In McMillan v. Barclay the patent upon which the present hill is founded was contested upon various grounds, all of which were fully considered hy the court, and a decree was rendered in favor of the complainant. No question is now made touching any of the specific defences set up in that case, but the patent is assailed for a reason not before.'suggested. It is urged that the patent, No. 63,-917, set up in this case, is a duplicate of patent 52,730, granted to the same persons and for the same invention.
On the twenty-third of July, 1855, the patent involved in this case was applied for, and was disallowed by the commissioner of patents on the twenty-fifth of August, 1866. After repeated efforts to obtain a rehearing, in the early part of 1867, the applicant amended his specification, and again pressed his application for consideration.
This amendment consisted in a modification of the claim, the body of the specification and the drawings remaining unchanged. A fe-examination was finally made, and a patent granted on the sixteenth of April, 1867. This patent is for “an improved application of steam power to the capstans of *723vessels,” and the claim is for “rotating a capstan placed on deck of a boat by means of an auxiliary engine, when said engine and capstan are placed forward of the steam boilers of said boat, substantially as herein before described, and for the purposes set forth.”
Without having amended the application for this patent, the complainant McMillan, on the twenty-fifth of April, 1865, filed an application for a patent for a “mode for working a capstan by steam,” which was allowed and issued February 20,1866, numbered 52,730. The claim in this patent is for “the arrangement of the wheel l, m, n, o, k, j, i, h, e, and d; shafts 6, 5, 4, 3 and B; capstan barrel p; heads q and r, and pins Q; the whole being constructed, arranged and operated substantially as heroin before described, and for the purpose set forth.”
Are these patents, then, for the same invention?
1. In McMillan v. Barclay, 5 Fisher, 189, the import of the claim of the patent of 1867 was hold to be “operating the capstan of a steamboat by certain mechanical means, actuated by steam derived from an auxiliary engine, where both the engine and the capstan are stationed on the deck of the boat forward of the steam boilers. The mere effect indicated is not claimed, for that would be clearly unallowable, but it is this effect produced by means substantially as described and employed under the conditions stated.” In other words, it is for a method of producing a useful result to be practiced by the use of mechanism described in the specification, under conditions therein prescribed.
2. The claim of the patent of 1866 is for an arrangement or combixiation of specific devices, which embraces only some of the devices described in the patent of 1867, and adds others which are not therein described. The specification contemplates the use of this combination in practicing the method described in the patent of 1867, and it is treated and described throughout as an improvement in the arrangement and combination of the mechanical appliances by which that method is to be effectuated by securing new and better re-*724suits. The claim is limited to a specific mechanical combination, and hence the patent can have no broader scope.
3. In the astute and able argument of the respondents’ counsel the claim of the patent of 1867 is erroneously assumed to cover every means of accomplishing the proposed result. So broad a definition of it must certainly be discarded, for reasons fully discussed in O’Reilly v. Morse. As already stated, it is for a method of producing a certain result by means of mechanism described in the specification, under certain essential conditions. The mechanical agencies required to transmit the power of the auxiliary engine to the capstan are component parts of the method, and it cannot therefore be said that the use of connecting mechanism substantially different from that'described in the specification would be an infringement of the patent of 1867. The same result may be produced if different means to that end are employed, without invading the patented method. Hence it is only the mechanical instrumentalities described and referred to in the claim, and those merely colorably different from them, which are within the protection of the patent.
4. The patent of 1866 is clearly for a combination of mechanical elements indicated in the claim," arranged and operating as described. The claim does not embrace all the mechanical devices described as constituents of the method covered by the patent of 1867, and' it includes others not described in the latter patent. In a mechanical or a patentable sense, then, the two patents do not cover the same invention.
5. But conceding that the invention claimed in the patent ot 1866 is within the scope of the patent of 1867 as one of the means adapted and intended to be used in effectuating the method claimed in that patent, is the patent of 1867, therefore, void?
The scope of a patent must be determined by its claims, and it does not necessarily cover all the descriptive matter in the specification. The effect of the specification is to describe the invention claimed in such full, clear, and exact terms that any one skilled in the act to which it appertains can *725make and use it, and thus that it may be available to the public after the expiration of the patent. Where the invention consists of a mechanical combination, it is not only proper, but requisite, to describe the construction and arrangement of its component parts, the purpose for which it is to he used, the mode of its operation, and its relations to other mechanism with which it is intended to co-operate. This mechanism may be indispensable to the operativeness of the combination, but it does not constitute a part of the invention unless it is embodied in the claim. In Forbush v. Cook, 2 Fisher, 669, Mr. Justice Curtis says: “If inclined wires are necessary to the action of the combination specified, so are many other parts of the machine; and all parts necessary to the action and combination specified might be said to enter into the mode of combining and arranging the elements of the combination, but need not be, and ought not to be, included in the combination claimed.”
Can there be any doubt, then, that any one night lawfully practice the method of rotating a capstan described in the patent of 1866, simply substituting other mechanism for the combination therein claimed, or what is the same thing, forming a new combination, by discarding some of the elements of the patented one ? And why? Because only the combination claimed, and not the integral method described, is covered by the patent.
With this limitation of the scope of the patent of 1866, might not a valid patent be subsequently granted as that of 1867, for an integral method, of which the invention claimed in the first patent is one of the elements ? I know' of do sufficient reason why this may not be. The patents are incompatible. In a patentable sense their subjects are different, and hence they do not cover the same invention. It is well settled that a mechanical device, adapted and intended to be used in combined connection with other devices, may he patented by itself, and that the combination of which it is a constituent part may also be separately patented. For what reason? Because both patents do not cover the same invention.
*726Upon the same principle, if the complainant had built an engine of novel construction, describing it as specially fitted and intended for use in his method of rotating a capstan, he might 'patent it alone, and he might also obtain a patent for his method as a whole, describing his previously patented engine as one of its component parts. I am, therefore, of opinion that the patent of 1867 is not invalidated by reason of anything contained in the patent of 1866.
6. These views are not impugned by the cases referred to in the argument, as I understand them. A brief reference to two of them, as representing the whole class to which they belong, will sufficiently illustrate this. The fundamental principle of all these cases is that two patents for the same invention cannot co-exist; and that where two are issued to the same person the last one is void.
In O’Reilly v. Morse, 15 How. 63, the order of the patents in this case is inverted. Morse had obtained a patent broadly covering his own invention. He afterwards obtained a patent for a new arrangement or combination of the telegraphic mechanism by which the electro-galvanic current might be passed along the whole line without interruption by the local circuits. The court held the latter patent to be valid, saying: “Nor can its validity be impeached upon the ground that it is an improvement upon a former invention, for which the patentee had himself already obtained a patent. It is true that under the act of 1836, § 13, it was in the power of Professor Morse, if he desired it, to annex this improvement to his former specification, so as to make it from that time a part of the original patent. * * Nor is he bound in his new patent to refer specially to his former one. All that the law requires of him is that he shall not claim as new what was covered by a former invention, whether made by himself or any other person.” The evident reference here is to the scope of the two patents under consideration, and its import is that, although Morse might have included in his first patent the arrangement or combination described in his second patent, yet as it was not claimed in the former, and was not, *727therefore, covered by it, he might make such arrangement the subject of a new claim and obtain a patent for it.
In The Suffolk Co. v. Hayden, 3 Wend. 315, a patent for an invention which was described, but not claimed, in a previous patent was sustained upon the ground that the invention, although described, was not covered by the first patent, and that this omission to include it did not operate as an abandonment of the improvement to the public.
There is another feature in the case which brings it into notable resemblance to the present case. Before the second patent above referred to was granted, an application by the same inventor was on file in the patent office for a patent for improvements in the interior arrangements of an elongated trunk, previously in use for cleaning cotton, among others, in the screen. Pending this application he applied for a patent for the improvement in the screen, in both the specifications filed similarly describing this improvement. A patent was granted upon this application, and sometime afterwards a patent also upon the original application. In reference to these two patents the court say that if they are for the same improvement the last one issued, and not the first, is void; obviously not for the reason that the description in both specifications was the same, but that the allowance of the last patent was futile, and that the original application for it was superseded and abandoned by the successful application for idle first. But the court says further that the last issued patent does not apj)ear to be for the same improvement covered by the first, Init for a combination with it of other devices, thus clearly implying that the patent is good.
Now these cases, with others of like character, establish:
1. That of two patents for the same invention the one last granted is void, although it may have been first applied for.
2. Whether two patents cover the same invention must he determined by the tenor and scope of their claims, not by the description in the specifications.
3. That separate patents for severable parts of the same invention may be granted, although the whole invention is fully described in each of them to explain the purpose and *728mode of operation of the parts covered by the claims in such patents. '
4. That the connection or combination of a patented device or improvement with other devices may be the subject of a valid subsequent patent,
Accordingly, then, the patent of 1867 must be held to be valid, and, as the defendant is proved to have infringed it, there must be a decree for the complainant, as prayed for.
Note. — Bee Babcock v. Judd, ante, 408.