This is an appeal from an interlocutory decree in favor of the plaintiff relating to United States letters patent, No. 821,123, granted to C. F. Roper, May 22, 1906. The defenses are anticipation and noninfringement. It was found in the court below that claims 1 and 9 of the patent were valid and infringed, and that claims 12, 13, and 18 were invalid, not being sufficiently definite. The present appeal of the defendant is from so much of the interlocutory decree as held claims 1 and 9 valid and infringed. No appeal was taken by the plaintiff from the interlocutory decree as to claims 12, 13, and 18.
[1] The plaintiff’s patent is for an improvement in filling-exhaustion-indicating mechanism “for detecting and indicating substantial exhaustion of the filling in the running shuttle, and thereupon to cause either automatic loom stoppage or automatic replenishment of the shuttle without stoppage.”
In the hearing before us only one of the prior art devices relied upon in the court below was urged in opposition to the finding of validity; that device is shown in United States letters patent No. 698,579, granted to W. I. Stimpson, April 29, 1902. It is true, as the defendant contends, that the feeler members of this device are brought in contact with the filling or the filling and its carrier on each alternate beat of the lay, the same as is the case in the plaintiff’s device; but the distinction between the two devices lies in the respective means provided and their mode of operation for effecting a change in the filling mechanism. In the plaintiff’s device one of the feelers is pointed, so that it sinks into or penetrates the filling, when there is filling on the bobbin, which feature is essential to the working of the device; while in that of Stimpson, although one of the feelers is narrower than the other, it is not pointed, and its mode of operation is not dependent upon its possessing a penetrative capacity. On the contrary, such capacity would seem to be detrimental to its operation. Then, again, the operation of the plaintiff’s feelers is dependent upon the softness or penetrability of the filling mass, as compared with the harder material of which the bobbin is composed, a feature in no way material in the operation of the Stimpson mechanism. In the plaintiff’s device, when the filling on the bobbin has become so reduced as to exert no pressure upon the impinging member, the penetrating member then contacts with the hard surface of the bobbin, both members are moved forward in unison, and the change is effected; while in the device of Stimpson the change is effected only when one of the feeler members is moved forward relatively a predetermined distance by an annular enlargement on the barrel of the bobbin, which is so located with reference to one of the feeler members as to effect this result when the filling has been sufficiently exhausted to permit it.
In the court below it was found that the “patented device effects in operation a substantial saving, by diminishing the amount of the filling left on the carrier at each loom stoppage or replenishment, to an extent not attainable by any prior mechanical exhaustion-indicating device. It embodies therefore a useful improvement, and one which I also find to have been patentably new, notwithstanding that it was *550made * * * in á ‘refined’ art.” We think the proofs fully sustain the findings as to utility and patentable novelty.
[2] The question of infringement as to claims 1 and 9 remains to be considered. These claims read as follows:
“1. In a loom, a shuttle to contain a carrier having a mass of filling wound thereon, filling-replenishing mechanism, and means to control the time of its operation, said means including two adjacent and yieldingly mounted, relatively movable members one of which is adapted to penetrate, and the other to impinge against, the filling mass in the shuttle, to effect relative movement of said members, substantial exhaustion of the filling permitting the penetrating member to engage the carrier and move in unison with the impinging member, to cause the operation of the controlling means.”
“9. In filling-exhaustion-indicating mechanism for looms, in combination, two adjacent and relatively movable members one of which is adapted to penetrate, and the other to impinge against, the filling mass in the shuttle to effect relative movement of said members, substantial exhaustion of the filling causing the members to move in unison, and means operated by or through such unison movement to control the operation of the loom.”
Upon this question it was found in the court below that the defendant’s feeler device, like that of the plaintiff,- consisted of two members movable relatively under certain circumstances and movable in unison under certain other circumstances, depending upon the filling on the carrier, and adapted to co-operate with the filling mass or its carrier at each alternate beat of the lay; that both members are moved forward against spring pressure,at each contact; that one of the members “presents a blunt surface at its point of contact,” and is no more adapted to penetrate or sink into the filling mass than the “impinging member” of the patent; that the other member is differently shaped at its point of contact, “and while rather better adapted than its fellow to ‘penetrate’ or ‘sink into’ the filling mass, it is still by no means so well adapted for that purpose as is the penetrating member of the patent”; that “there is a distinct difference in penetrative capacity” between the two members of the defendant’s device; and that “without such difference the mechanism would neither work so well, nor effect so much saving in filling.” In other words that it embodies the penetration idea of the patent.
It was further found that the mode or order of operation of the defendant’s feeler device in effecting a change was a mere reversal of that of the patent, and it was ruled that its mode of operation, to a substantial extent at least, was the equivalent of that of the patented device and that the defendant infringed.
In discussing the question of infringement it is important to keep in mind that the distinctive feature of the plaintiff’s patent is the penetrative or sinking-in character of the pointed member of its feeler. In Stimpson’s device of the prior art, the feeler members measure a predetermined distance from the surface of the yarn on the bobbin to the rim of the annular enlargement on the barrel of the bobbin, before calling a change. In the plaintiff’s device the feeler members, on calling a change, measure a predetermined distance from the surface of the yarn to the bobbin; and this is true, likewise, of the defendant’s device. The feeler members in all three devices move relatively to one another in reaching this predetermined measuring point. The new *551idea of the plaintiffs device consists in the patentee having laid hold of the principle of the relative density of the yarn and the bobbin and provided a feeler mechanism one member of which was adapted to penetrate the yarn and the other to impinge upon it and, when the yarn was sufficiently reduced, to measure the predetermined distance, which, upon being followed by a unison movement of the members, would call a change.
In determining whether the defendant’s device infringes that of the plaintiff it is necessary to ascertain whether the defendant makes use of the relative density of the yarn and the bobbin and employs as meajas a feeler, one member of which penetrates or sinks into the yarn, while the other impinges upon it, to effect the predetermined measurement necessary to calling a change.
In the plaintiff’s feeler device the penetrating or sinking-iu member is provided with two sharp points. Between them is a recess in which is located the impinging member with a blu;nt head. The end of the penetrating member normally projects rearwardly further than the end of the impinging member, the distance between the two being the predetermined distance or measurement to be made between the surface of the bobbin and the yarn when the change is called. Both are arranged to move relatively within a limited scope and then in unison.
In the defendant’s feeler there are two members located in the same, horizontal plane and but a comparatively short distance apart. Member No. 1 has a round, blunt end. This member, when the mechanism is at rest, projects rearwardly one-eighth of an inch further than No. 2. Member No. 2 has a square, blunt end measuring three fourths of an inch across it. Both members move relatively to one. another within a limited scope and then in unison, depending upon the amount of pressure exerted upon the respective members. Their relative movement is undoubtedly essential in two stages of the operation of the machine; one, in measuring the predetermined distance and effecting a change when the yarn on the bobbin is substantially depleted; and, two, in-immediately thereafter resetting the feelers in an inoperative position on the beat of the lay following replenishment of the shuttle. This resetting of the feelers in an inoperative position is due to the length of their movement forward by the stroke of the newly replenished bobbin, which causes the cam Ilf on the controller 7 to come in contact, with the governor 16 and raise the point 71 on the controller out of engagement in the slot 5 in feeler No. 2. In the conical stage of the yarn on the bobbin, intermediate the two above-named stages, this relative movement would seem to he nonessential.
The defendant contends that the relative movement in conjunction with the cam and the governor is essential and operative not only for resetting the feelers after replenishment, but also throughout the full yarn stage of the bobbin; that during that stage, it keeps the point 71 out of slot 5 in feeler No. 2 so that a change is not called.
The plaintiff, on the other hand, says that the cam and the governor perform no useful function whatever in the operation of the device except to reset the feeler members in inoperative position after replenishment ; that thereafter, even during the full stage of the bobbin, the *552rearwardly projecting point on member No. 1 sinks into the yarn on the bobbin o!n the beating up of the lay to such an extent over that of member No. 2 as to permit point 71 on the controller to remain in position on feeler No. 2 (without engaging slot 5), so that no change is called. In support of this contention the plaintiff introduced in evidence certain experiments made with the defendant’s device with three bobbins. After the experiments were had the bobbins were still well filled with yarn. The plaintiff’s expert testified that he used the three bobbins in four ways, one when the rounded feeler was used and the resetting cam H and governor 15 were operative, and that under those conditions the loom ran with no repeated engagement of the cam 11 and the governor 15. If this statement — that under those conditions the loom ran with no repeated engagement of the cam 11 and governor 15 — is correct, it would fully support the plaintiff’s contention that the cam and governor perform no useful function in the operation of the device after the feeler members, on replenishment of the bobbin, are reset in inoperative position. But the force of this testimony seems to us to be much weakened when the circumstances surrounding the operation of the device are considered and the conceded fact is taken into consideration that, after calling a change, immediately upon the beating up of the lay on replenishment of the bobbin the forward movement imparted to the feelers is sufficient to cause the cam 11 to engage the governor 15 and release the point 71 from the slot 5. It is the full stage of the filling on the bobbin that causes the prolonged stroke, and it is inconceivable that the movements following the first long one, which concededly makes the cam and the governor operative to reset the feelers, should not continue to be sufficiently long, during the well filled stage of the yarn, to cause the cam and governor to contact and keep the feelers inoperative.
Furthermore the defendant’s expert testified that he did not recall that his attention was directed to the cam as not touching the governor at the time the experiments were made; that, owing to the speed with which the lay beats up and the suddenness of the movements of the parts, a great deal of the feeling action is practically invisible; and that the cam might touch or pass over the governor sufficiently to prevent the controller rest from dropping without detection even when watched for.
The third use made of the bobbins in these experiments of the plaintiff was when the rounded finger was used and the resetting cam and governor were put out of commission by cutting a recess in the governor so that the cam could not engage it. Under these conditions the loom continued to run, and the explanation given by the plaintiff’s expert was that it was due to the sinking of the “pointed finger into the yarn,” which reduced the relative forward movement of the pointed finger No. 1 so that the point 71 on the controller rest was permitted to remain in position on feeler No. 2 and not engage in slot 5. In answer to this the defendant's expert, who was present at the time the experiments in question were had, testified that the tests were made only with three bobbins and that the loom was operated by one of the plaintiff’s skilled workmen; that the small number of bobbins thus *553tried and the brief extent of the trial were insufficient to demonstrate what would be the results of continuous running of the loom and feeler mechanism throughout a day, or of actual use under ordinary mill conditions and by ordinary mill operatives; that “the action of the feeler mechanism with this notch [recess in the governor] in use would be so uncertain and unreliable in any attempt at continuous running in a mill as to render the mechanism practically inoperative and unacceptable for practical use; that “this was shown by the fact that when, upon the conclusion of Mir. Browne’s tests [meaning the plaintiff’s tests] the loom was run again under the same conditions' for the purpose of enabling me to make further note of results, the feeler mechanism called the change prematurely in the case of two bobbins out of three;” and that “a feeler mechanism that will call replenishment at the full diameter stage is neither practicable nor acceptable.” These bobbins were put in evidence. The bobbins as to which the change was called prematurely were marked “2P” and “3P.” Bobbin 2P contains a large amount of filling and represents the stage spoken of as the intermediate or conical stage; bobbin 3P represents the well-filled or full diameter stage, and bobbin 1 the exhaustion or final stage.
The plaintiff introduced no evidence to refute the testimony that during the experiments with the cam and governor out of commission the feeler mechanism called a change prematurely on the bobbin which earned practically a full load of filling and did the same with the bobbin on which the filling was in the conical stage. It may be true that, during the full stage of the bobbin, the rounded feeler member No. 1, to some extent at least, sinks into the filling more than its companion member. But from the evidence presented, taken in connection with the elements that enter into the operation of the machine, we are unable to find that the depression caused by the rounded member at this stage plays any useful part in the operation of the device. It seems, rather, that the cam and governor, and not any depression caused by the feeler, are the operating means during this period which keep point 71 of the controller from engaging slot 5 and calling a change.
During the conical stage of the bobbin, the defendant’s rounded feeler does not depend upon penetration or depression to permit the device to remain inoperative, but upon the conical shape of the yarn, which permits the rearwardly projecting rounded feeler, when in contact with the yarn, to extend further rearwardly than member No. 2, so that the two members are moved forward in unison and no change is called.
In the final or operative stage of the device, when the yarn is substantially removed from the bobbin, there is no penetration by the defendant’s rounded member and there is no depression that can be useful in the operation of the device, the difference in diameters of the yarn on the bobbin at the points where the feelers contact with it permitting them to remain inoperative.
But the plaintiff contests this proposition and in answer thereto relies on experiments made with defendant’s feeler device, (1) when the rounded member had a square cap upon it, and (2) when the cap was removed; and because more yarn was removed from the bobbin when *554tésted with the rounded member after the same bobbin had been discharged on the test with the capped feeler, it contends that this shows that in this stage of the operation of the device the rounded feeler penetrates or sinks into the yarn and thus delays the relative action of that feeler before calling a change. In considering this contention, it is to be borne in mind that the rounded feeler with the square ended cap upon it, when contacting with the filling, would be moved forward sooner than when the cap was removed, due to the fact that the square inner corner of the cap would come in contact with the filling sooner than if the inner corner was round, the winding of the yarn on the bobbin being conical rather than cylindrical, and consequently would call a change sooner.
In view of these facts we are of the opinion that the defendant’s device in its mode of operation does not make use of the penetrative or sinking-in capacity of' the plaintiff’s penetrating member and does not infringe.
■ The decree of the District Court as to claims 1 and 9 is vacated, and the case is remanded to that court, with directions to enter a decree dismissing the bill as to said claims, with costs to the appellant.