SEVERENS, Circuit Judge.
The plaintiff in this case complains of the infringement 'of its patent No. 639,891, founded on an invention by John C. Duner, of “improvements in sand-boxes for cars,” and granted December 26, 1899. In a general way the invention may be described as one relating to sand-boxes located in front of the wheels of cars for the purpose of sprinkling sand on the track to increase the friction between the wheels and the track; and the substance of his invention was the appropriation of a “choke-valve” to be used in such boxes. Choke-valves were familiar to the arts of mechanics, and consisted of a hopper or reservoir, a pipe, or conduit extending below, and a basin or container below the end of the pipe to hold the contents as they descended into it. As the basin fills to the end of the pipe, it chokes the opening and prevents the further flowing of the contents. It is obvious that its principle of operation would be applicable to use in any case where the contents have a partial quality of fluidity; that is, where the particles have a moderate size and modified capacity of flowing upon one another. To open the valve and set the contents flowing, the basin must be moved away, or the pipe must be moved to one side.
The defendant is a street railway company and uses such valves on its cars. It denies that the plaintiff's patent is valid, and it denies infringement. It denies its validity because, it says, the invention is nothing else than another use of an old device, and further that the device had already been employed for the same purpose and in the same art; that is, in sprinkling sand on railway tracks.
The complainant's sand-box consists of a, hopper, a discharge pipe, or “passage,” an oscillating container consisting of a cup-shaped pocket [864]*864under the feed-pipe, a valve chamber or casing, and a discharge spout. These are the elements of the combination of the third claim, which is the only one involved; and it is manifest that they were all found, in one form or another, in other earlier structures wherein choke-valves were used, unless it be the casing, or “valve chamber,” as D.uner calls it, which is nothing more than a ho'using and bearings for the trunnions on which the'pockets are revolved. But there is a peculiarity in the complainant’s patent which gives rise to some questions which should be considered. The “pocket” which the specifications describe is not merely a single'pocket, but another pocket is included and is arranged alongside of the other with a partition part way between them. Thus, they are not movable upon one another. The construction is such that, when the first pocket fills, the contents flow into the side pocket, and provision is made for a discharge from the side'pocket into the chamber below.
Figures 3 and 5 illustrate the construction. Figure 3 shows the main pocket, 26, into which the sand descends through the pipe, 18, from the hopper. 30 is the side pocket, and 29 is a partition which extends back only part way, so as to leave an opening between them, through which opening the sand passes from the main pocket to the side pocket as the former fills. 32 is the edge of the side pocket, over which the sand is discharged into the chamber, and thence through the discharge spout, 15. At 47 is an opening in the main pocket, and, when it.is desired to pass the sand directly down, the pocket is revolved until the sand passes through the opening, 47. When the pockets are in this position, no sand will pass over the edge, 32, of the side pocket, but will flow back into the main pocket and be discharged through the opening, 47. Upon these facts counsel for the appellant contends that there are thus provided means which make only the main pocket necessary and render the side pocket inert, and he quotes the following from the specifications to show this:
“The apparatus as thus constructed is devised, as hereinbefore set forth, for the production of an intermittent feed, supplying a definite quantity of saud each time the operating pin, 45, is depressed. If it is desired, however, to produce.a. continuous feed, or one which will permit the continuous.flow [865]*865of the sand as long as the pin, 45, is hold depressed, I provide in the bottom, 27, of the enp or pocket, 2(5, an aperture, 47, so located that, when the valve is in the closed position (shown In Figs. 3, 5, and (5), said aperture is above the end of the feed pipe, 18; but when the valve is moved to its other position, as shown in Fig. 7, said aperture is opposite and immediately below the end of (lie feed pipe, 18. When the parts are in the position shown in Fig. 7, it is obvious that a continuous flow of sand will lie produced uni 11 the pressure on the operating pin is removed and the weight returns the valve to its closed position, whereupon the flow of sand will cease. When this particular form of valve having the aperlure. 17, is employed, the cup or pocket, 30, may be dispensed with, although iis enrployment is still desirable as a means for insuring a supply of sand in case the aperture, 47, becomes clogged or obstructed.”
—and then reads claim 3 upon this organization as a distinct and separate invention. But we cannot agree that the patentee and his counsel are correct in this, for while, in the circumstances supposed, it might be said that in a sense the second pocket is inoperatively inert and unnecessary, it is not altogether so. The patentee nowhere suggests that the partition, 29, between the main and side pockets, shall or may extend all the way between the pockets. He expressly says it shall not, but that an opening shall be left through which the sand may pass through from one to the other. We have therefore no right to assume an entire partition. That a claim which is construed by bringing into it, by reference to the specifications, a feature there found, must he regarded as equivalent to a claim having in itself such feature seems to be a logical deduction, and is in accord with the observation of Mr. Justice Blatchford in Fay v. Cordesman, 109 U. S., at pages 420, 421, 3 Sup. Ct. 236, at page 244, 27 L. Ed. 979, where he said:
“'The claims of the patents sued on in this ease are claims for combinations. In such a claim, if the patentee specifies any element as entering into the combination, either directly by the language of the claim, or by such a reference to the descriptive part of the specification as carries such element into (he claim, lie makes such element material to the combination, and the court cannot declare it to be immaterial.”
While the law permits us to read the specification into a claim upon a reference thereto in the latter, it would be an abuse of such liberality to first alter the specification and then read it, thus altered, into the claim. Such a practice is wholly inadmissible. If we were to do so for one purpose, we might do so for another, and so on, endlessly. The patent would become nebulous, and all particularity extinguished. If now, the side pocket were removed, the sand would run out of the opening into the chamber uncontrolled. Moreover, the sand collected in the side pocket would not he available; and, further, the sand in the side pocket would at no time during the operation oppose the leaking out in that direction of the sand in the main pocket and so, to some extent, relieve the entire want of a partition through the opening. If the facts supported the appellant’s contention, we see no reason why the conclusion might follow. If the matter specified furnishes the data, we suppose the patentee might assemble distinct elements, less than all, in combinations, which would represent distinct inventions, and also claim them all in another combination. This is a frequent practice. He may also claim separately each element in the combinations, supposing, of course, in every such case the element or combination of [866]*866elements indicate invention; but if we cut out the side pocket and suppose the opening in the partition, 29, healed, we should have a structure differing in no respect in principle from others contained in the former art. Modifications of old structures might be found, but not modifications having the dignity of inventions. If we are right in this, we need go no farther, for the defendant’s sandbox has but the one pocket and, of course, does not infringe the complainant’s, which has two pockets, or, speaking with more exactness, a double pocket.
But three former patents are adduced by the defendant to prove the Duner invention to have been anticipated, and we think they are sufficient. The first in date is a patent to McPherson, granted in 1895, No. 535,260, for an automatic feed for pulverizing mills, and having for its object the provision of a mechanism to effect a uniform feed of “oreor other substances.” The mechanism of the invention was a magazine of ore being pulverized having an opening at the bottom through which the contents passed out upon a plate, the accumulation on which choked back the material in the magazine. The receptacle was flat, because, the nature of the material being coarse, the sides did not need to be turned up to form a pocket. Then there were the Richard’s patents, Nos. 559,210 and 615,195, in which the choke-valve was employed in weighing and delivering grain and other material. In these were the hopper, the tube or conduit from the bottom of the hopper, and the curved, oscillating pocket, in which the contents spread out around the bottom of the tube and stopped the flow. When the pocket was oscillated, the contents of the cup were discharged, either over its depressed edge, as in the Duner side pocket, or perhaps, through a hole in the cup registering with a hole below, say, the upper end of the discharge spout. Later than these was a patent granted to A. W. Ham, September, 1897, numbered 590,220, for a sand-box for car tracks. The essential feature of this was, as in the Duner patent, the choke-valve. There was a hopper with an opening at the bottom and under that a cup or circular basin, in which the sand would pile up and cease to flow. An oscillating paddle hung on trunnions turning in the casing when the valve was in operation, carried the contents over the edge of the pocket into a chamber whence they were carried to the track by a discharge spout. The most material difference between this and the Duner valve having a single pocket (for it is with such a device we are now making comparison) consists in the 'fact that, while in Ham’s construction the pocket is stationary and the paddle oscillates over it, in Duner’s the tube is stationary and the pocket is oscillated under the tube; but this makes no difference in the principle, and there is no invention in making the change.
We held, in Campbell Printing Press Co. v. Duplex Printing P. Co., 101 Fed. 282, 293, 41 C. C. A. 351, that merely making the type bed stationary under traveling cylinders, instead of making the cylinder stationary and the type bed movable under it, did not amount to invention. Like decisions upon similar facts were made by Mr. Justice Curtis in Sargent v. Larned, 2 Curtis, 340, Fed. Cas. No. 12,364, by Judge Blodgett in Abbott Machine Co. v. Bonn (C. C.) 51 Fed. 223, and by the Supreme Court in Machine Co. v. Murphy, 97 U. S. 120, [867]*86721 L. Ed. 935, to which may be added Devlin v. Paynter, 64 Fed. 398, 12 C. C. A. 188.
The conclusions on which we rest our judgment are these: Conceding that the Duner patent is valid to the extent that it covers the double pocket valve, the defendant does not infringe the Duner patent so construed. Assuming that claim 3 of the Duner patent might be sustained as being one for a valve having a single pocket, about which we have much doubt, the invention, so restricted, was anticipated by earlier patents, and especially by that granted to Ham, in 1897.
The decree of the court below dismissing the bill will be affirmed, . with costs.
NOTE — The following is the opinion of Sater, District Judge, in the court below:
SATER. District Judge (sitting by designation). The defendant is using sand boxes constructed in accordance with the Haddock patent, No. 739,803, issued on an application filed January 14, 1903. The complainant is the owner of the Duner letters patent, No. 639,891, for sand boxes for cars, issued on December 26, 1899.
The complainant charges that the sand boxes used by the defendant infringe its letters patent, and has elected to stand upon the third claim therein mentioned, which is as follows: “In an apparatus of the character described, the combination with a hopper or reservoir having a discharge pipe or passage, of an oscillating valve having a cup-shaped pocket, into which the lower end of the feed-pipe extends, and a valve chamber completely inclosing said valve and provided with a discharge spout, substantially as described.”
Hoppers, feed pipes leading therefrom, choke-valves, inclosing chambers, and discharge spouts were well-known structures or devices, and were all used long prior to the granting of either the Duner or the Haddock patents. Duner, in so far as the record discloses, was the first to apply the choke,-valve to sand boxes for cars. Its use, however, in connection with other granular substances, was previously well known; a good illustration being found in tho Richards patent, No. 559,210. I shall not, however, determine whether the new use of the choke-valve by Duner is so analogous to its former use in the manipulation of coal, sand, grain, seed, and many other granular and pulverulent products that the applicability of his device to the new use would occur to a person of ordinary mechanical skill or not.
Complainant’s Exhibit “Early Model Duner Box” is constructed in accordance with the above-mentioned claim. Complainant also offered in evidence another exhibit, “Duner Single Dipper Box,” alleging that it also is constructed In accordance with the patent and that the sand boxes used by the defendant infringí» such device. His valve, as exemplified by that exhibit, consists of but a single pocket extending through less than 180 degrees, but has no aperture therein through which the sand may flow to the discharge pipe. Instead of a feed pipe, ii has a chute, apparently not so wide as that in the Haddock device, hut whoso, walls are more nearly perpendicular. It secures a continuous feed only, which feed is effected by so oscillating the valve as to swing It substantially beyond the wall of the chute, thereby permitting the uninterrupted flow of sand into the discharge pipe and thence to the track. Below the chute in the Haddock device is an oscillating gate, on which the sand flows from the hopper through the chute until it chokes. When the gate Is swung aside, it permits a continuous flow of sand to the track. The patent does not specify, and the Haddock device does not require, a passageway from the hopper to the valve in the form of a pipe, such as is shown in the Duner patent, to conduct the sand from the hopper to the gate or valve.
In determining what the Duner patent covers, the rule must be observed that patents are to be construed in the light of all the circumstances and of the apparent purpose of the parties to them. As Duner was required to specify and point out the part, improvement, and combination which he claimed as his invention, his specifications and drawings may he considered for the pur[868]*868.póSe óí better understanding the meaning of his third claim, not, however, for the purpose of changing it and making it different from what it is. Howe Machine Company v. National Needle Company, 134 U. S. 394, 10 Sup. Ct. 570, 33 L. Ed. 963; Walker on Patents, § 186. The words “discharge pipe,” “pas’sage,” and “feed pipe,” as found in the claim, are used interchangeably, and are synonymous. In his specifications he designates this same means for conveying sand from the hopper into the valve as a “feeding tube or passageway,” ‘feeding pipe,” “feed pipe,” and “feed spout.” The feed pipe contemplated is a fixed, rigid, downward-projecting obstacle, the inclination of whose extremity furthers the breaking up, as the valve is moved, of any packed or clogged mass which may form at the mouth of the feed spout, The valve mentioned in the claim, into which the feed pipe extends, has a cup-shaped pocket. A “pocket,” as defined by the Standard Dictionary, is a “small bag or pouch; * * * a cavity, opening, or receptacle.” The valve called for must have, therefore, a cavity or receptacle into which the sand flows until it chokes, and to render-the device operative it must have.such depth as will hold the sand and prevent its overflow. The pocket is cup-shaped. The term “cup-shaped,” in so far as the research of counsel and of myself .discloses, has not'been defined by the courts or lexicographers. In Knight’s American Mechanical Dictionary “cup” is defined as “a hollowed portion or object to hold a liquid.” The cup-shaped pocket or cavity, therefore, must necessarily have vertical or substantially vertical sides. Their angle of inclination and height must in any event be considerable. Duner, recognizing this' fact in his description of his device, specifies two vertical sides for his valves, and a curved bottom rising to the height of the vertical wall, 29, and thereby forming its other sides. The height of his cup-shaped pocket equals the radius of the circle, through about 270 degrees of which his curved bottom extends. To render his device operative, as described, not only is such pocket, cavity, or receptacle necessary, but it is also essential that his feed pipe should extend into and below the pocket’s vertical sides; otherwise, the sand as it flows through the feed pipe would escape over such walls and flow to the track. The valve which he describes would be practically useless, if sand were conducted to it through a chute, such as is shown in the Haddock patent.
His peculiarly designed valve, with a feed pipe projecting therein, resulted from his desire to provide a single apparatus having both a continuous and an intermittent feed. I-Iis means of securing the latter is described at great length. He says his first thought was a dipper. If he means such a dipper as is shown in his “Duner Single Dipper Box,” it is remarkable, in view of its effectiveness and its simplicity, as compared with the peculiarly shaped valve which he describes so minutely, that he made no mention of it and failed to use the word “dippex"” at any place in his letters patent He states that a straight discharge through a hole was his first idea, that the second pocket was an afterthought, and that he attached a second pocket to his first. How and why he proceeded to attach it to a dipper, such as is shown in his single dipper box, if that is what he first conceived and made, is not stated. The fact is that his earliest device shown is his “Early Model Duner Box,” which shows a valve such as he specifically describes in his letters patent. If the intermittent feed was an aftex*thought, it became predominant, and was manifestly uppermost in his mind.. Is not the explanation found in his statement that, when his first track sander was made, railroad men leaned toward the intermittent discharge, and he thought an intermittent box would be the easiest to market? At the conclusion of his lengthy description of'his apparatus for securing an intermittent feed, he adds: “If it is desired, however, to produce a continuous feed, or one which will permit the continuous flow of the sand as long as the. pin, 45, is held depressed, I provide in the bottom, 27, of the cup or pocket, 26, an aperture, 47, so located that, when the valve is in the closed position (shown in Figures 3, 5, and 6), said aperture is above the end of the feed pipe, 18; bxxt when the valve is moved to its other position, as shown in Figure 7, said aperture is opposite and immediately below the end of the feed pipe, 18. When the parts are in the position shown in Figure 7, it is obvious that a continuous flow of sand will be produced until the pressure on the operating pin is removed and the weight returns the valve to itclosed position, whereupon the flow of sand will cease. When this particular [869]*869form of valve, having the aperture, 47, is employed, the cup or pocket, 30, can be dispensed with, all hough its employment is still desirable as a means for insuring a supply of sand in case the aperiure, 47, becomes clogged or obstructed.” He is still speaking of the peculiarly shaped and minutely described valve. His purpose was to put forth a single mechanism, which, as occasion for its use arose, railway employes might utilize in such way as deemed expedient; i. o„ for either the continuous or the intermittent flow of sand. Although ho states that the second cup-shaped pocket may be dispensed with in caso a single continuous feed alone be desired, he adds that Its retention is desirable as a means of sanding in case the aperture in the larger pocket becomes clogged or obstructed. He still had in mind his specifically described valve, or he would not have advised its retention as a means of relief in case of the clogging of the aperture in the larger pocket.
To sustain its contention that boxes constructed like the Duner single dipper box fall within the terms of the patent, the complainant directs attention to Dimer’s recital in his letters patent that the details of construction may be modified without departing from the principle of his invention. He illustrates by stating that operating mechanism, for instance, other than that which To bas devised and shown, may be applied, whereby a spring may be substituted for a weight to return the valves to their normal or closed positions, and adds that he does not wish to be understood as limiting his invention to the preciso construction previously described and shown in the drawings. He still contemplated the use of both valves, as he has spoken of them in the plural. His statement that he does not wish to be understood to limit his invention to the “precise” construction described and shown in the drawings indicates that he did not contemplate a wide departure from such form of construction. It may be doubted whether his reservation of the right of modifying the details of construction was necessary, in view of the law of equivalents ; but his language may properly be considered in construing the patent. There is no suggestion of a modification in the valves or cup-shaped pockets themselves. Evidently the double purpose of his device and the preference previously expressed for the retention of both pockets or valves, whatever the construction might oe, was still present in his mind. The extent to which the use of his peculiar form of valve was riveted in his mind is evidenced by the identification of the complainant’s Exhibit “Early Model Duner Box,” as his first specimen of a sand box of the continuous feed type. I am satisfied from the record that his first single dipper box was not made until about June or July, 1903, and that his first conception of a box of that character then originated. He preceded his claim of both a feed pipe and a valve having a cup-shaped pocket, as elements of his combination, with the words “in an apparatus of the character described,” and followed it with the words “subsiantially as described,” and thereby effectually limited the application of lis claim to the particular kind of pipe and valve described in the specifications ; for, if there be a doubt as to whether the Haddock device is the mechanical equivalent of Duner’s, that doubt must be resolved against Duner so long as the claim contains the words “substantially as described.” Hobbs v. Beach, 180 U. S. 400, 21 Sup. Ct. 409, 45 L. Ed. 586. Duner and his successor in title must accept the feed pipe as an element in his combination so long as a valve having a cup-shaped pocket is claimed, into which the lower end of the feed pipe extends. The one cannot be accepted and the other rejected. Duner specified a particular form of device as the means by which the effect of his invention is produced, and there is, therefore, no infringement by the defendant, because the Haddock device has no feed pipe and no valve even remotely approaching that described in the Duner patent. Walker on Patents, $ 863.
The bill is dismissed, at the complainant’s costs.