LURTON, Circuit Judge
(after stating the facts as above). The Deitz coupler is one of a large class of patents for improvements upon the automatic car coupler of the type known as the Master Car [850]*850Builders’ Standard. The patentee, in his specifications, says that his invention “has for its object to provide means to throw open the knuckle simultaneously with the movement -which unlocks the knuckle, and to do this with the locking-pin in the simplest and most efficient manner.”
The first claim, the only one in issue, reads as follows:
“1. Tlie combination with a draw-head, and a horizontally-swinging knuckle pivoted thereto, of a vertically-moving locking-pin normally obstructing the path of said knuckle, but not when partially raised, while the knuckle is closed, is prevented from automatically descending until the knuckle opens, and means whereby completing the raising of the pin throws the knuckle open.”
The court below, after a careful review of the history 'of the whole art and a consideration of this claim, in connection with the specifications and drawings of the patent, reached the conclusion that, while the Deitz patent embodied a patentable improvement, it was a narrow patent, and must be limited to substantially the device shown by the specifications and drawings; and that, thus limited, the Major coupler did not infringe.
After a most careful consideration of the whole matter, we find ourselves in unanimous agreement with this result. The opinion of Judge Sater, who heard the case below, goes so- fully into the history of the art, and makes such a satisfactory comparison of the two dev vices in conflict, that we only deem it necessary to summarize the points upon which we assent to an affirmance of the decree.
The older forms of such coupler stopped with an unlocking device by which the locking-pin might be raised out of the locking position and mechanism by which, when the engagement with the opposing coupler was complete the locking-pin dropped in front of the swinging knuckle to couple the cars together.
Thus, it was old and the subject of hundreds of patents to mechanically lock and unlock by means of a single device operated by a lever and chain from one side of the car; but, before this operation of mechanically coupling could take place, one of the couplers had to be in a wide open position, so that the coupler of the other car could enter the coupling head and become engaged therewith. This required an opening by hand; the brakeman being then, in some conditions, exposed to danger. The next step in the art was therefore to add to the locking and lock-opening mechanism some device by which the pin might be left in a lock-set position before the cars were drawn apart and the knuckle thrown wide open with the operation of pulling the cars apart. What Deitz has done was to incorporate in the old Janney type of automatic couplers devices for lock-setting and knuckle-opening, whereby, he claims, that, without multiplying devices, he has devised a coupler which locks, lock-sets and lock-opens.
By “lock-setting” is meant setting the locking-pin in such position, before the cars are pulled apart, as that it is not in the path of the knuckle and keeping it in that position until they are drawn apart, when it drops into its normal vertical position in the path of the coupling knuckle, ready for another engagement.
The peculiar form of his locking-pin is shown in the drawings and described in that part of his specifications already set out, and need [851]*851not be repeated. The operation of lock-opening — that is, opening the knuckles when the cars are pulled apart so as to admit of another engagement — is fully described in the specifications set out. The parts of the locking-pin and knuckle are so formed as to coact, resulting in a peculiarly constructed kicking- cam, which throws the knuckle into its wide-open position.
1. It must he conceded that the device of Deitz has three functions: That of locking or coupling automatically; that of setting the locking-pin in an unlocked position; and that of kicking the knuckle into its wide-open position, when the cars are drawn apart, ready for another coupling.
2. But it was not new to set the locking-pin in a lock-set position. This is shown in many of the prior patents, some of which are referred to in the opinion of Judge Sater. This lock-setting was done in some of them by supporting the pin upon the tail of the knuckle, and, in others, upon some part of the draw-head. The fourth element of the claim is for “devices whereby the pin, if partially raised while the knuckle is closed, is prevented from automatically descending until the knuckle opens.” The specifications and drawings show that, when the pin is partially raised, it will fall back and rest upon the, end of the knuckle, and ride thereon until the knuckle is pulled open. There was nothing new in so lock-setting, nor in the manner of doing it.
3. Neither was it new in this art to provide some form of cam, or spring action, for throwing the knuckle into its wide-open position, when the cars were pulled apart. A number of patents for improved car couplers, including lock-opening devices, are in evidence, and some of them are referred to in the opinion of the court below. The novelty, if any, is in the particular construction of the kicker which is shown in the Deitz device; but it certainly did not, when Deitz made his invention, require any exercise of inventive faculty to put a kicking cam upon any kind of coupling knuckle.
4. Neither was Deitz the first to combine the function of lock-setting and knuckle-opening in the same coupler. The Buhoup patent of January 26, 1892, is an illustration of such a combination. That patent shows a horizontally sliding lock pressed forward either by a weight or by a spring which locks the knuckle. When the lock is moved back enough to clear the tail of the knuckle, it is in set position. To automatically open the knuckle, a further movement of the pin or lock in its opening direction sets in operation a kicking lever and throws the knuckle open. The knuckle opener of Deitz’s patent is a cam, and in Buhoup, a lever, and both methods-for knuckle opening were well known. The Buhoup locking-pin moves horizontally, and not vertically, as in Deitz’s, and Bulioup employed a separate lever for throwing his knuckle open; but vertically-moving locking-pins were well known in the old art. Deitz, himself, in his earlier patent, shows vertically-movable locking-pins, and means for lock-setting on the knuckle tail. So did those devices show an outside lever for throwing open the knuckle. That Deitz, in the patent m suit, dispenses with any separate outside mechanism for throwing the knuckle open, by arranging a cam kick by the coaction of his pin [852]*852and knuckle, is the primary reason for conceding to him a patentable improvement in the particular method by which he has done this;but the old art taught him how to provide for knuckle opening by placing a kicking incline, or cam, upon any kind of lock.
Quite a number of instances in which lock-setting and knuckle-opening were combined in the same coupler might be referred to. Among them are the patents to Buhoup, No. 573,961; to Ludlow, No. 488,769; to Whipple, No. 562,871; to Washburn, No. 556,936. Deitz’s earlier patents, already referred to, show couplers in which devices for lock-setting and lock-opening are combined.
The patent to Aabel was upon an application made after that of Deitz. It is a closer anticipation than anything in the art. We lay it upon one side. The question of whether the invention of Aabel or Deitz was first in point of time is too doubtful, upon the evidence, to justify resting our judgment upon that patent. For much the same reason, we shall not consider the value of the second Buhoup patent, as a limitation upon the field of invention open to Deitz.
6. Among other reasons which tend to confirm us in regarding the Deitz patent somewhat narrowly is the fact that, although his invention was made in 1896, it has never gone into commercial use. The complainants own the patent, and they make couplers; but they do not make the device of the Deitz patent. This may be due to a number of causes. One, which the expert evidence seems to establish, is the fact that the structure described by Deitz, and which seems essential to the proper operation of his arrangement for lock-setting and knuckle-opening, is that there must be a cavity behind the knuckle in which his locking block or pin is required to swing fore and aft to accomplish the operation. The evidence tends strongly to show that couplers are frequently brought together with great force, producing a shock and a rebound of the knuckle after being closed. This, it is testified, is likely to result in- at least occasional failure in locking and lead to separation of the cars by jarring the pin into its unlocked position. But whatever the causes — and they do not seem to involve any financial inability to put the device in use — the fact remains that this case involves the interposition of this long unused patent to suppress the actual use of the defendant’s coupler, a coupler which appears to us to have been a valuable improvement over any other shown in this transcript, and one which lias gone into large practical use.
The Climax coupler, the coupler made by the complainant company, is made under paténts to Clinton Tower, issued in 1903, numbered 728,049 and 728,182. That it embodies some of the principal features of Deitz is obvious, and, if the Deitz patent was one of primary character, the anticipation might be serious. There are plain differences, however, between the two inventions — quite enough to make the claim now made that the Climax is an embodiment of the Deitz quite unsatisfactory in view of the limitations which we must impose on Deitz to support his patent at all.
The whole ground, for the purpose for which we use the fact of no commercial use, is covered by the admission of Mr. Deitz that no couplers like the patent in suit have ever been put into actual serv[853]*853ice. The mere fact that a patent has not gone into practical use does not defeat it, nor deprive the patentee of relief in equity against an infringer. The patentee may, if he will, reserve the invention for his exclusive use, or he may suppress it if he elect. It is his private property for the time of the monopoly. Heaton Peninsular Company v. Eureka Specialty Co., 77 Fed. 294, 25 C. C. A. 267, 35 L. R. A. 728: Paper Bag Cases, 210 U. S. 405, 28 Sup. Ct. 748, 52 L. Ed. 1122. The use ,we make of the fact that the device has never gone into actual service is in the construction or interpretation of the patent. We are justified, in view of the facts of this case, in exercising much caution in attributing to this patent anything more than is plainly shown and distinctly claimed. Bradford v. Belknap Motor Company (C. C.) 105 Fed. 63; Crown Cork & Seal Co. v. Aluminum Co., 108 Fed. 845, 48 C. C. A. 72. This inference from nonuse, under the circumstances, is the converse of the inference drawn in respect of a doubtful patent when a showing is made that it has gone into large use and has displaced other devices. It is an inference against utility from the fact of long nonuse, unexplained by want of means or opportunity.
Concluding, as we must, that the state of the art requires that the claim in issue shall not be construed broadly as for a primary invention, we come to the claim itself to see what limitations must be put upon it to save the patent for any purpose, and what limitations the patentee has himself imposed upon his claim by his manner of drawing it.
The claim to begin with is for a combination. The first element is a draw-head. Now this, by reference to the specifications and drawings, means a draw-head of the M. C. B. type. It means also a draw-head with a cavity in the rear of the knuckle to permit of the ore and aft swinging movement of its locking-pin. Unless there is such a cavity neither his lock-setting nor knuckle-opening device will operate. The Major coupler draw-head has no such cavity, but is provided with a vertical wall behind its knuckle to prevent any possible fore and aft swinging of the locking-pin. Neither device could operate in the draw-head of the other.
The second element of the claim is “a horizontally-swinging knuckle, pivoted therein.” Broadly, this covers the swinging knuckle of the whole class of M. C. B. Standard couplers; but, if we turn to the specifications, we find that this swinging knuckle differs from that found in all other couplers, as well as from that used by defendant. To operate, as Deitz intended his device to operate, he must differentiate it from all others, or there can be no coaction with his vertically-operating locking-pin. Thus, he must provide the upper surface of its tail with a raised rib to retain the locking-pin in its lock-set position. Its tail must be narrow, to permit the locking-pin to drop off in the rear of the knuckle when the knuckle is open. The under side of the tail is provided with an inclined groove, with which an -upwardly inclined lip on the locking-pin must interlock as a key in order to throw the knuckle open. These features are not found in the knuckle of the defendant.
[854]*854The third element is “a vertically movable locking-pin normally obstructing the path of the knuckle, but not when partially raised.” This pin differs most substantially from the locking-pin or block of the Major device, and also from the locking-pin of the devices of the old art. It is a pin pivotally supported on the top of the draw-head, and is adapted to swing freely backward and forward within the cavity of the draw-head. The very law of its operation is defined by the requirement that it shall not obstruct the path of the knuckle “when partially raised,” and that it shall do so both when the knuckle is locked and when in wide-open position. That it shall not obstruct the path of the knuckle when in its lock-set position is because when in that position it is resting on top of the tail of the knuckle.
This claim is so drawn as to define the precise characteristic which distinguishes the Deitz invention from all others. The pin does not obstruct the path of the knuckle when partially raised, because the effect of partially raising it, and thereby unlocking the knuckle, is that the pin falls back and rests upon the tail of the knuckle, where it remains until the cars are drawn apart. When the cars are pulled apart, the pin rides off the knuckle end, “and drops,” as the patentee states, “down in its normal position, ready to couple automatically.” When so in its normal position, it obstructs the path of the knuckle by being across the path of its inward movement. To accomplish a coupling, it must be removed, and this is done by the knuckle tail striking it, and swinging it backwardly and upwardly out of the way, when the pin falls back after the knuckle has made its inward sweep, it falls again across its path and locks it in position. From this position in front of the path of the knuckle it can only be removed by lifting it up by means of a lever and chain upon the side of the car.
That we do not too narrowly interpret the Deitz patent in respect to what is meant in this element by “normally obstructing the path of the knuckle,” we refer again to his specifications, where he says:
“The purpose of this (the éngagement of the shoulder, D', with the lip, B') is to provide that when the locking-pin is lifted and let fall back again before the cars are pulled apart, permitting the knuckle to be opened, the pin will fall back and rest upon the end of the knuckle and ride thereon until-the knuckle is pulled open, whereby the pin rides off the knuckle end and drops in its normal position ready to couple automatically. The pin, when down in its normal position, locking the knuckle, is shown in Fig. 2, and when in this position and the knuckle is. open and a coupling is made, the pin swings hack, to free the end of the knuckle, to the position of the dotted lines in Fig. 2, the lip L of the pin engaging the projection P of the main casting, and thereby acting as a center about which the pin revolves when automatically coupling, and when the knuckle is home the pin swings back to its normal vertical position between the end of the knuckle and the side wall of the casting, thereby locking the knuckle.”
Upon his examination as a witness for the complainants, Deitz said:
“XQ. 306. In the coupler of your patent, No. 561,541, and also of your patent in suit, No. 576,094, the locking-pin normally obstructs the path of the knuckle at all times, does it not, excepting when the locking-pin is partially raised, and is lock-set on the tail of the knuckle? A. It does.”
“XQ. 308. In both of the couplers in question, when the knuckle is in “its open position and the locking-pin is in its lowest and normal position in rear of the tail of the knuckle, the knuckle, when rotated to its closed position, will [855]*855strike the looking-pin and swing it rearwardly until the tail has passed or moved laterally to one side of the locking-pin, and thereupon the latter will swing by gravity into its normal position again, and thereby automatically lock the knuckle. Is this correct? A. It is.”
This element of the Deitz claim cannot be read upon the device of the defendants without discarding the limitations which the patentee has imposed upon himself, as well as the limitations which are required by the state of the art.
Upon this subject we quite agree with the court below, who, in differentiating the device of the defendants from that o[ Deitz, said:
“In both the Deitz and the Major coupling the locking-pin obstructs the path of the knuckle when the knuckle is closed. This feature is common to other couplers having vertically-movable locking pins. Were it not present the coupler would he inoperative and worthless. As the knuckle on the Deitz coupler opens, the locking pin rides off of the'knuckle and falls in the rear of its vertical wall in its normal position ready to couple automatically and hangs vertically in the path to be traveled by the knuckle in closing. In closing, the knuckle is driven hack against the locking-pin, knocking it rearward out of its vertical position into the recess in the draw-head, from which recess, by the force of gravity, it then swings hack to its vertical normal position, where it is caught between the sliouldor of the knuckle tail and the opposite side of the draw-head, thus locking the knuckle. Complainant’s position is that, when the locking-pin is dropped from its lock-setting position, it. is wholly immaterial as to the location of the pin, provided it is in position to lock the knuckle when it closes; hut interpreting the first claim of Deitz patent in the light of the specifications renders this position untenable. Its contention is that, by the language of the specifications and of the claim, the locking-pin obstructs the knuckle only when it is closed and locked.’.’
In conclusion: The art to which the patent in question belongs is so crowded with devices intended to accomplish the same purpose as to leave little room for the inventor. That there has been a slow, step by step, advance in the direction of the ultimate automatic coupler and a final result is as much as can he said. That the combination of Deitz involves some advance we fully concede; but that he has made such an improvement as to entitle him to any considerable range of equivalents we cannot concede. Each improver, of which Deitz’s and Timms’ and Tower’s are the last to which our attention has been directed, has done something; but when we concede to each his own particular form of device, so long as it differs substantially from those which have gone before, and does not include them, we shall have protected each as far as they are entitled to protection in a field already filled with the efforts of thousands struggling for the same result.
The case is therefore distinguishable from the long line of opinions by this and other courts where the meritoriousness of the advance has justified more latitude of construction and a considerable range of equivalents.
The result is that the decree dismissing the bill because no infringement has been shown is affirmed.
NOTE. — The following is the opinion of Sater, District Judge, in the court' below:
SATER, District Judge. The defendants are manufacturing the Major coupler under the Timms pat ents, numbered 678,145, 685,802, and 734,099. The complainant, the manufacturer of the Climax coupler and the owner of the [856]*856Dietz patent No. 570,094, issued on the application therefor filed September 2, 1896,- alleges that the defendants infringe the first claim of such patent. The claim is made on a combination of five elements, which for convenience are designated numerically, and is as follows: “The combination, with (1) a draw-head and (2) a horizontally swinging knuckle pivoted therein, of (3) a vertically-movable locking-pin normally obstructing the path of said knuckle, but not when partially raised, (4) devices whereby the pin, if partially raised while the knuckle is closed, is prevented from automatically descending until the knuckle opens, and (5) means whereby completing the raising of the pin throws the knuckle open.”
Since the adoption of the Janney type of coupler as a standard by the Master Oar Builders’ Association, in 1887, the form of the outer or hook end of the knuckle has been substantially fixed; but the other end, sometimes called the “tail end,” utilized as it is in locking, lock-setting, and knuckle opening, has been made to assume by inventors many forms and variations. The claim under consideration imposes no limitation as to the form of construction of such end, but, standing alone, is broad enough to cover every form of knuckle tail. The form and construction of the vertically-moving locking-pin which normally obstructs the path of the knuckle, except when partially raised, its mode of operation, the devices for lock-setting, and the means which, completing the raising of the pin, throw the knuckle open, are not indicated; hut the language employed indicates a claim comprehensive enough to cover any kind of vertically-moving locking-pin, any device for automatically lock-setting which involves the partial raising of the pin while the knuckle is closed, and any means of knuckle throwing caused by completing the raising of the pin. The complainant insists that the claim, which is of a sweeping nature, should be broadly construed. The defendants contend that it must be read in the light of, and limited by, the specifications and the prior state of the art. The patent must be construed as a whole, and due effect he given to all its parts, and if the claim be wanting, as charged, in particularity, the patent may be interpreted with reference to the other parts of it, such, for instance as the drawings and specifications. In determining the scope of the Deitz patent, therefore, the specifications and prior state of the art will be considered.
From Deitz specifications and drawings it appears that A designates a draw-head of the Janney type; B the knuckle thereof pivotally secured by a pivot pin to the draw-head and extending into a recess therein. The outward arm of the knuckle constitutes its nose; the inward, the tail of the knuckle. On the knuckle tail projecting upward is a lip, Bi. The end of the knuckle tail is cut away on an incline, B2, leaving an inclined way, Bs. The locking-pin, D, is so made as to serve the purpose of automatically locking the knuckle, and also to throw the knuckle open when the end of the car to which it is attached is not coupled to another car. The locking-pin extends upward beyond the upper wall of the draw-head, and has at its upper end a cap with a forward projecting lip, B, which rests, when the pin is in its normal position, on the top of the casting, but on P, a projection of the main casting or draw-head, when the pin is swung back to free the end of the knuckle. In the pin’s forward edge is a slot, <2, consisting of two vertical slots located in different planes and connected by a diagonal slot. At the lower end of the main body of the locking-pin is a shoulder with a projecting lip. Di, designed to engage the lip, Bi, on the knuckle tail. On the lower end of the locking-pin is a projection, D2, with an upward projecting lip, D’. designed to engage Bs. The locking-pin, when in its normal position (which is vertical) locking the knuckle, is as shown in Figure 2, between the end of the knuckle tail and side wall of the casting or draw-head; i. e„ the vertical shoulder at the rear edge of the'knuckle is engaged by one side of the pin and the inner wall of the guard arm of the draw-head is engaged by the other side. If, when the knuckle is locked, the locking-pin he raised, it will, on account of its engagement of the projection, E, on the front wall of the casting with the upper portion of the slot, (2, move vertically until the shoulder at the lower end of the main body of the locking-pin has been raised slightly above the upper face of the knuckle tail, or to its unlocked position. The pin being still further raised, the diagonal portion of slot <2 engages the-[857]*857projection, E, causing tlio locking-pin to tilt laterally, and,-if the knuckle is in contact vritli another coupler, the pin being permitíed to fall back, the lip. iu. engages the lip, ID, on the knuckle tail. The locking pin is thus supported and" loekset in its partially raised and unlocked position, and as the adjoining car draws away from the one on which the pin is loekset the knuckle is pulled open. As this occurs, the pin rides off of the end of the knuckle tail and drops into its normally vertical position, ready to couple automatically. If the car on which the locking-pin is being raised, or is lock-set, is not coupled to another, and it be desired to open the coupler, a further raising of the pin, causing the lip, D3, to engage the inclined way, B3, by cam action forces the knuckle end outward and open. When the knuckle is open, D3 having cleared B3, the locking-pin drops back into its normally vertical position in the rear of the knuckle tail and is ready to couple automatically. When the knuckle closes, its rear or tail end strikes the locking-pin, which hangs in the course of the knuckle while closing, and pushes or drives it backward, whereby ihe lip, L, is caused to engage the projection, P, of the main casting, and acts as a center about which the pin revolves, and from which it swings back and forth when automatically coupling. When the knuckle is home or closed, the pin, by the action of gravity, swings forward to its normally vertical position between the end of the knuckle and the side wall of the casting, and, being caught between the two, the knuckle is locked.
The opening of the knuckle by cam action was not novel when Deitz made application for his patent, as appears from Winternight’s patent, No. 441,624, Browning's. No. 254.106, Barnes & Barnes', No. 337,650, Parsons’, No. 395,492, Burden's, No. 388,396, and Flohr's. No. 445,245. In the Barnes & Barnes device the knuckle, when unlocked, slides down the incline or cam by the force of gravity into an open position. In the Flohr and Parsons devices the knuckle is kicked open by an arm of a lever, which serves to lock the knuckle in position when closed. In Winternight’s the knuckle is swung open by the chain, which first unlocks U. In Burden’s ¡.lie incline or wedge on the locking-pin, coming in contact with the tail of the knuckle, rotates it into an open position and is the equivalent of a kicking lever. In none of the foregoing parents, however, is there a cam or lip attached to the locking-pin to cooperate with a cam or incline on the knuckle to open it; but in the following patents each of the several patentees affix to their respective locking keys and a single-arm or tailless knuckle a given shaped cam to throw the knuckle open: Barnes & Barnes, No. 448,852; Wells. No. 471,702; Clark, No. 481,997; Tower. No. 487,649; Dickey, No. 509,208; White, No. 514,296; and McCord, No. 543,158. In the Kirwan & Kirwan patent, No. 470,579, March 8, 1892, aii Incline on the locking-pin engages an incline oil the under side of the knuckle tail and kicks or throws the knuckle open. In each of the eight devices last named, when the incline or cam on the locking-pin. is operated to open the knuckle, it engages the incline on the knuckle, as does that of the defendants, but, not being lii>-sliaped, does not interlock with the knuckle incline. The locking-pin in all of them is vertical, but has no swinging motion. In the Dietz parent the piu differs in form and in extent of motion, is vertical, suings back into a recess in the draw-head and thence forward lo its normal vertical position, and has a lipped cam which interlocks with a groove on the incline or cam on the knuckle head; hut the normal position of his pin and üie office to be performed thereby are not unlike those of each of the others.
Locksotiing of the locking-pin also antedates Dietz’s patent. In the Wine-man patent. No. 292,724, January 29. 1884, the locking block is made to lock-set on ihe draw-head, and as ihe knuckle opens the block seats itself on the knuckle, where it remains until the knuckle closes, when it falls in front and locks it. In Thurmond’s patent, No. 420,709, June 21, 1890, the locking block, whan partially raised, is tilted by gravity on account of being hung on one side of the vertical center and locksets on the draw bar, and, as the knuckle opens, it lakes its place on the tail of the knuckle and remains there until the knuckle closes, when it falls and locks it. In McKeen’s patent, No. 430,743. Jane 24. 1,891). the piu loekseis, on the draw bar. As the knuckle opens, the pin adjusts itself thereon and falls into its locking position when the knuckle rotates to its closed position. Substantially the same operation occurs in Me-[858]*858Keen’s patent No. 430,744, June 24, 1890. The lockset in the above devices resembles that of the defendants, which complainant alleges infringes its own.
In the Hinton patent. No. 534.284, March 26, 1895, the locking-pin is made to lockset on a shoulder of the draw-head. When the knuckle opens the pin drops to the rear in the path of the knuckle's movement, and when the knuckle closes the incline or cam on it engages the incline on the pin and raises it, and after the knuckle has passed beneath it the pin drops into its locked position. The dropping of the pin in the rear of the opening knuckle, and its consequent obstruction of the knuckle’s path and inward movement, are features embodied in the Dietz patent. Similar loeksets are found in Dietz’s two former patents, No. 561,541, June 2, 1896, and No. 561,542, June 2, 1896.
Loeksets and knuckle openers in the same device preceded Dietz’s patent. In the Buhoup patent, No. 467,680, the locking device is horizontal and different in many respects from that of Dietz’s; but it is so far similar in operation that if the ears are pulled apart when the locking bar is set it returns to its normal position, and as the knuckle closes it obstructs the inclined surface of the bar, forcing it backward horizontally until the knuckle passes it, when the bar automatically resumes its normal position and locks, the knuckle. In the Whipple patent, No. 562,871, issued on his application of January 28, 1895, there is shown in a single piece both a lockset and a knuckle opener. Other patents are referred to in the briefs for devices in which both the knuckle opener and the lockset are found.
Considering the specifications, the details of construction of defendant’s device, and the state of the prior art, it appears that the knuckle tail, essential to the operation of the Deitz device, must be in the particular form specified by him, that his is but one of the methods whereby a horizontally swinging pivoted knuckle can be made to perform the functions necessary to make an automatic coupler operative, that there were methods other than his whereby a vertically-movable locking-pin, normally obstructing the path of the knuckle, is used to lock and unlock such knuckle, and the lock-setting may be effected by means other than those adox>ted by him. The claim in question must be limited by what is shown in the specifications and drawings.
Much of the language of Judge Taft in St. Louis Car Coupler Co. v. National Malleable Castings Co. (C. C.) 81 Fed. 706, a car coupler case, is appropriate. He said: “We must begin the consideration of the questions in this ease, therefore, with the full understanding that couplers of the general contour of the patent in suit were old before it was applied for; that the forked draw-head, with one arm to act as a buffer, and the other for the purpose of pivoting a coupling head or knuckle, having two arms, one to hook and the other with a tail, which should lock the latch in the interior of the hollow draw-head was old; and therefore that the only possible patentable novelty of a coupler of the Master Car Builders’ type must be found in the shape of the tail of the coupling head, in the relation of the tail to the hook of the outer arm, and in the locking device. * * * It is possible that the adjustment of the parts in the comxfiainant’s patent, their contour, and their varying shape, lead to a better general result, including all the benefits stated by the complainant’s expert. But the parts are so clearly a reproduction of similar parts used in other couplers of the same kind for the same pulpóse, and with the same functions, that no patentable novelty can be successfully asserted to exist in the complainant’s combination, except in the exact form in which it appears in the .specifications and drawings. It is a patent which, if valid at all, is entitled only to the narrowest construction, and any variation in any of the parts of the combination will prevent infringement. It is only one in a series of improvements, all having the same general object and purpose; and in construing the claims of the patent, they must be restricted to the precise form and arrangement of parts described in the specification and to the purpose indicated therein.”
That the Deitz device proved to be of such slight commercial or practical use, if it has any at all, is another reason for denying a broad construction to his patent. Granting the ease with which experts, who undertake to prove their adversary’s xn'oeess a failure, may score a success, it is nevertheless true that it is not shown that any couplers constructed under the Deitz patent have gone into commercial use, notwithstanding the great de[859]*859maud for life and limb saving coupler devices and the requirements of statutes, state and federal, exacting the use by railway companies of automatic couplers and impelling a quick detection and use of a successful coupler of that character. Even if the nonuser of the Deitz coupler were attributable to Ms failure to introduce it, such fact would be entitled to some weight as reflecting on its want of utility. Crown Cork & Seal Co. v. Aluminum Stopper Co., 108 Fed. 845, 48 C. C. A. 72. Some obvious defects in the coupler are pointed out in the evidence. II s locking-pin, when locked, is not held against any fore and aft swinging movement. When pulling strains are released from any cause, it is not unusual for locks having such movement to swing backward and thereby cause a separation of the train. In switching services cars are frequently brought together with such great impact as to cause the knuckle, when closed, to rebound so quickly as not to catch the locking-pin or block and lock the coupler. To lock the coupler, the rebound of the pin, after it is struck by the inward moving knuckle, must precede that of the coupler. To give a broad construction to the Deitz patent, in view of its doubtful utility and its never having gone into practical nse, would operate rather to the discouragement than the promotion of inventive talent. Deering v. Winona Harvester Works, 155 U. S. 286, 295, 15 Sup. Ct. 118, 39 L. Ed. 153; Electric Smelting & Aluminum Co. v. Pittsburg Reduction Co. (C. C.) 111 Fed. 742, 757. In Walker on Patents. § 376, it is said: “An invention which was never useful enough to be used in any productive business cannot be dragged across the road which leads towards success, and thus be made to prevent the progress of a useful art along that road.” See, also, Bradley Pulverizer Co. v. Bowker Fertilizer Co. (C. C.) 111 Fed. 537.
For the purpose of comparison, a consideration of the Timms patent, No. 685,802, becomes necessary. In his device the locking block or pin, 7, is confined and operated within and mounted to move in a chamber, 8; no portion of the block extending beyond the outer -wall of the draw or coupling head. There is no recess to the rear of the coupling block, because the swinging back and forth action found in the Deitz device is not present in the Timms decide. Attached to the upper portion of the pin is a chain which extends upward through an opening in the coupling pin, and also through an opening in the coupler head. The opening in the coupling pin is located at one side of the center of the block, to impart to it, when raised, a lateral as well as a vertical pull, and thereby tip it to an inclined position. At 11 the block is chambered, producing the beveled shoulder, 12, to engage the beveled seat, 19, on the coupling head, so as to retain the block in its elevated position against accidental displacement. The vertical groove, 13, on the block receives a tongue, 14, on the knuckle tail, whereby the block locks the coupler. By recessing the block at 15 the shoulder, 17, is formed, and below the recess is the inclined or beveled lip, 16, through which above-named recess and between the Up, 16, and the shoulder, 17, the knuckle freely passes when in motion. If the cars are coupled together, and it be desired to unlock the knuckle without opening- it, and to lockset the block, the block, by means of the chain, is raised until the tail of the knuckle and the recess in the block are in alignment, and as the chain, when pulled, exerts both a vertical and a lateral motion, the block is tilled, and its shoulder, 32, is pulled on the shoulder, 39, of the draw-head, where the block is seated. The knuckle is now unlocked and the block is lockset. The pulling apart of the cars opens the knuckle, which passes through the recessed portion of the block, and supports the latter during the opening and closing movement. If the cars are not coupled together, and it be desired to open the knuckle, an upward pull of the chain affixed to the Mock frees ihe vertically grooved portion of the block from the tongue of the knuckle, aligns the recess in the block with the knuckle tail,' and by a continued pull on the chain the lip at the lower -end of the Mock is drawn in an oblique direction against the beveled face of the knuckle tail, when the cam action of the lip against the knuckle tail thro-ws it into an open position. "When the pull of the chain ceases, the shoulder, 3.7, of the block, is found resting on the knuckle tail, where it remains until the knuckle, tail is closed, at which time it drops in front of and locks it. In the Timms device, the knuckle tail rearward from the tongue, 14, to a point, 3it, is approximately horizontally fiat, the purpose of which is [860]*860to permit a part of the turning opening movement of the knuckle without interference with or dislodgement of the block when lockset; but from the point, 5a, on the knuckle tail to its rear end the knuckle is inclined upwardly for thepurpose of engaging and dislodging from its seat the lockset block and pf setting it on the knuckle tail during the residue of its turning opening movement. Between the pivot pin and the outward curved edge of the knuckle tail the surface of the knuckle is inclined slightly upward, with a curved edge to the pivot pin, so as to engage the shoulder, 17, on the coupling pin, and lift it from its seat at 19, when lockset, the lip, 12, being carried outside of the seat or hook, 19, and in the closing or locking action of the knuckle, the locking block being wholly supported on the knuckle, the lip. 12, descends to a plane outside of that of the seat, 19, thereby enabling the block to fall to its locked position without again locksetting.
Does the defendants’ coupler infringe the third element in Deitz’s first claim of combination? In both the Deitz and the Major coupler the locking-pin obstructs the path of the knuckles when the knuckle is closed. This feature is common to other couplers having vertieally-movable locking-pins. Were it not present, the coupler would be inoperative and worthless. As the knuckle on the Deitz coupler opens, the locking-pin rides off of the knuckle and falls in the rear of its vertical wall “in its normal position ready to couple automatically,” and hangs vertically in the path to be traveled by the knuckle in closing. In closing, the knuckle is driven back against the locking-pin, knocking it rearward out of its vertical position into the recess in the draw-head, from which recess, by the force of gravity, it then swings back to its vertical normal position, where it is caught between the shoulder of the knuckle tail and the opposite side of the draw-head, thus locking the knuckle. Complainant’s position is that, when the locking-pin is dropped from its lock-setting position, it is wholly immaterial as to the location of the pin, provided it is in position to lock the knuckle when it closes; but interpreting the first claim of Deitz’s patent in the light of the specifications renders this position untenable. Its contention is that, by the language of the specifications and of the claim, the locking-pin obstructs the knuckle only when it is closed and locked.
The construction to be placed on the terms “normal position,” “normal vertical position,” and “normally obstructing the path of said knuckle, but not when partially raised,” is readily determinable from Deitz’s own language. In his specification he says: “The purpose of this [the engagement of the shoulder, Di, with the lip, Bi] is to provide that when the locking-pin is lifted and let fall back again before the cars are pulled apart, permitting the knuckle to be opened, the pin will fall back and rest upon the end of the knuckle and ride thereon until the knuckle is pulled open, whereby the pin rides off the knuckle end and drops in its normal position, ready to couple automatically. The pin, when down, in its normal position, locking the knuckle, is shown in Fig. 2, and when in this [i. e., its normal] position and the knuckle is open and a coupling is made the pin swings back, to free the end of the knuckle, to the position of the dotted lines in Fig. 2, the lip, D, of the pin engaging the projection, P, of the main casting, and thereby acting as a center about which the pin revolves when automatically coupling, and when the knuckle is home the pin swings back to its normal vertical position between the end of the knuckle and the side wall of the casting, thereby locking the knuckle.” The third element of the combination is couched in this language: “A vertieally-movable locking-pin normally obstructing the path of said knuckle, but not when partially raised.”
Deitz testified on cross-examination: “XQ. 306. In the coupler of your patent, No. 561,541, and also of your patent in suit, No. 576,094, the locking-pin normally obstructs the path of the knuckle at all times, does it not, excepting when the locking-pin is partially raised and is lockset on the tail of the knuckle? A. It does.” “XQ. 308. In both of the couplers in question (Dietz patent, 561,541, and the Deitz patent in suit, 576,094), when the knuckle is in its open position and the locking-pin is in its lowest and normal position in rear of the tail of the knuckle, the knuckle when rotated to its closed position will strike the locking-pin and swing it rearwardly until the tail has passed or moved laterally to one side of the locking-pin, and thereupon [861]*861the latter will swing by gravity into its normal position and (hereby automatically lock the knuckle. Is that correct? A. It is.” The pin, according to Reitz, is in its normal vertical position, and normally obstructs the kmickle, both when the knuckle is oqjen and when it is closed, but does not obstruct the path of the knuckle when partially raised.
The word “obstruct” does not necessarily Imre so radical a meaning as that of blocking the passage or stopping it. The obstruction of the pin to the closing knuckle may be slight; but, nevertheless, with whatever of weight it has, it hangs in the path of the closing knuckle, and whatever force is required to swing it therefrom must be imparted by the closing knuckle. In United States v. Williams, Fed. Cas. No. 16.705, it was said: “Speaking etymologically, to obstruct (‘ob-struo’) is to build or set up something in the way. i; * * In a more critical acceptation ‘obstruct’ implies opposition without active force, and does not imply that the opposition was in the end effective. * * * Thus it may be that an officer of the law was obstructed in his duty and hindered perhaps for a time, but not finally prevented from performing it. So, too, he may have been obstructed: but, surmounting or avoiding the obstruction, he may have been not even hindered.” The locking-pin suspended in the rear of the knuckle is a something in its way. Its obstructive force is not active or effective as regards the closing knuckles, because the knuckle surmounts the obstruction and is driven home. And again, in Chase v. City of Oshkosh, 81 Wis. 319, 51 N. W. 562, 15 L. R. A. 553, 29 Am. St. Rep. 898: “An obstruction is a blocking up: filling with obstacles or impediments : an impeding, or embarrassing, or opposing the passage along and over the street — and, to constitute it such, it need not be such as to stop travel.” See. also, State v. Leaver, 62 Wis. 387, 22 N. W. 576.
The term “obstruct” etymologically, is not so strong a term as “obstacle” or “impediment” or “prevent.” It is frequently used in a comparative sense, as the above authorities indicate. To obstruct does not require a stoppage of travel, of which fact Reitz, in the use of the term, took cognizance. The lodgment of the locking-pin in the rear of the knuckle, and in the path to be traveled by it after the pin has ridden off of it, is essential to the succession operation of Reitz’s device, because his locking-pin cannot otherwise be. brought into a position ultimately to lock the coupler. His patent suggests no means of bringing the locking-pin into position between the tail of the knuckle and the wall of the draw-head, to lock the knuckle, other than knocking the pin out of its normal position, by the rear vertical wall of the knuckle tail, and the subsequent return of the pin by force of gravity thereto, and to effect this result he has given the knuckle tail the particular form shown in his specifications. To accomplish the same result with the locking-pin differently located would require a device differently constructed from that which he patented.
The defendants’ device differs from that of Reitz in that its locking-pin or block never obstructs the rear wall of the tail of the knuckle. When the locking-pin of the Reitz coupler is partially raised and lockset on the knuckle, it does not obstruct the path of the knuckle. The upper surface of the tail of the knuckle is flat, and moves in a horizontal plane. As it rotates it supports, but does not move, the pin. In the defendants’ coupler the lock, when in its lockset position, rests, not on the tail of the knuckle, but on a seat or lip on the wall of the draw-head. When the knuckle rotates to an open position the upward inclined surface of the knuckle tail, engaging the lock, lifts and dislodges it from its seat on the draw-head. The lock then seats itself on the tail of (he kmickle and remains there, not only while the knuckle is opening, but until it rotates to its closed position. When the knuckle of the defendants’ coupler is open, the locking-pin does not, like that of the Reitz coupler, obstruct the knuckle’s path. The lock of defendants coupler does, however, obstruct the knuckle tail in its outward movement to the extent, though it may be slight, of frietionally contacting with the upward inclined surface of the knuckle tail while moving outward. In this respect the defendants’ device again differs from that of Reitz, in that its lock, when partially raised, obstructs the movement of the knuckle.
The third element of the Deitz claim is a vertically-moving locking-pin normally obstructing the path of the knuckle, both when locked and unlocked, but [862]*862not when the pin is partially raised. The defendants’ device has a vertioallymovable locking-pin normally obstructing the path of the knuckle when locked, and also when the knuckle, while the pin is partially raised, is moving outward from the draw-head. The third element of Deitz’s combination is not found in the defendants’ coupler, and there is consequently no infringement. In Cimiotti Unhairing Co. v. American Fur Refining Co., 198 U. S. 399, 410, 25 Sup. Ct. 697, 702, 49 L. Ed. 1100, Mr. Justice Day said: “In making his claim the inventor is at liberty to choose his own form of expression, and, while the courts may construe the same in view of the siiecifieations and the state of the art, they may not add to or detract from the claim. And it is equally true that, as the inventor is required to enumerate the elements of his claim, no one is an infringer of the combination claimed unless he uses all the elements thereof.”
The defendants’ coupler has supplied a want, and is of commercial and practical value. Having successfully accomplished the desired end, it cannot be said to be anticipated by the Deitz device, which has failed in that respect. Farmers’ Mfg. Co. v. Spruks Mfg. Co., 127 Fed. 691, 62 C. C. A. 447. The fact that defendants’ coupler promptly went into extensive use is insufficient to sustain its patent, if its device is clearly wanting in novelty. Novelty, in my judgment, is not wanting. But, viewing it in the most unfavorable light, it must still be said that the question of its novelty is fairly open under the law. This being so, the fact that the trade by such use has attested its superior utility and value, which use necessarily implies the exclusion from the trade, if not actual displacement therein, of couplers previously patented and used, is persuasive evidence that it involved invention and had patentable merit. Kinloch Tel. Co. v. Western Elec. Co., 113 Fed. 659, 51 C. C. A. 369; Barbed Wire Patent Case, 143 U. S. 284, 12 Sup. Ct. 443, 36 L. Ed. 154. The Deitz device should not be employed to thwart the valuable benefits of the defendants’ coupler, now in extensive use and concerning which there is no evidence that Deitz had the slightest conception. General Elec. Co. v. Brooklyn Heights Co. (C. C.) 118 Fed. 154.
Conceding that Deitz’s combination was patentable, the fact that his coupler has not been a commercial success, has supplied no want, has excluded none other from use, has failed to meet the demand for greater protection of life and limb by means of automatic couplers, while the defendants’ coupler has gone into extensive use, a radical difference in operation and mechanism between the two couplers is strongly, if not conclusively, suggested. Campbell Printing Press Co. v. Duplex Printing-Press Co. (C. C.) 86 Fed. 815. The defendants’ devices for locking, lock-setting and knuckle opening are altogether different from those described in Deitz’s first claim and specifications. The end sought to be effected is the same in both methods, but the devices are not the same. If, for instance, the cams on the knuckle tail and on the lower extremity of the Deitz locking-pin were like those on the defendants’ device, the Deitz patent, on account of the pin’s freedom to swing, would be fatally defective. Such an inclosure of the locking-pin would be necessary as would prevent its swinging motion and hold it in place while contacting with the knuckle tail in opening. To so inclose it would prevent its riding off and dropping in the rear of the knuckle tail, and this in turn would require "a reconstruction of his entire device. To make the pin effective as a knuckle opener it was necessary to provide it with the upturned projection, D», and the knuckle tail with the groove, B-, and also to provide for the projection, E, to contact with the pin in slot d, as the phi is raised. Considering the state of the art, as shown by the earlier patents in evidence, and the slight utility and the want of commercial and practical use of Deitz’s device, it is not entitled to protection as a pioneer invention covering the achievement of the desired result in its widest form, unlimited by specific details. Conceding that his claim should be sustained, it should nevertheless be restricted to the particular devices described in the specifications, and under such a construction the defendants’ device cannot be deemed to infringe.
Other points urged by counsel, pro and con, are riot considered. The bill is dismissed, at complainant’s costs, and an order may be drawn accordingly.