Opinion
No. 116.
December 23, 1941. Writ of Certiorari Denied March 16, 1942. See 62 S.Ct. 908, 86 L.Ed. ___.
Appeal from the United States District Court for the Southern District of New York.
Action by the Arcadia Knitting Mills, Inc., against the Princeton Knitting Mills, Inc., for a judgment declaring invalid all four claims of Patent No. 2,123,847, issued on July 12, 1938, to Harry Fleisher, wherein defendant filed a counterclaim for infringement. From a judgment of the District Court for the Southern District of New York, 39 F. Supp. 153, declaring the patent invalid and dismissing the counterclaim, defendant appeals.
Affirmed.
W. Hastings Swenartor., of New York City (Benjamin Jaffe, of New York City, of counsel), for appellant.
Herman Seid, of Syracuse, N.Y., for appellee.
Before L. HAND, SWAN, and CHASE, Circuit Judges.
The defendant appeals from a judgment declaring invalid all four claims of Patent No. 2,123,847, issued on July 12, 1938 to Harry Fleisher, and dismissing its counterclaim for an injunction forbidding the plaintiff to infringe them. Judge Goddard's opinion in the District Court may be found in 39 F. Supp. 153; to it we refer for a statement of the general nature of the controversy; our discussion assumes an acquaintance with it.
If Claims 2 and 4 were anticipated, Claims 1 and 3 are also invalid, because whatever useful novelty there was in the invention lay in what is known as the "one and one arrangement"; that is, that in each course the added "inlaid" yarn should "float" across only one "wale" and should be "inlaid" around the top of the next "wale." It is this which is supposed at once to give the protection against "runs" and the opportunity for "napping," in itself a well known practice before the date of the invention. Claim 1 prescribes that in every "repeat" of six there shall be three successive yarns in which the sequence of "float" and "inlay" is the same so that the succession for the whole sextette will be: — a, b; a, b; a, b; b, a; b, a; b, a. ("a" indicates a "float"; "b" indicates an "inlay.") Claim 3 includes all that Claim 1 covers and more; in it there must be in each "repeat" successions which are the same; thus it adds the following possible variants: — (1) a, b; a, b; b, a; b, a; b, a; b, a; (2) a, b; a, b; a, b; a, b; b, a; b, a; (3) a, b; a, b; b, a; b, a; a, b; a, b. Claims 2 and 4 include all the forms of Claims 1 and 3 and one more; i.e. that in which there are no successions having the same sequence of "float" and "inlay"; thus: — a, b; b, a; a, b; b, a; a, b; b, a. There is no reason to suppose that any popularity which the fabric enjoyed depended upon the successions of Claims 1 and 3; that is, that they made a more acceptable fabric than that which Claim 2 added. The specifications do not suggest any superiority of one over the other; and it is plain that all are open to any one's choice, provided that the "one and one arrangement" itself is open. Certainly after Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 62 S.Ct. 37, 86 L.Ed. ___, we should not be warranted in hanging a patent upon so slender a thread. Therefore, as we have said, the validity of all the claims depends upon that of Claim 2.
Judge Goddard found (Finding No. 3) that Claims 2 and 4 were anticipated by three fabrics "Arcadia 1838," "Arcadia 2294a" and "Arcadia 2362"; unless that finding is "clearly erroneous," the case is at an end. We may confine ourselves to "Arcadia 1838" which was sold in very substantial quantities before the application date of the patent in suit, February 23, 1938. The plaintiff put in evidence a "folder," bearing date, November 16, 1936, which contained a fabric anticipating Claim 2; if it succeeded in proving sales of this, it was enough. The outside of this folder bore the number "1838"; inside was the date, and under the heading, "Quality," were the words and figures "1838 new setup." After November 16, 1936 and before the date of application of the patent in suit the plaintiff sold very substantial quantities of a fabric billed as "1838"; it also sold five pieces before November 16, 1936, of an aggregate value of less than $20; these earlier sales were on September 9, October 15 and October 26. It claims to have proved that the fabric so sold was the same as that now attached to the folder; and there would seem to be no reason to doubt this except for considerations we defer for a moment. Its practice was to make up folders for all new fabrics, to put a date upon them, and to give them a number when they went into production. The folder here in question describes the weave as "1 long, 1 short" which is what it should be, and, as a document, it bears every mark of authenticity. The defendant was allowed access to all the plaintiff's files, and did not produce any other folder for "1838." Its attack upon the conclusion to be drawn from it rests upon the fact that the five pieces we have mentioned were sold before November 16, 1936, and that the folder bears the words "1838 new setup" on the inside. From these it argues that the fabric sold as "1838" could not have been that now found in the folder. The testimony of the witnesses was somewhat uncertain as to when the date was customarily put upon folders, though plainly the company may have wished to try out small swatches before putting a new line on the market; but it would not affect the result even if we should feel doubt whether there was not a change in the weave between September 9 and November 16. The only relevant matter is whether the fabric sold as "1838" after November 16 was what is now in the folder. It is to the last degree improbable that if the weave was changed, the weave sold as "1838" after November 16 was the earlier weave of which the company did not have the usual folder, and that it should have kept a record of the changed weave which it never sold in quantity though it gave it a number. As to the words "new setup," if there was a change in the weave between September 9 and November 16, those words may conceivably refer to it, but it would make no difference if they did, rather it would strengthen the proof. Certainly some fabric must have been in the folder on November 16, 1936, and that fabric must have been what was sold as "1838" at that time and later, not the suppositions earlier fabric for which it was substituted and of which we have no trace. Of course it is still possible that this fabric of November 16, 1936, sold as "1838," was not what we now have; there may have been a second change. But there is not a tittle of support for such a doubt, and besides, unless the folder is a forgery, whatever fabric was in the folder on November 16, 1936, was a "one and one arrangement." The proof of a prior use must indeed be convincing beyond reasonable doubt, but we are not justified without any evidence in imputing to a party the perpetration of a crime.
Judgment affirmed.