This patent has been the subject of marked vicissitude. It was at first sustained by Judge Dallas in’ this court in Hanifen v. E. H. Godshalk Co. (C. C.) 78 Fed. 811, but upon a rehearing, on account of certain expert evidence, by which he felt himself controlled, he decided against it. On appeal, however, he was reversed, and the patent upheld, although the court of appeals was not unanimous, Judge Butler dissenting from the views of Judge Shiras and Judge Acheson, who constituted the majority 28 C. C. A. 507, 84 Fed. 649. It came up again before Judge Gray in Hanifen v. Lupton (C. C.) 95 Fed. 465, where the validity of the patent was conceded, the suit being defended on other grounds. It next appeared in the Second circuit, and was sustained by Judge Townsend in a well-considered opinion (Hanifen v. Price [C. C.] 96 Fed. 435); but he in turn was reversed by the court of appeals *846of that circuit in an opinion by Judge Shipman, and the patent declared invalid (42 C. C. A. 484, 102 Fed. 509). On account of these conflicting decisions in the two circuits, the supreme court allowed a certiorari in the latter case, and it was supposed that the matter would be thus put at rest. But again there was a serious difference of views, which resulted in an affirmance by an equally divided court. Such an affirmation establishes no precedent or principle (7 Am. & Eng. Enc. Prac. p. 44), and, so far as this court is concerned, the decision of the court of appeals of this circuit sustaining the patent therefore remains. With no new considerations advanced, the question of its validity cannot be regarded here as an open one. At the same time I have re-examined it as though it were, and, with the benefit of all that has occurred since it was rendered, I see no occasion to vary from the conclusion so reached.
The patent was issued in 1887 to Eevi Bywater, and, according to the second claim which is the one in controversy, the invention is declared to be “a knitted fabric, composed of face and back yarns of different materials, the face yarn being looped at regular intervals and on alternate stitches of adjacent rows of the back yarn, and being matted and curly, and having a smooth back, whereby the said fabric has the appearance of looped or Astrakhan cloth as described.”
In the specifications which precede, the invention is said to consist of “a new and improved textile fabric having the face yarn thereof looped on the stitches of the back yarn; * * * the said face, which is formed of mohair, worsted, or other yarn, being beat up so as to present a wavy or curly surface, and the back, which is formed of woolen or other yarn, brushed so as to present a smooth surface, the fabric having the appearance of looped or Astrakhan .cloth.” In carrying out his invention the patentee declares that he employs a circular knitting machine, a partial description of which he gives, and in the operation of knitting the fabric he says that the thread by which the rough face or Astrakhan effect is produced is so placed upon the needles by the backing wheels as to be alternately in front of and behind two needles, the backing wheels being so set in a four-feeder machine that for successive rows of the fabric they alternately press back different needles, thus forming the loops on alternate stitches of adjacent rows. It will be thus seen that the patent is distinctly for a textile fabric of specific character and designated structure. It is not for the process by which it is made, nor the machine for making it, each of which is referred to merely to aid in describing it. The question, therefore, on which the validity of the patent depends is whether the fabric is new, or has been previously, in whole or in part, anticipated. On this question it is brought into comparison with the prior British patent of James Booth in 1881. There are other references, but, without stopping to discuss them, the case seems to turn on this one. Unquestionably imitation Astrakhan existed before either of these inventors; but it was the woven, and not the knitted, article, which Bywater was the first to actually produce. As said by Judge Dallas in his first opinion: “Knitted Astrakhan was created by Bywater, and this he accomplished not by merely applying the skill of the knitter to effect a change in either *847of their [i. e., prior] products, but by the exercise of his own inventive faculty.” That is the whole case in a nutshell, and it is abundantly sustained by a proper consideration of the matters involved. Booth did not aim to knit Astrakhan, and his patent, unaided, was not calculated to do so. What he claims to have invented was simply a novel description of looped fabric of ornamental appearance, whatever that might mean. Looking to the process by which if was made, we find that he employed for the back or body the ordinary wool yarn, capable of being afterwards felted, and for the face a worsted or long-fibered yarn that would not felt. This face yarn, which ultimately constitutes the loops, is laid in between the needles in any desired order, and tied to the body by the tie thread used in fleecy-backed hosiery, known as “stockinet.” The fabric so produced is then subjected to the process of fulling or felting, by which the back or knitted portion is shrunk or felted together, and the face yarn, “being laid in straight and tied,” is thereby caused to project from the body of the fabric in loops, producing, as he says, a very ornamental appearance. Did this disclose knitted Astrakhan? It is not so claimed by the inventor, by whom we must assume that the invention would be given its widest possible scope; and the suggestion of Judge Shipman in Hanifen v. Price, 42 C. C. A. 484, 102 Fed. 509-512, that he had in mind to make a knitted fabric which resembled the woven article ■ (unless there was something in that record of which I am not aware), would seem to be an entire misconception. The ornamental effect is all that he indicates, produced by the loops running in longitudinal or diagonal stripes, or with mixed irregularity in longer or shorter floats. Unless knitted Astrakhan is necessarily embraced by and involved in the fabric so described, it cannot be said to be disclosed by the patent. It is no answer that it has been produced by other knitters following the patent. We have no evidence of that kind in this case, however it may have been shown in others; Turtle merely testifying that he produced knitted Astrakhan in 1885 from a piece of the manufactured fabric which Booth had previously made. But even so, the additional skill by which this was able to be brought about, if it was, is an element of which we have no means of judging at this time. It is to be remembered also that, as a foreign patent it is not so much what could have been made out of it as what was inherent in it,—not its possibilities, but what it substantially diplayed,—that is to govern. Seymour v. Osborne, 11 Wall. 516, 555, 20 L. Ed. 33; Hanifen v. E. H. Godshalk Co., 28 C. C. A. 507, 84 Fed. 649. Neither will it do to say, as in the case referred to in the Second circuit, that the loops of Booth, being made of long fibered worsted, would necessarily twist and curl. That is not claimed for them in the patent, and we cannot assume that it was desired or intended; and, even if it was, to meet the Bywater patent, as well as to imitate Astrakhan fur, the loops must both mat and curl so as to present the shaggy appearance of the animal. This is an imposed effect, due not only to the yarn used and the way it is laid in, but to the subsequent treatment of the fabric, and there is nothing suggested by Booth calculated to bring it about.
*848What is there, then, to stand in the way of the invention of By-water? It is said that, even if Booth did not anticipate Bywater, still Bywater’s fabric is merely an advance or improvement, the result of a choice of yarns, which would suggest itself to any skilled knitter seeking to produce this special effect, and therefore not involving the exercise of inventive faculty. “It is doubtless true,” says Judge Shipman in Hanifen v. Price, supra, “that a distinction can be traced in the two patents as to the purpose that was in the mind of each patentee; the earlier patentee having in mind to make a knitted fabric which resembled Astrakhan in having a looped face varied according to the fancy of the knitter, and the latter patentee having in mind to closely imitate woven Astrakhan. But, in our opinion, that fact gives to the Bywater improvement nothing of a patenable character.” It is also admitted by Judge Shipman at another place in his opinion that not all long-fibered yarn will curl, so that whether it would or not in any given case would be problematical. It consists simply “in the selection of yarns to produce a particular effect or particular style of goods, and, after Booth had told the public (if, indeed, he told them anything of novelty) how to produce a knitted loop fabric, there was nothing of an inventive character in the selection of yarns to make a looped curly fabric more curly and matted. The validity of the Bywater patent should not rest upon the fact that the Booth patent said nothing about curly wool, but its validity depends upon the question whether, after Booth’s improvement had been described, there was anything of an inventive character left. Bywater, by a wise choice of yarns and continued mechanical improvement, succeeded in presenting to the public an attractive fabric, and had the great merit of being patient in the work of mechanical development, but the inventive idea was absent.” And again: “We do not regard the question of the Booth anticipation as the controlling one in the case. Indeed, the importance of the Booth patent consists in its bearing upon the question of patentable invention, and we coincide in the view of Judge Butler that upon the admitted facts in the case the work of Bywater was the mechanical work of an intelligent spinner, and was destitute of the element of invention.”
In this arraignment of the patent there are concessions, large enough, as it seems to me, to produce a very different result. The fact that there are distinctions between the work of Booth and that of Bywater, and that Bywater, by a wise choice of yarn and patient work in mechanical development, produced the fabric which he did, goes a long way to meet the charge of the want of invention which is made. What is there'’ that makes up the inventive faculty beyond this? The error into which I can but feel that the learned court fell consists in the assumption that Booth not only had it in mind to make, but actually produced, a fabric which resembled knitted Astrakhan, and that Bywater merely improved upon it. The difference between the work of the two inventors is of no such incidental character. It may be that by adaptive manipulation an Astrakhan effect can be worked out with the Booth loops, but, taking the patent as it reads, it is clear that it will not be produced naturally. The only provision for creating loops is by the felting of the back or body by *849which the face yarn between the tie points is loosened and protrudes. That these may curl is possible if a curly or crinkly yarn happens to be used; but that they will both mat and curl, which is the essential feature of the Bywater fabric, is riot suggested. It will not do, therefore, to characterize what Bywater did as built upon the work of Booth; or to say that Booth made loops which were a little curly, and By-water only made them more so. Neither can the result achieved by Bywater be said to depend simply on the selection of yarns calculated to produce a special effect. We must look at what he did as a whole, and so judge of it, analyzing, rather than dividing and dissecting, it. As already said, what he invented was a textile fabric of designated structure, produced by a specified method, and having an intended imitative Astrakhan effect. Generically, this may be of the same class as the style of fabric designed by Booth, both providing for looping the face yarn on the back or body regularly or irregularly according to a prearranged design. But the means em ployed by each is different; nor can either be said to suggest the other. Booth uses a long-fibered yarn or worsted, so that it will not felt; Bywater, mohair or luster, which will mat and curl. In Booth the face yarn is laid in straight, and the loops are produced by the felting process to which the fabric is subsequently subjected. Bywater makes his loops in the process of knitting, and controls their character from the start. In the Booth the face yarn is fastened down by the tie thread at each end of the proposed loop, which stands in the way of its curling. In the Bywater it is simply tied to the back as a part of the stitch, leaving it free to mat and curl as desired. You cannot make Astrakhan by the Booth patent unaided, or (as I am almost prepared to say) except as you in fact depart from it. By the Bywater it was effectually and designedly and for the first time produced.
No doubt it is not every slight advance in the art, such as is constantly being made by mere mechanical skill and adaptation, that is to be considered as invention. The design of the patent laws, as it is said in Atlantic Works v. Brady, 107 U. S. 199, 2 Sup. Ct. 231, 27 L. Ed. 438, “is to reward those who make some substantial discovery or invention which adds to our knowledge, and makes a step in advance in the useful arts.” It is at the same time recognized that the impalpable something which is said to distinguish invention from simple mechanical skill is at times not easy to discriminate or to define (McClain v. Ortmayer, 141 U. S. 419-427, 12 Sup. Ct. 76, 35 L. Ed. 800); and in the attempt to judge of it after the long lapse of years we are likely to be misled by the increased knowledge intermediately attained. That is the danger here. No knitter had produced Astrakhan cloth before Bywater, and how, then, can it be affirmed that any skilled knitter could? He could if he had the inventive genius to conceive it as Bywater did, but he could not without it. One criterion of invention is that others have sought and failed, even when the process is so simple, when discovered, that many believe they could have produced it if required. Walk. Pat. § 26. And why may that not be applied here? If, as it is stated, a gold medal was awarded Bywater by the Crystal Palace Wool Exhibition in *8501881 for his curled stockinet,- which was nothing more or less than Astrakhan, such a recognition by those most competent to judge, occurring at the time, ought to afford speaking proof of his real achievement. Conscious, therefore, as I am of my own inexperience in such matters, and having the highest regard for the opinion of the court from whose views I am compelled to differ, I am nevertheless unable to see why so material an advance in the textile art can be regarded as devoid of invention. To characterize it as amounting to no more than a selection of yarns, or a matter of mere mechanical skill, is a want of appreciation, as it seems to me, of what it really was,—a new textile fabric of great merit, which others had sought to produce in vain.
With the question of infringement—to which, by the way, the evidence of the defendants is mainly directed—I have no serious difficulty. The structure of the fabrics manufactured by them is displayed in the exhibits which have been produced, where the stitching of each is given in exaggerated form. These have been of the greatest assistance to me in the disposition of this part of the case. But notwithstanding the opinion of the experts which accompanies them, I am convinced that the great majority of these fabrics, if not in fact all, fall within the terms of the patent, and offend against it. As already seen, the inventor specifies with regard to his fabric that the face yarn is to be “looped at regular intervals, and on alternate stitches of adjacent rows of the back yarn, and, being matted and curly, and having a smooth back,” has the appearance of looped or Astrakhan cloth as described. That the greater portion of the defendants’ goods in outward appearance infringe upon the patent can hardly be denied. Some one or two in which the face yarn is of different colored threads, by which the diagonal lines of the looping is brought out, may not seem so much to do so. But it is the structure, rather than the appearance, which determines the infringement; or, rather, the appearance and the structure combined, the latter being the controlling feature. Turning, then, to the patent again, and analyzing its terms, we find that the looping of the face yarn must be “at regular intervals, and on alternate stitches of adjacent rows of the back yarn.” Admittedly these terms are not technical, and are to be construed according to their ordinary meaning, in which the opinions of experts are of little aid. By “regular intervals” we are to understand intervals that conform to a prescribed rule. This does not apply to the loops, be it noted, but to the intervals between; and by no means does each interval have to be a single stitch. Hence the loops may be of any desired irregularity, and pass over any number of stitches. Neither do I see why a regularity of intervals is not preserved where one set of loops of a certain fashion follows on after another, even though the interval between two immediately successive loops may not be the same. The intervals are not required to be equal, but regular, and this is satisfied by any repeated group or pattern. It further appears as an essential that the looping of the face yarn shall occur “on alternate stitches of adjacent rows of the back yarn.” This does not mean “every other stitch of successive rows,” as the defendants contend ; but, looking to the process of knitting and the’ course of the *851back thread which necessarily controls, the face thread is to be looped or tied to stitches which alternate with each other, as the successive rows of the back yarn come around after a complete revolution. This fabric is machine-made, as is specified, and the possible employment'of a four-feeder is recognized,—a circumstance which has got to be taken into consideration in construing the directions given. Where, therefore, a single thread machine is not used, the words “adjacent rows” mean what they would mean to the knitter,—the next row formed by the back thread after it has gone once around. Without departing in any respect from the terms used, this part of the claim may be rearranged and paraphrased thus: “The face yarn being looped at regular intervals in one row of the back yarn and to an alternate stitch of said back yarn in the adjacent row.” Here row is compared with row, as is manifestly intended, the back thread in one revolution forming one row and in the next another, and each being the adjacent row to the other for that particular thread. The word “adjacent,” even in its strictest sense, means no more than “lying near, close, or contiguous, but not actually touching.” Webst. Diet. There are degrees of nearness, and when you want to express the idea that a thing is immediately adjacent you have to say so, and that is not what is said here. If this, then, is the construction to be. given to the second claim of the patent, the defendants’ fabrics clearly infringe upon it in the main. It would be interesting to go over each of them, and point out wherein that is the case, but I do not think it would serve any useful purpose to do so. The matter will have to be taken up in detail by the master, and to a certain extent it is better to leave it open to him. It may be that some of the fabrics on further examination- can be distinguished and saved, but, the charge of infringement having been made out with regard to any of them, the extent to which it goes in others is not at present material.
Let a decree be drawn upholding the validity of the patent, and referring the case to a master to take an account.
Specially assigned.
1. See Courts, vol. 13, Cent. Dig. § 316.