This cause is returned to this court for hearing on a supplemental record in pursuance of a mandate_ which directed the taking of proof as to two aspects of the case, to wit:
“(1) To take proof as to whether the Holmes garments like, or substantially like, Defendant’s Exhibit G for identification, offered as a specimen of these garments, had been in use or on sale prior to the patent in suit; and
“(2) To take such testimony, if any, as it may deem advisable, to explain the construction and mode of operation of Exhibit G or similar garments, and thereupon to return the proofs to this court, with the opinion of the District Court as to (1) and as to (2), if there be any conflict in the testimony as to the manner in which the Exhibit G or similar garments were actually *523constructed, or as to the inode’ of operation of such Exhibit G or similar garments.”
The opinion of the court on the appeal on the original record is reported in 234 Fed. 886, 148 C. C. A. 484.1 We there stated:
‘‘With garments made under the Muller British patent, 8,766 of 1896, the Cook United States patent, 615,632 ot 1898, the Xtocheite British provisional specification, 1,343 of 1864, and the Tiehy Austria patent, 27,283 of 1907, laid on a table before the man skilled in the art, we think there was so little opportunity for inventive ability that patentability on the record as it now stands is doubtful. We do not confine our references to those just recited. They are merely examples of the prior art, all of which wo have regarded as pertinent in arriving at our conclusion. s * This brings us to a consideration of Exhibit G for identification, which was excluded by the trial court. Tills was offered by defendant as a specimen of the-so-called Holmes garments. We regard this as a closed crotch garment. To call it otherwise, in considering the prior art, is, we think, again drawing too fine a distinction. With this garment properly in evidence in connection with the other prior art, we think that no invention was needed to devise the particular construction shown in the patent, and hence we are satisfied that claim 1 would be void for lack of patentable novelty. Whether, therefore, this exhibit and the testimony relating to it were correctly excluded, becomes an Important question in the case.”
It will lie noted that we referred to Exhibit G for identification as a specimen garment, it having appeared from the original record that Exhibit G for identification was offered as a sample of a type and not as an original garment. Testimony has now been adduced as to prior art garments substantially like Exhibit G for identification, and the supplemental record also contains testimony inter alia in respect of garments known as Exhibits Y, Z, and ZZ.
Claim 1 contains two elements in combination: (1) A permanently closed crotch; and (2) a. posterior opening extending near the waist line to a point below the crotch in one leg only. Plaintiffs now endeavor to give a special meaning to the word “only”; but, clearly, the word “only” merely means (in connection with the context) that tlie extension shall be to a point below the crotch in one leg, and not in two.
Tlie first question is whether the prior art disclosed (a) garments with closed crotches; and (b) «garments with the extension to a point in one leg only. The latter, we held in our former opinion, was demonstrated by the Austrian patent to Tichy, dated January 25, 1907, and that conclusion we repeat. We also stated that we regarded Exhibit G for identification as a closed crotch garment, and, after consideration of the present contentions of plaintiffs, we remain off the same opinion.
We think that the testimony of Eacher, O’Brien, and Mrs. Cook fully establishes the existence of garments like Exhibit G for identification in the prior art. We are also satisfied that Exhibits Y and 'Z of the present record were in the prior art and are closed crotch garments. None of the foregoing had the extension to the leg. There is also testimony as to the manufacture of the garment Exhibit ZZ from *524a date prior to the patent date, Marsh, the manager of the Yale Knitting Company, successor of the Holmes Knitting Company, testifying:
“What we call the old Holmes type (Exhibit Z) we manufactured I should say two or three years from 1904; then we manufactured the white Exhibit (for identification ZZ)'. There was no gap between the manufacture of these garments, there was no gap in the manufacture from 1904 to 1914.”
This is clearly a closed crotch garment, but we are not satisfied that the opening extended partly down the leg.
The final question then is whether, in view of the existence of garments respectively exemplifying each element of the claim, it was invention to combine them in one garment. We have not overlooked the argument that the prior art followed what counsel now call the crotch tab principle, but we think that this is an artificial distinction. Obviously, in an art of this kind the effort is toward gradual improvement suggested by experience. As the trade gradually learns the requirements of customers or the defects in practice of garments of this character, the manufacturer endeavors to meet the situation, so as to improve the comfort and practical usefulness of such garments.
To give to this claim the broad construction contended for, it is necessary to hold that the combination is for a permanently closed crotch with a posterior opening extending from some point near the waist line to some point below the crotch in one leg. We think such a construction cannot be accorded to claim 1, and, thus construed, that the prior art negatives invention. The most liberal view which could be taken is that the claim goes no further than is illustrated by the patent drawings, and, from that aspect, the defendant’s device does not infringe.
But we do not place our conclusion on such narrow ground. We are of opinion that, in view of the prior art, the patent in suit (which seems to us not as efficient as plaintiffs’ commercial garment) accomplished no more than was to be expected of the man skilled in the art and did not rise to the dignity of invention.
Decrée reversed, and bill dismissed, with costs.
. In the reported case, 1900 (as the date of the patent) is a typographical error, and should be 1910.