The proofs in the cause fail to establish any individual liability for the matters charged in the bill upon Mr. Marvin, the individual defendant, and it is practically conceded that it should he dismissed as to him.
The defenses to the charge of infringement are: (1) That the *174claims sued upon are invalid for want of novelty and invention, or are limited to the construction shown in the patent drawings, and admitted to be different from the defendants’ construction; (2) that such claims are for nonpatentable aggregations; (B) that by limitation imposed by the patent office, and accepted by Craig without appeal, these claims are restricted to the construction shown m the patent drawings; (4) that a cup embodying Craig’s alleged invention was publicly used and sold mbre than two years prior to his application; (5) that the defendants do not infringe.
The second of these defenses it is 'not necessary to discuss.
It is not claimed that the lubricator described in the letters patent is a pioneer invention,, and it is clear that, in its general appearance and the principle of its operation, it strongly resembles, if it is not identical with, prior devices, for some of which Craig had obtained patents, and from several of which he had taken parts, and brought them into a combination which he claims is patentable. Lubricators of this general type have been so often the subject-matter of litigation within the last 20 years that it is unnecessary to enter, into a full description of the patented and alleged infringing devices. The induct pipe, the educt pipe, the condenser, the oil reservoir, the sight-feed, the glass tube and observation chamber, and up-drop and down-drop, and an equalizing pipe connecting the steam delivery with the oil exit, are-old. In this condition of the art, the first question is, has Craig added to it, either by a new and meritorious combination of familiar parts or the. addition of a new feature to mechanism in use? Craig’s departure from previous manufactures consists in locating the pipe which connects the steam inlet and oil exit within the condenser in a straight line between the two, instead of placing it outside of the lubricator, as in previous constructions. That this is the extent of his improvement under the letters patent sued upon is clear from their specification, which, after a general description of the device, states:
“The above-described lubricator is essentially like that exhibited in letters patent No. 277,464, dated May 15, 1883, and granted to me. I have made additions to it for the object or purpose hereinbefore mentioned; that is to say, I have provided the condenser with a pipe or conduit, p, to lead from it to the boiler, in order to conduct steam from the boiler into the condenser, such pipe having in it a stopcock, q.”
Seibert, in 1876, Baker, in 1880, Harvey, in 1881, and Holland, in 1882, employed this pipe externally for the same purpose. There could be no invention in this arrangement of parts, which, upon Craig’s own admissions, constitutes the sole feature which distinguishes his last from former patented devices. ’
The prior patent to Craig is as effectual to avoid the latter as' if issued to another. James v. Campbell, 104 U. S. 356; Roller-Mill Co. v. Coombs, 39 Fed. 25, 38. The patent sued upon seems to me clearly void for want of invention in view of the state of the art, and the limitation which the inventor has imposed upon himself in his specification as to the nature and extent of his improvement.
2. But the magnitude of the interests involved compels the consideration of other defenses presented, which, whatever opinion may *175be held of the patentability of Craig’s improvement, equally avail, if valid, to repel the complainant’s rights to relief. The patent, if sustainable, must, under the proofs, be strictly construed, and limited to the precise construction it shows. Craig filed his application June 1, 1885. It fully described his lubricator, stating that the “oil is intercepted by the steam passing from the condenser downward through the pipe g therein.” His first claim included “a conduit to lead steam from the boiler into the condenser, and with a passage to lead steam into the oil-discharging conduit.” This was in substance repeated in claims 2 and 9, which were held by the examiner to be alike, and for that reason were rejected. In this, he lays claim to the equalizing pipe. The alleged invention was held by the examiner, June 0. 1885, as anticipated in patent to Hodges & McCoy, of November 18, 1884, and the patent to Holland, of August 15, 1882. Then began a contest between Craig and the patent office, which waged until February 26, 1889, when his patent was issued, during which Craig amended Ms original claims no less than 14 times, in many instances canceling and recasting them entirely, and substituting wholly new matter for that discarded. The struggle on Craig’s part was to maintain a broad claim for the equalizing pipe, and on the part of the examiners to keep him within the field of invention, and hold him to the device which he had described and illustrated. Again and again his amendments were rejected as unwarranted by his descriptions and drawings. Disclaiming the Holland lubricator (patent No. 262,774), he asked and obtained an interference with the Hodges & McCoy patent, of November, 1884. Upon the issue presented by this interference, the examiners in chief awarded priority to Hodges & McCoy, and their decision was affirmed on appeal by the acting commissioner, who, December 6, 1887, denied, after full argument, a motion for rehearing. That: officer, however, May 19, 1888, granted a motion to reopen the case, and January 19, 1889, vacated his former decision, and declared priority of invention to be with Craig, and that he had not forfeited the same by public use, as claimed by his adversaries. During the four years of contention with the patent office, and upon the interference. from Craig’s first application, June 1, 1885, to his final amendment, January 91, 1889, when he withdrew his entire specifications and claims, and substituted therefor an altogether new set, in which he made a last effort to obtain the permission of the patent office to locate the pipe g either within or without the condenser, as he might prefer, notwithstanding his persistent efforts to expand and vary his claims, Craig was rigorously held to a construction which described and located the pipe g as “another conduit within and to lead steam from such condenser into the said oil-discharging conduit.” This phraseology he varied in the second claim of his last specification by substituting for the words “within * * such condenser” the phrase “another conduit wholly within the lines of the lubricator.” In his persistence to enlarge his claims and depart from the construction set forth in his application, he had so involved the record in the patent office by ingeniously worded amplifications of his claims and contentions for alternative forms of *176construction, which were as persistently rejected, that the examiner notified him as early as January 19, 1886, that if he still persevered in inserting such objectionable matter, which had been so often rejected, his only remedy was by appeal, “before which, however,” the examiner added: “In view of the several erasures and interlinea-tions throughout the specification and amendments, applicant is advised, in order to avoid confusion, to rewrite the description and claims.”
Craig took no appeal from the rulings of, the patent office restricting him to the location of the pipe g within the condenser. He was, however’, permitted to describe it as located “wholly within the lines of the lubricator,” instead of designating its position in terms as “within the condenser”; but in the light of the history of his application, as disclosed by the file wrapper, his own construction of his claims. before the examiner of interferences, and upon his successive appeals' to the examiners in chief and the acting commissioner, and exhibited in the drawings and specifications submitted in support of his application, which excluded the form of construction now claimed to be covered by his patent, the phrases “within the lines of the lubricator” and “within the condenser” must be held mérely equivalent expressions, which restricted him to the very arrangement of parts which his application described and delineated. Morgan Envelope Co. v. Albany Paper Co., 152 U. S. 425, 14 Sup. Ct. 627; Sargent v. Lock Co., 114 U. S. 63, 5 Sup. Ct. 1021; Roemer v. Peddie, 132 U. S. 313, 10 Sup. Ct. 98; J. L. Mott Iron Works v. Standard Manuf’g Co., 4 C. C. A. 28, 53 Fed. 819; Temple Pump Co. v. Goss Pump, etc., Co., 7 C. C. A. 174, 58 Fed. 196; Williams v. Shoe Co., 49 Fed. 245; Shaw Stocking Co. v. Pearson, 48 Feu. 234. To hold otherwise would be to give to the preferred form of expression a scope and elasticity which entirely contravene the construction which Craig accepted as the condition of his grant. Moreover, the second claim of the patent locates the conduit g “wholly within the lines of the lubricator”; and the phrase “as set forth,” at the close of that and claims 4, 5, 6, and 7, by necessary implication, refers to the construction described in the specification, and qualifies each claim. It makes interior location of that pipe not merely a preferred form of construction, but the sole form permissible under the patent. Corn-Planter Patent, 23 Wall. 181, 218; Brown v. Davis, 116 U. S. 237, 251, 6 Sup. Ct. 379; Burr v. Duryee, 1 Wall. 579, 581. He cannot now obtain by construction what the patent office repeatedly denied, nor can he evade his own limitations upon his claims. His patent cannot thus be enlarged. Burns v. Meyer, 100 U. S. 671. Manufacturing Co. v. James, 125 U. S. 447, 463, 8 Sup. Ct. 967; Railroad Co. v. Mellon, 104 U. S. 112, 118.
Despite these rulings, it is argued in Craig’s behalf that he is entitled to a broad construction of the patent, and should have the benefit, as against infringers, of the doctrine of equivalents. Waiving the assumption that his patent has been infringed, its character repels this pretension. The" principle is well settled that “wliei-t an invention is one of a primary character, and the mechanical functions performed by the machine are, as a whole, entirely *177new, all subsequent machines which, employ substantially the same means to accomplish the same results are infringements, although the subsequent machine may contain improvements in the separate mechanisms which go to make up a machine. Machine Co. v. Lancaster, 129 U. S. 273, 9 Sup. Ct. 299. This case cited McCormick v. Talcott, 20 How. 402, 405, where the court, recognizing the doctrine just stated, as positively declared the rule applicable to the construction of patents for improvements in well-known devices as follow's:
“But if tlio invention claimed be itself but an improvement on a know* machine by a mere changing of form or combination of parts, the patentee cannot treat another as an infringer who has improved the original machine by use of a different form or combination performing the same functions. The inventor of the first improvement cannot invoke the doctrine of equivalents to suppress all other improvements which are not mere colorable evasions of the first.”
This doctrine is stated still more positively in Knapp v. Morss, 150 U. S. 221, 230, 14 Sup. Ct. 81, where Mr. Justice Jackson, delivering the opinion of the court, said:
“If the Hall patent was a valid pioneer invention, the doctrine of equivalents might be inyoked with regard to the sliding blocks and rests, and thus a different question would be raised; but, being confined to the specific elements enumerated by letters of reference, it is neither entitled to a broad construction, nor can any doctrine of equivalents be invoked so as to make the appellant’s device an infringement of the second claim in controversy.”
See, also, Wright v. Yuengling, 155 U. S. 47, 52, 15 Sup. Ct. 1; Boyd v. Hay-Tool Co., 158 U. S. 260, 15 Sup. Ct. 837.
This rule must be applied to the Craig patent in view of the state of the art as evidenced notably by the prior patents granted to complainant, one of which confessedly is essentially reproduced with but a trifling variation, which was known and used by earlier inventors, whose devices long anticipated those of Craig. If any additional consideration were needed to confirm the construction herein given to the words “within the lines of the lubricator,” it is found in the testimony of defendant’s expert, Jesse Smith, who well says that, “if the claims of the Craig patent are to be construed broadly, * * * so as to cover the defendant’s device, it will also cover the device of the Mitchell patent as well as the device of the Clark patent:, both of which are wholly within the Craig invention.” The same may be said of the Seibert patent, which long antedated Craig’s cup.
It is urged by defendants with much force that all of the claims here sued upon would be directly infringed by a lubricator constructed under letters patent No. 340,486, issued to Craig May 20, 1886, which under the case of Miller v. Manufacturing Co., 151 U. S. 186, 14 Sup. Ct. 310, avoids the patent of February 26, 1889, or, at. least, limits its novelty to the precise construction which it describes. Without deciding that the last patent is invalid because all of its claims might have been made in that of 1886, the claims and essential features of the lubricator in the earlier patent constitute a strong argument for the strict construction of the so-called *178"improvement” in the patent of 1889. Craig’s lubricator, therefore, if patentable, must be limited to the form which he has given it. Railway Co. v. Sayles, 97 U. S. 554. It is clear, beyond doubt, that, thus construed, the defendants do not infringe. Their device differs materially in appearance, structure, and parts, lacking essential features of the Craig lubricator, and possessing others not found in it.
3. Was there a public use or sale of Craig’s cup two years before his application for the patent therefor? Hodges & McCoy’s patent for locomotive lubricator bears date November 18, 1884. Their application was filed September 30, 1884. Craig’s application was filed June 1, 1885. In order to successfully meet the issue upon the interference, it was necessary for Craig to carry his invention back of September 30, 1884, the date of the Hodges & McCoy application. The issue in' controversy between these parties was defined by the examiner of interferences in his decision of March 12, 1887, as follows:
“The combination of a lubricator provided with a sight-feed or observation chamber in which oil rises through water in its passage to the discharging conduit for leading such oil to the part or parts of the engine to be lubricated, with a conduit to lead steam from the boiler into the condenser of such lubricator, a.nd with another conduit within and to lead steam from such condenser into the said oil-discharging conduit.”
Craig testified that he conceived the invention in issue in the latter part of 1882. His idea and aim were to make “a lubricator that could be adopted for use in an ordinary stationary engine, and one that could also be adapted for use upon an engine when there were differential pressures to guard against.” He says that he completed such a lubricator in February, 1883. In support of this claim, he put in evidence, before the examiner, a drawing which he states was made October 5,1882, showing a device capable of being-made to embody the issue by the removal of a plug at the top of the condenser chamber, and the substitution therefor of a pipe connection with the boiler or steam pipe. He states that for the invention he obtained letters patent No. 281,241, July 7, 1883, on an application filed June 13th of that year. The lubricator of this patent and that of No. 277,264, he testifies, operate upon substantially the same principles. It will also be remembered that he admits the essential similarity of that patent to the one here sued upon, the difference between them, as above remarked, consisting in the addition of the pipe or conduit p to lead from the condenser to the boiler, in order to conduct steam from the boiler to the condenser. In support of this contention that he had completed in January or February, 1883, the lubricator upon which the issue was joined, he put in evidence the bills for work done thereon by parties whom he had employed in its construction. He satisfied the examiner that he had perfected and reduced it to practice in March, 1883, and operated it upon the Pillsbury engine at Lawrence, Mass.; and, on the evidence, the commissioner awarded him priority of invention, affirming not only his conception of the device, but its actual reduction to practice.
*179The proofs show that on March 13, 1883, Craig attached to the Pillsbury engine a lubricator made under letters patent No. 281, 241, except that it had in addition to some minor and immaterial connections the pipe p leading from the top of the condenser to the boiler, in order to conduct steam from the boiler into the condenser. This addition made it identical with the lubricator described in letters patent No. 398,583, upon which this suit is founded. Craig insists that this was merely an experimental use, but he admits that it was placed on the engine under his agreement that when he had got through experimenting with it, if Pillsbury wanted to buy it, he would sell it to him. He further admits that the lubricator ran for a week or two (other testimony fixes the period at three weeks, at least), when he had it removed from the engine, and plugged up the interior pipe B which corresponds to the pipe g in the patent here involved. It is not claimed that this was done because of the inefficiency of the lubricator, or that its use had developed the necessity for any alteration, but Craig’s purpose was to see how it would operate in that condition, which he stated “would make a different device of it.” It thus ran successfully for some time. Having thus satisfied Pillsbury’s engineer of its utility and effectiveness in both forms, April if>, 1883, he made out a bill for it to Pillsbury, who testifies that he paid the same on that day. No other change than this was made in the lubricator after it was attached to the engine, and Craig admits that its working did not seem to be materially changed thereby. There was no secrecy in this use, and the place where it was had ivas open to the tenants of the building. No complaint was made against the working of the cup before the closing of the interior steam tube B, and, if Craig is to be credited, no motive prompted the change but his desire to experiment, with a view to “make a different device of it.” While it is possible that this change was made as Craig testified, it is highly improbable. The story is open to the suspicion that it is prompted by the necessity of avoiding the effect of this use and of the sale to Pillsbury. Craig says that in February, 1883, when he completed the lubricator, he was advised by his solicitor that it comprised two separate and distinct inventions, which could not be incorporated in one patent. A cuj> for use upon locomotives was evidently the most valuable; yet he did not apply for a patent for it until June 1, 1885. Tf he then had the invention which he now claims, this is remarkable, if not incredible. In the interval between its completion, in February, 1883, and June 1, 1885, he took out three patents for improved ■lubricators, neither of which, as he claims, embodies the improvement to which he alleges title under this patent. This approaches a demonstration that prior to June 1, 1885, he had not made the invention which he here claims. James v. Campbell, 104 U. S. 356.
In this condition of the proofs, the examiner in chief, to whom Craig had brought the issue by appeal, held that the earliest date which could be assigned to Craig for the conception and disclosure of this invention was June 1, 1885, when he filed his application. The file wrapper shows that the first presentation in the patent *180office’ of Craig’s claim for a lubricator operative under differential pressures was April 6, 1887, when he filed his amended specification, making oath that he was the original and first inventor of the improvement or invention as described and claimed in the above amendment (the locomotive lubricator), in addition to what was embraced in his original application. This at first was rejected, as inconsistent with the original application, but was subsequently permitted to come in under that instrument. This latter circumstance of itself suggests, at least, that when he filed his application, June 1, 1885, a fortiori in February, 1883, Craig had not completed, if he had conceived the idea of, such a lubricator as he now claims. But if we give him the benefit of the doubt against all of the oppbsing circumstances which tend to discredit his claim to invention, and assume that as early as February or March, 1888, he had made a lubricator capable of the double use which he describes, and took it to his attorney for the purpose of obtaining a patent for it, we must conclude that his device was then perfected, and “had received from its inventor every element necessary to its operation,” to use the language of Mr. Justice Matthews in Manufacturing Co. v. Sprague, 123 U. S. 249, 8 Sup. Ct. 122; and that the sole purpose of attachment to the Pillsbury engine was to demonstrate its efficiency to Pillsbury, under the agreement that the latter should purchase it if its use established its utility. This use was for the purpose of trade and profit, and the test was merely ancillary to the sale. This, of itself, would defeat Craig’s patent, as it constituted a public use of the invention (if it existed) two years before his application. If there is any doubt, however, as to the character of its use, and its effect upon the patent, there can be none whatever on the admitted fact of the sale to Pillsbury. The test had demonstrated the operativeness of the device, and the lubricator had met the conditions upon which Craig proposed to sell, and Pillsbury agreed to buy, and he bought it because it had met those conditions. It was upon these grounds that the examiner in chief, after a careful review of the transaction in the light of evidence, held that Craig’s right to a patent was defeated under the statute (Rev. St. § 4886), although they awarded priority of invention to Craig. The assistant commissioner of patents, in an elaborate opinion, affirmed the conclusions of the examiner in chief. More than six months afterwards, he changed his views, and reversed his former decision. The reasons expressed for this action are not satisfactory. It was the contention in Craig’s behalf in the various proceedings of interference that the use of his cup on stationary engines should not defeat his claim, as his lubricator was designed for use on locomotives. This is inconsistent with the claim of invention which specified the completion of a cup capable of use as a single or double connected lubricator. If it were true that only prior use on a locomotive could avail to anticipate Craig, it is difficult to see on what ground he should be accorded priority of invention as of March 13, 1883, the date of its use on the Pills-, bury engine. It was upon this ground that the acting commissioner held that public use had not been proven against Craig. The examiner in chief had held that the cup used at Pillsbury’s did not *181embody tbe present invention, “the steam pipe leading down from the condenser being plugged up and without function; and, moreover, the engine being stationary, and not locomotive, no such functions were ever accomplished or carried out, and the invention was not made.” This overlooked the three weeks’ use of the cup prior to the closing of the interior pipe B. ‘-'Of course,” said the examiner in chief, “Craig cannot now urge this identical Pillsbury experiment as proof of complete invention March 13, 1883. * * * Craig cannot blow hot and cold with the same breath. He cannot plead that he had not the invention when public use is in question, and by the same evidence show that he had it when priority of invention is in question. ITe was given the benefit of the doubt on his own qualifications and distinctions, hit res in a gis valeat quam pereat,’ to save forfeiture, but the invention cannot now be expanded to save priority.”
Craig’s position here is as clearly incongruous as that taken in the patent office. To make good his claim to invention, he insists that he completed and reduced to practice on the Pillsbury engine a device embodying- his invention as early as March 13, 1883. To meet the defense that Ms alleged invention had been in public use and on sale more than two years before his application for a patent, he denies that the cup put on the Pillsbury engine contained the invention here claimed, because the interior pipe was plugged and functionless on the stationary engine. The fact, if it be a fact, that, after his cup had been in successful use for three weeks with its interior pipe open, Craig closed that pipe, in search of a different device, and that, thus changed, the lubricator was thereafter used by his vendee, does not avoid the effect of the first use. The article sold embodied both inventions, if he had two, and was sold without restriction or condition. There was nothing in the transaction which limited Finsbury’s use of it, with or without the plug. Piere is no evidence that Pillsbury knew of the change. Had he removed the plug, and used the cup as a double connection lubricator, he could not have been held an infringer. He had not bargained for the device which Craig’s experiment might develop,but for that which he commended as his finished production. It became his property absolutely. It was a sale of the device in the course of Craig’s business as a manufacturer. A conditional sale or a sale on approval as an offer to sell makes the device “on sale,” within section 4886. Henry v. Francestown Co., 2 Fed. 78; Kells v. McKenzie, 9 Fed. 284; Lyman v. Maypole, 19 Fed. 735. In order to constitute a public use of invention, it is not necessary that more than one of Che patented articles should he publicly used. “One well-defined case of such use is just as effectual to annul the patent, as many.” McClurg v. Kingsland, 1 How. 202; Fruit-Jar Co. v. Wright, 94 U. S. 92; Worley v. Tobacco Co., 104 U. S. 340; Egbert v. Lippman, 104 U. S. 333; Manufacturing Co. v. Sprague, 123 U. S. 249, 257, 8 Sup. Ct. 122.
In Egbert v. lippinan, supra, it is said, by way of illustration:
“Nor instance, if the inventor of a mower, a printing press, or a railway car makes and sells only one of the articles invented by him, and allows''the vendor to use it for two years without restriction or limitation, the use is just as public as if he had sold or allowed the use of a great number.”
*182Nor is the publicity of the use dependent on the number of persons to whom it is known where the device is given or sold for use without limitation or injunction of secrecy. A use or knowledge of the use, if confined to one person, is fatal to the patent. See, also, Manning v. Glue Co., 108 U. S. 462, 2 Sup. Ct. 860.
In Elizabeth v. Pavement Co., 97 U. S. 126, it is said that while abandonment of the invention to the public will not necessarily follow its public use or sale within two years before the inventor’s application, “yet if the invention is in public use or on sale prior to that time, it will be conclusive evidence of abandonment, and the patent will be void.”
The rule of judgment applied to cases where the question of public use arises is stated in Manufacturing Co. v. Sprague, 123 U. S. 249, 8 Sup. Ct. 122, as follows:
“In considering the evidence as to alleged prior use for more than two years of an invention, which, if established, will have the effect of invalidating the patent, and where the defense is met only by the allegation that the use was not a public use in the sense of the statute, because it was for the purpose of perfecting an incomplete invention by tests and experiments, the proof on the part of the patentee, the period covered by the use having been clearly established, should be full, unequivocal, and convincing.”
The condition imposed by section 4S86, Rev. St. U. S., does not require for the defeat of a patent, because of the sale or use of its subject-matter with the inventor’s consent two years prior to his application, that such sale or use must have been continued during all that period. It is enough if the inventor has sold an article or permitted its use without restriction at any time over two years before he applied for a patent. Andrews v. Hovey, 128 U. S. 267, 274, 8 Sup. Ct. 101; Id., 124 U. S. 694, 719, 8 Sup. Ct. 676; Egbert v. Lippmann, 15 Blatchf. 295, Fed. Cas. No. 4,306.
The object of section 4886, which before the revision of the statutes was section 24 of the act of 1870, as said by Mr. Justice Blatchford in Andrews v. Hovey, supra, was to “require tbie inventor to see to it that he filed his application within two years from the completion of his invention, so as to cut off all question as to the defeat of his patent by the use or sale of it by others more than two years prior to his application, and thus leave open only the question of the priority of invention.” Under these rulings and under the proofs in this case, it is clear that there was such a public use and sale of the Craig lubricator as to avoid his patent.
For these reasons, without discussing other questions, the bill must be dismissed, with costs.