We are told by evidence that use has shown three marked advantages in plaintiff’s patented apparatus: (1) Utilizing a flexible connection dispenses with rigid, accurate, and expensive pipe-fitting; ■ (2) the passage of volatile fluid through a long tube external to the heated chamber cools it; and (3) a thermosensitive bulb containing a trap effectually prevents anything but liquid as distinguished from- vapor pressing upon the diaphragm of the collapsible member.
It is unnecessary to say whether we accept this evidence in its entirety or not. The point is that on these facts an argument is based to the effect that, since these beneficial uses were inherent in the patented device, the claim in suit should, and indeed must, be read so as to cover any apparatus which obtains the same beneficial and useful results by substantially the same means.
[1] Strictly speaking, infringement of a patent is an erroneous phrase; what is infringed is a claim, which is the definition of invention, and it is the claim which is the cause of action.
One may appropriate many of the ideas or concepts suggested by specification and drawing, but it is the claim that measures both the patented invention and the infringement thereof. This rule obtains whether the patent be properly spoken of- as great or small, primary or secondary. Walker Pat. (5th Ed.) § 186.
[2] A patentee may describe something that he does not claim or claim that which he has not described; his grant of privilege is construed to cover only that which is both described and claimed, no matter how broad the claim-language may be. Loraine, etc., Co. v. General, etc., Co., 202 Fed. 216, 120 C. C. A. 615; Lovell v. Seybold, etc., Co., 169 Fed. 288, 94 C. C. A. 578. Description may limit a claim, which must always be read in the light of the prior art (a phrase amplified in Pelton v. American, etc., Co., 239 Fed. 321, 152 C. C. A. 308); but it can never expand it (Fowler v. McCrum, etc., Co., 215 Fed. 909, 132 C. C. A. 143). So that a patent (i. e., a claim) can never be given *581a construction broader than its terms in order to cover something which might have been claimed but was not. Universal, etc., Co. v. Sonn, 154 Fed. 665, 83 C. C. A. 422.
[3] The drawings may help out an ambiguous description, but never can they supply the entire absence of any written description of a feature of tíre invention (Windle v. Parks, 134 Fed. 381, 67 C. C. A. 363); the remark relied on by appellee from Columbia, etc., Co. v. Duerr, 184 Fed. 906, 107 C. C. A. 215, must be read in conjunction with the note at the bottom of page 907 of 184 Fed., at page 229 of 107 C. C. A.
[4] Within these limits patents are to be construed liberally “so as to effect their real intent” (Bossert, etc., Co. v. Pratt, etc., Co., 179 Fed. 387, 103 C. C. A. 45); but what their intent is must, be obtained from the specification and measured by the claim, for the present “condition of the patent law * * * leaves no excuse for ambiguous language or vague descriptions” (Merrill v. Yeomans, 94 U. S. 573, 24 L. Ed. 235), although it is not legal ambiguity when “the subject-matter is incapable of exact expression in terms of measurement,” and a skillful man with no measurements given can follow the directions of the patent (Eibel, etc., Co. v. Remington, etc., Co., 234 Fed. 624, 148 C. C. A. 390).
This case as put by the plaintiff is one of appropriation of ideas. That roaj excite sympathy, but no court can do more than discover whether there is infringement of claim.
[8] Without going over the very full prior art, we are of opinion, with the lower court, that the combination defined in the claim in suit and embodied in the device pictured and described in the specification possesses patentable novelty, because to the enumerated old elements is added “a trap connection with said bulb.” The idea of using volatile liquids in bulbs communicating with valves outside the heated chamber and actuating that valve 'by the ebb and flow of the volatile liquid was old, and indeed it is present in defendant’s No. 8 apparatus.
The functional difference between plaintiff’s apparatus as described and defendant’s No. 8 is the prolongation of the tube, pipe, or conduit inside the bulb' with a crooked end at the bottom thereof.
That this is the “trap connection” meant by the claim in suit seems very plain. This arises from the meaning of the word “trap” as well as from a well-known legal rule of construction.
[8] A “trap” by definition is a seal of water or other fluid to prevent the passage of gases through a pipe with which it is connected. That water seals of many kinds are very old is common knowledge, and when such a phrase as “trap connection” is used in a claim founded in the specification before us, it must refer back to the novelty just described.
[7] The legal rule is this: A claim is to be construed so that it may survive rather than perish, and, if the phrase “trap connection” in this claim be used to mean airy kind of positioning of a tube that will hamper or minimize the escape of gases therefrom, then the claim reads directly upon defendant’s No. 8 device, which is admittedly prior art.
[8] With the claim so read as to validate it the question whether de*582fendant’s present device infringes rests upon the contention that defendant’s No. 8 had no conduit, and that to add to it a long-cooled conduit consisting of a flexible tube is in and of itself infringement.
The language of the claim is a “conduit opening into said” vessel, and it is obvious that such claim would be infringed (if read literally) by any conduit no matter how short or long, hot or cold, it might be. We agree with appellant that the construction contended for would read into the claim the phrase, “long-cooled conduit”; which cannot be done. The contention is also opposed to the ordinary meaning of the word “conduit.”
Nowhere does the specification state that patentee’s tube is to be long; doubtless it was to be as long as convenient; but that its length had anything to do with cooling is a concept neither expressed in nor suggested by the specification.
If such concept had been described and its embodiment pictured, the claim would not be bettered under McCarty v. Lehigh, etc., Co., 160 U. S. 110, 16 Sup. Ct. 240, 40 L. Ed. 358. To substantially’insert the necessary words would be going further than to substitute the specification for the claim, and that certainly cannot be done. Safety, etc., Co. v. Gould, etc., Co. (D. C.) 230 Fed. 850. We are driven to the conclusion that the coolness of the connecting tube arising from its length is an afterthought. Whether it is a good one or not makes no difference; it cannot control or change claims as written. McBride v. Kingman (C. C.) 72 Fed. 913, affirmed 97 Fed. 217, 38 C. C. A. 123. Thus the case is one for applying the rule that this function now so fervently urged was not set forth in the specification, whereas another function, i. e., the trap so carefully described, was set forth and claimed; and this “is significant proof that [that] which has not been disclosed by [the patentee] to the public is not his invention.” Electric, etc., Co. v. Gould, etc., Co., 158 Fed. 617, 85 C. C. A. 439.
If the meaning be sought of the word “conduit” as set forth in books of approved authority, the result is the same. That word means a channel, including a pipe or tube for the conveyance of a fluid. That which is conveyed is “conducted,” a verb which has the same root as “conduit.” The short tube of defendant’s No. 8 machine is said to have been but an inch long, yet it conveyed or conducted fluid just as much as if it had been yards in length; the difference is in degree and not in kind. We conclude, therefore, that the claim in suit is valid when read as above indicated, but when so read is not infringed.
The decree below is reversed, and the cause remanded, with directions to dismiss the bill, with costs in both courts, for noninfringement.