Appellant filed its bill in the District Court to restrain infringement of claims 3, 4, S, 7, and 9 of patent No. 799,012 issued to appellant on September 5, 1905, as assignee of Alexus C. Lindgren, for an improvement in disk harrows. On final hearing the District Court found against appellant on the question of infringement and dismissed the bill for want of equity, which finding and decree is assigned for error.

The claims sued on read as follows:

“3. In a disk harrow, the combination with the frame, of a disk gang adjustable with relation to the frame transversely of the line of draft, and movable around a vertical axis, á hand-lever .mounted on the frame in ad-’ vanee of the disk gang, and rearwardly-extending -connections between the hand-lever and disk gang for turning the latter on its vertical axis, said, connections being adjustable at the rear and relatively to the disk gang in the direction of the transverse adjustment of the gang.

“4. In a disk harrow the combination with the frame, of a disk gang adjustable therein transversely of the line of travel, and movable around a vertical axis, a hand-lever movable longitudinally, a connecting-rod extending longitudinally rearwardly therefrom; and adjustable connections between the rod and disk gang.

“5. In a disk harrow the combination with a frame, of a disk gang adjustable therein transversely of the line of travel and mounted to turn on a vertical axis, a hand-lever for controlling said axial movement, a rod extending rearwardly from the hand-lever, and a link having its inner end jointed to said rod and its outer end connected adjustably with the disk gang.

“7. In a disk harrow the combination with the frame member having front and rear frame-bars, of a saddle-plate extending, longitudinally of the machine between said bars and adjustable vertically around a fore-and-aft axis extending longitudinally of said plate, and provided with a vertical bearing-socket,a disk-yoke formed with a vertical trunnion seated in the socket, means for connecting said parts together, and disks carried by the yoke.

“9. In a disk harrow the combination with the front and rear frame-bars, of a horizontally-arranged saddle-plate sustained by said bars and rotatably adjustable at its outer end around a fore-and-aft axis extending longitudinally of the machine, a clamping device at the opposite end of the plate for holding it in its different adjusted positions, and a disk gang carried by the saddle-plate.”

The .device of the patent relates to a reversible disk harrow, largely used in cultivating the soil in orchards and other lands where the space between the rows is narrow and where different adjustments are required. It consists of a rectangular or transversely-elongated frame, to the widely spaced front and rear bars of which a tongue is rigidly attached — saddle-plates on either side of the tongue, which are respectively mounted pivotally, and consequently tiltingly, on and within the front and rear bars, and which carry means for horizontally swinging the disk carriers, levers, and other devices for positioning the disk carriers and disks, and an operator's .seat. It is appellant’s contention that claims 7 and 9 cover the frame, while claims 3, 4, and 5 have ref*729•erence to the other elements of the patent. The defenses set up are lack of patentable novelty in view of the prior art, and noninfringement.

The harrow consists of two distinct parts, one on either side of the tongue. These and the operation of them are exactly alike, and it w-ill be necessary to treat with only one of them. In order to simplify the description of the device, figure 1 of the patent in suit is here reproduced:

It was not new with Lindgren to provide a disk harrow producing substantially the results attained by the device of the claims in suit, so that the invention must be sought in the means employed by him.

Appellant’s harrow has, generally speaking, the following features in common with the prior art: (1) Its disk gangs are so arranged as to be capable of being adjusted in different angling positions with reference to the forward line of travel of the harrow; (2) the gangs may be tilted toward or away from the center of the harrow; (3) the gangs may be bodily advanced along the front and rear bars toward or away from the tongue or center of the harrow. In the means provided for the accomplishment of these, three results, the patentee claims he has *730achieved invention. In the first place, he contends that the transversely-elongated frame 1, above shown, with its widely spaced front and rear bars, within which the disk gang and the operating means are mounted, and its capacity for enduring the strain without reinforcing rods was new. He also claims as new and incident to the use of the frame the saddle-plate 17, bridging the space between the front and rear frame-bars with means for laterally adjusting the same upon the frame-bars as tracks, and having horizontally pivoted bearings or axes on and wholly within the bars. He also claims as new the arranging of the vertical pivot upon the horizontally pivoted member in such a way that its axis intersects that of the horizontal pivot within the plane of the frame, and the special form of adjusting mechanism for manipulating the gang on its vertical pivot which compensates for lateral adjustment of the gang and maintains the draw-rod in its correct line of movement. Among the other advantages claimed for the rectangular frame are that its two wide apart frame-bars, front and rear, afford adequate and rigid anchorage for the tongue without the use of brace-rods, and that it furnishes the mountings for a saddle-plate in the plane between the front and rear bars, which carries the vertically pivoted trunnion, number 14 in figure 4, which in turn carries the yoke and disk gang, and that it serves -as a support for all the other operating parts. In order to prevent unnecessary torsional strain upon the draw-rod when the.disk gang has been positioned at an angle transverse to or different from the line of travel of the plow, and especially when the saddle has been laterally removed to its outermost position, the patentee has provided links 29 and 30, with which the draw-rod 26 is connected by a hook formed at its rear end, which passes through the overlapped converging ends of said links at a point in the line of draw of said rod. These links extend divergingly to detachable union with the pins 70 at the end of the two arms 31 and 32 of the vertical trunnion 14 on the saddle plate, shown in drawing 4 of the patent here reproduced:

These links are made adjustable by forming therein a number of holes, 60, 61, etc., as shown in figure 4,t by means whereof the length of the links may be adjusted to compensate for the lateral removal of the disk gang.

*731[1,2] Assuming that the prior art discloses no rectangular frame, no saddle-plate between the front and rear bars of the frame, and no mounting of the operative parts thereon as shown in the claims in suit, and- no compensating mechanism for avoiding tortional .strain upon the draw-rod, the claims are deemed valid. The art is an old one, and hardly admits of radical improvement, while the wide range of uses for these plows makes welcome every inventive improvement, however simple.

Turning to the prior art, we find a number of patents deemed by appellee to be pertinent to the present issue. Of these we need only examine into the teachings of Corbin patent, No. 325,224, issued August 25, 1885, for a disk harrow, patent to Settle, No. 609,560, issued August 23, 1898, for a reversible disk harrow, patent No. 609,989 issued to Little August 30, 1896, for a reversible disk harrow, patent No. 670,070, issued to Willis and Porteous March 19, 1901, for a disk harrow, and patent No. 797,289, issued to Kennedy and Sharp August 15, 1905, for a disk harrow or cultivator. It will be noted that this last-named patent was issued 25 days prior to the issuance of the patent in suit, upon application filed March 21, 1904. The date of filing the application for the present patent was August 27, 1904. In this situation appellant undertook to establish Lindgren’s date of invention as prior to that of Kennedy and Sharp. In, that behalf appellant introduced evidence to establish the fact, • and now claims that the record shows that the Lindgren invention was fully completed and embodied in an operative machine, like that of. the patent, as early as January 1, 1903, that date being at least 14 months prior to the date of the application for the Kennedy and Sharp patent.

The patentee testified that he had first tried out his patent prior to January, 1903. He produced on the trial below the first shop order book of complainant, in which he pointed out six entries dated January 17, 1903, calling for the making of 100 Moline reversible harrows, identical with those of the claims in suit. The witness Johnson testified that he made those entries, and that they were made in the regular course of business, and on January 17, 1903. The witness Douglass identified the entries made on page 479 of a second order book, containing a list of all the parts entering into the construction of the harrows, as having .been made by him, and stated that the book indicates that the entries were made in January, 1903. The witness Howlett, at .that time head of the cost department of appellant’s business, recognized book B in his handwriting, referring to the Lindgren device. He stated that it was the custom to number and catalogue the parts. This seems to show that the patentee was working on the experimental machine in 1902. Howlett also produced pencil specifications which, in his own handwriting, read, “To shop, 1-28-03 How-lett,” and swore the entry was made at the time of its date. Howlett also identified “desk book 1” in his own handwriting, listing the cost of each of the parts — one entry being dated January 16, 1903 — and stated that the entry denoted the date when the complete machine was first delivered to him. He testified that the parts so entered by him were identical with those shown on page 210 of appellant’s 1904 cata-*732logue, and that the machine so delivered to him was shown on page 64 of appellant’s 1904 catalogue. There were placed in evidence by-appellant the regularly kept hooks of appellant, designated as the first shop order book, the second shop order book, original 'catalogue number book B, desk book 1, and original pencil specifications. These books were fully identified. They malee cross-references to one another. We are of the'Opinion that the identity of these early machines with that of the 1904 catalogue is established. Where one seeks to carry the date of invention back of the date of an anticipating patent, he assumes the burden of proof, and must establish an earlier date, “by evidence so cogent as to leave no reasonable doubt in the mind of the court, that the transaction occurred substantially as stated.” Deering v. Harvester Works, 155 U. S. 286-301, 15 Sup. Ct. 118, 39 L. Ed. 153; Columbus Chain Co. v. Standard Chain Co., 148 Fed. 622, 78 C. C. A. 394; Barbed Wire Patent, 143 U. S. 275, 12 Sup. Ct. 443, 450, 36 L. Ed. 154; Coffin v. Ogden, 18 Wall. 120-124, 21 L. Ed. 821.

[3] In cases where the invention may be exhibited in a drawing or in a model, it will date from the completion of such a model or drawing as is sufficiently plain to enable those skilled in the art to understand the invention. Walker on Patents, § 7, citing Doom Co. v. Higgins, 105 U. S. 594, 26 L. Ed. 1177; Deering v. Harvester Co., supra, and other cases.

[4] In order to constitute an “invention” in the sense in which that word is employed in the patent act, the party alleged to have produced it must have proceeded so far as to have reduced his idea to practice, and have embodied it in some distinct form. Gayler v. Wilder, 10 How. 498, 13 L. Ed. 504.

[5] Tested by the foregoing, was the proof, as to prior, invention above referred to, sufficient to fix the date of Eindgren’s invention as of or prior to January 1, 1903? Piad appellant produced one of the disk harrows from the 100 said to have been ordered and manufactured prior to January 1, 1903, priority of invention would have been conclusively established. No such evidence has been produced nor has its absence been accounted for. In Mueller v. Glauber, 184 Fed. 609, 106 C. C. A. 613, we held that in attempting to show a prior use the defendant was required to produce the best evidence available or account for its absence. In that cáse, prior use was sought to be proven by oral evidence alone, which was not deemed sufficient. In the present case the appellant produced order books, stock books, original specifications, catalogue number book, and others, and also appellant’s catalogue for the year 1904, together with oral evidence, all of which, taken together, can and do leave no doubt in our minds that the books and entries were made prior to January 1, 1903, and that Eindgren had invented and reduced to practice his harrow prior to that date. The- evidence can be accounted for on no other theory. We therefore, hold that the Kennedy and Sharp patent, -for the purposes of this suit, is not available as a part of the prior art.

The Corbin patent, relied on by appellee, lacks the rigid frame of Eindgren and, consequently, Eindgren’s mounting of the operative *733parts. It makes no provision for preventing the deflection laterally of the draw-rod when the disk gang is shifted with regard to. the direction of travel or by lateral removal, and thus entails heavy strain, in moving the gangs obliquely. It has means for lateral adjustment, of the gangs radically different from Lindgren’s. It has nothing equivalent to Lindgren’s saddle-plate supported between the front and rear bars of the frame, nor any yoke supported from a saddle-plate. It lacks, and therefore does .not anticipate, Lindgren’s combination.

The Settle patent has but one frame-bar. This bar carries an angle plate which affords a socket for the vertical pivot extending upwardly from the gang yoke. It has no rigid frames supporting a saddle-plate mounted on horizontal axes journaled in the front and rear frame-bars. A limited vertical tip can be given the disk gang by manually adjusting certain screws. It has no saddle-plate at all, or anything like it; nor any horizontal axis for such a plate. It affords no compensating means for retaining the draw-rod in straight adjustment, with its lever arm connection whenever the gangs are threatened with deflection by reason of the positioning, of the disk gang at an angle transverse to the line of travel of the plow, or by lateral, removal from its normal position with reference to the lever handle.' Nor has it any means for adjusting the disk gang in respect to an independent connection between it and the rear end of the draw-rod. For these and other differences in combination it does not anticipate the claims in suit.

The Little patent shows a frame consisting of two bars set edgewise and arched at the center, and has no end bars. It malees no provision for the protection of the draw-rods from frictional resistance in case of threatened deflection thereof. The connection between the two bars is made by means of a block which carries the vertical axis for the disk gang and is incapable of carrying the saddle-plates of Lindgren. No part of this patent could be adjusted to the Lind-gren device. It is not available for purposes of anticipation.

In the Willis and Porteous .patent the adjustment at the rear end of the draw-rod is entirely independent of the lateral adjustment for the gangs. In Lindgren, these two adjustments are mutually dependent upon each other. It ignores the necessity for lengthening or shortening of the members which afford transmission of movement from the rear end of the draw-rod to the gang mounting as shown in Lind-gren.

Thus an examination of the above fairly representative patents of the prior art fails to disclose any disk harrow containing the elements combined in the harrow of the claims in suit. The claims must therefore be held valid. In view, however, of the condition of the art at. the date of invention, they may not be held to be entitled to a broad construction.

Appellee’s harrow has the rectangular frame and saddle-plate of Lindgren arranged as in the patent in suit. The saddle-plates are slid-able along the front and rear frame-bars as in Lindgren, ánd adapted to be likewise clamped upon the bars in any desired position of lateral *734adjustment. The saddle-plate is mounted to turn on a horizontal axis spanning and within the frame-bars. The tilting feature is present as in Rindgren. All the advantages of Rindgren’s widely spaced bearing are obtained. The axis of the vertical pivot does not exactly intersect the axis of the horizontal pivot, but does so practically, thereby obtaining that advantage of the Rindgren construction. The rigid frame is used alone, for support of the parts as in Rindgren. Appel-lee also provides a device to prevent deflection of the draw-rod in case of lateral removal or transverse positioning of the gangs with reference to the forward travel of the harrow. The injection of the bell crank lever into this means of adjustment does not relieve appel-lee from liability as an infringer when using the adjustable link in connection with a bell crank, if the use of the links would otherwise constitute an infringement.

We are of the opinion that appellee’s harrow embodies the substance of appellant’s invention as set out in each of the claims sued on, and that such use should be restrained, and that it was error to dismiss the bih. '

The decree of the District Court is therefore reversed, with directions to' grant appellant the relief prayed for in the bill.