PER CURIAM.

The district judge held that the defendants did not infringe claim one of Patent No. 1,964,911, because their compressed cotton pad was not "convoluted" as the patent defined that word (page two, lines 86-88): "folded in curves or tortuous, sinuous windings, that is, bending in and out in a serpentine, wavy or accordion folded form." When the application was filed on May 22, 1933, this passage was not in it; but another passage was (page one, lines 90-97) which for convenience we shall call "the second passage" and which read as follows:

"The next step in the manufacture of the device is to fold the pad 11 into convolutions such as shown in Fig. 2 to reduce the length thereof. This folded convoluted pad is then placed into a die and pressed under high pressure from both the sides and the end to form a hard cylindrical core 13 with the withdrawing member 12 extending therefrom."

Figure Two showed regular zigzag folds all the way through the strip. The examiner rejected the claim because he did not like the use of the word, "convoluted", which to him meant folded like a leaf, "vernated". He preferred the word, "accordion", and said that "accordion folds are the natural result of compressing a long strip * * * into a tight core of considerably lesser length." To meet this objection Haas kept the word, "convoluted", in the claim, but defined it as we have said. If that was all, it might be difficult to say that the introduction of the definition had circumscribed the claim to the kind of "convolutions" described in "the second passage"; it really did no more than incorporate "the second passage" and that was already implicit in the word, "convoluted" as Haas had used it. No new term was introduced to secure allowance. However, the claim did not have its origin in the patent in suit at all; it went back to Haas's earlier patent, No. 1,926,900, the application for which was filed on November 19, 1931. That application did not contain either "the second passage" or Figure Two; but it did specify that the pad should be compressed under high pressure, though it did not disclose how this was to be done. During the course of this application through the Office Haas proposed a claim — No. 17 — which was substantially the same as claim one in suit except that it did not contain the concluding phrase: "said core being considerably shorter than said strip." The examiner rejected this claim and claim 18, because "convolutions" were not described and did not "necessarily result" from compressing "a long strip * * * into a tight core." After unsuccessfully trying to secure the two claims by adding the concluding phrase just quoted, Haas cancelled them on May 22, 1933, reserving, however, his right to claim them in another application, which he filed at the same time and which became the patent in suit. Into this application he introduced Figure Two and "the second passage."

If Haas had secured the claim by the introduction of Figure Two and "the second passage" into the first application, we cannot doubt that there would have been a "file-wrapper estoppel." I.T.S. Rubber Co. v. Essex Rubber Co., 272 U.S. 429, 444, 47 S.Ct. 136, 71 L.Ed. 335; Schriber-Schroth Co. v. Cleveland Trust Co., 311 U.S. 211, 221, 61 S.Ct. 235, 85 L.Ed. 132. He would have failed to get a claim based upon such "convolutions" as were inevitable by forcing a long strip into a short core; and in its place he would have secured a claim into which he introduced the element that the "convolutions" must be regular zigzags. Obviously it made no difference that this limitation was incorporated into the claim only by reference; and it also made no difference that the claim was itself transferred from the earlier application to another, co-pending with it. The doctrine of "file-wrapper estoppel" depends upon the fact that, when an applicant has accepted the rejection of a broad claim he may not later assert that another claim, deliberately restricted to secure its allowance, is its equivalent; and that argument is as cogent when the claim is transplanted into a new application as when it is amended in the original. If Haas wished to insist upon all kinds of "convolutions" which might result from high pressure, he was not free to abandon haphazard "convolutions" and then to assert that the claim covered them. Indeed, he apparently supposed that zigzags had some useful purpose; at least he said so in the specifications (page two, lines 46-49; line 78). The defendants do not make up their pads as described in "the second passage;" the cotton is tied in the centre by a thread and then doubled upon itself, after which the pad so made is crushed into a die with such resulting folds as chance may make. It results that when one draws upon the thread, the strip does not "straighten," and is not "withdrawn in the elongated condition of Fig. 1" (page two, lines 48, 49).

Little need be said about the second patent — No. 2,024,218. As to it also we agree with Judge Campbell. Any advance which Haas made beyond Scrutton's British patent ((1924) No. 218,725), Ellis (No. 1,131,349), Roberson (No. 1,191,736) and Haas (No. 1,642,950)) was trifling, and quite insufficient to support a patent.

Judgment affirmed.