Suit by United Shoe Machinery Corporation against the Day Wood Heel Company for infringement of patent to Russ, No. 1,-536,691, dated May 5, 1925, for a “Heel Blank Breast-Shaping Machine” and of patent to Russ, No. 1,528,345, dated March 3, 1925, for “Heel and Method of Making tho Same.” The defenses were noninvention, anticipation, and prior use and knowledge. Claims 1, 3, 5, and 6 of patent No. 1,536,691, herein called the machine patent, and all claims of patent No. 1,528,345, herein called the product and method patent, wore in suit. The court sustained the machine patent as valid and infringed, and defendant appealed. The court held the product and method patent invalid because of both prior use and non-invention, and plaintiff appealed.

Product and Method Patent No. 1,528,-345: Claim 1 for the product is typical. 1 Claim 3 for the method differs in no substantial respect from claim 1. Both claims are for the same invention, if there is invention, whether for the product or process, and need not be separately considered.

Dong before 1910, the wood shoe heel waá in commercial use. During the period from 1910 to 1912, there was a demand for a type of wood heel for women’s shoes called the “Cuban.” In the terminology of the manufacturers, that part of the heel secured to the shoe is the “seat,” and that part contacting with the ground is the “top.” As the heel is attached to the shoe, the front portion is called the “breast” as distinguished from the rounded side and back portion called the “contour.” The earlier Cuban heels were made with a straight breast, but the fashion demanded a change to the transversely concave breast. This was easily provided by forcing the heel back straight across the cutting edge of a barrel saw, but in this operation the intersection of the concave breast with the surface of the rounded contour presented slightly curved edges when viewed from the side. The trade asked that these edges be made straight, and Ross made them straight upon his machine which cut the transverse concave breast and by tho same operation made the breast convex longitudinally. His conception was that a breast so designed and properly proportioned would terminate at its intersection with the contour in straight edges when viewed from either side. This signified more than mere mechanical skill. It was not an obvious thing. Its conception was difficult, and the actual achievement is not now easy of explanation, but the heels thus made were a commercial success, and we think they amounted to patentable invention both in process and in product.

We do not think defendant has shown prior use or knowledge of such heels or the method of producing them beyond a reasonable doubt. Witham, whose testimony is uncorroborated, made wood heels with a concave *898 breast and straight breast edges in 1916 or 1917. He claims to have straightened the breast edges by applying a buffing roll, but it is quite certain that he knew nothing of heels having straight breast edges produced by the Russ machine or method. Tisdale, likewise uncorroborated, testified that he made wood heels with a curved breast and straight breast edges about 1901 or 1902; that he made the straight edge by applying a buffing roll, but he frankly admits that the concave breasts did not have longitudinal convexity.

Sanders, a manufacturer, having the common interest of the industry in the re-suit of the suit, testified that on November 7, 1927, whieh was while the suit was in progress, he made two heels which he exhibited. These transversely concave heels are also slightly convex longitudinally and have straight breast edges. Although they were made by Sanders to be exhibited with his testimony, no one connected with the plaintiff was present at the experiment. ' No one was present at all except Dickinson, whose testimony was not taken. Sanders testified that he made these heels by the use of a bevel guide attached to a barrel saw machine, whieh caused the heel block to pass the saw in the are of a circle; that by the use of this same attachment he somewhere between 1909-1913 made similar heels for a customer whose name he does not remember. He does not claim to have intentionally produced the straight breast edge by means of the neutralizing effeet of a heel breast transversely concave- and longitudinally convex. His straight breast edge was a pure accident, and novelty is* not negatived thereby. Tilghman v. Proctor, 102 U. S. 711, 26 L. Ed. 279; Pittsburgh Reduction Co. v. Cowles Elec. S. & A. Co. (C. C.) 55 F. 301, 307; see also Diamond Patent Co. v. S. E. Carr Co., 217 F. 400, 402 (C. C. A. 9).

Day, Treasurer' of the defendant company, testified that for about three years beginning in 1910, and while doing business at Haverhill, Mass., as the Fred A. Day Corporation, he made several thousand dozen Cuban heels similar to heels covered by plaintiff’s patent and by processes similar to the method of plaintiff’s patent by the use of a tapering metal guide attached to the saw table of a barrel saw and whieh forced the heel block by the saw in a circular path. Neither this guide nor' any of these heels are in evidence. According to Day, he made these heels without any particular urge or reason. It is fair to say that Day is corroborated by the testimony of Ms then employees, Felch, Post, and Owen, to the extent that some such heels were made about 1910 by the Day Corporation, but he is uncorroborated- as to the quantity of the output. Felch thought that the quantity was possibly more than a hundred dozen, and'Owen testified that they were . “not much of a go.” Day and Felch left the Day Corporation in 1912, and were employed for about three years by the Maple Wood Heel Company at Newburyport, Mass. Felch testifies that Day also made the anticipating heels at the Maple Company, but Day has no recollection of it. Day testified that- the principal customer of the Day Corporation was the Fred W. Mears Company; that “he must have sold them these straight breast line Cubans in very large quantities, thousands of dozens, perhaps”; but Wentworth, who was foreman of the Mears Company in 1911 and 1912, testified that during that period he never saw nor heard of such Cuban heels; that he was employed by the Maple Heel Company from September, 1913, until February, 1914, and that during that time none of the anticipating heels were made, and that he wquld have known it if they had been made. Lovejoy also testified to the same effeet.

Anticipation cannot be proven alone by the number of witnesses introduced to support it. Barbed Wire Case, 143 U. S. 275, 12 S. Ct. 443, 36 L. Ed. 154; American Bell Tel. Co. v. People’s Tel. Co. (C. C.) 22 F. 309. It must be determined by a consideration of the evidence in the aggregate, and, so treating it, the ease is not free from reasonable doubt. Giving due weight to the testimony of these witnesses taken 12 to 15 years after their purported knowledge of the anticipating heels, it is entirely too uncertain whether the heels and method for making them were sufficiently similar to plaintiff’s patent to constitute anticipation. Deering v. Winona Harvester Works, 155 U. S. 286, 301, 15 S. Ct. 118, 39 L. Ed. 153.

Day left the Maple Heel Co. about 1915 and established the defendant, the Day Wood Heel Company, at Cincinnati, and later established plants at Columbus and St. Louis, at wMch points, the company has-since operated. It has never since used the tapering bevel metal guides. That it did not use them from 1915 to 1924 is not remarkable, because during that period there was but slight, if any, demand for heels having curved breasts and straight breast comers, but it is significant that in 1924 the Day Company with the return' of the fashion, straightened the breast edges by means of the old buffing roll *899 as used by Tisdale and Witham rather than by the tapering bevel metal guide. Plaintiff’s application for bis process and product patent was originally filed May 15, 1924. The patent was granted March 3, 1925. On June 30, 1926, defendant inserted in “American Shoemaking” the following advertisement: “Improved Wood Heels Made by New Machine — Straight Edge Lino Breast — With the aid of a new machine wo have recently perfected, we now offer neat-appearing Cuban Heels. Our latest models have added snap and style, due to the now method of curving. Let us show you samples today.” We think it quite improbable that defendant would have published this announcement touching “improved wood heels” made by a new maeliine “straight edge line breast. * ’ * Our latest models have added snap and style due to new method of curving,” if it had really known of such heels in 1910 and of the method for making them.

Machine Patent No. 1,536,691: We think it likewise disclosed invention. Claim 6 is printed. 2 The elements are old, hut in combination a machine was evolved which mechanically, quickly, economically, and by one operation produced the straight edge curved breast Guban-heels designed by Russ, and for which there was a trade demand. It conception and construction involved something more than the obvious. It largely eliminated the human element present in all prior machines used for any similar purpose.

We do not think this machine patent was anticipated by patent to Conner, No. 1,415,-921 (1922). Conner’s idea was to cut the breast edges of the heel with a flexible or flexing knife. His device could not be successfully operated upon a wood heel, and does not in any way disclose Russ’s conception.

Nor was it anticipated by patent to Marshall, No. 1,510,009 (1923). The Marshall patent was designed to cut a groove or recess in the heel at the lower extremity of the heel breast. Modification was necessary before it would perform the function of the patent in suit. Topliff v. Topliff, 145 U. S. 156,161,12 S. Ct. 825, 36 L. Ed. 658; Babcock & Wilcox Co. v. Springfield Boiler Co., 16 F.(2d) 964, 969 (C. C. A. 2). The same observation applies to Sanders, patent No. 1,213,271, issued in 1917. The undisputed evidence is that it would bo impossible for the Sanders device to produce a heel breast concave transversely and convex longitudinally. The remaining patents relied upon as anticipations, to wit, Stevens, No. 190,542 (1877)., Osten, No. 128,243 (1872), and Luttrell, No. 951,-784 (1910), are for stave sawing machines, an entirely different art, and have little or no relevancy. Nor was there anticipation by the machine made by Draper in 1920 or 1921. Admittedly the heel breast made upon the Draper machine was concave longitudinally, not transversely, and its edges were curved.

Infringement: Proof of infringement is clear. In patent No. 1,528,345, the claims read upon defendant’s product and method, and in patent No. 1,536,691, the claims in issue read upon defendant’s machino.

The decree of the District Court is reversed as to patent No. 1,528,345 and affirmed as to patent No. 1,536,691.