BOOTH, Circuit Judge.

This is a suit for alleged infringement by appellee (hereafter called defendant) of patent No. 1,345,870, granted to Robert Nash and George Henry Alexander on July 6, 1920. In the trial court decree went for defendant, on the ground of noninfringement. Validity of plaintiff's patent was not adjudicated.

The complaint is in conventional form, alleging ownership of the patent, infringement thereof by the defendant, and praying for an injunction and an accounting. The answer denies infringement, and sets up invalidity of the patent by reason of certain prior patents alleged to be anticipations, also by reason of lack of patentable novelty in view of the prior art.

Plaintiff's Patent and Its Scope.

The patent is for a "fluid pressure measuring instrument," and the specifications state that the parties have invented certain new and useful improvements relating thereto. Such instruments are old in the art. They find wide application at the present time in indicating the depth or quantity of gasoline in the tank of an automobile, and the devices of both plaintiff and defendant are used for that purpose — the indicator of the device being placed on the dash of the automobile.

The specifications of the patent in suit recite:

"In its rudimentary form such apparatus when employed, for example, for indicating the level of fluid in a tank, consists of a vessel (herein termed a displacement chamber) containing entrapped air or other gas, said vessel being in communication at its lower end with the lower end of the tank, and at its upper end with one end of a U-tube containing the indicating liquid. When the air in the tank is under pressure, the other end of the U-tube is connected to the air space in the tank, otherwise it is open to the atmosphere."

Figure 1 of the drawings is here reproduced:

The underlying principle of such devices in general is that the pressure exerted by a body of liquid upon a body of air entrapped in a bell or displacement chamber varies in direct proportion to the depth to which the bell is immersed in the liquid. A change of the depth will be reflected in a change of pressure. Referring to the figure, it is apparent that when the tank a is filled with gasoline, some of it will pass through the pipe c into the displacement chamber b, and rising therein, will entrap a volume of air. The pressure of the gasoline in the tank being communicated to the entrapped air will be further communicated through the pipe g to the liquid in the U-tube or gauge, depressing the liquid in the left leg, and causing it to rise in the right.

The gauge being so constructed as to indicate the difference in the levels of the gasoline in the tank and the gasoline in the displacement chamber, it is apparent that the real depth in the gasoline tank can be ascertained correctly from time to time by this method, only if the level in the displacement chamber remains constant. The level of the gasoline in the displacement chamber is called the "datum level," and, unless this datum level is constant, the reading on the gauge will necessarily be incorrect. The constancy of this datum level depends upon the uniformity of the volume of entrapped air in the displacement chamber. The object of the invention of the patent in suit, is to enable a practically constant datum level to be maintained, and to be conveniently restored, when occasion requires, in a simple and convenient manner. The action of plaintiff's device is thus described in the specifications:

"To compensate for absorption or leakage of air from the displacement chamber air can be blown in through the pipe e periodically and restore the original datum level. Or air can be blown into the chamber through a cock i. Excess air then escapes through the pipe e. The latter thus serves both in determining and in restoring the datum level. It is also to be noted that the inner open end of the pipe e is flush with the normal datum level of the chamber b so that in case the level falls below normal it will be automatically returned to or constantly maintained normal by liquid flowing into the chamber from the tank a while, if the datum level rises above normal, air may be blown into the pipe e to restore the original datum level."

The patent has but one claim. It reads as follows:

"In fluid pressure measuring instruments, the combination comprising a closed displacement chamber in communication with a body of fluid and containing a quantity of entrapped air, the height of the air space above the fluid datum level in the chamber being small relatively to the diameter of the chamber, a fluid gauge communicating at the end with the air space in the displacement chamber, and a pipe in communication with the fluid in said chamber, said pipe serving to determine the datum level and also as a passage for air to or from the chamber for restoring such level when required, substantially as described."

The elements of the claim are:

(1) A closed displacement chamber ( b) in communication with a body of fluid ( a) and containing a quantity of entrapped air, the height of the air space above the fluid datum level in the chamber being small relatively to the diameter of the chamber;

(2) A fluid gauge ( h) communicating at one end with the air space in the displacement chamber; and

(3) A pipe ( e) in communication with the fluid in said chamber, said pipe serving to determine the datum level and also as a passage for air to or from the chamber for re-restoring such level when required, substantially as described.

This claim being a combination claim, and no one of the elements thereof being claimed as an invention, all of such elements are conclusively presumed either to be old in the art or not patentable. Richards v. Chase Elevator Co., 159 U.S. 477, 486, 16 S. Ct. 53, 40 L. Ed. 225; Hay v. Heath Cycle Co. (C.C.A.) 71 F. 411, 413; Campbell v. Conde Implement Co. (C.C.) 74 F. 745; Johnson Automobile Lock Co. v. Noser, etc., Co., 9 F.2d 265 (C.C.A. 8). It follows as a matter of law that the contention of plaintiff that the vent pipe e in his patent is not found in the prior art, and is a patentable addition to the combination, cannot be sustained; nor is it true as a matter of fact that the vent tube or its equivalent is not found in the prior art. Such tube is found in the Murphy patent, 867,994; the equivalent is found in the French patent to Remy, 453,970, which has a three-way cock; the tube is also found in the Payne patent (British), 14,811, and in Parks, 1,131,412 — although in these various patents the arrangement of the tube and the functions which it performs are not the same as in plaintiff's patent. Plaintiff's contribution, if any, to the prior art, was in the particular arrangement of the old elements and in the proportionate size thereof; for example, the first element has the displacement chamber so constructed that "the height of the air space above the fluid datum level in the chamber [is] small relatively to the diameter of the chamber," and in the third element the vent pipe e is so arranged in size and in position that it serves "to determine the datum level, and also as a passage for air to or from the chamber for restoring such level when required."

In our opinion, these words which we have quoted from the claim itself are not words of mere description, as contended by plaintiff, but are also words of limitation. We think this is shown by the language of the specifications, and also by the disclosure of the file wrapper. The specifications contain the following statement as to the displacement chamber:

"Some arrangements hitherto proposed have been of little or no commercial value because the length or height of the displacement chamber has been made greater than the diameter. This involves comparatively large fluctuations of level (herein termed the datum level) in the displacement chamber and in consequence, especially when the air in the tank is put under variable pressure, it becomes impossible to accurately calibrate the U-tube in a manner which will suit all levels and pressures in the tank."

And as to the vent pipe:

"The invention comprises the combination with the displacement chamber, of a pipe which serves both to determine the datum level and also as a passage for air to or from the chamber for restoring such level when required."

The file wrapper discloses that after the claim had been rejected upon the reference to Murphy, 867,994, and to Rohm, 1,195,736, the attorney for the inventor wrote to the Commissioner of Patents as follows:

"The specification has also been amended to bring out more fully the advantages of the pipe e, which, as pointed out to the examiner in a recent interview, is not shown by the references now in the case. For this reason the original claim 1 is submitted for reconsideration, and it is hoped that unless there are other structures in the prior art showing the subject-matter of this claim, that the claim will be allowed. It is to be noted particularly that Murphy does not show the arrangement set forth in the specification regarding the advantage of pipe e and upon which point allowance of the claim is requested."

And thereupon the specifications were modified by inserting a statement as follows:

"It is also to be noted that the inner open end of the pipe e is flush with the normal datum level of the chamber b, so that in case the level falls below normal it will be automatically returned to, or constantly maintained normal, by liquid flowing into the chamber from the tank a, while, if the datum level rises above normal, air may be blown into the pipe e to restore the original datum level."

These excerpts from the file wrapper, we think, show that it was the arrangement of the pipe e; that is, its positioning relative to the other elements, that was stressed as the important factor. This vent pipe e was to be so positioned that it would perform three functions: determine the datum level; act as a passage of air to the displacement chamber; act as a passage of air from the displacement chamber.

We think that, when an inventor in a crowded art has used limitative language in setting forth his claim, such language should not be rejected as surplusage, but should be considered as narrowing the patent. The trial court construed the claim in the way we have indicated, and we think properly so. Keystone Bridge Co. v. Phoenix Iron Co., 95 U.S. 274, 278, 24 L. Ed. 344; McClain v. Ortmayer, 141 U.S. 419, 425, 12 S. Ct. 76, 35 L. Ed. 800; Weber Elec. Co. v. Freeman Elec. Co., 256 U.S. 668, 678; Adams Elec. Ry. Co. v. Lindell R. Co., 77 F. 432, 449 (C.C.A. 8); I.T.S. Rubber Co. v. Essex Rubber Co., 272 U.S. 429, 47 S. Ct. 136, 71 L. Ed. ___; Deubener v. R. Brownson Co., 20 F.2d 324 (C.C.A. 8).

In Stirrat v. Excelsior Mfg. Co., 61 F. 980, 984, this court said:

"The statute requires the inventor to `particularly point out and distinctly claim the part, improvement, or combination which he claims as his discovery.' Rev. St. § 4888. When, under this statute, the inventor has done this, he has thereby disclaimed and dedicated to the public all other improvements and combinations apparent from his specifications and claims that are not evasions of the device and combination he claims as his own. The claims of his patent limit his exclusive privileges, and his specifications may be referred to to explain and to restrict, but never to expand, them."

In Lakewood Engineering Co. v. Stein (C.C.A.) 8 F.2d 713, 715, the court said:

"An inventor may choose his own language in describing his invention and in stating his claims, but that description is the basis of the government's action in granting him patent monopoly, and must be construed in accordance with its plain, usual, and ordinary meaning, the same as if written into any other contract. * * * When a claim is clear and distinct, a patentee may not go beyond the words of his contract for the purpose of establishing infringement, and the range of equivalents must be measured by what is both described and claimed."

That the claim of a patent may be limited by words therein, attributing a special function to one of the elements, by reason of its form or position; and that an accused device will not be held to infringe, though it has a similar element, if that element does not and cannot produce that special function, see Gruendler Co. v. Hussman Co., 16 F.2d 571 (C.C.A. 8).

Does the Defendant's Device Infringe?

Defendant's Exhibit 9, here reproduced, may be taken as illustrating defendant's device:

It is true that defendant's device has three elements of similar character to those in plaintiff's device; but it is to be noted that the first element (the displacement chamber) of defendant's device does not fulfill the requirement of the first element of plaintiff's device, viz. "the height of the air space above the fluid datum level being small relatively to the diameter of the chamber," this being one of the essentials of the first element of the claim of plaintiff's patent, and the evidence shows that the third element (the vent pipe) of defendant's device does not fulfill the requirement of element 3 of plaintiff's device, viz.: "Said pipe serving to determine the datum level and also as a passage for air to or from the chamber for restoring such level when required," — these being essentials of the third element of the claim of plaintiff's patent. The datum level is determined in defendant's device by the small hole, known as the "air inlet," in the plate between the top bell and the bottom bell of the displacement chamber, in connection with the vacuum tank which operates to suck a supply of gasoline from the main tank to the vacuum tank.

Defendant's device is intended for use in automobiles employing a so-called "vacuum feed." When thus installed, the action of the device, as shown by the evidence, and which may be more readily understood by reference to the figure, Defendant's Exhibit 9, is as follows:

A vacuum is created in the vacuum tank intermittently, and gasoline is thereby drawn from the main tank of the automobile through the gasoline tube, the open end of which projects into the bottom bell of the displacement chamber. The air pipe has its upper open end near the upper part of the gasoline tank, above the surface of the gasoline, and its lower open end near the bottom of the bottom bell. When the vacuum tank operates to suck gasoline from the main tank to the vacuum tank, air is sucked down through the air pipe into the bottom bell. Some of this air travels along the under side of the bell plate, and if the supply of air in the top bell has become depleted by absorption or otherwise, so that some of the gasoline has passed into this top bell through the opening in the bell plate, some of this air passing along the under side of the bell plate will pass through the opening into the top bell and force the gasoline out of the top bell, thus maintaining the supply of air in the top bell and in the connection leading to the U-tube. Surplus air drawn into the bottom bell passes through the wire gauze at the edge of the bell and bubbles up through the gasoline in the main tank.

Defendant's vent pipe, according to the testimony of Professor King, one of the witnesses for defendant, does not operate to determine the datum level. It does not act as a passage of air to or from the displacement chamber; but it is a mere safety device to protect the gauge from having its liquid either blown out at one end or sucked out at the other. If for any reason, such as filling the gasoline tank very rapidly, gasoline should be forced through the air inlet into the upper bell of the displacement chamber, it would, except for the defendant's vent tube, increase the air pressure in the air tube leading to the gauge and push the liquid out of the gauge; but owing to the presence of the vent tube, and owing to the fact that it has a cross-sectional area very much larger than that of the air inlet, this gasoline, being pushed into the upper bell of the displacement chamber, goes out through the vent tube, thus allowing the air in the upper chamber and in the air tube to remain constant. The result is that though a vent tube is present in the combination of defendant's device, yet, inasmuch as it performs neither the function of fixing the datum level, nor of allowing air to pass into or out of the upper bell of the displacement chamber, this vent tube element is not co-operatively present in producing the same result as is produced by plaintiff's device. Inasmuch as it is not co-operatively present in producing the same result, there is no infringement, since the test of infringement of a combination claim is that the accused device must employ the same elements, they must produce, when functioning, the same result, and must produce it in substantially the same way as the device of the plaintiff's patent.

The accused device in the case at bar lacks "that identity of means and identity of operation which must be combined with identity of result to constitute infringement." Kokomo Fence Machine Co. v. Kitselman, 189 U.S. 8, 24, 23 S. Ct. 521, 47 L. Ed. 689; Werner v. King, 96 U.S. 218, 229, 24 L. Ed. 613; Anakin Lock Works v. Dillon Lock Works, 292 F. 45, 48 (C.C.A. 8); Johnson Automobile Lock Co. v. Noser, etc., Co., supra.

Furthermore, the function which the vent tube in defendant's device does perform is one which the vent tube in the plaintiff's device cannot perform. This is stated by the witness King, and the reason given is that the position of the vent tube in plaintiff's device, and the very insufficient size of the vent tube as compared with the pipe c prevents it from performing the function which is actually performed by the vent tube in defendant's device. Our conclusion is that plaintiff's patent is not infringed by defendant's device.

As this conclusion on the question of infringement disposes of the case, we do not deem it necessary or advisable to discuss the validity of plaintiff's patent. It is sufficient to say that we have assumed it to be valid, if construed with the limitations that we have pointed out.

Decree affirmed.