkinson, 138 Fed. 56, 70 C. C. A. 482; Baker v. Duncombe Co., 146 Fed. 744, 77 C. C. A. 234; Bullock Co. v. Gen. Elec. Co., 162 Fed. 28, 89 C. C. A. 68.
Topliff v. Topliff, 145 U. S. 156, 12 Sup. Ct. 825, 36 L. Ed. 658; Keystone Co. v. Adams, 151 U. S. 139, 14 Sup. Ct. 295, 38 L. Ed. 103; Potts v. Creager, 155 U. S. 597, 15 Sup. Ct. 194, 39 L. Ed. 275; Hobbs v. Beach, 180 U. S. 383, 21 Sup. Ct. 409, 45 L. Ed. 586; Kellogg Co. v. Dean Co. (C. C. A. 6) 182 Fed. 991, 105 C. C. A. 545.