Burroughs Adding Mach. Co. v. Felt & Tarrant Mfg. Co.

SANBORN, District Judge.

These patents are for improvements on adding machines, for which Eelt had taken out earlier patents expiring before this suit was begun. The chief question is whether the mechanism employed by defendant, different from that illustrated by the patents in suit, is an infringement. The improvements covered by the patents are many of them meritorious and valuable, and give the machine much greater flexibility and rapidity of action than the earlier forms. As to most of the improvements the Burroughs Company has sufficiently departed from the patented form to escape infringement, but as to one of them, particularly, there is question.

The principle of action of an adding or calculating machine is not difficult to understand, but there is much complication in the actual machine. Aside from the recording of results on paper, and restoring the machine from one operation so that another may be begun, known as “clearing the machine” or canceling the former operation, an adding machine depends on three points. These are:- (1) The use of the numeral-wheel and column-actuator; (2) the carrying mechanism, by which the turning of the units wheel beyond a full revolution carries one to the next wheel, and so on through a hundred, a thousand, ten thousand, etc.; and (3) the employment of stop mechanism to prevent the wheels from overrunning, and thus turning up a greater number than the one struck. ,

The Numeral-Wheel Mechanism. On the periphery of the numeral-wheel the digits 1 to 9 and the cipher are printed, and there are nine keys, marked 1 to 9. When a key is depressed the wheel will turn up to view the number which the key bears. Thus if the 6 key is touched the number 6 will come up, and if then the 2 in the-same bank or column is touched the number 8 will appear, each key having the ability to turn the wheel that many tenths of a revolution corresponding to its own value. Depressing the 3 key turns the wheel three-tenths around, the 8 four-fifths, etc. If the 9 is struck and then the 1 the wheel will complete one revolution, and turn up the cipher. All this is accomplished by the relation of the key bars to the column-actuator and the numeral wheel, as shown by the accompanying cut, being plaintiff’s Comparative Drawing No. 22, which needs only the number keys to give a clear idea of the latest form of calculating machine. With a bank of keys over each column-actuator the picture would be complete; but it may readily be understood that if the keys are so placed that they can depress the column-actuator, number 1 key being nearest the wheel and number 9 farthest away, the same depression of the latter may be made to depress the actuator 9 times as far as the. former, since 9 is near the fulcrum of the actuator and 1 is farthest away.

*863

*864Carrying Mechanism. When a column of figures is added with paper and pencil the tens, hundreds, etc., are carried from the units, tens, etc., columns. The same thing must be done mechanically in an adding machine by turning the tens wheel one-tenth of a revolution every time the units wheel makes a full revolution, two-tenths when it makes two revolutions, etc. In this way the values of all the operated keys are added in one common sum or total upon the series of numeral-wheels. ” When the units wheel completes a revolution the tens wheel moves forward one step, two steps when it completes two revolutions, and there is the same result in respect to the tens and hundreds, hundreds and thousands, etc., for as many banks of keys as the machine has. This carrying operation is brought about in the Felt construction by a ratchet'and pawl construction, and in defendant’s by a planetary gear. This is the point at which the main question of infringement comes up, the defendant claiming that it makes use of different means and operation.

The following figures will serve to illustrate plaintiff’s carrying devices. Figure 5 is taken from the Felt patent 366,945, and 5a from appellant’s brief.

The snail cam shown in different positions in both figures is attached to the hub of all the numeral-wheels except that of the highest order. When a lower order wheel completes a full revolution its cam, turning counter-clockwise, pushes out the pawl M', registering with the next higher wheel, by means of its shoulder m2, into the position shown by Figure 5a, and when the tail of the cam clears the rear end of the *865shoulder the spring operates on the pawl and rotates the higher order 'wheel one step, so that the figure 1 will show on the higher wheel and the cipher or some figure between that and 9 on the lower. Thus the carry is made, no matter what keys of the lower order wheel may be operated.

To illustrate this carrying operation suppose the 9 key in the units column is struck 12 times so that the sum will be 108. On the first stroke the units wheel turns nine-tenths of a revolution, and the tens wheel docs not move because its pawl M’ has not yet been actuated by the units wheel cam and the tens wheel pawl spring. But when the 9 is struck again the units wheel revolves another nine-tenths, and turns up an 8, and the carry has been made so as to move the tens wheel one step, and show 18 on the top of the two wheels. When the thircl stroke is made the units wheel will show 7 and another carry occurring the tens wheel will show 2. So with each successive stroke the units wheel goes nearly around, and the tens wheel advances one step until the eleventh, where it remains stationary because the units wheel was then starting as at the beginning with the cipher uppermost; hut when the twelfth stroke is made the units wheel cam and the tens pawl move the tens wheel one step, and the tens wheel (having now completed one revolution) has by its cam and the hundreds wheel pawl operated on the hundreds wheel to move it one step, so that the final result of all the strokes on tlie 9 key is 1 on the hundreds wheel, a cipher on the tens and an 8 on the units, or 108.

A novel improvement on this carrying device, covered by claim 29 of the first patent in suit, No. 762,520, presents the chief question in the case, and will be explained after adverting to the third point, the stop mechanism.

The Stop Mechanism. The third requisite of an adding or calculating machine is to provide automatic catches or stops to prevent overrun of a wheel. Without such stops the rapid striking of a key might rotate the wheel so far as to produce an unauthorized carry. Thus if 99 was simultaneously struck in the units and tens columns both wheels might overrun so as to make a carry in both the tens and hundreds wheels, and show a total of 110 instead of 99. The most important of these stops operate on the column-actuators, and will be referred to later.

Improvement in Carrying: Claims 29 and SO. Referring to the cuts 5 and 5a it may be seen that there would be a loss of carry if, when tlie carry is being made, both the lower and higher order wheels are moving. As described by plaintiff’s counsel:

“The thousands wheel might at one and the same time bo receiving both an impulse from a key of its own and an impulse from the carrying mechanism of tlie hundreds wheel; and if there were not some special means for preserving both impulses the lesser or carrying Impulse might be swallowed up in the greater or key-driven impulse that such thousands wheel was receiving.”

The patentee refers to this feature as follows:

“In order to prevent the loss or swallowing of the carrying movements in the other and generally larger movements of the numeral-wheels received from *866the impulse of the keys, which would occur if the carrying took place simultaneously with the key movements, I have devised means whereby the opera» tion of the carrying mechanism of the different denominations is caused to take place between the key-strokes in the same denominations and after such strokes have been completed and the numeral-wheels have moved in accordance therewith.”

In securing this Mr. Felt also obtained another important result highly useful and entirely new, called the duplex key action or fluidity of key movement. This was the capacity of the machine for striking a number of keys in different denominations at the same time, without striking them absolutely together. That is, the key-action may be independent, or irregular and overlapping, thus giving the machine an elastic or fluid action. It is possessed by both plaintiff’s and defendant’s machines alike, and has been called by plaintiff’s witnesses and counsel, “fluidity of keyboard action.” How this new result is obtained by both parties will now be described; the vital question being whether defendant, in using different means and a different operation, has really taken an equivalent.

In order to show how Felt attained these objects modified figures of the patent, as shown in defendant’s brief, are here reproduced. The mechanism is described by Mr. Felt with sufficient clearness:

“The sleeve 44 is also provided at opposite sides, as shown at Fig. 26, with two projections 71, each adapted to engage the under edge of a latch 73, pivoted on the cross-rod 48a. The latch is extended over and rests on the column-actuator of the denomination to which the carrying is to be done, and pref: erably on the pin 67 on the actuator, and a spring 99, already mentioned, draws the latch over onto the actuator. Normally the latch is out of engagement -with the projections 71; but when any key of the column to which the carrying is done is struck the actuator of that column moves down, so that the latch drops into position, where it must engage the first or nearest one of the projections 71 as soon as the sleeve begins to turn. The sleeve is thus arrested before it gets fairly started or has performed any function, and continues to be held by the latch until the column-actuator has fully completed its upstroke after being depressed by a key. When the actuator thus returns to its normal position and as it arrives at the same, it lifts the latch through the contact of the pin 67 with the end of the latch, so that the engagement with the sleeve is terminated, leaving the sleeve free to turn under the power of the carrying-spring 43 and through the mechanism already described to operate its numeral-wheel through a one-tenth revolution.” „ "

The inventor, having conceived his idea of improving his old machine by having the higher order actuators suspend the carry from the lower order wheels, and allow it to occur only after the driving stroke of the next higher actuator had been made, drew a number of specific and one broad claim in order to secure the legal benefit of his invention. Each of these claims contains five elements, the first four being substantially the same, and being old, but the fifth represents the new conception. Claims 29 and 30 follow:

“29. The combination of a series of denominational numeral-wbeels, a series of column-actuators operating said wheels, a series of keys for each actuator, and a carrying mechanism for each wheel, and means whereby the several carrying mechanisms may be temporarily controlled by the next higher actuators.

*867

”30. The combination of a series of denominational numeral-wheels, a series of column-actuators operating said wheels, a series of keys for each actuator, and a carrying mechanism for each wheel, and means whereby the actuator of a higher denomination may delay the carrying from a lower denomination until it has completed any movement imparted to it by tbe keys.”

Claim 30 specifically represents the actual invention described in the specification, and claim 29 may be properly construed to cover all variations and equivalents of the five elements going to make up the real invention.

*868The mechanism of the delaying latch of the Felt machine has been already described, in connection with the patent drawings. The “inventive concept,” or underlying and fundamental principle of the invention, was thus previously described in the specifications. After referring to his own prior art machines, Mr. Felt proceeds:

“A third objection to this class of calculators has been that they are liable to add incorrectly if the operator strikes two or more keys of different denominations simultaneously, because under such circumstances the carrying is often lost in the movement imparted directed to the register-wheels by the keys, and by reason of this fact it has not heretofore been possible to operate the machines by striking a plurality of keys simultaneously. Obviously a machine in which correct results can be obtained by striking two or more keys in different denominational series at a time will enable' the operator to increase the speed of his calculations very materially, and by my present invention I not only obviate all danger of miscalculations from this cause, but produce a machine well adapted to permit the habitual striking simultaneously of a plural* ity of keys.”
“In order to prevent the loss or swallowing of the carrying movements in the other and generally larger movements of the numeral-wheels received from the impulses of the keys, which would occur if the carrying took place simultaneously with the key movements, I have devised means whereby the operation of the carrying mechanism of the different denominations is caused to take place between the key-strokes in the same denominations and after such strokes have been completed and the numeral-wheels have moved in accordance therewith. This feature of the invention will now be set forth.”

And the patentee then proceeded to describe the mechanics of the invention in the words quoted on an earlier page.

[ 1 ] Since defendant does not use the delaying latch the question is thus presented as to the effect of describing the inventive concept or idea, claiming it as thus described, as in counts 28 and 30, and then attempting to broaden it by a more general count, as in claim 29. This situation has often come before the courts, and the governing rule well expressed by this court, and by Judge Colt, speaking for the Court of Appeals for the First Circuit. In Mossberg v. Nutter, 135 Fed. 95, 99, 68 C. C. A. 257, 261, Judge Colt said:

“In approaching a patent, we are to look primarily at the thing which the inventor conceived and described in his patent, and the claims are to be interpreted with this particular thing ever before our eyes. In confining our attention too exclusively to a critical examination of the claims, we are apt to look at them as separate and independent entities, and to lose sight of the important consideration that the real invention is to be found in the specification and drawings, and that the language of the claims is to be construed in the light of what is there shown and described.”

In this court, in State Bank of Chicago v. Hillman, 180 Fed. 732, 104 C. C. A. 98, Judge Grosscup said:

“The question of law presented, then, is this: Can the patentee rightfully include in his claims something that does not emerge from the description? Can a patentee describe something to the world in his letters patent that means just that thing or its equivalents and nothing else, and, having claimed that, claim in addition something not thus described and not its equivalents? We think not. The description is required to set forth the invention in such full, clear, concise, and exact terms as to enable any person, skilled in the art to which it appertains, or with which it is most nearly connected, to make and *869■use the same; and the claim is to enable the public to know the bounds and scope of the invention Thus disclosed’; but ‘any claim which is broader than the described invention is void, even where that invention is valuable, and could have supported a valuable claim.’ Walker on Patents (4th Ed.) § 177, citing Edison v. American Mutoscope Co., 114 Fed. 934, 52 C. C. A. 546.”
“There is nothing in Winans v. Denmead, 15 How. 830, 14 L. Ed. 717, or the Paper Bag Patent; Case, 210 U. S. 405, 28 Sup. Ct. 748, 52 L. Ed. 1122, brought to our attention since the argument, nor In any of the rules of law cited (‘that the claims of every patent should be construed, if possible, to cover and protect the actual invention made by the patentee, and should not be restricted to the particular form of device disclosed in his patent, if other forms may embody it.’ or that ‘the patentee’s claim is the “measure of his invention,” ’ or that ‘where the claims of a patent are clear and unambiguous, there is no room for construction’) that contravenes what has just been said; for what is sa.id in both of these cases, and in all of these rules, is based on the fact that the inventive concept is disclosed in the description, whatever may have been the mechanical form that such concept subsequently took. Certainly it was not intended by these cases or these rules that an inventive concept, that is separate and apart from the one embodied in tile description, should become a part of the patent simply by being included in the claims.”

The Hillman Case was approved by this court in Stevens v. Mitchell, 220 Fed. 455, 136 C. C. A. 283, and the Supreme Court announced the same doctrine in Snow v. Lake Shore & M. S. R. Co., 121 U. S. 617, 7 Sup. Ct. 1343, 30 L. Ed. 1004. To the same effect are Celluloid Co. v. Arlington Co., 52 Fed. 740, 3 C. C. A. 269; Whitaker Cement Co. v. Huntington Dry Pulverizer Co., 95 Fed. 471, 37 C. C. A. 151, and Jewell Filter Co. v. Jackson, 140 Fed. 340, 72 C. C. A. 304.

Thus the gist or spirit of the invention was to cause the carry to occur only between actuator strokes, and to be held up and delayed until the next higher actuator has come to rest after an impulse of-its own. This is the “law of the machine” and the prime object of the invention; but the operation is entirely foreign to defendant’s device, in which the carry is not delayed, but occurs as well during a higher actuator movement as at any other stage.

Plaintiff’s position is that claim 29 covers any and all means by which the carrying mechanism is temporarily controlled in any degree by the next higher actuator; and that unless this meaning be given to claim 29, the latter cannot be distinguished from claim 30. In support of this plaintiff’s counsel appeal to the well-understood theory of claim differentiation described by Judge Baker in Damson Consolidated Store Service Co. v. Hillman, 123 Fed. 416, 59 C. C. A. 510, to the effect that separate claims are not to he construed as identical unless fairly unavoidable. “To construe claim 8 as being the same as either claim 10 or claim 11 would be to declare claim 8 void as a duplication. Such a course should not be pursued unless it cannot fairly be avoided.” Baker, C. J., in Kennicott Co. v. Holt Ice & Cold Storage Co., 230 Fed. 157, 144 C. C. A. 455.

[2] The rule is of course well established that the courts will always endeavor to distinguish the several claims of a patent, one-from another, and that where a patent contains two similar claims the differT ence in the wording of the two claims will be given effect if possible, and the more broadly stated claim will be construed more broadly than the narrowly stated claim, if it can be done within the limits of the *870patentee’s invention as described in his specification and illustrated in his drawings.

But this rule of construction of patent claims is entirely subordinate to the fundamental and controlling rule that a patentee’s broadest claim can be no broader than his actual invention, no matter how it may be expressed or what other claims his patent may contain. When a patentee has fully and clearly described his actual invention in tire specification and drawings of.his patent, and has fully covered that invention by the broadest claim to monopoly which the law will allow him, he cannot then, by merely including ,in his patent a more broadly or more vaguely stated claim, cover and monopolize something more than and different from his real invention. Now, while it is a very common thing for the courts to construe a claim of a patent by reference to some other claim of the same patent, and to apply the rule of construction to which plaintiff refers, it is equally common for tire courts to wholly disregard and ignore such rule of construction whenever it comes into conflict with the fundamental and controlling rule that a patentee’s claim can be no broader than his actual invention, and they have not hesitated to hold a claim invalid where it could not be restricted by construction to the patentee’s actual invention as disclosed in the specification and drawings of his patent. ,

That is what this court did in the Hillman Case, and what was done .by the Circuit Court of Appeals of the Eighth Circuit in tire Jewell Filter Case.

It is also to be noted, however, that there are differences between claims 29 and 30, sufficient to so distinguish them, and to hold that both are valid. Claim 29 counts on control of the “carrying mechanism” by the “next higher actuator,” while claim 30 counts on delay of the “carrying” by the “actuator of a higher denomination.” It will not do, in this difficult and intricate matter, to make the absolute statement that the fifth element of both claims is identical.

[3] Defendant’s Mechanism. It has been stated in the opinion that defendant’s mechanism, which it is now necessary to describe, does not include means for a delayed carry, or any equivalent mechanism. The object of both constructions is to enable one complete revolution of a lower order numeral-wheel, say the units, to transmit to the next higher or tens wheel a one-step carry, or one-tenth revolution. Each numeral-wheel (except the first and last) has a dual function. Each is a lower order wheel to its next higher neighbor, and a higher order wheel to its next lower. Each must transmit the carrying impulse to the next and receive it from the lower one, and in so doing neither must interfere with the other.

In order to do these things rapidly and correctly all the mechanism must be connected like a watch in the most exact and perfect way, and within a very small space. Between the units and tens wheels (also the tens and hundreds, etc.) there is placed a set of small cogwheels or pinions which wind up a spring in'order to make the one-step carry. There is also a cam, an escapement, lever, catch, etc. So when the units wheel turns from 1 to 9 it winds up the spring ready *871to be tripped when the wheel reaches the cipher position and to turn the tens wheel one step. Some device, also, must be arranged to prevent the one step carry being smothered or swallowed up by the larger movement of the next lower wheel, as described by Mr. Felt in the language quoted. Felt does this by a ratchet and pawl construction and the delaying latch 73 controlled by the next higher actuator, and defendant by the transmission mechanism referred to and the planetary or differential gearing shown in the accompanjdng cuts.

HIGHER ORDER WHEEL AND ACTUATOR—DEFENDANT'S CARRYING MECHANISM.

*872DEPENDANT’S DEVICE.

Higher Order .Actuator, Backstop, Driving Pinion, Pinwheel and Overrunning Pawl.

The foregoing drawings represent a planetary gear and some additional devices such as the higher order actuator, driving pinion, pinwheel, backstop and pawl to prevent overrunning, all attached to a higher order numeral-wheel, the actuator shown being the “next higher” one referred to in claim 29. The sun gear is normally stationary, is not turned by the movement of the actuator, but only by the carrying spring attached to the transmission devices, and wound up by the movement of the lower order wheel not shown.

It may be difficult to describe the characteristic motion of a planet gearing, but it is just like the differential of an automobile. Its fundamental principle is that a cogwheel turned by two other cogwheels moving simultaneously will take on the sum of both movements. The *873planetary movement has long been well known, and is an accelerated motion due to force transmitted from two points into one. That is, when the driving pinion alone moves the planet gear turns. When the sun gear alone moves the planet gear also turns. When both driving and sun gears move together the planet gear takes on a more rapid motion as the result of both, and the planet gear is the last step in the train which makes the carry by turning the higher wheel one step.

Now as the lower order wheel through the transmission and spring turns the sun gear, and the higher order actuator turns the driving pinion which in turn drives the higher order numeral-wheel, it is obvious that the difficulty described by Mr. Felt as the loss of carry by the swallowing up of one movement by the other cannot occur, because the planet gear will transmit to the higher wheel the motion it receives from the sun gear just as well when the numeral-wheel is turning from a key impulse given by the actuator as it will when the actuator is at rest. When.the actuator is making its rising or driving stroke it will be seen that the planet gear is being carried idly around the sun wheel by the internal gear N. If at this time a carry comes over from the lower wheel, by the spring turning the sun gear contra-clockwise, this motion of the sun gear will be transmitted through the planet wheel by the internal gear N to the numeral-wheel N1, and move it one step in the opposite or clockwise direction.

Carry-Control by Next Higher Actuator. Claim 29 counts on “means whereby the several carrying mechanisms may be temporarily controlled by the next higher adulators.” So a further question that may be thought material is whether defendant’s carrying device is within this claim. And since claim 29 is clearly valid because it was new to control the carrying operation by the next higher actuator, it will conduce to a better understanding of the case to inquire whether or to what extent defendant uses the temporary control of claim 29, and if so whether defendant’s carrying device is an equivalent of plaintiff’s.

The notion of control of carry in the Burroughs’ device is based upon a fundamental principle applying to all differential gearing, which is that a pinion or cogwheel placed between two other cogwheels cannot turn either one unless the other is either held stationary or driven in the proper direction. Thus the driving pinion I1 in the first cut must be held stationary or be moving forward at the instant that the planet wheel is turning the numeral-wheel N1 by the influence of the sun gear M". The second cut shows how this driving pinion is normally held from backward motion by the backstop pawl K when the column-actuator P'x is making its downward movement, also how the backstop will rido over the pinwheel when moved clockwise by the upward or driving stroke of the actuator. Those who understand automobile operation will know why the driving pinion cannot be allowed to turn backward while the planet gear is being operated by the sun gear. When a car is coasting down hill, in gear, but with the spark turned off, the movement of the hind wheels is propelling the engine through the differential and main shaft. During this operation, if one of the hind wheels should leave the ground while rounding a curve *874the engine would stop, because it is absolutely necessary that both wheels shall be on the road, or the differential pinion, corresponding to the planet gear, will idle about the large gear wheels connected with the rear axle shafts in the differential, and exercise no driving power on the main transmission shaft. The same thing happens if one wheel of the car is lifted from the ground and turned by hand. This will operate the engine if sufficient force be used, but if both wheels are off the engine will not turn over. This illustrates the point that the driving gear must either be stationary or held from backward movement while a carry is being made.

Bearing this in mind, the cut may be explained by saying that the actuator shown is the “next higher” one in mesh with a gear wheel not shown, and which is supposed to be rising and making its driving stroke at the exact instant when a carry is coming over from fhe lower order'wheel. In the cut, however, the parts are shown in their normal, inactive position. The pinwheel and driving gear are held from backward movement by the backstop K pressed against the pinwheel by its spring, and from forward movement by the actuator itself. The driving gear and pinwheel are rigidly connected.

Suppose now the actuator is depressed by its own key-movement. The latch J will be pulled by its spring downward away from its registry with the pinwheel, and thus allow forward movement of the pinwheel and driving gear, but which are still held from backward movement by the backstop K. So if a carry is coming over during the descending or idle stroke of the actuator, it will be properly made, because the driving gear is “on the ground,” held immobile by its backstop. But when the actuator begins its upward or driving stroke the backstop, by the turning forward of the pinwheel, is made to slide over the cogs or pins and thus prevented from holding back the driving gear. Of course such gear cannot be held back and go forward under the lead of the actuator at one and the same time. It is at this instant of a rising actuator that the'latter is said to temporarily control a carry if one then happens to be coming over.

By referring to the diagram the operation of a driving or rising stroke may be clearing understood. The actuator forces the backstop out of commission and itself steadies and holds firm the pinwheel and attached driving pinion by pushing them forward. It is like the pilot assuming the wheel on an incoming steamship; he takes control from the helmsman and assumes it himself. So it is plain that it is accurate to say that when both K and / are out of commission during the driving stroke the only thing which can steady the driving gear is the actuator, and if at this instant a carry is made the actuator, by pushing the driving gear forward, prevents it from going backward, and so in a sense “temporarily controls the carrying mechanism.” But this is manifestly quite a different control from that positive prevention of a stroke which occurs when plaintiff’s latch 73 is made to absolutely tie up the carrying mechanism until the rising actuator stroke is completed.

We have not lost sight of defendant’s argument that the Burroughs’ higher order actuators do not prevent.loss of carry, but in that con*875iieciion we are satisfied, even if the argument is sound, that these actuators do “temporarily control” the carry in the sense that they prevent the driving pinion from going backward by the operation of driving it forward.

The case is thus brought squarely within the rule of the Hillman Case that when an inventor has described and claimed his real invention he cannot add to it by adding a broader claim. Felt’s fundamental idea was the delayed carry temporarily governed by the next higher actuator descending and putting in operation the delaying latch 73. This he described most clearly, and. fully covered in claims 22 to 28 and 30. He could not by another claim cover a distinct inventive idea, not described or even intimated in his specifications as his conception, namely, the concept used by defendant, of the nondelayed, simultaneous afrry, even though both operations are individualized to the next higher actuator. The inventive notion of a nondelayed simultaneous carry is contained in the. prior art patents of Cook, No. 503,946 and Webster, 484,887, but as individualized to the next higher actuator it is not actually used by either one. Felt was the first to bring into actual use this individualized control, with a highly beneficial result in securing the duplex key-action and fluidity of key movement above referred to. But this individualization of the control was expressly confined by him to control through delaying mechanism, as an improvement on his own older machine, in which simultaneous striking of the keys, either in unison or overlapping, was impossible without loss of carry. That was his invention: there is nothing even to suggest that the actual invention was broader than the invention as described by him in his patent, unless it be the language of claim 29. And this claim, fairly construed, both independently and as compared with tho language of the other claims, does not support the broader construction. While the inventor must have all due credit, if, as we hold, his real invention is not used by the defendant, there can be no infringement.

Stop Device Claims, 1-10 and 16-21, No. 762,520. At the reargument it was thought that aside from claim 29 of the first patent in suit there was no infringement, on the theory that both parties were equally entitled to improve upon the old Felt machine, and that each had done so in its own way. As these stop mechanisms are quite complicated it will be enough to compare the most limited of the respective claims of the Felt and Horton patents.

Felt claim 3:

“Tlie combination with the keys and the column-actuator operated by the keys and provided with coarse spaced teeth, projections or shoulders, ol! two hinged levers arranged alongside of the actuator, one adapted to be depressed by the odd keys and tho other by the even keys, and two pivoted stop devices adapted to be swung each by one of the levers, and thereby to be forced into engagement with said shoulders, the acting portions of said devices being located in different planes so that one may engage later than the other.”

Horton claim 10:

“In a machine of the character described the combination of an adding wheel; an actuator lover therefor having two confronting series of shoulders, those of one series staggered with relation to those of the other; a swinging *876stop adapted to engage said shoulders; a bar connected to said stop and having two series of cams, those of one series alternating with those of the other and oppositely inclined relative thereto; a series of depressible keys to act upon the lever different distances from its pivot and having studs to act upon the cams of said bar, the odd keys acting upon one series of said cams to swing the stop one way, and the even keys acting upon the other series of cams to swing the stop the other way; and a spring applied to the stop to centralize it.”

Comparing the two devices one has five stop shoulders and' two stops, and the other nine stop shoulders and one stop. One has two hinged levers or catches and the other a hinged bar. Defendant’s device is the more simple and easy to understand. The only invention shown in either device is in decreasing the number of shoulders and increasing their size, and in this respect defendant employs different means and operation. Each party is entitled to its own form. The devices differ in principle, operation, mechanical construction and arrangement of parts, and there is no infringement.

Other Stop Devices, Claims 11-15, 38, 39, 53, No. •762,520. The devices referred to are the two pivoted stop devices referred to under the pi-eceding division for stopping the downward movement of the actuator, a rod across the machine for stopping its upward movement, and a spring latch timed to prevent overrun of the numeral-wheels after the rod had arrested the upstrokes. The arrangement of stops is simply a matter of mechanical skill, and there could be no invention except in the precise form employed. The prior art abounds in all kinds of stop devices, and there is a mere change of form from the first Felt patent, now expired. Defendant does not use just the same combination, and does not infringe. Claims relating to other stops are 40, 41, 63, 70, 71 and 73, but they all stand on similar grounds, being simply specific arrangements requiring mechanical skill only. The claims are narrowly valid but not infringed.

Canceling or Zero Claims, Patents No. 767,107, and 960,528. These claims relate to the process of “clearing the machine” by swinging a lever forward and back, known as canceling or zeroizing, by which all the numeral-wheels are lined up with the cipher of each uppermost. The two devices are shown in the first diagram on an earlier page, being Comparative Drawing No. 22. All the column-actuators move in plaintiff’s machine, but only the units actuator in defendant’s, and the mechanism is different, as well as mode of operation in other respects. Canceling mechanism is old, and is found1 in Felt’s prior art machines. No infringement of these patents is shown.

Claims 7, 8, and 16, No. 762,521. These claims cover the carrying springs described in the discussion on claim 29, in combination with their numeral-wheels, which are being constantly revolved against the tension °of the spring, and restrained by it. This patent was not mentioned on the oral reargument. It is simply an ingenious triplex coiled spring, required by the necessity of storing carrying power. Defendant does not use it or anything like it.

Claim 1q, No 762,521. This claim relates to plaintiff’s numeral-wheel with a celluloid face. Defendant’s wheel is quite different. The celluloid face element' is not insisted on by plaintiff. ■ "

*877The decree appealed from is reversed, with direction to enter a decree sustaining all claims in suit but finding them not infringed, with costs.

Reversed.