ANDERSON, Circuit Judge.
In this patent infringement suit, the court below found no infringement, and dismissed the bill. The patent sued upon is the Jennings patent, No. 15,637, a reissue — on an application filed May 5, 1923, and issued June 26, 1923 — of No. 1,447,854, issued March 6, 1923, on an application filed June 15, 1917.
The patent covers a combination of old elements, and relates to a so-called wet vacuum heating system, through which a partial ■ vacuum is established in the return pipes from the radiators in a steam-heating system. In outline, such a vacuum system consists of a low pressure boiler, in which the steam is developed; pipes connecting the boiler with the radiators; return pipes, trapped at their outlets, to permit the condensed steam, or water, to fiow back; a tank or receiver for the returns of water and air; and pumps to create the requisite vacuum and to force the water back to the boiler.
In the early days of steam heating, the circulation was attained by pressure from the boiler. The vacuum system, by which the steam is drawn, instead of forced, through the ■ pipes, was years ago found to give a more uniform and more easily regulated heat in the radiators, to lessen troubles with leaks in the system, and to be more economical.
In the earlier development of the vacuum type, reciprocating pumps were commonly used, but they were found unsatisfactory. _ The returns are made up of gritty water, mingled with air and other gases. Gritty water is obviously destructive of the ordinary reciprocating pumps; they wear, become noisy, and are expensive to maintain. To avoid these difficulties, various attempts were made to use rotary air and water pumps, but, for various reasons, without complete success, mi til Mr. Jennings made his invention, which was so successful that over 8,100 installations of his apparatus were made in about six years prior to the trial, which began in November, 1924. His claim that it has revolutionized the vacuum steam-heating art is, as the court below found, fairly supported by the evidence.
A sketch of the genesis of the Jennings invention and patent will be a convenient approach to the present problem:
About 1908, Jennings, who had been educated as a mechanical engineer, became connected with the Nash Engineering Company, which was then starting in a small way to develop the inventions of Lewis H. Nash, among which was a rotary air pump or compressor, using water for its pumping action. For this device Nash applied for a patent February 24, 1910; it was granted March 31, 1914, No. 1,091,529. This Nash rotary air pump or compressor is a wheel in an elliptical casing, partly filled with a liquid — in common practice, water, though mercury might probably be used. The wheel is divided by radial plates, thus making buckets or displacement chambers. When the wheel is rotated rapidly, the resultant centrifugal forcé carries the water outward towards the periphery of the wheel, so that, as the wheel passes the longer diameter of the ellipse,’ the inner portions of the opposite displacement chambers are left void, and then, as these displacement chambers approach the smaller diameter of the casing, the water is forced back, tilling the displacement chambers. Inlet ports or .apertures are so placed that air will be drawn [719]*719into the partially emptied chambers as they pass the longer diameter of the ellipse, and be forced out on the other side of the pump through outlet ports, as the chambers fill with water in passing the shorter diameter of the ellipse. The water in these displacement chambers thus operates like a piston in the ordinary pump. The result is that, as the pump whirls very rapidly, the air is drawn in on one side and forced out on the other side, and the pump can be used either as a suction pump or as a compressor for air. On this record, it is not an efficient pump for water.
In 1913 or 1914, Jennings undertook to put this pump to use in a wet vacuum steam-heating system. He found, as his evidence shows, that the chief difficulties with the pumps then in use were that they wore, became noisy, and also required too much power to operate. At about that time electrical motors were coming into common use for such purposes. One of his chief objects was to reduce the amount of power required. Jennings found that, in the returns in a vacuum system, the air was in a. much larger proportion than the water; that the water came intermittently or in slugs, while the air required practically constant pumping. He testifies that, under the methods then in use, the practice was to “discharge both air and water against boiler pressure”; that it occurred to him that, if he could develop a unit which would have the air pump relieved of the discharge head, there would be a substantial resultant economy and efficiency. He accordingly experimented with combining on a single shaft, electrically operated, two pumps— a centrifugal water pump and the Nash air pump. He attained an operative device of this combination in 1915; but he had trouble with the operation of his centrifugal pump, in that it would become “air-bound” when there was no water for it to pump. But in 1917 he overcame this difficulty by connecting the water pump with the tank, some inches above the bottom, so that the tank always retained sufficient priming water. Details of the priming arrangement are not now material. The general result was success. His invention is well described in the language of his original application, filed, as above noted, on June 15, 1917. In the specification he says:
“This invention relates to a wet vacuum pumping apparatus. The object of the invention is to provide an apparatus in which the air and water are separated, and the air exhausted by means of a pump and delivered into the atmosphere; the water being withdrawn by a separate pump and discharged against any desired pressure. Automatic means may be provided whereby the operation of the pumps is controlled according to the vacuum or the quantity of water returned, or by both. * * * Briefly stated, my apparatus comprises a receiver into which the returns of air and water are discharged, the water collecting in the bottom and the air in the upper portion of the receiver. A pumping unit is provided, which comprises a centrifugal water pump and a hydro-turbine air pump. The latter is connected to the top of the receiver, to exhaust the air from the same and discharge it into the atmosphere. The water pump is connected to the lower portion of the receiver, and withdraws the water from the latter and discharges it where desired. Means are provided whereby the operation of the water pump is controlled by the quantity of water in the receiver, and the operation of the air pump is controlled by the air pressure in the receiver.”
After a lengthy description of the apparatus, with references to the accompanying drawings, and a detailed statement of the functions performed by each part of the combination, the specification ends as follows :
“Although I have described a specific arrangement of the receiver, pumps, and piping, it is very apparent that my invention is not limited tp the particular embodiment shown and described, but that the details thereof may be varied within wide limits; the only. requirement being that there shall be a receiver for the returns, a water pump and an air pump each separately taking its respective flwid from the receiver.” (Italics supplied.)
This sentence simply emphasizes what otherwise appears in the drawings and description of the apparatus found in the specification and in the account there given of the method of operating the apparatus, as well as in all the claims made in the original application. In this original application were 11 claims, the first of which may, for present purposes, be taken as typieal of all and is as follows:
“A water and air pumping system, comprising a receiver for air and water, a pump for withdrawing water from said receiver, and a second pump separately connected to said receiver for withdrawing air therefrom.”
In every one of these claims both pumps are referred to as pumping from the receiver — the . centrifugal pump taking the water [720]*720from the receiver to the boiler, and the hydro-turbine or Nash pump the air from the receiver. The original specification, the drawings, and all the claims excluded any pumping of the returns into the tank. Nothing’ could be clearer than that the applicant provided that all the pumping of the returns was to be done after they had flowed into the tank.
This patent had a long and rather stormy passage through the Patent Office. Most aspects of the'matters there in controversy have no bearing* on the present issue's. The Examiner originally disallowed most of the claims, on the Mohn patent for a condensing apparatus, No. 1,005,997. There was an appeal to the Board of Examiners in Chief, and an elaborate argument, mainly on the contention that the substitution of the Nash pump for reciprocating pumps, together with other elements, all old, entering into efficient combination, involved’ invention. The applicant contended that, “in the few years in which the apparatus of the patent * * * has been in use over 5,000 plants have been installed in the' United States alone.” -Presumably the argument of "commercial success went far to convince the Patent Office that' Jennings had made a real invention'.' During these lengthy proceedings, the claims were restated, elaborated, and increased from 11 to. 19, but no material changes were made in the.specification.
We find nothing in the history of the Patent Office proceedings indicating that that office ever considered, or that Jennings ever asked it to consider-, any alleged invention relating to a vacuum heating apparatus, in which both pumps were not connected with the tank and operated substantially as already 'stated, and as was‘Clearly and explicitly described in the specification and in all of the original claims.
When the patent was finally granted, on March 6, 1923, 7 of the 19 claims (the ones now sued upon) did not, in absolutely explicit words, limit the air pump to pumping air from the receiver. All the other 12' claims do, in unmistakable terms, so require. The 7 claims now sued upon are as follows:
“2. The combination, with the separating chamber of a vacuum steam-heating system, of a rotary gas .pump employing water for its pumping action and having a rotor provided with displacement chambers, said pump being connected and operating to exhaust the gas 'from the system by the suction and displacement action of the water in said chambers, and a water pump connected and operating to Withdraw "water of condensation from'the'separating1 Chamber.'
“3. The combination, with the separating chamber of a vacuum steam-heating system, of a rotary gas pump employing water for its pumping action and having a rotor provided with displacement chambers, said pump being connected and operating to exhaust the gas from the system by the suction and displacement action of the water in said chambers, and to retain all the water in the system, and a water pump connected and operating to circulate the water of condensation back to the boiler.
“4. The combination, with the separating chamber of a vacuum steam-heating system, of a rotary gas pump employing water for -its pumping action and having a rotor provided with displacement chambers, said pump be-, trig connected and operating to exhaust the gas from the system by the suction and displacement action of the water in said chambers, connections for supplying the gas pump with water for its operation from the system, and á water pump connected and operating to withdraw water of condensation from the chamber.
“5., The combination, with the separating chamber of a vacuum steam-heating system, of a rotary gas pump connected to the chamber and employing water for its pumping action, and a water pump connected to and operating to withdraw the water of condensa-, tion from the separating chamber, the connee-.' tion between the water pump and the separating chamber being at a higher level than the connection between the gas pump and the separating chamber, whereby the gas pump will contain its operating water, and will continue in operation when the water pump has. exhausted the water in the separating chamber down to the level of its connection therewith. ..
“6. An apparatus for use in connection with the return main or mains of a vacuum steam-heating system, comprising a gas pump for exhausting the gas from the system, a water pump for withdrawing the water of condensation, and connections for relieving the water pump from gas.”
“13. A liquid and gas pumping apparatus, including in combination a receiver for the liquid and gas, a centrifugal pump for withdrawing the liquid from the receiver disposed at a lower elevation than the normal level of water in the receiver, said pump having a liquid pipe connecting its suction chamber with the receiver and a gas pipe affording communication between the top of the suction chamber and the top of the receiver, for the purpose set forth.”
[721]*721“19. A liquid and gas pumping apparatus, including in combination a receiver for liquid and gas, a pumping unit comprising a liquid pump and a gas exhausting pump, the gas pump being of the hydro-turbine type, and the liquid pump being of the turbine type, said unit being disposed at an elevation below the normal water level in the receiver, a conduit extending from the receiver below said water level to the gas pump to supply make-up water, and a gas pipe providing communication between the top of the suction chamber of the liquid pump and the top of the receiver.”
Meantime, Benjamin Skidmore, Jr., of Chicago, on an application filed August 21, .1919, and granted September 13, 1921, obtained a patent on “a fluid-displacing apparatus or rotary pump.” This pump was also in combination put, in 1921, on the market as a wet vacuum pumping apparatus.
The question of infringement, then, arises out of the plaintiff’s present contention that these seven claims, supra, are (when construed in the light of the specification as it reads in the reissue) broad enough to cover the defendant’s apparatus. In the defendant’s apparatus the pump _ producing the vacuum is attached between the returns and the receiver; it therefore pumps from the returns into the receiver all the air and water, the air is then allowed to escape from the receiver into the atmosphere, and by another pump the water is withdrawn from the receiver and forced into the steam boiler. The water is thus pumped twice — both times against pressure.
Passing for the moment the question as to whether the Skidmore pump reads on the claims now in suit,, it is clear enough, that the prospective competition from the Skidmore Company was one, perhaps the controlling, motive in inducing the plaintiff to apply for a reissue of the Jennings patent, in its attempt so to narrow his specification as to permit a broad and liberal construction of the claims now in suit, and to cover such devices as the Skidmore apparatus, in which the pumping of the returns is “from the system,” and is not specifically limited - to pumping from the receiver.
When the plaintiff first learned, in 1921 or 1922, of the prospective Skidmore competition, the Jennings patent had not yet been granted, but it was granted on March 8,1923. Shortly thereafter, on May 15, 1923, Jennings applied for a reissue, which was allowed, apparently pro forma, on June 26, 1923. In the affidavit accompanying Jennings’ application for a reissue he asserts that the patent “[is] possibly inoperative, for the reason that the specification thereof is possibly defective, and that such defect consists particularly in the fact that the paragraph in the specification of said patent, lines 83-92, inclusive, page 4, is possibly defective, in that it is ambiguous, uncertain, and not in accordance with the scope of the claims allowed.”
He therefore asks a reissue, eliminating from the specification the last sentence thereof, supra, containing the statement that “the only requirement being that there shall be a receiver for the returns, a water pump and an air pump, each separately talcing its respective fluid from the receiver,” and making, also, a so-called correction of the statement in lines 57-60 on page 1 of the patent. In the original patent the language there used was:
“A pumping unit is provided, which comprises a centrifugal water pump and hydro-turbine air pump. The latter is connected to the top of the receiver to exhaust the air from the same, and discharge it into the atmosphere. The water pump is connected to the lower portion of the receiver, and withdraws the water from the latter and discharges it where desired.”
In the reissue this was changed to read:
“A pumping unit is provided, which comprises a centrifugal water pump and a hydro-turbine air pump. The latter is connected to exhaust the air and discharge the same where desired.”
No changes were made in the claims.
The obvious — the avowed — purpose of this reissue was to ground an argument that the Jennings patent was broad enough to cover a wot vacuum pumping apparatus like the Skidmore device, in which the air and gas are pumped from the returns into the receiver, and not from the receiver, as in all the Jennings installations.
It is elementary that claims must be construed in the light of the specification and drawings. I. T. S. Rubber Co. v. Essex Co. (C. C. A.) 1 F.(2d) 780, 783; Walker on Patents (5th Ed.) § 186, and cases cited; Bickford Co. v. Merrill (C. C. A.) 268 F. 540, 541; U. S. Corporation v. Safety Car Heating & Lighting Co. (C. C. A.) 261 F. 915.
In effect, the plaintiff’s learned counsel concedes that the claims now in suit, if construed in the light of the specification and drawings in the original patent, do not cover, the Skidmore apparatus.
The plaintiff has no ease, unless the claims can be read broadly, so as to cover pumping of the air. and gas “from this system” and [722]*722without limitation derived from the language of the original specification.
We agree with the court below that the attempt by this reissue thus to broaden the scope of the Jennings patent must fail, and for two reasons:
(1) Even after the deletions, the specification of the reissued patent shows that Jennings never conceived any combination which did not involve connecting the inlets of both pumps to the receiver, and thus putting the receiver into that portion of the apparatus which was under partial vacuum, and pumping the water but once.
We need not support this conclusion by lengthy and detailed analysis of the apparatus, as shown in the specification and drawings, and its method of operation in the reissue, No. 15,637. It is enough to advert to the language on page 1, lines 59-67, immediately following the language changed in the reissue as noted above. It is there stated:
“The water pump is connected to the lower portion of the receiver, and withdraws the water from the latter and discharges it where desired. Means is provided whereby the operation of the water pump is controlled by the quantity of water’in the receiver, and the operation of the air pump is controlled by the air pressure in the receiver.”
Manifestly the control of the air pump by the air pressure in the receiver contemplates that the air was to be pumped from the receiver ; otherwise, there would be no variation of the pressure in the receiver with which to operate the automatic control of the air pump.
Again, on page 3, beginning with line 120, Jennings says:
“In the operation of my apparatus, the first returns through the. pipes 10 flow into the receiver 12, and the water flows from the pipe 49 to supply the air pump.”
In the Skidmore device the returns do not “flow into the receiver.”
Again, page 3, line 129, to page 4, line 5, Jennings sets forth:
“Thus when the switch 53 is operated by the float 54 the pumps will be started, and the centrifugal water pump will immediately discharge water to supply the boiler, and the air pump will operate to exhaust the air from the receiver, thus reducing the pressure within the- latter.”
This, particularly when read in the light of the context and the. drawings, shows that practically all the essential devices of the Jennings apparatus operate on the theory that the returns flow into the receiver and are pumped therefrom. The make-up water, the automatic devices for the stopping and starting the pumps, all are arranged for an apparatus in which the pumping is from the receiver and in no part into the receiver.
Moreover, all drawings and the detailed references thereto show an apparatus identical in every particular (so far as present issues are concerned) with the apparatus disclosed in the plaintiff’s original application of June 15, 1917, in which, as already noted, every one of the 11 claims expressly refers to pumping both water and gas by separate pumps from the receiver.
We cannot read the reissued patent in its entirety without reaching the conclusion that the plaintiff’s combination was in its main features a Nash pump and a centrifugal water pump attached to one shaft, electrically operated, and each connected by their inlets with the receiving tank, so that each pump was “separately taking its respective fluid from the receiver.”
The argument of plaintiff’s learned counsel that this phrase is ambiguous we find not persuasive. The phrase means exactly what it says; it states the very essence of Jennings’ invention.
(2) If the claims in the reissued patent are not to be construed as thus limited, we see no escape from the conclusion that the reissue was invalid.
As already sufficiently indicated, we find it impossible to construe the original patent as covering any wet vacuum pumping apparatus in which both pumps do not draw their respective fluids from the receiver, instead of pumping either gas or water, or both, into the receiver. It is elementary that the statute as to reissue (Rev. St. § 4916 [Comp. St. § 9461]) limits the reissue to the same invention. as that covered by the original patent. Walker on Patents (5th Ed.) § 233; I. T. S. Rubber Co. v. Essex Rubber Co., 1 F.(2d) 780; Mahn v. Harwood, 112 U. S. 354, 5 S. Ct. 174, 6 S. Ct. 451, 28 L. Ed. 665; Parker & Whipple v. Yale Clock Co., 123 U. S. 87, 8 S. Ct. 38, 31 L. Ed. 100; Russell v. Dodge, 93 U. S. 460, 23 L. Ed. 973.
This ease falls flatly under the doctrine thus stated in Russell v. Dodge by Mr. Justice Field, 93 U. S. 463, 23 L. Ed. 973:
“And as a reissue could only be granted for the same invention embraced by the original patent, the specification could not be substantially changed, either by the addition of new matter or the omission of important particulars, so as to enlarge the scope of the invention as originally claimed. * # * The original patent was not inoperative nor invalid from any defective or insufficient speci[723]*723filiation. Tlie description given of the process claimed was, as stated by the patentee, full, clear, and exact, and the claim covered the specification; the one corresponded with the other. The change made in the old specification, by eliminating the necessity of using the fat liquor in a heated condition, and making in the nqw specification its use in that condition a mere matter of convenience, and the insertion of an independent claim for the use of fat liquor in the treatment of leather generally, operated to enlarge the character and scope of the invention.”
If we were to adopt the plaintiff’s present construction of the reissued patent, this language would he applicable; for, as we have already stated, it is impossible to construe the original patent as covering any wet vacuum heating apparatus in which both pumps do not have their inlets in, and draw from, the receiver.
This would leave plaintiff with no patent. But our general conclusion is in accord with that of the court below — that the reissue was for the same invention, and is to he construed as not covering the defendant’s device.
While these differences between the two devices are, as the court below held, enough to dispose of the case, it may be well to add that, on the weight of the evidence, the Skid-more pump is a different device — not merely in appearance but in method of operation— from the plaintiff’s Nash pump. It is unnecessary to attempt a detailed description of the Skidmore pump, concerning which so experienced and learned a patent expert as Mr. Joseph P. Livermore, called by the plaintiff, said:
“I find that pump very difficult to explain. I think I can understand it, but I do not know as I can make anybody else understand it.”
Doubtless in its operation it uses water, and is, in that aspect, like the .Nash pump. But on this record the Skidmore pump does operate effectively in pumping both the air and the water from the returns into the receiver. It is at least doubtful whether the plaintiff’s Nash pump would function, if put in the same position, to perform the same work as the defendant imposes on the Skid-more pump. At any rate, the plaintiff’s own testimony and that of his expert, Livermore, is explicit to the effect that, if the Nash pump were “hooked up” to pump the returns from the pipes into the receiver, as does the Skid-more pump, “it would be very inferior to the present arrangement.”
Moreover, to pump water twice, when once is sufficient, is an obviously wasteful process; and one of the merits, soundly grounded, claimed by Jennings for his apparatus, was reduction in the amount of power required to produce the desired vacuum in the system, as well as to put the water from the returns, without appreciable loss, back into the steam boiler.
It follows that, however the situation is viewed, the Skidmore apparatus is different in design, in construction, and in operation, from the apparatus described in the plaintiff’s patent.
The decree of the District Court is affirmed, with costs to the appellees.