This is the ordinary suit for the infringement of a patent, issued on December 11,1928; to Elmer K. Cole, No. 1,694,683. The subject of the invention is a shackle for holding the end of a metal cable, made up of three strands of twisted wire, of the kind which has now become very common as a guard on highways. The cable enters one end of the shackle; to the opposite end a shank or bolt is fixed hy which the cable may be anchored to a post or mooring. The shackle is of two parts, a clamp and a wedge; the clamp, a conical piece of metal, hollow, with an opening at the apex about the size of the cable to be used. Immediately fixed to the base, but of smaller diameter, is a threaded collar axially identical with the clamp, through which its interior may be reached. The inside of the clamp is rifled with three helical grooves, corresponding to the pitch and diameter of the twisted strands of the cable, and therefore capable of receiving them upon their separation. The wedge too is a cone, with three helical grooves, also of the same pitch and diameter as those of the cable. In operation the end of the cable is put through the apex of the clamp; the wedge is inserted through the collar and between the strands of the cable. It is then tapped or otherwise forced forward, separating the strands, each of which finds a place between a groove in the wedge and a groove in the clamp, which together substantially embrace it. Thereupon the bolt or shank is screwed down into the collar, passing through it and against the end of the wedge which it drives home and holds in place. The result is that a strain on the cable causes the strands to be still more tightly held in the grooves, but the pressure instead of tending to flatten the strands, is generally distributed over their whole periphery. The five claims are broadly of two classes. The first two do not, as we shall show, contain the shank or holt, or any other means for driving and holding the wedge in place; but this element is present in the last three, in claim five being specific to the disclosure. Claim one specifies grooves in both clamp and wedge; claim two in the clamp alone; it is the most general of all.
The supposed infringement was of very similar general kind; that is to say, it had a clamp and a wedge, and in operation the cable was put through the apex of the clamp and held in place by driving the wedge between the strands. The differences were these. There were no grooves on the inside of the clamp, and the grooves in the wedge were not helical, and not therefore of the same pitch as the strands of the cable; on the contrary they were parallel to the axis of the wedge. (An earlier form had had grooves set at an angle to the axis, hut these were abandoned before suit brought.) Finally, though there was a screw threaded collar fixed to the clamp, instead of being directly incorporated into its base, it was joined to it by two legs about five inches long. The result of this last detail was to leave the opening in the base of the clamp more accessible, and to require a much longer descent of the shank through the collar before it reached the wedge. It was indeed still possible to drive home the wedge by this means, but the defendant did not sell shanks, and, so far as appears, its customers have never used them for this purpose. There is therefore no evidence of contributory infringement of the last three claims. The District Judge held that this element was also incorporated into claims one and two; in the first, by the phrase, a “screw threaded bolt for said opening”; in the second, by “means for forcing the strands of said cable into said grooves.” *687We do not agree. There might be a screw threaded bolt in the opening which was not used to force in the wedge and in fact the defendant’s shackle is used with a bolt, as we have said. Again, the wedge alone forces the strands into the grooves. Thus, although we concur with the reasoning and the result as to claims three, four and five, claims one and two remain.
Claim two originally stood without the' phrase “of substantially the same diameter and pitch, as the strands of the cable to be clamped.” It then read merely as a claim for a clamp with helical grooves and means for forcing the strands into them, and in that form it was rejected on references that are not in the record and of which we cannot therefore know the pertinence. Cole amended the rejected claim by inserting the language we have mentioned, and although claim one had been allowed as it stood he introduced the same phrase into that as well. His argument now is in effect that it may be disregarded under the doctrine of equivalents; and that certain shallow scorings made in a soft coating laid upon the inside of the defendant’s clamps are the equivalent of his helical grooves, though these scorings are of neither the same pitch nor anywhere near the same diameter as the strands. They are produced by heavy tension upon the cable; they are markings, not of the strands as strands, but of the wires which chance to touch the surface of the clamp. Apparently they act in some degree as a cushion and possibly prevent the strand as a whole from being crushed out of shape quite as much as it would otherwise be. However that may be, claim two is certainly not infringed under well settled law; an amendment introduced to escape a reference becomes an essential of the claim which cannot be later ignored. Hubbell v. U. S., 179 U. S. 77, 83, 21 S. Ct. 24, 45 L. Ed. 95; I. T. S. Rubber Co. v. Essex Rubber Co., 272 U. S. 429, 443, 47 S. Ct. 136, 71 L. Ed. 335; Smith v. Magic City Club, 282 U. S. 784, 789, 790, 51 S. Ct. 291, 75 L. Ed. 707.
There remains therefore only claim one which was voluntarily amended. Assuming for argument that this makes a difference, still we cannot wholly disregard the language so introduced; indeed if we were to read the claim as it stood originally, it would not verbally cover the defendant’s shackle. Though the scorings on the inside of the clamp be “helical grooves,” the grooves in the wedge are not “helical” at all; at least we should have to disregard that element of the claim in toto to find any infringement. No doubt the interpretation of patent claims depends more upon the advance made by the inventor than upon the words used, and in spite of protestations to the contrary, courts do at times play fast and loose with them as they do not with other formal documents. It is therefore always proper, and generally necessary, to look at the prior art in order to learn how closely it presses upon the disclosure. Cole’s invention came by no means out of the blue; as far back as 1877, nearly fifty years before he filed his application, Collier and Thomas, No. 194,991, had disclosed a conical shackle for holding a hempen rope whose fibres were forced apart by a conical wedge driven in from the other end and then soldered in place. Gottsehalk, No. 1,466,127 repeated the idea in 1923, this time with a wire cable, the wedge driving the wires apart and the whole being made solid with solder. Very few additions to this were necessary to comprise all that the defendant does; the wedge need only be grooved, the clamp lined with a soft thin coating, and the shackle used for a cable of several strands of twisted wire, instead of a single strand. The first of these elements had been already disclosed by Marehand, No. 985,915, in 1911. This was a shackle for a single strand of twisted wire with a core. There was a clamp without grooves, and the co-operating element was a conical wedge, set forth in several forms, one of which had grooves, not strictly helical it is true, but diagonal to the axis, and so nearer to the patent in suit than the defendant’s present form. The core was to pass through an axial hole in the wedge, which was half split, and the surrounding wires were to be placed in the grooves. If the cable was made of strands set in corresponding grooves the defendant’s device would have been reproduced, except for the soft coating. It scarcely seems to us that, given the disclosure of a single strand so shackled, it took invention to adapt it to a cable of several strands; the necessary adjustments were inevitable as soon as it became desirable to use such cables. More may be perhaps said for the soft coating as an invention, though it would be going far to hold that Marchand’s disclosure did not extend to the use of a soft metal for his clamp, or to the rust-proofing of a hard metal. Besides, we may not disregard all limitations deliberately imposed upon a claim, and read it as broadly as the prior art would allow the inventors to have drawn it. Even if it would be an invention to coat the clamp with such metal, that was not claimed when the grooves were described as of the same pitch and di*688ameter as the strands. To say that the cable is held by substantially the same means seems to us therefore untrue; really all that the defendant does is to increase the friction of an uneoated clamp and ease the pinch on the wires a little. That is a different means from embracing the whole strand between the two members. The art was too close to give much room for equivalents; if the patentee meant to claim a combination in which the wedge was grooved and the clamp was not, perhaps the field was open; but we should not care to hazard a guess as to whether the Patent Office would have allowed a claim over Marehand merely because of a soft coating on the inside of the clamp. While the claims are valid within their scope, they do not, we think, cover the supposed infringement.
Decree affirmed.'