The action in the court below was to restrain the infringement of Letters Patent No. 328,099, dated October 13, 1885, issued to George Allen for a device for converting motion in oil pumping apparatus.

The substantive portions of the Letters Patent and drawings are as follows:

*608“My invention relates to an improvement in eonvertin¿ motion In oil-pumping apparatus. In Letters Patent No. 313,907, granted me on March 17, 1885, I showed and described an upright rotary shaft provided with a crank and a disk loosely mounted on the wrist-pin of the crank, said disk being adapted to the attachment of pump-actuating rods leading in any desired direction therefrom. In a second application for Letters Patent allowed me on August 10, 1885, I showed and described a modified form of the construction shown in my former patent, in which the pump-actuating rods were loosely secured directly on the wrist-pins of the crank.

“While the above constructions have proved eminently practicable in use, and well adapted to satisfy the needs of a great majority of the cases where a greater or lesser number of wells are to be pumped by a single-actuating shaft, yet there are instances in which it might be desirable to modify the means for attaching the pump-actuating rods to the driving-shaft.

“The object of my present invention is to provide means for attaching the pump-actuating rods to the driving-shaft at one or several points of elevation and in groups, leading in opposite general directions from each of two points of elevation, or in any desired direction from the same point of elevation, as may be found most expedient.

“With these ends in view my invention consists in certain features of construction and combinations of parts, as will be hereinafter described, and pointed out in the claims.

“In the accompanying drawings, Fig. 1 is a view of a portion of the pumping apparatus embodying my invention, showing rods leading in any direction from the same point' on the driving-shaft, and Figure 2 is a similar view showing two groups of rods, the groups leading from different points of elevation on the drive-shaft and in generally opposite directions therefrom.

“A represents an upright shaft journaled in suitable bearings in a supporting-frame, and driven by bevel-gear Bb, connecting it with the engine-shaft 0.

“To the shaft A are rigidly secured one or more eccentric disks or wheels, D, provided with the eccentric straps or rings E, loosely mounted thereon. To the straps or rings E the ends of the pump-actuating rods F are secured, the rods leading in any direction therefrom, as shown in Figure 1, or one group of rods, F, leading to the right from one of the eccentric disks, and another leading to the left from .another of the eccentric disks or wheels, as shown in Figure 2. The latter construction admits of a more perfect balancing of the pumps, as those on the right will be all descending while those' on the left are lifting, or vice versa. This may be quite satisfactorily accomplished, however, when but a single eccentric is employed by using a little care in distributing the rods about the ring E. •

“Instead of one or two eccentrics, three or more may be used, and the capabilities of the pump be thereby increased in respect to the number of pump-actuating rods having the same or different lengths of stroke.

“Having fully described my invention, what I claim as new, and desire to secure by Letters Patent, is—

1. The combination, with an upright shaft and means for rotating it, of an eccentric rigidly secured on the shaft, á strap or ring mounted on the eccentric, and pump-actuating rods attached to the strap or ring, substantially as set forth.

2. The combination, with an upright shaft and means for rotating it, of one or more eccentric disks or wheels secured on the shaft, straps or rings loosely mounted on the eccentrics, and pump-actuating rods secured to the straps or rings, substantially as set forth.”

The patent thus issued was only a step in advance of Allen’s previous inventions. On March 17th, 1885, he took out Letters Patent No. 313,907, embodying all the features of the patent in suit, except that instead of an eccentric rigidly secured on the shaft, with straps and rings loosely mounted on the eccentric, he used a wheel or disk loosely mounted upon the wrist-pin of a crank se*609cured to the upper end of the shaft. The substantive portion of this previous patent is as follows:

“My invention further consists in a wheel or disk adapted to the attachment of pump-actuating rods leading in different directions, and mounted on the wrist-pin of a crank secured to the end of a vertical shaft.

“My invention further consists in certain features of construction and combinations of parts, as will be hereinafter described, and pointed out in the claims.

“In the accompanying drawings, Figure 1 is a plan view of the wheel or disk, showing pump-rods attached thereto, and Figure 2 is a vertical section through the shaft and wheel or disk.

"A represents a vertical shaft suitably mounted, and driven by bevel-gear a and B, the latter being secured on the engine-shaft b. The upper end of the shaft A is provided with a crank, O, having the wrist-pin c.

“D is a wheel or disk loosely mounted on the wrist-pin c, and provided with a series of radial slots, d, which extend from the upper side of the rim nearly but not auite to the lower side.

“The wheel D consists, preferably, of a hub, d', a rim, D', and a series of spokes, ds connecting the hub and rim, but may be solid between the hub and rim, if found more convenient, my improvement, so far as the wheel or disk is concerned, consisting, essentially, in the series of slots or recesses formed in the upper side of the rim, as described.

“The rods E, which lead from the wheel D to the pumps, and which, when reciprocated, actuate the pumps, are adapted to fit. loosely in the slots d, and are provided with enlarged heads e, which prevent the rods from drawing through the slots. The rods are placed in and removed from the slots at pleasure, the only lock against vertical displacement being their weight. They are free to assume different angles in a vertical plane, and are adjusted in the most advantageous positions in a horizontal plane hy placing them in their appropriate slots. As the shaft A rotates, the wheel I) is caused to describo a circular path in space, and the rods, no matter in what direction they lead, are thereby reciprocated. Moreover, as the wheel D is free to rotate on its axis, the rods E adjust themselves to each new position which they are required to assume during the rotation of the shaft A.

“Instead of providing the ends of the rods with enlarged heads or nuts, as shown at e, they may be provided with yokes, as represented at e', the yokes being adapted to embrace one of the elevated sections between two successive slots.

“The apparatus as thus constructed is particularly adapted to use in positions where a small number of wells are to be operated, and is capable of being set up and run at a moderate expense; but I do not wish to be understood as limiting myself to any restricted use, as it is capable of being operated upon either a grand or limited scale.

“I am aware that it is not new to provide means for connecting pump-rods leading in any desired direction, and do not therefore claim the same, broadly; but,

“Having fully described my invention, what I claim as new, and desire to secure by Betters Patent, is — ■

1. In an apparatus for pumping oil-wells, the combination, with an upright shaft and a crank secured to its upper end, of a wheel or disk loosely mounted on the wrist-pin for attaching pump-actuating rods, substantially as set forth.

2. In apparatus for pumping oil-wells, the combination, with a crank secured to the upper end of an upright shaft, of a wheel or disk loosely mounted on the wrist-pin, said wheel or disk being adapted to the attachment of pump-actuating rods in any direction, substantially as set forth.

3. In apparatus for pumping oil-wells; the combination, with a crank secured to the upper end of an upright shaft, of a w'heel or disk loosely mounted on the wrist-pin and provided with a series of radial slots for the attachment of pump-actuating rods in any desired direction, substantially as set forth.

b. A wheel or disk for attaching- pump-actuating rods, having a series of radial slots formed in the upper side of its rim adapted to receive the *610of pump-actuating rods from anv desired direction, substantially as set forth.”'

Later, on the 8th of September, 1885, Allen took out other Letters Patent, No. 326,008, which modified the former device by dispensing with the wheel or disk, and in its stead pivoting the actuating-rods directly to the wrist-pin, so that the revolution of the crank and pin would reciprocate the rods leading in all directions. The drawing of this patent clearly illustrates its working and is as follows:

*611

*612Whatever invention is contained in the Letters Patent in suit must be found,- therefore, in the substitution of the eccentric, with its strap or ring loosely mounted (as a base upon which to fasten the actuating-rods), for the disk fastening, or the pivotal fastening, previously employed. A correct appreciation of the scope and purpose of this change can only be obtained by a review of the actual conditions confronting the inventor.

The general purpose of oil pumping devices like those of the Allen patents is to convey force from a central power throughout a field of oil wells, so that the pump of each well may be operated from the central power. These fields, so far as they relate to pumping apparatus, present conditions unlike those to which pumping apparatus has otherwise been applied. The wells over a given field of territory, subject to the central power, are frequently numerous; they are not evenly distributed radially about the central power; they are not equidistant from the central power; their surfaces, where the pumps must be applied, are not, owing to the undulations of the ground, upon the same level, and they must be pumped continuously, for if an idle day or two intervene, it may take weeks to bring them back to their normal productivity. Our attention has been called to no other use of pumps where such conditions of inequality exist — conditions that make their reciprocal balancing especially difficult.

The oil pump, too, operates differently from other pumps, and from other arts in which shafts actuated by an eccentric have been employed. In pumps in other fields, and in other arts employing shafts and eccentrics, there is need for power exerting a thrust as well as a pull, but in the oil pump the downward motion is brought about, not by a thrust from the ppwer station, but by the weight of the pump valves, sucker rods, and column of oil resting upon the valves. In the oil field the power is applied solely to raise the pumps.

In the development*of Allen’s ideas the disk pump, chronologically, came first; then the pump with the pivotal fastenings on the wrist-pin; then the patent under consideration. But in the natural order of evolution, that has given to the patent under consideration its effective features, the pump with the pivotal fastenings at the wrist-pin is primary to- the disk pump. We will consider the Allen inventions, therefore, in the latter order.

It is manifest, upon an examination of the drawings of Letters Patent No. 326,008 (the pump with pivotal fastenings at the wrist-pin), that if but two actuating-rods, leading in different directions, were attached to the wrist-pin, the strain upon the wrist-pin, as it moved through the circle, would be, in each of the arcs of the circle, necessarily unequal. The crank is, in effect, the short arm of the lever, its attachment at the shaft being the fulcrum. The power being constant, the rapidity of motion, as the point of strain is approached, will change; or the rapidity of motion being constant, the power must change. In case the rapidity of motion is constant, the strain upon the fulcrum or crank at its connection with the shaft will always be greatest when the crank is most nearly at right angles -with a line drawn from the fulcrum to the pump; in other words, there would, on *613(¡very revolution of the crank, be two dead points, to overcome which, at a revolution of constant rapidity, the driving force must be adjusted. Between these dead points the driving force would be in excess of the requirements.

Of course the volume of power can not be modified automatically; so that what would occur in the case of two rods thus oppositely adjusted would be constant changes of rapidity of motion between the several points of heaviest strain, growing more rapid as the point of heaviest strain is left, until the point of lightest strain is reached, and then less rapid, until the next point of heaviest strain is reached. The practical results of such unequal distribution of strain would be numerous, the most serious and obvious of which would be the racking of the machine and the waste of power.' 'Changes of strain so abrupt and numerous would, in an apparatus like an oil well, have the effect of rapidly shaking it to pieces. This is accentuated by the fact that the driving shaft is an upright one, and that the crank is located much above the ground, so that recurrences of heavy strain, first upon one side of the shaft, and then upon the other, would tend, in addition to racking the apparatus, to pull it out of place. An equal balancing of pull in all directions tends to throw the whole weight to the base of the upright shaft; but an excessive pull in one direction will throw the strain of the -weight transversely to the shaft.

Now, if instead of two actúa ting-rods, reaching in opposite directions, there were many, reaching in all directions, all attached to a crank, as in patent No. 326,008, the changes of strain would still take place, but would, with the number of rods, be correspondingly subdivided; so that the changes, increasing in number, would decrease in intensity. By an equal distribution of many wells around the central power, the changes of point of strain might be1 so* modified as to be practically eliminated.

But an oil field is not adapted to such an equal distribution of actuating-rods. The wells can not be located with reference to the central power. They can not be distributed radially at equal angles around the power shaft — the fulcrum to the crank. With a fixed crank, and comparatively rigid fastenings, like that used in patent No. 326,008, and an impossibility of an uniform distribution of the wells, with reference to the fulcrum, there can be no adequate counterbalancing of the strains, unless the crank — the short arm of the lever— has some degree of elasticity.

The patent in suit provides practically a short arm to the lever approximately elastic. The loosely mounted ring to which the actuating-rods are attached shifts back and forth, so that there is no inexorable massing of the weights at the dead points. It may be that in practical operation the pull of two or more of the pumps may be found to be approaching, at the same time, the angle of greatest strain, and, if the fastening to the lever were rigid, would be carried en masse beyond that strain; but the susceptibility of the ring to retarded motion, during the period that this unusual weight is to be lifted around the point of heaviest strain, decreasing the motion, decreases the strain; for the force necessary to lift is in the inverse order of rapidity of *614motion. A lighter weight following, the ring would swing forward, thus increasing the motion, and compensating for the previous loss of motion. In other words, the loose mounting of the ring gives to the short arm of the lever a play, such as it would not have, if the rods were attached rigidly to the lever. The varying weights are, by this elasticity, or play, distributed, so as to fall more equally throughout the whole circle of the disk’s revolution..

The disk, whether it be a wheel revolving upon a wrist-pin of a crank at the wheel’s center, or an eccentric, is, with reference to the motive shaft, a mere crank. The strain upon an eccentric is always upon what is called its fat side. The disk in the patent under consideration could, without impairing the motion of the ring, be cut away, so that nothing would remain but the lean side and a long arm, the lean side being a guide way only, the long arm bearing the strain. The loose mounting of a ring upon such an eccentric, or upon a disk wheel mounted upon the wrist-pin of a crank, is, in effect, as if the disk or crank (for mechanically they are the same) were made of an elastic material, bending to increasing weight so that strain is partially overcome by correspondingly decreased motion, and then recovering, by increasing motion, its normal form under diminishing weight.

It is evident that in the balancing of many wells, each one of which contributes a varying weight, and two or more of which may, at times, mass their weights upon a crank or disk of that character, there would be less abruptness of strain than in a rigid crank whose revolutions were inexorably constant. A crank so elastic would adjust itself as nearly as is conceivable to the varying weight, and thus, by automatic counterbalancing, eliminate, partially at least, the changes of strain.

What has been said respecting the elasticity of the patent in suit applies, also, to the invention embodied in Letters Patent No. 813,907 (the disk wheel pump), and differentiates them both from the old style of pull wheel called to our attention. There are in these pull wheels none of the elements of elasticity. They each have advantageous features of their own; but none are adapted to overcome the massing of strain. In wells equally distributed about the central power there could be such adjustment ás to balance the strains, but in the actual fields, where such equal distribution is impossible, the pull wheel is inferior to the patent under consideration just to the extent that the disk of the patent in suit yields, by this practical elasticity, to the demands of changing weights.

Our attention has been called, also, to previous devices in allied arts, to show that eccentrics have been used for purposes somewhat similar. The Corliss engine is perhaps the best single illustration of the previous art. An examination of the Corliss patents shows that the eccentric disk has a rigid arm extending to one of the steam cylinders. Connected with this disk, by pivot joints,, are a number of arms, corresponding in length to the rigid arm, some of which are connected to the pistons of steam cylinders, and others to the pistons of pump cylinders. In- the operation of the engine the steam admitted to the steam cylinders, and pressing at dif*615ferent points upon the crank shaft, through the disk, causes the rotation of the crank shaft, which, in turn, causes the reciprocation of the pistons in the pump cylinders.

But it must be held in mind that in this Corliss contrivance the one arm is rigid. This holds the ring upon the disk to an unchanging revolution. There is no room for the play or elasticity found in the patent in suit; and, therefore, none for the balancing of weights which such elasticity affords. In just the respect that the patent in suit overcomes the difficulties of the oil situation it differs from the Corliss device. We agree with the court below, and with Judge Acheson in the Emery Suit, that a ring thus rigidly held by a pit-man is different essentially from a ring loosely mounted, as in the patent in suit.

The Weirick & Lathrope patents brought to our attention by the appellant are not as clear as we would wish. It is admitted that in the operation of one of the pumps manufactured under this patent the piston rod was rigid. While it is insisted in argument of counsel for appellant that none of the rods were necessarily rigid with the eccentric, it is admitted that one of the connecting rods is always so arranged as to stop the swing of the eccentric within certain limits. These limits are, at most, very narrow, and differ from the wide swing of the patent in suit that gives to it its elasticity. The power imparted in the Weirick & Lathrope patents being that of a thrust, we doubt if the machine would be practicable, unless one of the rods was substantially rigid. These patents do not appeal to us, therefore, as anticipations of the patent in suit in the feature of elasticity that gives to the latter its advantageous charactér.

The other patents brought to our attention can be dismissed with the observation that, compared with the patent in suit, they present the same differences.

The chief difficulty in the case under consideration is to distinguish the Allen disk pump from the Allen eccentric pump. Unquestionably, as we have already pointed out, the eccentric is functionally the mechanical equivalent of a crank. Pare down the fat side of the disk until only a slender arm remains (the lean side of the disk having no function except that of a guide way to the ring), and, barring some incidental differences, the strain would fall, and the ring would move, precisely as if the disk were untouched. It follows that in the controlling matter of counterbalancing the wells by means of an elastic crank, the disk and the eccentric contrivances are substantially identical. As stated by one of the experts for the appellee the eccentric device is, at best, an improvement only upon the disk device. The employment of the eccentric would, in our judgment, -be an infringement upon the mechanical feature that gives to the disk device its distinguishing characteristic.

But as an improvement upon Allen’s previous invention, may it not be patentable? In the disk patent there can be no bearings of the crank shaft above the crank carrying the disk. ' The aetuating-rods attached to such a disk would, in its revolution *616about the crank shaft, come in contact with the portion of the shaft projecting above the crank bearing. In the eccentric device the crank shaft can be prolonged upwards so as to find a bearing .in a frame work leading to the ground. It is evident that the strain upon an upright crank shaft, with no bearing above, would be much greater than if the strain fell between two bearings, and the racking ,of a disk apparatus having no upper bearing- would be correspondingly greater.

Then, too, the center of strain of the disk, falling upon the wrist-pin above the crank, is different from the center of strain falling in the center of the crank itself, as in the eccentric. This difference, as the preceding one, affects very appreciably the amount of racking the disk apparatus, compared with an eccentric, would suffer.

Another difference: In the disk apparatus there can be-but one crank; in the eccentric there can be many. There might be great advantage in the latter arrangement, both in equalizing the pull of the wells, and in affording space for actuating-rods, whereby the number of wells operated might be increased.

It is thus manifest that, the eccentric form of apparatus admits of a more solid frame work, and, to that extent, adds something to the diminution of strain — a result that is the-chief objective of the feature of elasticity that constitutes the merit of the disk invention. The eccentric form, in effect, carries ^forward to a nearer perfection the instrumentalities calculated to bring about the minimum of strain. The improvement found in the eccentric is, therefore, in the direct line of ascent from the feature that constitutes the gist of the disk invention. It makes the disk invention more available in respect to the very thing toward which it was chiefly directed.

That the eccentric form has largely supplanted the disk form is not disputed. Inasmuch as the rim of the eccentric, furnishing a guide way to the ring, is longer than the perimeter of the wrist-pin of .the disk device, there is correspondingly more friction, and it was feared when first brought to the attention of the expert that this would prevent an extended use of the eccentric; but ex- • perience has shown that the advantages of the eccentric more than counterbalance this disadvantage, and that wherever the disk form had previously been used, the eccentric has taken its place. Counsel for appellee do not deny this fact, but insist that the favor shown toward the eccentric is due to economy rather than to mechanical advantage — that it can be explained by the fact that the eccentric form can be set up as an entirety in the shops, and as such transported to the fields, while the disk form must be built partially in the shops and partially in the fields.

It is not in our power to nicely weigh these considerations. We cannot tell with accuracy how far the favor shown to the eccentric over its predecessors is due to economy of construction; to longevity, through the minimizing of strain; or to superior working facility. We know only that, for one or all of these reasons, it has *617found the oil public’s favor. The probability is that this favor is due, in some measure, at least, to each of the differences pointed out.

The conception, then, embodied in the patent in suit, consists in further improving the essential objective feature of the disk patent. The eccentric had never before been used for such a purpose. It never before had been made elastic, so as to adjust the varying weights, and thus balance them. The embodiment of this conception in mechanical form necessarily embraced the disk conception, and, to that extent, infringes it; but, though an outgrowth, it goes distinctly beyond the disk apparatus, substantially improving it, as we have already said, in those features that give to the disk apparatus its chief merit and purpose.

Why is not such improvement patentable? Its patentability will exclude no one who chooses from the use of the previous disk apparatus. It is a further step in the direction towards which the disk conception was the first long step; it is a decided and valuable addition to the mechanism used in oil fields, and, in connection with the disk form, for the purposes in view, it has no predecessor in the whole previous „uses of eccentrics. This consideration, in our judgment, raises it to the merit of invention.

It is stipulated by the appellant that before the bringing of this suit he made and sold machines in which there was an eccentric, or eccentrics, rigidly secured upon an upright shaft adapted to be rotated, and having a strap or ring mounted on each eccentric having a series of bolt holes around the same adapted to have connecting rods attached thereto; and that in use the connecting rods were pivoted to the straps or rings. This, in practice, is the eccentric apparatus of the patent in suit. Though there is no mention of the loose mounting of the ring, full play of the ring is evidently contemplated, else the ring could have no purpose. It gives to the apparatus the elasticity required for the balancing of the wells. It is a contribution by the appellant of a part of the apparatus, which, set up, operates precisely as the patent in suit. It is contributed with the intention that it should be united or used in connection with the other parts. This constitutes infringement. Rob. Pat. § 905.

The judgment of the circuit court, will be affirmed.