delivered the opinion of the court:
This is an appeal from the decision of the Board of Appeals of the United States Patent Office affirming the decision of the Examiner of Interferences awarding priority of the subject matter defined in the single count in issue to appellee, Ernest H. Shaff.
The invention in issue relates to a carburetor for use on automobile engines, and more particularly to means whereby the richness of the mixture of fuel and air produced by the carburetor may be automatically controlled in accordance with the temperature of the engine and the suction produced by the engine during its operation.
The interference is'between' appellant’s' patent No. 1^915^851-; issued June 27, 1933, on an application, No. 573,418, filed November 6, 1931, and appellee’s application No. 435,394, filed March 13, 1930.
The count in issue, which originated in appellant’s patent, sufficiently describes the invention. It reads:
1. In a carburetor, means forming a mixing conduit, said conduit including an air inlet, a fuel inlet, and a mixture outlet, a manually operated valve for said conduit, a valve for controlling the flow through one of said inlets to enrich or lean out the fuel mixture discharged by said carburetor, a heat responsive device capable of operation independent of suction for operating-said valve in accordance with the temperature to lean out the mixture upon an increase of temperature, and means for operating said valve to lean out said mixture upon an increase in suction in the carburetor, said last-named *1351means including a suction conduit connected to a point in the mixing conduit posterior to said manually operated valve.
Appellant is the junior party and the burden was upon him to establish priority of invention by a'preponderance of the evidence.
During the motion period, appellant moved to shift the burden of proof on the ground that the invention defined by the appealed count was disclosed in his patent No. 1,872,708, issued August 23, 1932, on an application, No. 358,700, filed April 27, 1929.
The question presented by the motion to shift the burden of proof was whether appellant’s prior application, which matured into patent No. 1,872,708, as originally.filed, disclosed,the last element defined in the appealed count; that is, “means for operating said valve to lean out said mixture upon an increase in suction in the carburetor, said last-named means including a suction conduit connected to a point in the mixing conduit posterior to said manually operated valve.”
That application, as originally filed, disclosed a thermostatic valve which was operatively responsive to heat. It was silent, however, as to whether the valve was “moved by suction towards closing position when the suction is increased”; that is, whether the valve was capable of operation by suction as well as heat.
In his original decision, the Primary Examiner stated that—
Ttois valve is very close to the fuel orifice and it is believed that the static suction as well as the suction created due to the flow, of the liquid will act on the valve with an appreciable force. This force would necessarily bend the thermostat more or less towards the orifice to close the latter, and in view of the length of the thermostat and the resulting leverage the effect of suction on the valve of patent 1,872,708 is believed to be appreciable also. This being an inferred function of the Ericson structure as originally disclosed, it is believed to be immaterial that the specification as originally filed did not specifically point out this feature.
The examiner concluded, however, as stated in his second decision made in response to a motion by appellant for reconsideration of his original decision, that the last element defined in the count called for “some kind of a suction actuator such as a piston, diaphragm, or the like [not disclosed in appellant’s application], * * * and a mere conduit for communicating suction to the valve proper [which was disclosed] will not answer the requirements of the count,” and, accordingly, overruled appellant’s motion to shift the burden of proof.
Appellant also presented that issue to the Examiner of Interferences, and, on appeal, to the Board of Appeals. Each of those tribunals held that appellant’s prior application (No. 358,700, which matured into patent No. 1,872,708) did not disclose the invention defined by the count in issue. Their reason for so holding, however, *1352was somewhat different from that relied upon by the Primary Examiner.
In his original decision, the Examiner of Interferences discussed at length the issue of shifting the burden of proof. We quote from that decision:
The earlier Erieson application as filed does not mention that the thermostatic number SO is suction actuated. The disclosure of this application was not amended until October 19, 1981 to indicate that the member 50 was moved by strong suction. However, even though this amendment was held to establish a conception of the invention on behalf of the party Erieson, he would not prevail, inasmuch as the date thereof is subsequent to the filing date (March 13, 1930) of the senior party Shaft While it is disclosed in the earlier Erieson application as filed that strong suction is present, the term “strong” is merely relative and of course is not indicative of the numerical value thereof. Moreover, there is no disclosure indicating the desired thickness, width, stiffness and length (or the size of jet hole) of the thermostatic material so as to show that a material was desired which would be sufficiently flexible. Even by referring to the drawing it is not clear that the thickness, width and length of the thermostatic element is such as to inherently render it movable by “strong suction.” The drawing, even though it appears from the objection on page 135 (Erieson Record) that it is not contended that it was drawn to scale, shows element 50 as being rather short, thick and wide, so as not to be clea/rly capable of being moved by suction. Of course if this element were not drawn to scale it cannot be seen how such a representation in the drawing could be seriously relied upon, when it might possibly be of larger in’oportions. It is therefore held that the drawing and disclosure of the earlier Erieson application does not show, describe or suggest a thermostatic member which is inherently capable of being moved by engine suction. [Italics ours.]
In case it be decided on appeal that the earlier Erieson application inherently discloses that the thermostatic member is actuated by engine suction it is necessary to determine whether this application also discloses a “means for operating said valve * * * upon an increase in suction in the carburetor, said last-named means including a suction conduit.” Assuming that the member 50 is so constructed that it could be actuated by suction, then the engine creating suction together with the passages leading from the engine through the carburetor and the conduits to deliver said suction to member 50 through jet 49 is believed to reasonably satisfy this means element of the count, since the count not only requires that the desired function be obtained but the last clause of the count specifically recites that a suction conduit is to be included.
It is therefore held that movement of thermostatic member 50 by suction is not inherent to the original disclosure of the Erieson application per se.
It appears from the record that considerable evidence was introduced by appellant for the purpose of establishing that the suction operation called for by the appealed count was inherent in the disclosure in appellant’s earlier application — No. 358,100, now patent No. 1,872,108.
It may be said at this point that appellant admitted in his testimony that, at the time he filed his earlier application, he did not know that the thermostatic valve disclosed in that application was responsive *1353to suction, and that, at that time, he did not know how to proportion the parts of the valve to make it responsive to suction.
It is contended here, however, by counsel for appellant, as it was before the Examiner of Interferences and the Board of Appeals, that appellant has established by testimony, physical exhibits, and documentary evidence not only that the involved invention was inherent in his first application, but that he had actually reduced the invention to practice during the year 1929, prior to the date (January 11, 1930) to which, it is conceded by counsel for appellant, appellee is entitled for conception of the invention.
The evidence referred to by counsel for appellant, with the exception of one exhibit (No. 55) which will be hereinafter referred to, was fully discussed by the Examiner of Interferences in his original decision, and, to some extent, was again reviewed in a second decision made pursuant to a motion by appellant for reconsideration of the original decision, and was held insufficient to establish either that the invention defined by the appealed count was inherent in the original disclosure in appellant’s application No. 358,700 (now patent No. 1,872,708), or that appellant had conceived and reduced the invention to practice during the year 1929 or at any time prior to the filing of appellee’s application — March 13, 1930.
It will be observed that the Examiner of Interferences called attention to the fact that appellant’s application No. 358,700 was amended on October 19, 1931 to indicate that “the [thermostatic] member 50 was moved by strong suction.”
It appears from the decision of the Board of Appeals that appellant’s first filed application was amended October 19, 1931-, about two weeks before the filing (November 6, 1931) of appellant’s application No. 573,418, which matured into the patent here involved (No. 1,915,-851) on June 27,1933.
Relative to the amendment to appellant’s first filed application (No. 358,700, now patent No. 1,872,708) the Board of Appeals said:
Referring to the showing in application No. 358,700 of Ericson which matured into the patent 1,872,708, we find this disclosure is directed to a thermostatic valve control for fuel inlet in a carburetor. As originally filed, the application said nothing of a control of this valve by the air suction in the carburetor. It is true that on page 7 of the specification is described a strong suction transmitted through the venturi 28 and the idling tube. Daring the prosecution of this application, in an amendment filed October 19, 19S1, the specification was amplified to describe the effect of this strong suction on the thermostatic member SO to cause it to be draitm to closed position and restrict the flow of fuel into the mixing conduit.
* * * * * * *
The examiner has discussed the testimony offered in an attempt to show that during 1929 there was discussion of the thermostatically operated valve disclosed by Ericson in his early application and some of these witnesses seem *1354to think such valve was operated, by suction as well as through changes in temperature. We are in agreement with the examiner that the proofs are insufficient. The fact that Ericson himself was not aware of the suction effect when he filed his early application in 1929 seems to us good evidence that others less interested in the development could hardly have had more knowledge as to the operation of this fuel control valve. None of the documentary evidence during this period referred to the suction action on the valves.
$ $ $ ‡ $ ijc
We have not overlooked the testimony that Ericson purchased very thin material for his thermostatic element in 1929, but inasmuch as the thinness of the material is not the only determining factor as to operation by suction, we cannot conclude that his thermostatically operated valve formed of this thin material would inherently be suction operated. [Italics ours.]
The board further stated that appellant’s application No. 446,899, filed April 24, 1930, which matured into patent No. 1,915,852, June '27,1933, was amended October 19,1931. It does not appear from the board’s decision, however, of what that amendment consisted. Relative to that patent the board said:
While Ericson had a disclosure in Patent.No. 1,915,852 (application filed April 24, 1930) which might support the count, yet as this application was not filed prior to the filing date of Shaff this matter need not be considered as it would not help Ericson in a showing of priority.
The board was of opinion that appellant’s application No. 358,700 (now patent No. 1,872,708), as originally filed, did not disclose the involved invention, that the evidence introduced by appellant was not sufficient to warrant a holding that it did, and affirmed' the decision of the Examiner of Interferences awarding priority of invention to appellee.
It is contended here by counsel for appellant that the evidence of record does-not establish what amendments, if any, were made on October 31, 1931 to appellant’s application No. 358,700, filed April 27, 1929, now patent No. 1,872,708, and that, therefore, the tribunals of the Patent Office had no authority to examine the file history for ' the purpose of determining what amendments, if any, were made to that application; that the holding of the Patent Office tribunals that the amendments to that application involved new matter was based upon evidence not of record in the case; and that this court should consider the disclosure in the patent without regard to any amendments which may have been made to the application as originally filed.
The question as to whether appellant disclosed the involved invention in his application No. 358,700, as originally filed, was raised by his motion to shift the burden of proof. In order to determine that question, it was, of course, proper for the tribunals of the Patent Office to determine whether there were amendments to that application, and, if so, whether they involved new matter. We hold, therefore, that *1355tbe tribunals of tbe Patent Office not only bad tbe right, but that it was their duty, to examine appellant’s application No. 358,700, and tbe •amendments thereto, in order to determine whether, as originally bled, it disclosed the involved invention.
In view of the fact, however, that neither the application, as originally filed, nor the amendments subsequently made thereto were made ¡a part of the record, and, therefore, are not before us, we must accept ■the statements contained in the decisions of the tribunals of the Patent Office as to the disclosure in the original application, as well as to the subject matter of the amendments subsequently made thereto.
It is further, contended by counsel' for appellant that the evidence of record establishes that appellant conceived the invention and actually reduced it to practice sometime during the year 1929, but that, should it be held that that evidence is insufficient to establish that appellant actually reduced the invention to practice during that period, it is, nevertheless, sufficient to establish conception by appellant as early as June 1929 or December 31, 1929, and that he was diligent in reducing the invention to practice at the time appellee entered the field (January 11, 1930), and continuously thereafter, until April 24, 1930, when appellant filed his application No. 446,899 (now patent No. 1,915,852); that that patent discloses the involved invention; that appellant is entitled to the date of the application upon which that patent issued for constructive reduction to practice; that he was the first to conceive the invention; and that, as he was diligent until he filed his application No. 446,899, April 24, 1930, he is entitled to an award of priority.
It will be observed that the Board of Appeals stated in its decision that application No. 446,899, filed April 24, 1930, was amended on October 31, 1931, the date on which appellant amended his first filed application — No. 358,700.
Application No. 446,899, as originally filed, was not made a part of the record. Accordingly, as neither of the tribunals of the Patent Office held that, as originally filed, that application disclosed the invention and as it appears from the decision of the Board of Appeals that it was amended (to what effect and for what purpose is not stated in the board’s decision), we are unable to hold that appellant is entitled to the filing date of that application for constructive reduction to practice.
In his decision, the Examiner of Interferences reviewed the testimony of appellant’s witnesses at some length, and concluded that it was insufficient to warrant a holding that appellant had either conceived or reduced the invention to practice prior to appellee’s filing date — March 13,1930.
*1356Those of appellant’s witnesses upon whose testimony appellant relies were employees of appellant’s assignee — Carter Carburetor Corporation. They were asked whether they had seen one or more carburetors like that disclosed in patent No. 1,872,708, and some testified that they had seen such a structure in 1929.
As illustrative of the character of testimony introduced by appellant in an effort to establish conception and actual reduction to practice of a carburetor conforming to the count in issue, we quote from the testimony of George M. Bicknell, who had been employed by the Carter Carburetor Corporation for twenty-four years, and, at the time of the giving of his testimony, was chief engineer of that corporation :
Q. 9. Did you ever have any thermostats made for carburetors as set forth in patent #1,872,708?
A. Yes.
Q. 10. Have you any drawings of such devices?
A. I know we have a sketch of the exact type of thermostat shown here. We have a drawing similar to and equivalent to the thermostat shown in the patent dated 1BS1-S9. [Italics ours.]
The witness, whose testimony was not corroborated either by appellant or any other witness, stated that the drawing, dated December 31, 1929, which was introduced in evidence as appellant’s Exhibit 55, was made by Tom McGann, a draftsman; that he recognized Mr. McGann’s handwriting on the exhibit; and that the exhibit was made pursuant to instructions from him. He then testified as follows:
Q. 20. What thickness of metal was used or indicated in this drawing for the thermostat?
A. .017.
Q. 21. Would you say that in view of the patent #1,872,708 and your knowledge as an engineer that you selected this material and thickness for the construction ?
A. No. We wired the Dole Valve Company of Chicago, and asked them to furnish us thermostat metal in .015 and they did not have it but they had metal in .017 and they sent us the samples out of that metal.
Q. 22. Then, you would have used .015 metal if it had been immediately available?
A. Yes, that is what we asked for.
Q. 23. Did you ever make a carburetor with the thermostat as shown in that drawing, or have it made?
A. I believe we did, either exactly as shown in this drawing or similar to it. I am quite sure we made samples exactly like that.
Q. 24. Did you have tests made of such carburetors?
A. Yes.
Q. 25. Did you find, as a result of such tests, that the thermostat operated in response to heat and suction to lean out the mixture up to a certain point?
A. Yes, it did.
Q. 26. Did you ever drive an automobile with that or a similar device on it’
A. I did.
*1357Q. 27. Did it operate to your satisfaction?
A. Very well.
Exhibit 55 does not contain any statement to the effect that the thermostatic valve would be operatively responsive to suction, and there is no testimony of record as to when carburetors conforming to the disclosure in that exhibit were constructed.
Appellant’s witness Morris Brown, employed by the Carter Carburetor Corporation, whose work at the time he testified was “Experimental work and calibration of carburetors with different passenger car motors,” stated that he was familiar with the “construction and operation of modern carburetors.” When asked if he was familiar with the carburetor shown in appellant’s patent No. 1,872,708, he replied that he was, and that he had seen such a carburetor sometime in 1929. He testified that the thermostat in that carburetor was “anywhere from .015 to .030” in “thickness”; that during 1929 thermostats of different thicknesses were tested by him and the witness Maclaran Sawyer, whose testimony was stipulated. The stipulated testimony-does not.add anything to Brown’s testimony.
The witness Brown after referring to certain tests made with carburetors conforming to the disclosure in appellant’s patent No. 1,872,708, testified as follows:
Q. 94. Did you make any tests which would have enabled you to say whether the thermostat was operated by suction?
A. No. The suction factor so far as I was concerned wasn’t known to me.
Counsel for appellant also rely upon the testimony of appellant’s witness Leonard D. Boyce for proof of the construction and successful operation in 1929 of a carburetor conforming to the count in issue.
The witness Boyce testified that he was an engineer in the employ of the Carter Carburetor Corporation, and that his duties consisted of calibrating carburetors for the various Chrysler made products. He was handed a copy of appellant’s patent No. 1,872,708, and was asked to state “just what the invention consists of.” In reply, he stated:
It consists of thermostatic bi-metal attached to the main jet in such a way that when the temperature is low, one end of the bi-metal will move away from the orifice. If the temperature goes sufficiently high, the bi-metal will move toward the orifice, closing it except for a small hole as .indicated in this patent drawing as 59. In other words have the effect of reducing the size of the orifice. [Italics ours.]
We quote further from the testimony of the witness:
Q. 17. The thermostatic element is indicated at 50 in that patent and in your answer to the previous question you have referred to the operation of this element by means of variations in temperature. Is the variation in temperature the only operation mentioned for that element? If not, please state how it may operate otherwise?
*1358A. The answer is no. It is possible if the metal or if the orifice 59 is brought close to the jet, the suction on this jet loill camse that bi-metal to go still closer even though the temperature is not high enough to hold, it against the jet.
Q. 18. To your knowledge, was any carburetor or carburetors built corresponding to the disclosure in that patent 1,87$,708?
A. I have seen one or more of such carburetors in the plant.
Q. 19. State if you know whether there was ever such a carburetor applied to an automobile and if so at approximately what time?
A. I installed one of these carburetors on a four-cylinder Plymouth ear in the summer of 1929.
Q. 20. State if you know whether Mr. Ericson owned an automobile in the early part of 1929 and state what kind of a car it was ?
A. He had an Oakland.
Q. 21. Do you know whether a carburetor corresponding to patent 1,87$,708 was ever applied to the Oakland? .
A. Yes, I was told by Mr. Ericson that he had one on the car on a visit that he made to my home in the summer of 1929.
Q. What do you mean by the summer of 1929?
A. Around May or June. [Italics ours.]
The witness also testified that at that time (May or June 1929) appellant “insisted that the thermostatic metal, regardless of whether the temperature was high enough to close off the jet, or not close it had given him leaner mixture as he increased the speed of the car”; that, at that time, he disagreed with appellant, but that since “seeing some flow curves” he agreed with him; that he had driven the Oakland car referred to in his testimony; that it operated satisfactorily, but that he did not make any tests to determine how the carburetor functioned.
There is other testimony, given entirely from memory by some of appellant’s witnesses, to the effect tliat a carburetor was constructed . in 1929 in conformity with appellant’s patent No. 1,872,708, and that the thermostatic element in that carburetor was, made of very thin material. We are in agreement, however, with the conclusion reached by the tribunals of the Patent Office that that testimony is wholly insufficient to establish either conception or actual reduction to practice of the involved invention at any time prior to the time appellee filed his application (March 13, 1980), and, for the reasons herein-before stated, we are unable to hold that the Examiner of Interferences and the Board of Appeals erred in holding that appellant’s application No. 358,700 (now patent No. 1,872,708), as originally filed, did not disclose the involved invention.
Accordingly, the decision of the Board of Appeals affirming the decision of the Examiner of Interferences awarding priority of invention to appellee is affirmed.