SWAN, Circuit Judge.

Infringement is not disputed, if the patent is valid; but appellant challenges its validity. It is contended that the alleged improvement described in the patent is lacking in patentable invention over the prior art; that no inferences favorable to the validity of the patent are to be drawn from plaintiff's commercial success, because that success is to be credited to the noncreasing quality of the lining material used in the commercial product rather than to the teaching of the patent; and that the patent is insufficient as a disclosure. These points will be considered in inverse order.

The claims of the patent are exceedingly broad and general. The four claims adjudicated read as follows:

"1. A necktie, comprising a body portion including a knot-forming part and a woven fabric resilient lining connected thereto, said resilient lining extending into the knot-forming part of the tie.

"2. A necktie, comprising a body portion, and a woven fabric resilient lining connected therewith by loose stitching.

"3. A necktie provided with a lining attached thereto and consisting of woven fabric cut on the bias.

"4. A necktie comprising a body having folds, a woven fabric elastic or resilient lining in the tie body, and loose stitching uniting the folds of the tie body and connecting the woven fabric elastic or resilient lining thereto."

They must, of course, be read and interpreted in the light of the specifications. After stating that his invention relates to improvements in neckties, and particularly to those of the four-in-hand type, the inventor recites that ties are necessarily subjected in ordinary use to pulling strains, which tend to distort the tie and break the stitching which unites the body material and the lining. The prior art, he says, has used inelastic linings stitched to the folded portions of tie, particularly throughout the neck portion, which passes between the folds of the collar, and has failed to overcome the stated objections. His object is:

"* * * To so construct a necktie of the four-in-hand type that the lining shall be sufficiently elastic or resilient in character, and so stitched to the body material as not to cause breaking of the stitching or distortion of the tie, and, at the same time, so that the lining shall be capable of withstanding the pulling strain to which it may be subjected after having yielded lengthwise with the body material to a limited extent, and so that, when the pulling strain shall have been relieved, the body material and lining will assume their original shape and dimensions."

He then describes his invention as consisting of novel features of construction: The body material, such as silk, will preferably be cut on the bias, so that it will be more or less elastic in character; the lining is loosely stitched to the folds of the body material and "* * * is made of woven fabric cut on the bias, so that it shall have limited elasticity or resiliency, and this constitutes a valuable feature of my invention. With the use of lining of woven material having limited elasticity, it will stretch with the stretching of the body material, and at the same time it will not be sufficiently more elastic than the elasticity of the body material as to cause breaking of the stitching, but said lining will suffice to relieve the body material from excessive stress or strain when the tie is subjected to great pulling force. It will be observed also that, when the tie is stretched, the loose stitching will slip, and this will assist in avoiding breaking of the thread."

Finally, he describes the result attained: Distortion will be prevented, as the stitching will not be broken, and the body material and lining will return to normal positions when the pulling strain is discontinued.

What the patentee has disclosed appears to us to be this: A tie having a suitable body material, preferably bias cut silk, with a woven fabric lining, bias cut and of limited resiliency, and a loose stitching uniting body and lining. The resilience of body and of lining must be so related that the stitching will not break when the tie is subjected to the strains of ordinary use, that the "lining will suffice to relieve the body material from excessive stress," and will return to its normal position, bringing the body material with it, when the pull ceases. The invention is in the combination of these things, not in any one of them.

It is argued that this is too indefinite to teach necktie manufacturers how to use the patent; that the material and texture of the lining is crucial to the combination, and is not disclosed. It is true that it leaves to the manufacturer an undefined latitude of choice; he is to select a lining that has the required qualities, and the claims cover any lining which does have them. The question is whether, given this disclosure, the tie maker, of ordinary skill in the art, could make the patented tie without resort to independent invention. See Eibel Process Co. v. Minn. Process Co., 261 U.S. 45, 65, 43 S. Ct. 322, 67 L. Ed. 523. That he must exercise a choice in his selection is not fatal to the disclosure, provided it gives him an adequate guide.

We think it does. First, he must select from among woven fabrics one which, cut on the bias, is resilient enough to return to normal after being stretched by service strains. Such fabrics were available. Next, he must select from among them one which has a relation to the body material. The relation is defined, on the one hand, as requiring the lining to give under ordinary strains. This give is disclosed as a means of preventing the breaking of the thread and the distortion of the tie, and it is clear that this refers to those cases in which, if the lining were too rigid, the silk under the ordinary grasp might pull over its surface, and not only break the thread, but, having only itself to resist the strain, become distorted. On the other hand, the lining must be stiff enough to reinforce the silk, so that together the two will not give so far as to break the thread, and fail to return to "their original shape and dimensions."

All this appears to us fairly described in the disclosure, and, while it requires independent choice, we think it furnishes a sufficient guide to the skilful artisan. Silks vary in elasticity, and it may require some experimentation to determine in each case what lining will do and what degree of looseness in the stitching. But this is inherent in the subject, and we fail to see how the invention can be more definitely stated, unless the inventor is to be required to describe a specific application of his inventive thought, to which he will be limited. Langsdorf did not attempt to say that the woolen linings of his commercial product would alone suffice; perhaps haircloth or other material of limited resiliency, properly related to the resiliency of the tie body, would do as well. The patent left undisclosed no more than was necessary, if the invention was to be claimed in general terms, and not so much as to forbid the possibility of practice. It is not fatal to a patent that the description is in terms of performance and result. That is a vice only when the ordinary artisan called upon to follow the disclosure has no means at hand to reach the result, or when the result may be reached by other means than that disclosed.

It is urged that there is in the patent no definition of the degree of resilience of the lining, except that it must function perfectly; that if the thread breaks, or the tie becomes distorted, under the strains of ordinary use, it was not the lining called for by the patent, while, if the combination works successfully, it was; and that this attempted definition of the resilience of the lining by the result of the use of the tie is within the vice condemned in Holland Furniture Co. v. Perkins Glue Co., 48 S. Ct. 474, 72 L. Ed. ___, decided by the Supreme Court May 14, 1928. As we understand that case, it merely holds that the inventor of one substance — a particular starch glue which will perform the function of animal glue — may not claim all others which will reach the same result. To allow this, as Justice Stone says, would extend the patent monopoly beyond the discovery, and would discourage rather than promote invention. But, when the other examples of the product patented do not need independent invention, and must use what the patentee has disclosed, we see no objection to claiming the whole class. The question really comes down to whether the combination disclosed is as such a contribution to knowledge, and requires in its practice no more than the common knowledge of the art. Unless this were true, the law would deprive the inventor of his contribution, either by limiting him to the instance he gives, or, if he gives none, by holding that he had contributed nothing.

Certainly Langsdorf gave to the necktie trade something new and immensely useful, which the trade at once seized upon and began to copy. The figures and facts which demonstrate the impressive success of his tie are set out in the opinion of the District Court and need not be here repeated. Franc-Strohmenger Cowan v. Siegman, 25 F.2d 108. But it is urged by appellant that the trade learned from Langsdorf's commercial product, not from the disclosure of his patent, and that plaintiff's commercial success is attributable to the merits of a wiry, woolen fabric, known as "Resiline," which was used for a lining, rather than to the merits of the construction taught by the patent. Five judges have passed upon this disputed question of fact. In the Sixth Circuit litigation, Judge Westenhaver in the District Court, and Judge Denison of the Circuit Court of Appeals, decided it in favor of the patentee; while Judges Moorman and Donahue, constituting the majority of the appellate tribunal, invalidated the patent, chiefly, we think, because they deemed it not entitled to any favorable inferences from the success of Langsdorf's product, for the reason that that success was attributed "mainly, if not entirely," to the extraordinary qualities of Resiline. Forchheimer v. Franc, Strohmenger Cowan (C.C.A.) 20 F.2d 553.

Judge Thacher in the instant suit found that the plaintiff's commercial success was not to be ascribed to that material, but to the teachings of the patent. Resiline was not a newly discovered fabric, though its use as a necktie lining was new. Langsdorf did not even know of it until after his application had been filed. The ties submitted with his application were lined with a woolen fabric of English manufacture. Two of his original ties are in evidence, and do not appear to differ essentially from the ties lined with Resiline. He substituted Resiline for the English fabric, because it was made at home and was cheaper. Similar materials had long been available, had the trade known enough to use them in the way Langsdorf taught. The fact that they were not so used before, and that the trade immediately seized upon their use as soon as Langsdorf had shown how to combine them with the silk body material of a tie, is cogent evidence of the patentability of his construction. See Kurtz v. Belle, 280 F. 277 (C.C.A. 2); Van Heusen v. Earl Wilson, 300 F. 922 (D.C.S.D.N. Y). We agree with Judge Denison's statement in his dissenting opinion in the Forchheimer Case, supra, at page 558:

"When for many years the practical art has been trying to get a result by efforts in one direction, and, by abandoning those efforts and turning in the opposite direction, some one, though in a very simple way, gets a new thing, which the whole trade at once accepts as satisfactory, and which largely displaces all competition in its immediate field, a court must be exceedingly sure of its ground in order to reverse the verdict of those long familiar with the subject in its practical aspect."

So we come to the conclusion that the disclosure of the patent is sufficient, and is entitled to favorable inferences of inventive patentability. The question remaining is whether the invention was anticipated.

The nearest patent of the prior art is Heath, No. 335,071. The specifications state that the interlining, because of its extended form, is longitudinally elastic, but is transversely inelastic, so that it will firmly hold and retain the outer facing. The tie made under this patent is of rigid construction, and shows a linen canvas lining cut straight. No purpose is disclosed in its slight longitudinal elasticity; it teaches no such practice as Langsdorf's.

The evidence of the alleged anticipation by S.S. Loeb Co. in 1920, in manufacturing ties with bias-cut, woven haircloth linings, was carefully analyzed and discussed by the District Court. It will suffice to say that we agree with his conclusion that it is too confused to stand for proof of a prior use. Loeb's own witnesses do not agree among themselves. If Loeb had made such ties, he certainly had not appreciated what was in the invention, and had abandoned them by October, 1921. The art gained nothing from him.

The all-silk ties made from a single piece, with the excess material folded in, such as the Tremlett scarf, we do not consider a lined tie, within the meaning of the patent, which implies a lining of a different material, or at least of a different degree of resiliency from the body silk. For the same reason, not to mention others, the silk-lined ties of the prior art were not anticipations. The ties with bias-cut cotton or Canton flannel linings were properly disposed of by Judge Thacher, because not resilient in the sense of the patent. The supposed prior use of the Eisenstaedt tie, with its bias-cut cotton canvas lining, also lacks the stretch and return capacity required by the patent. In none of the ties cited as anticipations was any attention given, so far as appears, to the relative resiliency of body and lining which Langsdorf taught. In the view of the majority of the court, the patent was not anticipated and should be sustained.

The decree is accordingly affirmed.

MANTON, Circuit Judge, dissents.