SMITH, Judge (writing for the majority in part and dissenting in part).

The rejection of appealed claims 1, 2, 3 and 81 under 35 U.S.C. § 103 is predicated on two references:

  British Patent     614,386     Dec. 15, 1948
    (hereinafter
    "British")
  Wheeler          2,812,304     Nov. 5, 1957

The invention defined by the appealed claims relates to an apparatus (claims 1, 2, 3) and a method (claim 8) for producing power from controlled nuclear fission. The claimed apparatus consists essentially of two members rotatable relative to each other, which are designated as the "rotor" and the "stator," on each of which is mounted a subcritical fissionable mass. The masses are so arranged that, during rotation, one of the fissionable masses is moved tangentially with respect to the other. The separate subcritical fissionable masses are so selected that when such movement occurs, they are brought into what is termed "juxtaposite position," the combined masses become hypercritical. This results in a chain reaction which produces what appellant refers to in the specification as "an intense nuclear spark." Further relative rotation of the one mass beyond this "juxtaposite position" carries it away from the other mass, eliminates the hypercriticality of the juxtaposed masses, extinguishes the nuclear spark and interrupts the chain reaction before a nuclear explosion occurs.

Appellant states that his nuclear reactor may be used in the propulsion of aircraft and rockets as well as in reactors for conducting a breeding reaction. Appellant points out in his specification that while a "complete nuclear turbojet" is shown in the drawings, "The novelty resides in the design of the reactor and the method of operation * *." He further explains the novelty as residing "in the construction of a rotary and fast reactor in which the energy is produced in a series of discontinuous pulses, instead of continuously as in prior designs."

There are two grounds of rejection under section 103 The first ground of rejection is based on British alone. In this rejection, the board relied on the teachings of British and the "well known" fact "that relative rotary motion between two parts is the simplest way to obtain periodic relative motion." The limitations in the appealed claims not disclosed in British were found to be "obvious" and the exercise of "mere engineering judgment."3 The second ground of rejection is based upon the combined teachings of British and Wheeler.

Appellant argues that while the British Patent issued on Dec. 15, 1948, its disclosures must be read as of its Convention filing date, May 2, 1939. Appellant argues that as of May 2, 1939, the disclosures in the British patent, relied on by the board in its decision, were "mere speculations" and not "actual engineering knowledge." Appellant admits that the alleged "speculative" statements first disclosed in 1939 "have proved to be true." Thus, appellant does not here challenge the accuracy of the "speculations" made in British.

The issue of obviousness must be decided according to the controlling provisions of 35 U.S.C. § 103. This requires, inter alia, that obviousness is to be determined as of the time the invention was made. On the record before us, that time must be taken as March 28, 1958, the filing date of appellant's application.

When appellant's argument is considered in the light of the issue here, its thrust is directed to the weight which should be accorded the teachings of British. Section 103 is not specific as to the weight to be accorded to a particular reference, but requires rather that we consider the totality of the teachings of the prior art to determine what would have been obvious to one of ordinary skill in the art at the time of making the invention in issue.

In the instant case, and even assuming that certain statements in British were "speculative" as urged by appellant and thus perhaps not completely credible when made, the unchallenged fact is that these statements have since been proven to be true. The statements in British were credible information and available to one of ordinary skill in the art at the time appellant's invention was made. We therefore perceive no reason why, in determining obviousness of the claimed invention, one of ordinary skill in this art should not be expected to recognize the statements in British as being valid teachings at the time the claimed invention was made. Thus, we consider as credible the teachings in British which were relied upon by the board for what they fairly taught one of ordinary skill in this art at the time appellant made his invention.

We therefore agree with the board that British "has the concept of imparting periodic motion to two subcritical masses, so that during the portion of the periodic motion in which the two masses have approached each other, they shall become hypercritical." British discloses the periodic and controllable liberation of energy by controlled intermittent nuclear fission, characterized by the examiner as a "pulsed neutron flux."

The teachings of British, especially in view of the terms "critical mass," "critical values," and "unlimited branching," indicates that each uranium body there disclosed constitutes a subcritical fissionable mass, but when they approach each other during their periodic movement, they reach a condition of hypercriticality characterized by the "unlimited branching of the chains." British does not contain any drawings or a description of any specific apparatus.

* * * The patent does not disclose any reactor comprising two relatively rotatable members each of which has any fissionable and subcritical mass associated therewith. Since the British patent does not disclose the relatively rotatable members, it of course cannot disclose the means for rotating one with respect to the other. The British patent fails to suggest that there is any predetermined velocity at which one member might be rotated with respect to the other so that two subcritical masses might become momentarily hypercritical to produce a chain reaction lasting for a time on the order of 10-⁴ seconds.

While the effective date of British pre-dates much of the subsequently acquired and subsequently disclosed knowledge concerning nuclear fission, appellant concedes that it "contains certain statements based on theoretical considerations which have proved to be true." Among such statements are those relating to time as a function in chain reactions and those in which it is recognized that time may be used as an element of control of the reaction. In addition to such theoretical considerations, British also refers to the use of distinct masses of "uranium" and describes how they may be periodically caused to approach and be separated from each other "to provide within limited and determined periods the conditions of the unlimited branching of the chains." As expressed in British:

The mass containing uranium can for instance consist of a fixed part and of a movable part which is periodically separated from and moved towards the fixed part by means of any suitable mechanical means.

The first embodiment of British * * * broadly suggests that two subcritical masses of uranium should be periodically brought together and separated. Relative motion of the two masses is clearly required. The mechanical details are so simple and obvious that there was no necessity for spelling them out. A large number of arrangements immediately suggest themselves, as any one consulting standard engineering textbooks and/or handbooks can determine. Among the simplest and therefore the most obvious are two bodies which have relative rotary movement * * * and two bodies having relative reciprocatory motion, like a piston in a cylinder * * *. Since each is a most obvious way of producing the periodic motion described in British, it should be unnecessary to decide whether the latter type of motion is more obvious than the former, or vice versa. It should be sufficient to a determination of the issue before the Court, if it is satisfied that relative rotary movement is an obvious way of producing the periodic motion contemplated by British.

However, as appellant points out, Wheeler "relates to a reactor which depends upon the use of a moderator whereas appellant's rotary reactor does not require a moderator." It is pointed out in appellant's brief that:

Wheeler discloses a reactor built on a rotary principle but for an entirely different purpose than appellant's rotary reactor. The plates of Wheeler are arranged to rotate through passages between the stationary moderator plates in the upper half of the device. After the plates emerge from the moderator passages they are rotated through a suitable collant such as ordinary water disposed in the collant tank. The rotating plates are composed of natural uranium having a hollow space between the sides for receiving the highly active uranium or plutonium isotopes. The generation of power in the Wheeler reactor occurs in precisely the same way as in ordinary stationary reactors by reason of the presence of abundant supplies of the highly active isotopes arranged in conjunction with the neighboring moderator plates of graphite or beryllium. There is no relative movement whatsoever between two subcritical masses of fissionable material into hyper-critical relationship. The only reason for the rotation of the plates containing the fissionable material is to permit them to be cooled by withdrawing them from the moderator spaces and moving them through the large mass of cooling water. The velocity of rotation is unspecified in Wheeler because speed is in no way critical. [Reference numerals omitted.]

Thus, while Wheeler discloses a reactor employing rotating parts, we agree with appellant that Wheeler does not contemplate controlling the power solely by the relative rotation of separate fissionable masses. It likewise is clear that in the Wheeler reactor, there is no generation of power because of movement between the two masses into hypercritical relationship; and there is no control of power generation by selection of a predetermined velocity of rotation of the separate subcritical fissionable masses to control the time during which the separate masses are so positioned as to become hypercritical.

From the foregoing, we think it apparent that one of ordinary skill in this art, at the time of appellant's invention, would have found it obvious from the teachings of the references to employ separate subcritical masses of fissionable materials and to so move them that their combined mass becomes critical and a divergent fission reaction is initiated, which is controlled by moving the separate masses out of juxtaposition to destroy the critical relationship and into positions where they are again subcritical.

Concerning claims 3 and 8, the court is in agreement that these claims define the invention in such generality as to embrace subject matter which we think would have been obvious to one of ordinary skill in this art at the time of appellant's invention. Accordingly, the rejection of claims 3 and 8 is affirmed.

Concerning claims 1 and 2, a majority of the members of this court affirm the rejection for the reasons stated in Judge Martin's opinion. I dissent from the affirmance of the rejection of claims 1 and 2 for the following reasons.

The express language of section 103 requires us to consider the differences between the invention claimed and the prior art before making a determination on the issue of obviousness. One such difference is found in appellant's claimed reactor which requires use of a gas turbine to effect rotation of the rotatable members at a predetermined velocity, as claimed in claim 2; and his recognition of the relationship between that velocity and the subcritical fissionable masses that, as claimed in both claims 1-2,

when the said masses are in proximity during said rotation, they become momentarily hypercritical and generate a divergent fission chain reaction lasting for a time on the order of 10-⁴ seconds.

Viewing appellant's invention as a whole against the broad teachings of the prior art, I find that claims 1 and 2 define a nuclear reactor which I find to be non-obvious in view of the prior art. Accordingly, I would reverse the rejection of these claims.

The judgment of the court is that the decision of the board affirming the rejection of claims 1, 2, 3 and 8 is affirmed.

Affirmed.

MARTIN, Judge (for the majority).

We agree with the fundamental approach of Judge Smith's opinion. However, a majority of the court holds that all the claims were properly rejected over the references, 35 U.S.C. § 103.

Claim 1 recites a "means of rotation," and claim 2 recites "a gas turbine" as that means of rotation. Appellant's recognition of the relationship between rotational speed and the proximity of the subcritical fissionable masses is recited in both claims 1 and 2 as:

* * * when the said masses are in proximity during said rotation, they become momentarily hypercritical and generate a divergent fission chain reaction lasting for a time on the order of 10-⁴ seconds.

Neither the means of rotation nor the velocity-time relationship are such differences from the prior art as to render the subject matter as a whole non-obvious.

That velocity (speed) is merely a way of describing and governing the time the two subcritical masses are in proximity is as fundamental as v=d/t. That it is necessary to keep the passing masses only momentarily hypercritical is equally fundamental. There appears to be nothing in the appellant's disclosure which indicates what range of time "on the order of 10-⁴ seconds" encompasses,1 or that such time is critical.2 Indeed, the specification does not attach any particular significance to that particular time value. At best, it merely suggests no more than the obvious — that one dare not keep the two masses together too long.

It is the chain reaction engendered upon the passing of the two masses in close proximity which lasts for "on the order of 10-⁴ seconds," not the passing itself. That recitation, being but descriptive of the result of the method of properly operating the apparatus, has not been shown on this record to confer patentability to the apparatus claims 1 and 2. In this regard the British reference states:

In order to regulate the average liberation of energy, one could influence, automatically or not, the time during which the chains are developing and/or the interval between two developments of the chain reaction. [Emphasis added.]

From the above considerations we do not find the added velocity-time relationship, insofar as it is an apparatus limitation, to render the basic stator-rotor apparatus, as defined in claims 1 and 2, as a whole non-obvious over the prior art within the meaning of section 103. The board's decision in respect to these claims is also affirmed.

WORLEY, Chief Judge, and RICH and ALMOND, JJ., join in the foregoing opinion.