brake heads, and that the co-action of these elements was effectual to prevent the rotation of the brakes upon the beam, I have no doubt. These elements are distinctly shown in the drawings and in the model produced at the trial. They are pointed out and clearly defined by the expert witness for the defendant, and very equivocally, if at all, denied by the experts of the complainants. It is true the function to be performed by this notch and lug are not described in terms in the first Hein patent, but their utility seems to me to be so obvious that their appearance in the drawings necessarily and quickly discloses it.

The defendant's beam does not show the exact notch of the complainants' device, but rather the notch of the first Hein patent, in this: that it is not distinct and separate from the slot of the tension rod, but rather a continued opening, extended from the point where the tension rod obliquely enters the hollow beam to the end of the beam itself; the end of the slot serving the same purpose, however, as the distinct and separate shallow notch found in the periphery of the beam of the patent in suit. Neither is the defendant's lug located in the same relative position in the cup of the brake head as that of the complainants. It more nearly resembles the lug of the first Hein patent than that of the patent in suit. The complainants' device, and the device of the first Hein patent, do not show exactly the same construction of lugs and notches, but they each, in my opinion, show the equivalent construction, namely, the Iug and the engaging notch or slot, and each employ these agencies to accomplish the same purpose, in substantially the same way. If, therefore, the defendant's construction is an infringement of complainants' patent, the first Hein patent is necessarily an anticipation of complainants' device. That which is an infringement, if later, constitutes an anticipation, if earlier. Knapp v. Morss, 150 U. S. 221, 14 Sup. Ct. 81, 37 L. Ed. 1059.

The evidence and argument of counsel have taken a wide range on this point, and many other considerations are called to the attention of the court which might dispose of these third and fourth claims, but I deem it unnecessary, in view of what has already been said, to devote any more time to their consideration. They are, in my opinion, in so far as they relate to the engaging of the lug and notch to prevent rotation of the brake head upon the beam, void for want of patentable novelty. If any special claim of novelty is made for the provision of the third claim, which provides for the perforation of the beam for the passage of the truss rod into it, in lieu of having a continuous slot from the point where the truss rod enters the beam to the end of the beam, it is enough to say that the defendant's device does not infringe any such element of the claim. The other element of claims 3 and 4, namely, the cupped recess for the reception of the end of the beam, will be considered in connection with other claims of the patent.

The next and only remaining claims of this second Hein patent are the fifth, sixth, and eighth. They are as follows:

"(5) A trussed, hollow, or tubular brake beam, having a camber, and having one or more notches at its ends, in combination with brake heads having

cupped-shaped recesses, inclined bearings and one or more projections, and the truss rod, passing through and holding said heads in position.

"(6) A trussed, hollow, or tubular brake beam, having a camber, in combination with brake heads having cupped-shaped recesses, a truss rod passing through and holding said heads in position, and the inclined bearings for supporting the nuts on the ends on the truss rod."

"(8) In a trussed brake beam, the combination with a cambered compression member of brake heads having cupped-shaped recess for the reception of the cambered compression member, and a tension rod, substantially as and for the purposes specified."

Eliminating from these claims the elements found in other claims already disposed of, there remains in them substantially one feature only for further consideration, and that is the feature of the cuppedshaped recesses or sockets for the insertion of the ends of the compression member or beam into the brake head. These claims, also, in so far as they involve this cupped-shaped recess, had a troublesome experience in the patent office. The claims, as originally filed, were for either a cupped-shaped or hollow socket. As so claimed, they were rejected on references to which I need not now refer, and subsequently, by divers changes, the claims were modified and amended, so as to appear as now found in the claims under consideration. As so modified and amended, they were again rejected by the examiners on references to prior patents, and afterwards allowed on appeal. The final allowance of the claims was made after argument by the patentee's attorney showing, among other things (what is quite apparent from the claims themselves, as finally allowed, and the descriptions and drawings of the patent), that a cupped-shaped recess, as here claimed and allowed, implies a bottom or closed end. In an argument made September 30, 1889, the attorney for the patentee declared the particular advantage or function of a construction embodying the cupped-shaped recess of the claims in question to be that, "by reason of the ends of the recesses of the brake heads being closed, these heads, of course, cannot slip or be drawn along the beam." In an argument made by the attorney on April 10, 1888, he states that the citation of the first Hein patent is avoided by the closed ends of the cup, which the attorney says Hein's first patent did not show.

Without pursuing the argument or proceedings in the patent office further, it may be confidently said, as a result of the evidence in this case, that the hollow brake head or recesses generally, as first claimed, were abandoned on references made by the patent office, and that a voluntary limitation was accepted by the patentee, namely, a cupped-shaped recess alone, which implies, as is admitted, a bottom or closed end, and that the special function or new use to be served by the device of the claims is to prevent the brake heads from slipping or being drawn inwardly along the beam. The defendant's device which is alleged to infringe these claims is one in which the recess for the brake head is not cupped shaped,--that is, it does not have the closed end, but is simply a hollow perforation, just big enough to permit the entry of the cylindrical end of the beam, with no obstacle like the closed end to prevent its sliding through the recess; and it is urged that the defendant employs a device consisting of a cap and washer at the outer open end of the recess, through which the

end of the tension rod extends exactly as it does through the bottom of the cupped-shaped recess of the claims in question, and that by this device the same function is discharged as that which results from the cupped-shaped recess of the claims; in other words, that the defendant's device consists merely of two parts instead of one, as in the claims of the patent under consideration, and is therefore the mechanical equivalent of the device of such claims. It is urged by the complainants' counsel, and many authorities are cited to the proposition, that infringement is not avoided by simply making in two pieces what was formerly made in one piece only. This proposition cannot be denied, provided the device with the two pieces performs the same function and produces the same result as the prior one-piece device. But is it true that the defendant's device performs the same function as the device of the claims in question? I think not. I find nothing in the defendant's device, apart from the operation of the other elements of the tension rod and nuts (which are not now involved), which would prevent the sliding of the brake head inwardly along the beam. Unless tightened up by the tension rod, the brake head proper can slide along the shaft without any physical interruption by the cap and washer. They simply separate-that is, the cap and washer separate from the end of the tubular beam, while in the device of the claims it is physically impossible for the brake head to slide inwardly along the shaft of the beam, even without the effective operation of the tension rod and nut. The resistance of the bottom of the cupped-shaped recess or socket at once prevents such motion. But whether this difference of function or result exists in fact or not, it cannot, in my opinion, be now questioned by the complainants. They are estopped from so doing by the enforced amendment of their claims, accepted by the patentee, which limit them to a particular construction. Brill v. Car Co., 33 C. C. A. 213, 90 Fed. 666, and cases cited. They are also estopped from so doing by representations made in the patent office to secure their patent. Their patentee secured the grant with much difficulty, after many citations, and after divers changes, modifications, and amendments, to avoid such citations, and then only upon the representation, voluntarily and repeatedly made, that his device, the socket or recess of the brake beam, was made in one piece, and so made with closed ends, like the bottom of a cup, in order that it might not slip along the shaft inwardly. Under such circumstances, the construction of the patentee's brake head in one piece, in order thereby to perform the particular functions claimed, is a self-imposed limitation, which complainants are not at liberty to question or disregard. Westinghouse v. Power-Brake Co., 170 U. S. 537-558, 18 Sup. Ct. 707, 42 L. Ed. 1136. In fact, it appears to me conclusively from the proof in this case that, if the complainants' invention is not so limited, it was anticipated by several of the references made by the patent office, as disclosed by the file wrapper. It results that, conceding to the complainants the full benefit of their invention, as found in claims 5, 6, and 8, and conceding the validity of the claims, notwithstanding the strenuous attack upon their patentability made by the defendant's counsel, the defendant is not shown to have infringed either one of them.

The next patent for consideration is No. 430,755, granted to H. B. Robischung, of date June 24, 1890. Its second claim only is alleged to be infringed by the defendant. It is as follows: "A clamp or yoke having flanges provided at or near their extremities with inwardly projecting lugs, which limit the closing of the flanges, substantially as and for the purposes specified." This claim relates to a yoke clasped around a brake beam, with extended flanges, through which a bolt is inserted upon which to hang the safety chain. In actual use, the operating nut at the end of the bolt is screwed up, and has a tendency to cause the flanges to bind or rigidly hold the safety chain, and to cause the chain itself, when so rigidly held and subject to the jolts and jars of service, to loosen up the bolt. The object to be accomplished was to go regulate and control the tightening of the bolt as to always permit the chain to hang freely over it, unbound by the flanges of the clip. To accomplish this object, lugs are provided, one on each of the inner sides of the flanges, at their ends below the bolt, so that, in the process of tightening up the bolt, the lugs, being first brought into contact, interpose effective limits to prevent the flanges from impinging upon or binding the chain. This is all plainly disclosed by an inspection of Fig. 5 of the drawings of the patent, as follows:

The figure "3" here represents the lugs, and the figure "5" represents the bolt passing through the flanges.

There is no pretense that the yoke, flanges, or bolt of this claim are new, or that the purpose served by them is new. The invention of the claim consists simply in the intervention of these lugs at the end of the clip, to limit the effect of tightening the bolt, and thus always keeping the flanges sufficiently apart to prevent their squeezing the chain hung over the bolt. The question is whether this expedient rises to the dignity of "invention," as the term is employed in the patent law. I do not believe that every little expedient resorted to, not in an abstruse art or pioneer invention, but in ordinary physics, for the purpose of more effectually applying the well-known laws of force, should be easily monopolized. Without any consideration of the prior art, it seems to me that the employment of the lugs of this patent for the purpose stated was a suggestion of common experience, and so obvious and appropriate to prevent the squeezing of the safety chain as to be naturally and rationally seized upon by a skilled mechanic at all familiar with the requirements of railroad service. If so, it did not involve invention. Hollister v. Manufacturing Co., 113 U. S. 72, 5 Sup. Ct. 717, 28 L. Ed. 901. Given the necessity of holding the

flanges apart, so as not to squeeze the chain, and at the same time keep the rod over which the chain hangs reasonably tight; can anything be more obvious than to interpose some solid substance, like lugs, to prevent too much impingement? I think not. But, when the prior art is considered, we are not left to the sole resort of obviousness. The lugs of this patent are shown in the cant-hook clip in evidence, and the mechanical equivalent of the lugs is, in my opinion, found in the Preston patent, No. 369,383, and the first Hein patent, No. 361,009. In these patents the flanges of the clip are prevented from approaching too closely together by the interposition of the tension rod, located so as to pass between the flanges of the clip. In my opinion, claim 2 of this patent is void for want of patentable invention and novelty.

The next and last patent requiring consideration in this case is No. 466,984, granted to H. B. Robischung, of date January 12, 1892. The first and second claims only are alleged to be infringed. They are as follows:

"(1) In a cambered trussed brake beam, the combination, with the compression member and the strut or post, of an interposed key or wedge, substantially as and for the purposes specified.

"(2) In a cambered trussed brake beam, the combination, with the compression member, of a strut or post having a collar for the compression member, said collar provided with key slots and a key or wedge, substantially as and for the purposes specified."

The first claim, it is noticed, relates exclusively to the key or wedge. The second claim relates to the same key or wedge, and also a slot in the collar of the strut for the insertion of the key. The invention of these claims may therefore be considered together. It consists in providing a slot for the insertion of a wedge in the collar of the strut which clasps the beam, for the purpose of taking up the slack between the collar of the strut and the beam, and for the purpose of overcoming lost motion. The complainants' device involving these claims is readily seen in the following cut, Fig. 3 of the patent:

The circular or bent key, a', b', represent both the slot and the key called for by the two claims in question.

By reference to the above drawing, it is observed that the key of the claim is of a peculiar circular or bent shape, and follows a tortuous channel or slot made for it transversely across the beam. In other words, this device presents a peculiar and unusual con