back plate, the second part of which was bent out into a lateral plane so as to go slightly past the center of the back plate and then back again to the edge of it, forming a hinge leaf, and the remaining part of which was then developed into a half ring in a plane approximately at right angles to the planes both of the first and second parts. When two of these wire sections were opposed and in suitable engagement provided at the point where they met, they constituted a toggle-joint operating like Pitt. This construction is shown in the drawing herewith reproduced and is recited in claim 1 shown in the margin.1

It is not claimed that Schade is anticipated in the complete sense by the prior art, but the defense is that, in view of this art, the change made by him did not involve invention. Rightly to apprehend this art, further structures must be described. By an English patent, Lindner had shown a continuous wire half ring, leaf off-set and edge pivot of a form very close to Schade's corresponding parts; but the two leaf off-sets did not reach each other, and were held in one position by a sliding clamp instead of being held in alternative positions by a spring pressure. Blackmer and Robson also showed two wires, each developed into a half ring, and each containing an off-set portion like Schade. In this, as in Lindner, the two off-set portions did not touch each other, but, unlike Lindner and like Schade, the off-set portion served as a lever which, under spring pressure, held the ring open or closed. An encasing spring member formed an abutment against which this lever rested. Normally, it held the ring closed, and when the ring was forced open, the lever passed the point where the spring would throw it back and the spring held it open. McMillan, instead of using Pitt's flat plates, carried his ring sections themselves to the point of meeting where they were united with a hinge or pivot rod, and he so mounted them in the edges of the base plate that they constituted the spring pressed toggle-joint. The following selected figures from Lindner, Blackmer and Robson, and McMillan illustrate the description that has been given.

11. "In a temporary or loose-leaf binder, the combination with a baseplate made of spring sheet metal, of separable eyes mounted pivotally directly on the edges of said base-plate and having bent wire arms, the arms of the two opposite separable eyes being engaged to form toggle-levers between the sides of the spring base-plate substantially as set forth."

[1] The patent of McMillan is No. 794,536, and was issued July 11, 1905, on application filed October 13, 1904. The Blackmer and Robson patent was applied for April 28, 1904, and issued August 29, 1905,

being No. 798,157. It is urged that these are not a part of the art prior to Schade whose application was filed January 17, 1905, after the filing of Blackmer and Robson and of McMillan, but before the issue of either. There is no evidence in the record as to the respective dates of invention, save that afforded by the usual printed copies of the three patents admitted by stipulation, and by the certified copy of the McMillan file wrapper and contents. While, in case of objection and insistence upon strict proof, it might be necessary to produce a copy of the Patent Office record in order to show the filing date, yet the practice of relying upon the date shown on the officially printed copy has now become so general that, in this circuit, the printed date is accepted as proof whenever objection is not made. Drewson v. Hartje Co., 131 Fed. 734, 738, 65 C. C. A. 548. We, therefore, assume that the filing dates of these patents are as above stated.

[2] This court held, in Drewson v. Hartje Co., supra, 131 Fed. at page 739, 65 C. C. A. 548, that a patent, the filing date of which antedated the filing date of the patent in suit, was, prima facie, anticipatory; and we have repeatedly accepted and applied that rule (e. g., Electric Co. v. Westinghouse Co., 171 Fed. 83, 87, 96 C. C. A. 187; Jackson Co. v. Adler, May 18, 1917, 243 Fed. 386, 156 C. C. A. 166; Twentieth Century Co. v. Loew Co., 243 Fed. 373, 156 C. C. A. 153, May 8, 1917). To rely upon these cases would be sufficient; but the contrary proposition is urged upon us so often and is observed to such an extent in current decisions that it seems to deserve re-examination. The cases, found cited to support the claim that such a patent is not part of the prior art, are collected in the margin. We are satisfied that the confusion arises out of a question of pleading, and has no foundation in matter of substance.

The patent statute (R. S. § 4920 [Comp. St. 1916, § 9466]) catalogues five supposedly independent matters of defense, and requires the pleading of those which are to be relied upon. The separation is not analytically perfect; the last clause [b] of "third" seems to belong to "fifth," and since the patentee could not possibly be the first inventor of something that, before his invention, had been patented to or described by another, the entire first clause [a] of "third" seems quite unnecessary. However, we have to deal with the statute as it is. The two classes now material are :

"Third [a] that it [the invention] had been patented or described in some printed publication prior to his supposed invention or discovery thereof, or [b] more than two years prior to his application for a patent therefor.

2 Bates v. Coe, 98 U. S. 31, 25 L. Ed. 68; Diamond Co. v. Kelly (C. C.) 120 Fed. 282, 287; Anderson v. Collins (C. C. A. 8) 122 Fed. 451, 458, 58 C. C. A. 669; Thompson-Houston Co. v. Ohio Co. (C. C.) 130 Fed. 542, 546; Eck v. Kutz (C. C.) 132 Fed. 758, 764; Ajax Co. v. Brady Co. (C. C.) 155 Fed. 409, 415; Union Co. v. Smith (C. C.) 173 Fed. 288, 291; Gray Co. v. Baird Co. (C. C. A. 7) 174 Fed. 417, 421, 98 C. C. A. 353; Gen. Elec. Co. v. Allis Co. (C. C.) 190 Fed. 165, 170; Johns Pratt Co. v. Freeman (C. C. A. 3) 204 Fed. 288, 122 C. C. A. 512 (by affirmance of [D. C.] 201 Fed. 356, 360); Horton Co. v. White Lily Co. (C. C. A. 7) 213 Fed. 471, 476, 130 C. C. A. 117; Alvord v. Smith (D. C.) 216 Fed. 150, 154.

"Fourth. That he was not the original and first inventor or discoverer of any material and substantial part 3 of the thing patented."

Since an invention is not "patented" and the patent is not "published" until issued, it is apparent that when a patent is pleaded under this third defense and the fourth defense is not involved, its application date is of no importance whatever. It is equally sure that where the tendered issue is whether the patentee was the first inventor (fourth defense) and it appears that another had made the invention, the date of that other's patent, or whether he ever had a patent, or whether the prior description was by printing or without, is immaterial; the date of his invention is the vital thing.

4

Under the famliar Patent Office rule, the filing of an allowable application is a constructive reduction to practice, and so there must be a presumption that the patentee had made his invention at the date of his filing. The cases where a substantial change in the invention is made pending the application are exceptional, and, of course, he who alleges an exception must prove it." The existence of the presumption is worked out and declared, with perhaps unnecessary care, by the first Circuit Court of Appeals in Automatic Co. v. Pneumatic Co., 166 Fed. 288, 293, 92 C. C. A. 206. Hence, it assuredly follows that if a patent in suit was applied for January 15th, and there is nothing to carry the patentee's invention back of that date, and if a patent disclosing the same invention was issued to another in July upon an application filed January 1st, this tends to show that the patentee of the patent in suit was not the first inventor. This has been distinctly held not only in Drewson v. Hartje, supra, but by the Seventh Circuit Court of Appeals (Barnes Co. v. Walworth Co., 60 Fed. 605, 606, 9 C. C. A. 154); by Acheson, C. J. (Westinghouse v. Chartiers Co. (C. C.) 43 Fed. 582, and cases cited on page 588); and by the Second Circuit Court of Appeals (Sundh Co. v. Interborough Co., 198 Fed. 94, 97, 117 C. C. A. 280); and it has been expressly recognized and applied by the Supreme Court (Pope Co. v. Gormully Co., 144 U. S. 238, 244, 12 Sup. Ct. 637, 36 L. Ed. 420); by Mr. Justice Bradley (Kearney v. Lehigh Co. [D. C.] 32 Fed. 320, 322, 323); and by the Second Circuit Court of Appeals (Hillard v. Fisher Co., 159 Fed. 439, 441, 86 C. C. A. 469; Otis Co. v. Interborough Co., 222 Fed. 501, 502, 138 C. C. A. 97).

Turning to the cases cited contra: Bates v. Coe, in using the language so often quoted (98 U. S. on pages 33, 34, 25 L. Ed. 68) is speaking solely of "evidence to sustain the second [third] defense," and this restriction has not been observed when it has been quoted as if applicable to all defenses. In Horton Co. v. White Lily Co. (C. C. A. 7), the court

3 Since the thing patented is a unit, to which "any material and substantial part of" is essential, and since no patent is found to be either anticipated or infringed except by comparison with devices containing every substantial and material part, the special force of the quoted words is not evident.

4 Duryea v. Rice, 28 App. D. C. 423; Computing Co. v. Standard Co. (C. C. A. 6) 195 Fed. 508, 511, 115 C. C. A. 418; Sundh Co. v. Interborough Co. (C. C. A. 2) 198 Fed. 94, 97, 117 C. C. A 280.

Webster Co. v. Higgins, 105 U. S. 580, 594, 26 L. Ed. 1177.

was considering two patents to the same inventor, a very different situation. In some, as in Anderson v. Collins (C. C. A. 8), it is carefully recognized that the ruling involves only the third defense. In others, the state of the pleadings does not appear. Some seem to depend upon a distinction between "anticipation" and "priority of invention," applying the former name to the third defense and the latter to the fourth, though we do not see why "anticipation" is not properly applicable to both. So far as any of the other cases may be distinctly inconsistent with the proposition of the preceding paragraph, we must think them er

roneous.

It was the common, if not universal, practice under the old equity rules for an answer to recite under one paragraph all the older patents relied upon, and then, under another, to repeat all the names of the patentees with their places of residence. In deference to the simplifying spirit of the new rules, the defendant here pleaded that the patentee was not the first inventor, "but that said invention was first made by the applicants of the patents recited in paragraph 9." In such cases, there is no question of pleading, but the pertinence of the earlier filing date is clear. We should hesitate to reach the other result, even when the fourth defense was not pleaded in terms. Since patenting to another before the invention of the patent in suit necessarily implies denial that the patent in suit was issued to the first inventor, to plead the third defense is usually itself to plead the fourth, and it would be unfortunate that an invalid patent should be held valid only because a good defense had been pleaded by a wrong name. Any amendments that might seem necessary would doubtless be allowed on request, with due regard to the prejudice, if any, thereby caused.

[3] When we consider this prior art, including McMillan and Blackmer and Robson, we are forced to conclude that Schade took no inventive step. His continuous wire, developed into three parts, each part having a function, can be found merged in Pitt's hooks and flat plates, yet there would be ingenious, and, we assume patentable, novelty in observing this continuous wire and developing it out of its merged condition, if it had been a new thing after it was developed; and so Pitt alone will not serve to invalidate. McMillan makes his hooks or rings out of wire, and extends the hooks themselves to a meeting point, making a toggle lever hinge, and pivotally mounts the rings on the edges of the back plate. The language of Schade's claim 1, if given the broad meaning which could naturally be attributed to the selected words, plainly reads upon McMillan; and yet this is not conclusive of the question of invention, since the language of the claim, by reference to the specification, might fairly imply that the "bent wire arms" of the claim referred to arms bent up laterally and to hinge leaves of the general type shown rather than those which were mere continuations of the hook in the same plane, like McMillan. Even Lindner, considered by itself or in association with the others, would not necessarily negative invention, since the wire off-set of Lindner, shaped like Schade's wire off-set, had no function as a spring pressed lever, and to combine the thought of such a shaped and formed continuous wire with Pitt's thought of spring actuated toggle levers might well be

156 C.C.A.-12