SHIPMAN, Circuit Judge.

This is an appeal from a decree of the circuit court for the Southern district of New York which dismissed the bill brought by the H. W. Johns Manufacturing Company against the defendants for an infringement of letters patent No. 257.167, dated May 2, 1882, and issued to Henry W. Johns, for an asbestus rope steam packing. In Judge Townsend’s opinion the full text of the specification is quoted (60 Fed. 900), from which it appears that the entire invention was an asbestus rope, with or without a central core of asbestus or hemp; the outside fibers of the rope being laid flat in the direction of the rope, and the interstices between the outside strands being filled or evenly built up by a sizing composed wholly or in part of asbestus. The specification adds that “the central strand may be saturated with a solution of India rubber, which I find imparts a degree of elasticity which does not interfere with the solidity of the packing.” An examination of the descriptive part of the specification, apart from the claim, would lead the mind to suppose that this addendum was not an integral part of the invention, but was a detail which was, .if used at all, to be employed in connection with the external paste coating of the rope, which apparently constituted the significant feature of the improvement. The third claim, however, is as follows: “The asbestus rope packing for steam joints, composed of a series of strands of asbestus, with a central core saturated with a solution of India rubber, as hereinbefore set forth.” The correspondence between the patent office and the patentee’s attorney shows that the patent office allowed this claim; the patentee having-pointed out that it was for a central core saturated with a solution of India rubber, and not for a solid rubber core which had been previously used. The defendants sell an asbestus packing not externally cemented, but having a central core, upon' which a sizing made of starch is used to cause the outside strands to adhere to the inside core. The question is one of infringement, and is whether starch can be considered an equivalent of India rubber, in this patent.

The complainant does not take the position that the adhesive propertv of the India rubber solution produced an unanticipated benefit, by causing the outside strands to stick somewhat to the central core, but its theory is that the claim specifies India rubber generally, — an article which has many known qualities, among which is' adhesiveness, — and therefore all materials which cause adhesiveness. are equivalents. Thus, the expert for the complainant understands *987that any solution which will be adhesive, but will not impart a degree of elasticity which interferes with the solidity of the packing. and will permit the separation of the strands, will he within the patent; and he further understands that the starch sizing, while adhesive, has not imparted a degree of elasticity which interferes with the solidity of the packing, and is therefore the equivalent of India rubber. The argument of the complainant is to the effect that India rubber has known qualities, and therefore any material which has any of the qualities of India rubber which have proved to be useful in rope packing is an equivalent. The specification does not permit this construction of the claim. Ko allusion was made to the adhesive quality of India rubber. It was to be used simply to impart elasticity, and not to impart adhesiveness; and the specification would he satisfied if the saturated core was not, and was so made as not to be, adhesive. Its language obviously mean! that the central strand is saturated with an India rubber solution, so as to produce elasticity in the core, and thereby elasticity in the rope; and the meaning of the claim is, an ashestus rope, its central core being saturated with an India rubber solution for the purpose named in the specification. It is true that an inventor can secure, by means of ids patent, the use of the patented combination for beneficial purposes which were unknown to him, and “is entitled to the benefit of all the uses to which it. [the patented machine] can he put, no matter whether he liad conceived the idea or not.” Roberts v. Ryer, 91 U. S. 150; Wells v. Jacques, 5 O. G. 364, Fed. Cas. No. 17,398. Thus, a person who had ascertained That a solution of India rubber was useful in an ashestus rope packing for another purpose thau that of producing elasticity, and who attempted to use it for that purpose alone, would be an infringer. .But the patentee cannot prevent the use of ingredients in the combination winch may accomplish the previously unknown result, but are not equivalents of the patented ingredient in the performance of the function which is specified and described in the claim; ami therefore, if the starch sizing produces no elasticity, the defendants’ packing is not an infringement, but is a new, though perhaps not a patentable, combination. When one combination is asserted to be an infringement of another patented combination, “a device in one, to be the equivalent: of a device in another, must: perform the same functions.” Rowell v. Lindsay, 113 U. S. 97, 103, 5 Sup. Ct. 507. The complainant therefore adhered, in its part of the record, to the idea that a solution which should produce elasticity was desirable, and even necessary, and discarded, to a certain extent, the theory of its expert (who did not understand that any oilier quality than the adhesiveness of the rubber is made use of in the solution referred to in the complainant’s patent), and attempted to prove that a starch sizing, which is adhesive, imparts also elasticity to an ashestus rope. One witness thought that it is necessary that some elastic and adhesive substance should he used, and he found that starch sizing produced elasticity, because, having caused the outside strands to adhere to the core, they can be laid more loosely together than otherwise would have been per*988missible.- This was-not the elasticity which the patentee stated in his specification when he said that a solution of rubber which saturated the core imparted elasticity, — in other words, made an elastic core. The other witnesses for the complainant added nothing to the general stock of knowledge in regard to the ability of a starch sizing to impart elasticity to a rope. .In our opinion, a sizing made of starch, to cause the asbestus strands to stick to the central core, is not the equivalent of a solution of India rubber which saturates the core to make it elastic. The decree of the circuit court is affirmed, with costs of this court.