After' the foregoing statement of facts, GROSSCUP, Circuit Judge, delivered the opinion of the court, as follows:
The decision of the. Circuit Court turned upon the question whether the patents sued upon embodied a pioneer invention. We concur with that court in the opinion that if the invention was not of a primary character — a .substantial departure from the machines of the previous art — -the appellees’ machine is so sufficiently differentiated from. the. patents .sued, upon that a claim of infringement can not be maintained. But if, on the other hand, appellants’ patents were the first to give to the world a workable, portable machine for weaving wire fences in the field — a machine distinctly creating a new.product — and aptly embody in their specifications and claims the mechanical arrangements that bring about such a result, the decree below is erroneous.
' To accurately comprehend the scope and merit of the appellants’ inventions it is necéssary to'feview the preceding art. 'This divides *655itself into two branches; that relating to loom machines, where the wire netting was made in the factory, and afterwards transferred to the field; and that relating to field machines, where the fence was constructed in situ.
The Middaugh and Wilcox machine was the most advanced of the field machines, and may be used as an illustration of the art in that direction. It was used to build a picket fence, the pickets of which were held in place by twists or weaves of parallel longitudinal wires. It employed mechanism that would give to such parallel wires the necessary twists, but did not involve a shilling, either of the wires or of the spools, such as is essential to the weaving of a diamond mesh. It did not. in fact, build a wire mesh fence, but a wooden fence, with wire supports. It thus differs from the appellants’ inventions, not only in the mechanism employed, but in the product turned out. The wooden wicket fence, with its wire-supports, is, in no just sense, a predecessor, in kind, of the diamond mesh wire fence. Field machines of this character, therefore, must be dismissed, as not anticipatory of appellants’ machines.
The two inventions in the loom machine branch of the art, most closely resembling the appellants’ patents, are the Nesmith, patent, and the William Smith (English) patent.
The Nesmith patent was taken out April 4, 1854, and seems to have remained unimproved upon until the Davisson patent, December 4, 1883. The patent is set out at large in the statement of facts, and there is no need, in this connection, of calling attention to the details, other than those connected with the mechanical means, generally speaking, whereby the twisting and the shifting, requisite to the weaving of the netting was brought about.
An examination of the model, drawings, and specifications shows that it contemplated but two border-wires; though intermediate stay-wires could have been provided, by the use of intermediate twisting-heads with a hollow core. Each of the twisting-heads — ■ not only those carrying the border-wires, but the ones intermediate —had radial slots through which were carried the feeding wires, being held in tension by appropriate mechanical means. A turning of these twisting-heads carried the respective wires on their opposite sides around each other, and, at the borders, around the border-wires, tiras giving the necessary twists. This having been accomplished, the twisting heads were brought into position where the slots of each were directly opposite to the corresponding slot of the -other, so that, by an appi-opriate shifting device, the feeding-wires were carried into the opposite slots, and thus placed upon the opposite twisting-heads. It thus happened that, when the tvvisiing-heads were again put in motion, the feeding-wires were turned in a direction opposite to that of the previous turn; and, by repeated revolutions of this kind, the diamond mesh was woven.
Two - distinctive features of this patent must be home in mind: First, that the changes of the feeding-wire, so that it would be rotated alternately in opposite directions, are brought about by the shifting of the wire, and not of the reels from which it is wound; and secondly, that there was necessitated a supplemental .machine, *656whose office was, by a series of opposite revolutions, to keep the wire in the rear of the twisting-heads from becoming itself twisted. These differences clearly dispose of the Nesmith patent as an anticipation.
. The ¡Smith patent is as readily disposed of. There is no proof that it ever built in the field a fence, and the description discloses nothing to differentiate it from the class of looms to which the Nesmith invention belongs; for, in speaking of the manipulation of the feeding-wires, Smith says: “I apply to the lower part of the machine feeding bobbins, supplied with the* wires or cables, which pass up to the work in suitable guides, and I cut a space out of the centre of one of the halves of the twisting wheels (which are always made in two parts), so that my wires shall pass thro,ugh; then when the divisions of the wheels are carried apart by the ordinary sliding bars my wires are pushed by guides out of the spaces in the wheels into slots in one of the sliding bars about midway between the turning wheels and on the opposite side of its half wheel, so that my wires may be held while the twist is made, and they are thus relieved from the action of the twisting wires which form the mesh.
An examination of the appellants’ patents will show that, beginning with the Davisson patent, the twisting of the wires was effected by mechanical means entirely different from those of Nes-mith or Smith. Davisson dispensed with the supplemental mechanism necessary to the Nesmith machine, and, so far as we know, to all preceding loom machines, by substituting revolving reels for the stationary ones. ■ Instead of passing the wires through the slotted twisting-heads (which necessitated the rear untwisting mechanism) Davisson projected forward from the revolving gear wheel spindles provided with a cross-head at their working ends adapted to alternately receive and discharge the reels from which the feeding-wire was uncoiled. The revolution of these spindles, with the reels thus mounted upon them, rotated the reels in such manner as to twist the feeding-wires around each other, and around the border wires. After as many such rotations as were needed to give the requisite twists, the reels, having been brought forward to a proper position, were shifted to the cross-heads of the adjacent spindles.
The two significant features of the Davisson invention are: That the diamond mesh is made by a shifting of the reels, instead of the wires; and that the supplemental untwisting mechanism is avoided. This was a distinct advance in the improvement of loom machines. By an ingenious mechanical change, — the substitution of shifting-reels for mechanism shifting the wires — it reduced the loom machine from a complicated and bulky structure to a machine at once simple and effective.
But the Davisson machine, nevertheless, remained a loom machine. It was wholly inadaptable to use in the fields. It gave to the farmer no walking, workable diamond mesh fence maker.
What problem, then, did Kitselman, in 1887, have before him; what available material was at hand; and what did he accomplish? The. most casual comparison shows, of course, that his machine — • *657and especially its product — is essentially different from the fence builders of the Middaugh and Wilcox design. It differs radically also from the Xesmith patent, in that interweaving is done by the shifting of the spools, instead of a clumsy shifting of the wires. Its nearest kinsman is the Davisson patent; and, although the Da-visson patent is owned by the appellants, it must, for the purpose of determining the scope of the Kitselman invention, be looked upon as an anticipation.
Kitselman utilized the Davisson hollow spindle, projecting forward of the rotating-gear, with its cross-head and transferable spool-carriers thereon, for the purpose of twisting the intermeshing wire, and imbedding the longitudinal wires along the central line; but from this point he made an important departure — a departure that took the fence builder practically from its belly, as a loom machine, and set it upon its feet as a walking, working field machine. It individualizes the passing of this art from its lower antecedents to its present state of comparative perfection.
The mechanical changes were simple, but the results were revolutionary. The spindles of the Davisson machine were arranged vertically; had, with reference to the Davisson machine, a longitudinal movement; and were alternately forced forward and withdrawn from the plane of operation by means of a shifting device that was necessarily bulky and impracticable for field use. The Kitselman spindles were horizontally placed; had no longitudinal or lateral movement; and were confined permanently to the plane of their rotation.
In the Davisson machine, as in the Kitselman, the spindle, with its reels, acts as the twisting agent, the spindle forming the rotator; but in the Davisson machine the spindle, where the warp wire emerges from its hollow center, stands well back from the reels, thus allowing the warp'wire to go unsupported to the plane of operation; while in the Kitselman machine the spindle extends clear forward to the twisting zone, to which it carries, through its hollow center, the warp wire, fully supported against side pulls. In the Davisson machine the spool-carriers are transferred to their adjacent spindles by means of an apparatus previously described as involving a longitudinal and transverse motion. Kitselman effects the transfer of his spool-carriers by shifting the one to the spindle above the spindle with which it has just operated, and the other to the spindle below. This shifting, being brought about by a longitudinal motion only, thus eliminating the necessity of a transverse mo lion.
In the Davisson invention a simple gear is set behind each spindle, but none of these gears engage each other, so that, when the spool-carriers are shifted to adjacent spindles and cross-heads al each successive twist, they are revolved about a practically new center. Kitselman provided each twister at its central section with a spur gear of sufficient diameter to engage the gears of the adjacent twisters, and by this means imparted a simultaneous motion to the whole series of twisters. The gear arrangement of the Davis-son machine tends to twist the opposing carriers out of alignment; *658subjects tbe central wire to certain deflection; and pulls tide woof wires somewhat from their intended direction. In the Kitselman machine the pull, incident to the twisting operation, is constantly equalized, the central portions of the twisters offsetting each other in the plane in which the strain comes.
These distinctions are, to a certain extent, subsidiary, but nevertheless important. Their results entered into the general result, whereby the machine is taken from the ground, and effectively- put upon its feet. The striking, effective change, however, is in the ■elimination from the Davisson machine of the transverse motion, -(the transverse motion with reference to the Kitselman machine) .and as a logical result, of the cumbrous machinery effectuating such transverse motion. The effect is as if each alternate spindle of the Davisson machine (those out of gear and standing back during.the ■act of twisting) were eliminated, and the reels were transferred directly to and from the remaining adjacent spindles. It is this feature that gives opportunity for short spindles; that obviates the clumsy transverse adjustment; that, in short, compacts the machine within a space that permits its being set up upon its feet.
The Davisson machine could be put to no other use than that of a factory loom; it could not, in the very nature of its structure, be ■set up within a narrow compass. The Kitselman machine, by reason of the changes already pointed out, can be compactly mounted in a narrow vertical frame. The Davisson machine is, as we have already said, a fence builder practically on its belly, with all the clumsiness and inertness of the lower species; the Kitselman invention, for the first time, put the fence builder on its feet — an easy going, adaptable, working machine of the highest species. It built, for the first time, in the field, a wire mesh fence. It became a farm implement — an implement before unknown — as much so as the first harvester or the first stump puller.' It turned"over to the farmer in the field the work that had been previously done by the wire weaver in the shop. Out of it practically came a new product —a product as common now as the barbed fence, and probably much more useful.
The mechanical features distinguishing the Kitselman patent from the Davisson machine — and, of course, from all the preceding fence builders — have been indicated with sufficient precision. The first, second, eleventh, and fifteenth claims of the patent, read in connection with the drawings and description, seems to us to adequately embody them. In construing a claim the whole specification is taken together, and if the terms in which the claim is stated are consistent with those'of the description, the latter with its accompanying drawings is treated as an amplification and explanation of the former, illustrating and applying its more concise and ■definite expressions. Eqb. Pat. •§ 523.
Thus amplified by the description, the claims disclose a fence builder which, for the first time, employing a series of sectional twisters different from any that preceded, gears these twisters together for simultaneous rotation (a conception not involved in- the .Davisson patent); and brings about the necessary shifting by a *659longitudinal movement; .thus eliminating the Davisson transverse movement with its concomitant disadvantages — disadvantages fatal to the machine as a walking fence builder. Kitselman disclosed to the world for (he first time a practical means of supplying the farmer with a highly useful fence. He has set forth these means in detail, in the descriptive portion of his patent, and lias framed the claims mentioned to correctly embody them as a mechanical unity. His work is the work of a primary inventor, certainly as much so as in the cases presented in The Corn-Planter Patent, 23 Wall. 181, 23 L. Ed. 161, and The Barbed-Wire Patent, 143 U. S. 275, 12 Sup. Ct. 443, 36 L. Ed. 154; and his Letters Patent, are, therefore, entitled to a liberal construction.
A liberal construction, in our opinion, includes the appellee's’ machine. The latter is different in many* features of mere mechanical structure; in the direction traversed by the spool carriers; in the attachment of the spindles; in the* mounting of the pinions; in the manner of supporting the spool-carriers, and in many other particulars — set out in the opinion below and in the argument of counsel. But all those subsidiary differences still leave the appel-lees’ machine within the underlying structural conception — the mechanical individuality — of the Kitselman invention. They are changes here and there — oftentimes to its betterment — of the details, but they do not make the appellees’ machine, any the less, the direct offspring of Kitselman’s thought, and its mechanical embodiment. Within this view, appellees’ machine const ilutes an infringement.
The decree of the Circuit Court must be reversed and the cause remanded with directions to enter a decree in accordance with this opinion.
WOODS, Circuit Judge, dissents.