A dog is an instrument for holding a log in position while it is being saw'ed into boards or planks. Formerly the log was supported on the ends of the carriage, and was held firmly down, and secured against lateral displacement by iron dogs hinged to the carriage, and having teeth, which were driven into the ends of the log. When the circular saw came into use the form of the carriage was changed. Instead of supporting the log at the end, it was laid so that its side rested on the side rail of the carriage, and was supported its entire length. To hold it in position, Iwo or more movable supports, called “knees,” were placed upon the carriage behind the log. To prevent the rolling of the log, two or more pieces of iron were hinged or hung to the cross-pieces of the carriage, having at their free ends a right-angled hook, which was driven into the log. These were also called “dogs.” After a time this arrangement was followed by teeth attached to the knees, and operated in various ways. Some were driven down into the timber by blows, and some by means of levers of various shapes. In some cases the teeth moved simultaneously in an upward and downward direction,—that is, into the upper and under sides of the log or cant,—and in others the action was separate. These teeth were not often locked in place, but in a few cases the adjustment of the lever locked them in position. These teeth were also called “dogs,” but the term was also applied to the whole device, which was placed on the knee of the carriage, and included all the mechanism for protruding and retracting the teeth. Under this stale of the art the patent in this suit was issued. The patentee, Rodebaugh, claims an improvement in these dogs, consisting of an eccentric lever, pivoted to the standard upon which the dog slides, the office of which is to raise and lower the reciprocating shaft, which carries the dog-head by means of a connecting strap between the eccentric lever and reciprocating shaft. By the use of this device the teeth of the dog are forced into and withdrawn from the log at pleasure. In order to ascertain the lull scope of Rodebaugh’s invention it is necessary to advert for a moment to the first claim of his original application. , It read as follows;
*884“(1) /The eccentric lever, E, E1, connecting strap, E, combined with the vertical shaft or bar, D, carrying the dog-head, substantially as described for the purpose specified.”
This was a broad claim for the use of an eccentric lever and connecting strap, combined with a shaft carrying the dog-head, for the purpose of raising and lowering the teeth, and was rejected by the patent-office, upon the ground that it had been substantially anticipated by patent 163,309, issued to C. It. Ely, fora head-block. This was a combination of a lever working with a cam, the object of which was to elevate and depress the dog-head, which was carried upon a cylinder in much the sam e manner as Rodebaugh’s. The ground of the rejection is not stated, but it is obvious that it must have been upon the theory that the eccentric lever of Rodebaugh was a substantial equivalent for the weighted lever and cam of the Ely patent. The operation of the two is admitted to be substantially identical; and, in our opinion, the action of the examiner in holding that there was no invention in substituting the one for the other was correct. Upon this intimation from the patent-office, Rodebaugh amended his claim, and restricted himself to the combination of the eccentric lever pi voted to a standard, with a connecting strap articulated to an arm of the reciprocating shaft carrying the dog-head; the parts being so arranged that a downward movement of the lever will imbed the dog into the log, and lock it there, the lever assuming a perpendicular position against the standard, out of the way. Defendants’ theory is that Rodebaugh, having assented to the rejection of his original claim, and having substituted another, is bound by the literalism of his new claim; and, inasmuch as in the defendants’ device the lever when locked does not assume a perpendicular position, as required by Rodebaugh’s claim, there is no infringement. If it be broadly true that a patentee who has reformed his claim under instructions from the patent-office is thereby debarred from the benefit of the doctrine of equivalents, his position is correct. It is true, there is an intimation in the case of Sargent v. Hall, 114 U. S. 86, 5 Sup. Ct. Rep. 1021, that the limitations introduced into an application after it was persistently rejected, must be strictly construed against the inventor, and in favor of the public, and looked upon as in the nature of disclaimers; but, on a careful consideration of this and other cases, we think nothing more is, meant than that wdiere, under the state of the art and the action of the office, a patentee of a combination has modified and limited his claim, he shall be held strictly to the combination as he has described it. Thus, in Leggett v. Avery, 101 U. S. 259, it is intimated that if an applicant, in order to obtain the issue of a patent, acquires in the rejection of a claim thereto, a reissue containing such claim is invalid. But there is no suggestion that the patentee of such reissue is not entitled to the benefit of the doctrine of equivalents. So, in Vulcanite Co. v Davis, 102 U. S. 222, 227, after disclaiming an assertion that the correspondence between the applicant for a patent and the commissioner of patents can lx; allowed to enlarge, diminish, or vary the language of a patent after it is issued, it is said that a patent, like any other written instrument, is to *885be interpreted by its own terms. “But when a patent boars on its face a, particular construction, inasmuch as the specification and claim are in the words of the patentee, it is reasonable to hold that such a construction may bo confirmed by what the patentee said when he was making his application. The understanding of the parties to a contract has always been regarded as of some importance in its interpretation.” So, all that was said in Fay v. Cordesman, 109 U. S. 420, 3 Sup. Ct. Rep. 236, is that, if the patentee specifies any element as entering into the combination, ho makes such element material, and the court cannot declare it to be immaterial. It is his province to make bis own claim, and his privilege to restrict it. If it be a claim to a combination, and be restricted to specified elements, all must be regarded as material, leaving open only the question whether an omitted part is filled by an equivalent device or instrumentality. There is nothing in any of these cases inconsistent with what had been previously regarded as well settled, or to lead one to believe that it was the intention of the court to debar the patentee from his right to pursue an infringer, who has endeavored to avoid his patent by the use of a well-recognized mechanical equivalent. AVe think, therefore, that we are bound to look to the state of the art at tiie time this patent was applied for to determine the limitations upon the claim in question.
Under the first clause of plaintiffs’ patent as amended, there are none of the so-called “anticipations” which are worthy of serious consideration, except that of Craney, which undoubtedly resembles Rodebaugh’s device iu all its impoitant particulars. There is a fixed standard, A, corresponding to plaintiffs’ standard, !», a guide-bar, B, answering to the reciprocating shaft, 1), and the same lever, E, though in one it is a “T” lever and in ilie other an eccentric; a connecting bar, I), performing substantially the same functions as the connecting strap, F, of the Rodebaugh patent, and a projecting or horizontal arm, C, which answers the purpose of plaintiffs’ arm, tí, projecting from the upper end of the reciprocating shaft. All these elements perform substantially the same functions as the corresponding elements of the Rodepaugh patent. There are, however, three distinctions in the two devices which it becomes necessary to notice before determining finally the question of anticipation:
1. In the use of a “T” lever, instead of an eccentric. This fact is relied upon by the plaintiffs as constituting the improvement which distinguishes the Rodebaugh patent, not only from that of Craney, but from all the others which are claimed as anticipations. But the testimony, even of plaintiffs’ expert, indicates very strongly that for the purpose of driving the teeth of the dog into the log, a “T” lever performs the same function, and in as effeoíual a manner, as an eccentric. The operation of the two is precisely the same. Indeed, a model was exhibited in which an eccentric-and a “T” lever were used to move coincidentally the same connecting rod, and their operation was shown to be practically identical. The one difference suggested is that eccentrics are not- affected by back pressure,—a distinction which is strenuously denied by defendants’ expertsj—while the lever may be moved backward and forward, *886unless it is locked. This, however, is of no importance, as it is always intended that the lever operating the dog shall either be locked or held, in place by a weight upon the end of the lever. We are clear in our opinion that there is no invention in substituting one for the other.
2. There is in the Rodebaugh patent a rearrangement of the entire combination of the -Craney patent, by placing the lever and locking device midway between the upper and lower ends of the standard, instead of at the bottom, and using it to depress the upper dog, instead of raising theloy^er dog'into the log. We do not understand, however, that the rearrangement of an old combination, where each element of the combination operates practically as before, is patentable, unless a new or greatly improved result is gained. Woodward v. Dinsmore, 4 Fish. Pat. Cas. 163, 169. The variations of the Rodebaugh patent from that of Craney are scarcely greater than those of defendants’ device from that of Rodebaugh, and it is entirely clear to our mind that defendants’ device differs from that of Rodebaugh only in the rearrangement of the combination, by which a connecting rod, operated by a thrust, and articulating with the movable shaft at the bottom, is substituted for the connecting strap, F, operating by tension, and articulated with the reciprocating shaft at the top <by means of the arm, G. In this particular the case is much like that of Ives v. Hamilton, 92 U. S. 426.
3. The feature of the Rodebaugh patent, by which the lever, when the dog is locked, assumes a perpendicular position against the standard, out of the way, is not found in the Craney patent, and is probably the novel arrangement which obtained from the patent-office the allowance of his first claim as amended. Now, this is either a mechanical equivalent for a lever which locks horizontally, or it is not. If it be such an-equivalent, then it is anticipated b}^ the Craney patent, but if it be not such equivalent, and be a patentable improvement upon the Craney patent, then it is not infringed by the defendants, since they do not use this feature of the claim. There was undoubted^ considerable mechanical ingenuity shown by Rodebaugh in readjusting the various elements of the Craney combination, and perhaps some improvement in its mechanical operation, and we think his device approaches very near the border line of invention; but'upon the point which has been most earnestly pressed upon our attention, that the eccentric is a patentable improvement over the “T” lever,—and the plaintiffs’ case was practically put upon this ground,—we have not been able to adopt their view.