[1] The validity of letters patent No. 914,426 granted to appellant as assignee of Charles W. King for improvements in metallic rod packing ring and their infringement by appellee are involved in this appeal from a decree dismissing the bill. Judge Sanborn held the patent valid, but not infringed.
*740The second of the two claims in the patent had' been disclaimed before suit. The first claim reads as follows:
“A rod packing ring of soft metal divided into two segments, each of which may be moved laterally over the rod to be packed and each having tapered ends, one with a convex end surface adapted to lie under, and the other with a concave end surface adapted to lie over the corresponding tapered ends of the other segment, said concave and convex surfaces being parallel to the axis of the ring and so disposed that, when assembled on the rod, the segments interlock with each other and the rod to prevent either segment from being moved laterally away from the' rod, while the segments may be readily moved together or separated by moving one segment relative to the other in a direction parallel to the axis of the rod.”
The purpose and function of the invention is thus described in the specifications of the patent:
“The object of the present invention is to provide a packing which will be effective in preventing the flow of fluid along the rod and will be simple in construction and composed of a relatively small number of parts and which can be readily assembled and disassembled.
“A particularly important feature of the invention is the formation of a soft metal packing ring in two parts, so shaped that the ring can contract readily to compensate for the wear of the rod or packing ring and thereby maintain a tight joint, while at the same time the parts interlock with each other and. the rod, when assembled on the rod, in such manner as to prevent a lateral movement of either segment away from the rod, while at the same time each segment of the ring may be laterally moved onto or off the rod when disengaged from the other segment and the two segments when on, the rod can be assembled together in the interlocking or normal position or disassembled by moving one segment relative to the other in a direction parallel to the axis of the rod. This I accomplish by the peculiar configuration which I give to the ring segments.”
The drawing of Figure 5, described as a perspective view showing the manner of assembling and disassembling the ring on the rod, will be sufficient.
The inventor further states:
“The concave and convex end surfaces, however, are parallel to the axis of the rod, so that the segments may be readily separated by moving one segment relative to the other in a direction parallel to the axis of the rod in the manner shown clearly in Fig. 5.
“These surfaces must, however, be of sufficient capacity and convexity and so disposed’ as to permit the proper contraction of the ring, while at the same time the segments are effectively locked together when assembled on the rod. At the same time, each segment must not have its rod engaging surface so great or its overlying part f* so shaped that the segment cannot be moved laterally over the rod to be packed. This means that each segment must not engage the rod over an are of more than 180 degrees.”
The defense of noninfringement is based, first, upon the fact that in defendant’s ring the meeting faces of the segments, instead of being strictly parallel to the axis of the ring, as well as to each other, form a slight angle about five degrees with the axis of the rod; second, upon the alleged limitation of claim 1 to this specific form by reason *741of the disclaimer of the broader claim 2; and, third, upon the grant of the Munich patent of April 28, 1904, No. 1,095,163, under which defendant makes its ring.
If, however, the King patent is to be construed broadly enough to cover defendant’s ring, then the defense of anticipation is urged primarily because of the Jerome, patent, No. 211,299, for stuffing boxes for steam engines.
1. Infringement. There is one, and only one, difference between defendant’s and the King ring. The meeting faces of the curved and tapered ends of defendant’s ring, while parallel to one another, are not parallel to the axis of the ring. They depart five degrees from mathema tical parallelism.
If King’s claim is to be limited to the requirement of exact parallelism, there is no infringement. Concededly, mathematical parallelism is neither possible nor essential in devices of this kind, when made for commercial purposes. With substantial parallelism, the very functional interoperation described in the patent will be attained.
Defendant concedes that the King claim is not limited to mathematical accuracy in the parallelism. The real question is whether, notwithstanding a departure therefrom to the extent of five degrees, substantial parallelism still remains and, the functions thereby designed to be secured, are still attained. If so, then in our judgment the King claim is to be construed as including such substantial parallelism.
Now a material departure from substantial parallelism of the contacting surfaces with the axis of the rod in a ring having the King features would prevent the lateral application of the ring segments to the rod, because in at least one of them the ends of the inner faces would approach each other by a distance less than the internal diameter of the ring and the rod. Because of the softness of the metal, an essential element of the claim, a slight departure, a departure only to the extent that defendant has made, will not, however, affect the lateral applicability. Due to the softness of the metal and to the fact that the tips of the ring are finely drawn out, such segments cannot only be laterally applied to the rod without destruction or functional distortion, but, when so applied, they interlock by sliding together axially, and function exactly like plaintiff’s with respect to packing for maintenance of tightness. It follows, therefore, that the substantial parallelism called for in the King patent covers a construction, like defendant’s, departing from mathematical parallelism by five degrees. Adam v. Folger, 120 Fed. 260, 56 C. C. A. 540.
That some advantages of plaintiff’s more nearly mathematical parallelism are thereby sacrificed — defendant’s segments are rights and lefts, while plaintiff’s are interchangeable, further, defendant claims that there is a slight distortion of its sections or a slight inaccuracy in the. fit — does not afford an escape from the charge of infringement; defendant’s result “is the same in kind and effected by the employment of plaintiff’s mode of operation in substance.” Winans v. Denmead, 15 How. 330, 14 L. Ed. 717. And even if, by reason of the extent of defendant’s departure from mathematical *742parallelism and the inclination of the segment faces in the same direction, certain more or less compensating advantages are obtained — an easier assembling of the sections and their wedging together when assembled — and the Munch patent covering these be valid as an improvement in this particular on King, a point we need not determine, yet inasmuch as defendant still adheres to and necessarily must adhere to a substantial parallelism, “it has none the less succeeded in appropriating all that was of value” in the King ring. Hoyt v. Horne, 145 U. S. 302, 12 Sup. Ct. 922, 36 L. Ed. 713.
Defendant’s is no new combination of either the same or different elements; it has all of plaintiff’s elements so combined as to perform the same functions in the same way. As in Elizabeth v. Pavement Co., 97 U. S. 126, 137, 24 L. Ed. 1000, defendant’s peculiar form may constitute an improvement on plaintiff’s and perform an additional office, but it none the less performs the office assigned to it in the King patent.
Clearly the second claim was disclaimed because it covered, not a ring, but an old combination of the ring and the other elements used in the stuffing box. The packing ring in this combination was described only as composed of interlocking segments and formed with a conical end. Obviously neither this claim nor its disclaimer can extend or limit the scope of the first claim, the only claim for a special ring construction.
[2] 2. Anticipation. Concededly the closest approach to the King ring is the ring shown in Jerome patent, No. 211,299, dated January 14, 1879, 30 years earlier than King. If the drawing alone were considered, it would seem fully to portray King’s ring. But invention involves conception of at least some function, as well as the selection of the means whereby that function can be operatively secured. And while Jerome would anticipate all later inventors to the full extent of his invention as disclosed by his specifications and claims read in conjunction with the drawings, and that, too, irrespective of whether he understood the scientific principles involved therein or appreciated all of the functions of his construction, he does not anticipate that which cannot be attained compatible with his real conception.
[3] While the number of parts into which an object is divided or the exact form it assumes may be of no importance whatsoever in determining anticipation or infringement, it may become absolutely determinative if a given number of pieces or a certain shape is essential to the' realization of one or the other conception. Now Jerome’s drawing shows a "ring of two segments, a preferential form, with tapering ends. But, far from pointing out any need of limiting the number of pieces to two, he expressly states that the ring mav be composed of any desired number of sections. Moreover, he nowhere indicates that the degree of tapering shown in the drawing is functionally required. Therefore if, to attain some function, a ring of two segments, no more and no less, and each tapered in a certain way, is absolutely essential, the conception of that function as attainable by these means cannot have been anticipated by Jerome; and to that extent, at least, a subsequent .inventor may have made a patentable *743improvement. Nor is it material that the function attained by these limitations is itself old.
Concededly, to secure the interlocking of the sections with each other and with the rod when assembled thereon, sections which when not assembled on the rod do not interlock, and no one of which,, without the other, interlocks with the rod, it is absolutely neceessary that the ring have only two sections, and that the ends taper no less than the drawing shows. The conception of this interlocking function of a segmental packing ring, essential to King’s invention, was therefore not a part of Jerome’s invention, and no one going to the Jerome patent alone would find therein any such disclosures.
[4] As was held in Brill v. Third Avenue Ry. (C. C.) 103 Fed. 289, quoted by us in Gray Tel. Pay Station v. Baird Mfg. Co., 174 Fed. 417, 98 C. C. A. 353:
“A patent for a mechanical combination is not anticipated by a drawing in a prior patent which incidentally shows a similar arrangement of parts, where such arrangement is not essential to the first invention, and was not designed, adapted, or used to-perform the function which it performs in the second invention, and where the first patent contains no suggestion of the way in which the result sought is accomplished by the second inventor.”
An examination of Jerome’s two segment sketch drawing discloses two contacting tapered faces that are not parallel either to the rod or to each other, but inclined in opposite directions toward the center of the rod, making it impossible to assemble, or disassemble them by moving them in directions parallel to the rod. Jerome specifies several times that the opposite tapering ends of each section must be on a curve coinciding with the periphery of the ring, and in one place he uses the expression “exactly coinciding”; the meeting faces of the lapping etids must therefore be cylindrical. But, if they are cylindrical, the only form suggested or contemplated by Jerome, and not helical, like a later King ring covered by a 1914 patent, then, when the two sections are made in exact conformity with the drawing, they cannot slide together or apart along the rod by a movement parallel with the rod, because they would absolutely interlock with one another, and not also with the rod, thereby preventing a movement of one towards or from the other and parallel with the rod.
It follows, therefore, that a ring made in exact conformity with the drawing, read in the light of the specifications, as we believe it must be read if its • teachings are to be discovered, would.be impracticable and inoperative for the use intended by either Jerome or King; if the cylindrical surface were adhered to, and not, as in the model purporting to follow the drawing, concededly departed from sufficiently to permit a slight helical movement, the segments with laps as shown in the drawings could not be assembled on the rod; it would be necessary to slip the ring over the end of the rod; if, however, they were tapered less than the drawing shows, they could not interlock with each other and with the rod, when assembled together. And, in fact, no Jerome rings have ever been made for practical use.
That Jerome may have been ignorant of some function disclosed by his patent would not affect it as an anticipation; on the other hand, *744the fact that the drawings might suggest some change therein, mechanically slight, but operatively yielding a new function, will not deprive the later inventor of his improvement.
In our judgment, neither Jerome’s drawing of Figure 3, the ring, considered even entirely separate from Figures 1 and 2, the stuffing box, nor the Jerome patent itself, anticipated King.
It is unnecessary to examine in detail the other patents. None of them discloses a packing ring of King’s construction or with its functions.
The decree must be reversed, and cause remanded, with directions to proceed further in accordance with the views herein expressed.