CITE:  415 F.2d 1393
CMON:  August 1969
PLAIN: In re Charles D. Prater and James Wei
DEFND: United States Patent and Trademark Office
COURT: United States Court of Customs and Patent Appeals
DATE:  August 14, 1969 


Appeal from decision of Patent Office, Serial No. 155,236, 56 CCPA.  The
United States Court of Customs and Patent Appeals, 415 F.2d 1378, entered
judgment and rehearing was granted.  On rehearing the United States Court
of Customs and Patent Appeals, Baldwin, Judge, held that claims 1, 6-9,
12, and 17-21 of application for patent for method of processing or
analyzing conventionally obtained spectrographic data to produce a
quantitative spectographic analysis of qualitatively-known mixture by
which unknown component concentrations may be determined with minimum
error were properly rejected for failure to point out and distinctly
claim the subject matter regarding as the invention, but that the
apparatus claim was patentable over objection that it was obvious.


  A patentable process is not limited to operations on physical substances
  or the means used in performing it.

Before WORLEY, Chief Judge, MCGUIRE, Judge, sitting by designation, RICH,
ALMOND, and BALDWIN, Associate Judges.


This appeal was originally argued {1} before this court {2} on May 9, 1968,
and decided {3} on November 20, 1968.  On petition of the Commissioner of
Patents, a rehearing was granted by this court {4} on January 16, 1969, under
the provisions of Rule 7 of this court.  Briefs were filed, {5} and the case
was reargued {6} before this court {7} on March 3, 1969.  This supersedes
the decision and opinion of November 20, 1969, though various portions of
the latter are repeated herein without specific references thereto.

This appeal is from the Patent Office Board of Appeals decision affirming
the examiner's rejection of claims 1, 6-10, 12, and 17-21, all of the claims
in appellants' application, {8} based solely upon considerations of law and
statutory construction. {9} No prior art references have been cited.

The Invention

The invention includes both a method {10} and apparatus {11} for the
processing, or analysis, of conventionally obtained spectrographic data to
produce a quantitative spectrographic analysis of a qualitatively-known
mixture, for example a mixture of gases, by which the unknown component
concentrations may be determined with minimum error.  The raw spectrographic
data is conventionally obtained in the form of spectrogram, typically a
continuous trace having a plurality of peaks. By conventional techniques,
there may be derived for each peak an independent first order linear
algebraic equation relating the peak height to the unknown component
concentrations. Typically however, there are more peaks from which equations
may be derived than are required to solve for the unknown component
concentrations; that is, a set of equations, greater in number than the
number of components, may be derived from the spectrogram. {12}  From that
set, any subset of equations, equal in number to the number of components,
may be selected {13} to solve for the concentrations. {14}  The selected
subset of equations may then be conventionally solved to produce a
quantitative analysis.

Appellants have made a discovery which lies at the heart of their invention.
Appellants have discovered: (1) that the different subsets of equations result
in varying degrees of undesired "error amplification" in transforming the
spectrographic data involving peak heights to the desired concentrations;
{15}  (2) that there exists a certain relationship indicative of such error
amplification; and (3) that that relationship is related to, and may be
expressed in terms of, the determinants of the subsets of equations, the
determinant of largest magnitude indicating the subset of equations involving
least error amplification. "As far as this record shows, this discovery
was new and unobvious."  {16}  Prior to appellants' work, there was no
systematic method for, or means of, selecting the subset of equations
generating least error amplification.

Thus, based on their discovery, appellants have invented a method, within the
otherwise conventional spectral analysis method, for selecting the optimum
peaks providing that particular subset of equations least susceptible to
error amplification. The essence of the method is finding the subset of
equations having the largest determinant amongst all the possible subsets
that might be chosen.  Applicants have also disclosed in detail a machine
{17} for carrying out their invention. While the disclosed method and
apparatus are described by reference to a special-purpose analog device,
such device is visualized by appellants as but one device having the inherent
capabilities to perform the desired functions; and it is disclosed that a
general-purpose digital computer might also be used. {18}

Claims 9 and 17 are typical of the method claims and read:

  9.  In mass spectrographic analysis where, from a given sample of material
  there is generated a spectrum function having peaks therein spaced along
  a mass scale with respect to which the relationship between concentration,
  contribution factor of each of the m constituents of the mixture and the
  magnitude of each of the n peaks in said spectrum is represented by a set
  of m linear algebraic equations and where n is an integer greater than m, 
   the method of selecting for analysis a set of m peaks least
  susceptible to error in concentration determination  which comprises

   dividing each  said  contributing factor  for each
  peak  by a normalizing function ,

   successively generating a determinant function  for each said
  set of peaks,

   successively generating output indications of the magnitudes of
  said determinant functions, and

   selecting therefrom the determinant function of greatest magnitude
   for identification of said peaks least susceptible to error.
  [Emphasis added.]

  17.  The method of determining  with minimum error from the
  spectra of spectral analysis  the concentration of the components
  of a mixture  where the components are known and the
  concentration-determining peaks of the spectral analysis are present
  in number exceeding the number of said components, which comprises

   generating physical representations of the magnitudes of the
  coefficients  of simultaneous linear equations defining the
  concentrations of said components as functions of the heights of said
  peaks of said spectral analysis,

   generating  from said physical representations of  the
  magnitudes of said coefficients  the magnitude of the determinant
   of a plurality of sets of said simultaneous equations, the number
  of equations of each of said sets being equal in number to the number
  of said components,

   comparing  said physical representations of  the magnitudes
  of said determinants  of said sets of equations  for
  identification of the set  of said equations  whose determinant
  has the largest magnitude, and

   generating physical representations of the concentration of each said
  component of said mixture from said physical representations  of the
  magnitudes of said coefficients  of said set  of simultaneous
  equations  having said determinant of largest magnitude  and
  from said heights of said peaks included in said last-named set of
  equations. [Emphasis added.]

Claim 10, the only machine claim, reads:

  10.  In spectrographic analysis where, from a given mixture of m
  constituents, spectral functions having peaks therein are obtained and
  wherein the relationships between the concentrations of said constituents
  and the peaks in said function correspond to relationships in a set of
  linear simultaneous equations,  the system for selecting  from
  said functions  the combination of  m of the n  peaks 
  therein  least susceptible to error in concentration determination,
  where n is an integer greater than m, which comprises

    means for generating a scalar function  representative of the
   determinant for a first set of said equations corresponding with a 
   first set of peaks in number equal to m,

    means for generating successive scaler functions  each
   representative of a determinant for sets of equations corresponding
   with different sets of m of said peaks, and

    means for determining that one of said scalar functions of greatest
   magnitude  for identification of said combination of m of the n
   peaks least susceptible to error.  [Emphasis added.]

   The Examiner's Rejection and Board's Affirmance Process Claims

In his Answer, the examiner restated the rejection of the process claims for
failure to comply with 35 USC 101, 102, and 112. He preliminarily reasoned:
(1) that, if the invention is within a statutory class of patentable subject
matter under 35 USC 101, it must be as a "process"; (2) that the appealed
process claims are readable upon a mental process with the only physical
limitation being the generation of physical representations, which may be
merely pencil markings on paper; and (3) that processes having only mental
novelty are unpatentable because they are outside the statutory "process"
class, citing In re Abrams, 38 CCPA 945, 188 F.2d 165 (1951). The Answer

   At this point the rejection can take alternative forms, depending
   upon an apparently unsettled point of law, i.e., whether a claim is
   unpatentable if it reads on non-statutory subject matter in addition
   to statutory subject matter. The first form is under 35 USC 101 and 102,
   and the second form is under 35 USC 101 and 112.

Under 53 USC 101 and 102

The examiner argued that, if a claim is readable on subject matter outside
the statutory classes, then the claim is unpatentable under 35 USC 101, it
being irrelevant that the claim can also be read on subject matter within the
statutory classes.  The examiner felt that each of the present process claims
can be read on mental calculations with the appropriate mathematics and, as
the only physical steps, writing on paper.  The examiner held that, to the
extent that those claims require anything physical, they are fully met under
35 USC 102 by the process of making marks on paper.  The examiner concluded
that  the only novel subject matter recited in the claims must reside in the
calculations which are outside the statutory classes of 35 USC 101 and which,
therefore, may not patentably distinguish the claims over the process of
marking on paper.Thus, the claims were rejected under 35 USC 101 and 102.

In regard to the examiner's 101-102 rejection, the board held that the
process claims "do not fall within the statutory definition of a process as
that term has been interpreted by the Courts over the years," citing In re
Yuan, 38 CCPA 967, 188 F.2d 377 (1951) and Cochrane v. Deener, 94 U.S. 780.
The board further stated:

   It * * * [is] beside the point that the solution of the mathematical
   problem can be done by machine. The claims has set forth nothing which
   cannot be performed purely as a mental exercise using appellants'
   discovery that the equations having the largest determinant are the
   ones to use.

Under 35 USC 101 and 112

Under the 101-112 rejection, the examiner assumed that a claim is within the
statutory classes of 35 USC 101 if it is readable on subject matter within
that section, even if it can also be read on subject matter outside that
section. The examiner argued however, that, if a claim covers or reads on
subject matter outside the statutory classes, as well as subject matter within
the statute, then the claim fails to particularly point out and distinctly
claim the invention as required by 35 USC 112. That is, said the examiner,
a reasonable interpretation of the process claims makes them appear to
cover subject matter outside the statutory classes of patentable subject
matter under 35 USC 101, and claims which read literally on unpatentable
subject matter do not particularly point out and distinctly claim the
subject matter which applicants regard as their invention as required
by 35 USC 112.

The board did not specifically comment in regard to the 101-112 rejection of
the process claims, apparently affirming the examiner. See Rule 196(a),
United States Patent Office Rules of Practice.

Apparatus Claim

The rejection of apparatus claim 10 was also in two forms.

Under 35 USC 101, 102, and 103

The examiner noted that, in addition to reading on appellants' disclosed
analog embodiment, claim 10 may also be read on a properly programmed
general-purpose digital computer. He argued that, having the principle of
mathematics, which is unpatentable under 35 USC 101 even though discovered by
appellants, it would be obviout under 35 USC 103 to program a general-purpose
digital computer which is old under 35 USC 102, such a computer having
been known and in public use more than a year before appellants filed.
Since, in the examiner's view, a properly programmed digital computer is
thus not patentable to appellants, claim 10 was rejected under 35 USC 101,
102, and 103.

The board viewed claim 10 as generally an apparatus counterpart of
claim 17 and thus considered claim 10 unpatentable, citing In re Yuan, supra.
That is, so held the board, the sole novelty in claim 10 resides in the
mathematical computations which are themselves nonstatutory subject matter
under 35 USC 101. Furthermore, the board felt that the physical "means"
could be, and are anticipated under 35 USC 102 by, such commonplace means
as pencil, paper, and ruler, since the essential novelty of the claim is
in the mathematical calculations.

Moreover, the board found that the essence of the rejection of claim 10
is that it defines nothing more than a general-purpose digital computer
programmed to perform the required mathematical operations and that the
programming would be obvious under 35 USC 103. The board agreed that such
a machine would be obvious.

Under 35 USC 101, 102, 103, and 112

Here, the examiner argued that claim 10 does not particularly point out
and distinctly claim appellants' invention as required by 35 USC 112 since,
in addition to reading "on a device (the analog embodiment) to which they are
entitled to coverage", (emphasis added) the claim also covers or reads on a
properly programmed general-purpose digital computer on which appellants are 
not entitled to coverage for the reasons set forth above.

The board did not specifically address itself to the 101-102-103-112
rejection of claim 10, again apparently affirming the examiner.

The Solicitor's Position

The solicitor, in his original brief, stated that:

  the Patent Office contends that all of the method claims read on a
  mathematician's use of pencil and paper, and that the "means" of
  apparatus claim 10 may correspond to those implements along with a ruler.

  * * * Both the examiner and the board were concerned over the breadth
  of the claims, {19} while conceding that appellants' discovery was
  apparently new and unobvious. [Emphasis added.]

Appellants' Position

In their original brief under the heading "The Facts Support Patentability",
appellants contended that:

   (1) They have made a discovery which has advanced the useful arts,
   one which the Board found to be "new and unobvious."

   (2) They have developed a process for utilizing their discovery.

   (3) They have invented a machine which makes possible the
   utilization of their discovery without human intervention.

   (4) They have disclosed that their process, evolved from their discovery,
   may be practiced by a properly programmed general-purpose computer without
   human intervention.

   (5) In the absence of prior art - and no references are relied upon by
   the Examiner or by the Board - all tests of patentability have been met.

Appellants further contended that the decisions relied upon below, namely,
In re Abrams, supra, In re Yuan, supra, and Cochrane v. Deener, supra, are
not here applicable.  Appellants have not argued against the soundness of
the results reached in those cases or of the so-called "mental steps"
doctrine (i.e., the non-patentability of mental processes) as it may be
based upon the Constitution, {20} statutes, or case law, as they interpret
them; rather they seek to distinguish the present case from the earlier
cases and the "mental steps" doctrine, as interpreted by appellants.

Appellants point out that in both Abrams and Yuan, as contrasted with the
present case, there was not disclosed a machine or apparatus for carrying out
the invention automatically and without human intervention.  Appellants argue
that, where no such completely automated machine or apparatus has been
disclosed, it is quite reasonable that certain steps of a process might be
considered purely mental in nature.  On the other hand, appellants urge that
their disclosure of apparatus for carrying out the invention without human
intervention precludes the process or any part thereof from being considered
purely mental.  In summary, they think that neither Abrams nor Yuan supports
the proposition that a claim, limited to a process performed by apparatus
 by construing the claim in the light of the specification, is
unpatentable simply because the process might conceivably also be carried
out within the human mind without the apparatus. Thus, as we view
appellants' position, they do not regard the appealed claims, when read
in the light of the specification, as covering a purely mental process
or a mental process coupled with pencil and paper markings, nor do they
seek such claim coverage. {21}

   Insofar as Cochrane v. Deener is concerned, appellants urge that that case
represented a reaffirmation or an extension, as opposed to limitation, of the
subject matter which may be patented.  Appellants feel that Cochrane v. Deener
merely made it clear that a broad area of invention may be protected by
process claims, not limited to any special arrangement of machinery, and
that the decision did not set forth the metes and bounds of subject matter
for which process claims may be granted.

   As to the apparatus claim, appellants argue that it matters not whether
the claim reads on a general-purpose digital computer, properly programmed to
perform their invention. They urge that they are entitled to such protection
inasmuch as such a properly programmed general-purpose digital computer was
neither old nor obvious at the time of appellants' invention.


Process Claims

We are in agreement with appellants that this case is not controlled
by Abrams, Yuan, or Cochrane v. Deener, as we view those cases.  Insofar as
we can determine, this is a case of first impression in that the claims in a
mechanical, as distinguished from chemical, case have been rejected because
of their breadth notwithstanding that no prior art reference has been
cited and that the claims read on admittedly patentable statutory subject
matter disclosed in the specification.

A more exhaustive analysis of Abrams may be found in Judge Smith's previous
opinion, supra; however, for the present purposes, Abrams need not be
discussed in such depth.  In Abrams, no prior art reference was cited in the
application entitled "Petroleum Prospecting Method," the claims being rejected
as failing to define subject matter properly within the terms of R.S. 4886,
the precursor of 35 USC 101 and 102, in that "the steps in the claims which
constitute the heart of the invention are purely mental in character".

The court referred to Abrams' specification in interpreting the steps of the
claims, but Abrams had disclosed no means whatever for performing, without
human intervention, two claimed steps of calculation and comparison.  No
analog device for carrying out the steps was disclosed in the Abrams
specification; and at the time Abrams filed (April 28, 1944), general-purpose
digital computers were still in the future.  Thus, Abrams disclosed and
claimed a process which could only be performed in the mind insofar as the
teachings of the application were concerned.  Abrams therefore presents a
significant difference from the factual situation in the present case in which
the teachings of the specification provide a full disclosure of at least
analog apparatus for carrying out the claimed steps  without requiring
any steps to be performed in the human mind.

Yuan, decided about a month after Abrams, contained the following oft-quoted

   This court has deemed it to have been thoroughly established by decisions
   of various courts that purely mental steps {22} do not form a
   process which falls within the scope of patentability as defined by
   statute. {23} [Emphasis added.]

But, as appellants point out, "Yuan's disclosure was the use made of
equations by pencil-and-paper with the mind of the operator at work to
interpret the results." Again, as in Abrams, insofar as the disclosure was
concerned, the process (or the critical step thereof) was one that
 required  the use of the human mind - indeed, a purely
mental process or step.

That a process was an "art" capable of being patented was already considered
beyond dispute {24} in Cochrane v. Deener, involving a process or method
of manufacturing flour.  The issue was infringement, the defendant
performing the process using apparatus of somewhat different construction from
that disclosed by the patentee.The Supreme Court found the patent valid and
infringed.  In the course of its opinion, the Court produced the following,
often-quoted passage of dictum:

  A process is a mode of treatment of certain materials to produce a given
  result. It is an act, or a series of acts, performed upon the subject matter
  to be transformed and reduced to a different state or thing.  If new and
  useful, it is just as patentable as a piece of machinery. In the language
  of the patent law, it is an art.  The machinery pointed out as suitable to
  perform the process may or may not be new or patentable; whilst the process
  itself may be altogether new, and produce an entirely new result.  The
  [Cochrane] process requires that certain things should be done with
  certain substances, and in a certain order; but the tools to be used
  in doing this may be of secondary consequence.

This passage has sometimes been misconstrued as a "rule" or "definition"
requiring that all processes, to be patentable, must operate physically upon
substances.  Such a result misapprehends the nature of the passage quoted as
dictum, in its context, and the question being discussed by the author of the
opinion.  To deduce such a rule from the statement would be contrary to its
intendment which was  not to limit  process patentability but
to point out that a process is not limited to the means used in performing
it. {25} See In re Ernst, 21 CCPA 1235, 71 F.2d 169 (1934).

Thus, it is clear that this case is not controlled by Abrams, Yuan, or
Cochrane v. Deener. However, we do not feel that the distinction which
appellants have very clearly pointed out is entitled to the significance which
they would attribute thereto.  Appellants feel that, since they have disclosed
apparatus for performing the process wholly without human intervention and
since they are seeking coverage only for the machine-implemented process,
{26} they have avoided the so-called "mental steps" doctrine.  Although in
view of our decision here we find it unnecessary to analyze and/or review
in depth the so-called "mental steps" doctrine, it would appear that the
disclosure of apparatus for performing the process wholly without human
intervention merely shows that the disclosed process does not fall within
the so-called "mental steps" exclusion. {27} Of course, we have already
pointed out that there is no dispute, as between the principal parties,
that appellants have disclosed a patentable statutory process.

However, that is quite another question from the one before us, namely,
whether appellants are entitled to claims of the breadth of those they seek
here.  As already noted, we view appellants' position to be that they are not
seeking patent coverage of any purely mental process or any mental process
coupled only with pencil and paper markings, {28} but they are seeking
coverage of the operation of a properly programmed general-purpose digital
computer performing their process, {29} as well as that of an analog device
of the type disclosed.  Appellants feel that they are entitled to a claim
of the scope of claim 9 since, when read in the light of the
specification, claim 9 does not cover a mental process. As to the
other method claims, appellants feel "that generation of physical
representations is conventional language in a step to be performed
by a machine and absent any function of the mind in its performance",
"that a physical representation includes more than mental operations",
and that "[this] takes it away from the purely mental concept."

   Claim 9

With respect to claim 9, the broadest claim, appellants acknowledged in
their original brief that:

  Viewed apart from appellants' disclosure this claim is broad enough
  to cover the method if practiced by the use of pencil and paper by an
  operator, the operator directing the pencil. Such a person may devide each
  contributing factor by a normalizing function.  He may solve for the
  value of the determinant for each set of equations. He may compare the
  results to select the set of equations having the greatest determinant
  for identification of that set of equations least susceptible to error.
  [Emphasis added.]

However, appellants urged that claim 9 cannot be read in a vacuum but instead
must be read in the light of the specification. We agree.

Nevertheless, "reading a claim in the light of the specification," to
thereby interpret limitations explicitly recited in the claim, is a quite
different thing from "reading limitations of the specification into a
claim," to thereby narrow the scope of the claim by implicitly adding
disclosed limitations which have no express basis in the claim.  This
distinction is difficult to draw and is often confused by courts; but
it is even more difficult for attorneys, attempting to work within the
framework of the former, not to cross over into the latter.

In our view, appellants would really like us to read a limitation of the
specification into the claims, not merely interpret the claims in the light
of the specification.  When read in the light of the specification, claim 9
does read on a mental process augmented by pencil and paper markings. We find
no express limitation in claim 9 which, even when interpreted in the light of
the specification, would support the conclusion that the claim is limited to
a "machine process" or "machine-implemented process." This is particularly
important in this case since the board noted that, in their brief before the
board, appellants acknowledged that "[though] not practical for most of the
needed applications, their method, theoretically, can be practiced by hand."

Inasmuch as claim 9, thus interpreted, reads on subject matter for which
appellants do not seek coverage, and therefore tacitly admit to be beyond that
which "applicant regards as his invention," we feel that the claim fails to
comply with 35 USC 112 which requires that "[the] specification shall conclude
with one or more claims particularly pointing out and distinctly claiming
 the subject matter which the applicant regards as his invention."
(Emphasis added.)  This is true notwithstanding appellants' disclosure
of a machine-implemented  process.  There are quite sound reasons why,
in an infringement suit on an issued patent, courts may sometimes "interpret
patent claims in the light of the specification" so as to protect only that
phase of the claimed invention that constitutes patentable subject matter
and thus do justice and equity between the parties. {30} However, this court
has consistently taken  the tack that claims yet unpatented are to be given
the broadest reasonable interpretation consistent with the specification
during the examination of a patent application since the applicant may then
amend his claims, the thought being to reduce the possibility that,
after the patent is granted, the claims may be interpreted as giving
broader coverage than is justified.{31}  In re Sweet, 55 CCPA 1191,
393 F.2d 837 (1968); In re Soderquist, 51 CCPA 969, 326 F.2d 1016 (1964);
In re Tibbals, 50 CCPA 1260, 316 F.2d 955 (1963); In re Henatsch, 49 CCPA
915, 298 F.2d 955 (1962); In re Lundberg, 44 CCPA 909, 244 F.2d 543 (1957);
In re Kebrich, 40 CCPA 780, 201 F.2d 951 (1953). We are not persuaded by
any sound reason why, at any time before the patent is granted, an applicant
should have limitations of the specification read into a claim where
no express statement of the limitation is included in the claim.

Thus, with respect to claim 9, appellants have not particularly pointed out
and distinctly claimed the subject matter which they regard as their invention
as required by 35 USC 112, it therefore being unnecessary for us to consider
35 USC 101 or its applicability here.  Accordingly, the board's decision as
to claim 9 is AFFIRMED.

Claims 1, 6-8, 12 and 17-21

Claim 17, supra, is representative of the claims falling within this group.

In arguing that claims 1, 6-8, 12, and 17-21 do not read on a purely mental
process or a mental process coupled with pencil and paper marking, appellants
further advise us that:

   Webster's Third New International Dictionary (1964) defines "physical"
   in the sense:

   "... of or relating to natural or material things as opposed to things
   mental, moral, spiritual, or imaginary ..."

Having given careful consideration to appellants' additional arguments
concerning the proper interpretation to be given to claims 1, 6-8, 12, and
17-21, we feel that the rejection must nevertheless be affirmed.  "Generating
physical representations" appears to us to be broad enough, even when read in
the light of the specification, to encompass pencil and paper markings which a
mathematician might make in documenting or recording his mental calculations.
Hence, since appellants do not seek such broad coverage as hereinbefore
discussed, we think that claims 1, 6-8, 12, and 17-21 must also fail under
35 USC 112 for the reasons expressed with regard to claim 9; with respect
to those claims also, the board's decision is AFFIRMED.

Apparatus Claim

Apparatus claim 10 presents quite a different question and, in our view,
requires a different answer.  As we see it, the underlying statutory basis for
the rejection of apparatus claim 10 is 35 USC 103 which precludes the grant of
a patent if and only if "the subject matter as a whole would have been obvious
at the time the invention was made to a person having ordinary skill in
the art".  Appellants' discovery, discussed in the second paragraph under
the  heading "THE INVENTION", supra, is, it seems to us, part of their
contribution to the art. On that basis, appellants' discovery should be
considered as part of "the subject matter as a whole" and
not part of the prior art. It is conceded by the Patent Office 
that that discovery is both new and unobvious.  Thus, based on the record
before us, we do not perceive any reasonable basis for concluding that
"the subject matter as a whole", as defined by apparatus claim 10, would
have been obvious at the time of appellants' invention.

We have carefully considered the basic position of the Patent Office that
it would be obvious to program a general-purpose digital computer to
practice appellants' invention and that apparatus claim 10 reads on such a
computer, as well as the disclosed analog device.  We find that position
fatally defective in that it, in effect, assumes the existence as prior
art of appellants' discovery that the relationship indicative of error
amplification "is related to, and may be expressed in terms of, the
determinants of the subsets of equations, the determinant of largest
magnitude indicating the subset of equations involving least error
amplification." {32}  Perhaps today, after reading appellants' disclosure,
the public dissemination of which the patent system fosters and encourages,
it might be obvious to program a general-purpose digital computer to practice
the invention. But 35 USC 103 requires an analysis of the prior art at the
time the invention was made to determine whether the invention was obvious.
Graham v. John Deere Co., 383 U.S. 1 (1966). Assuming the existence, at the
time of the invention, of general-purpose digital computers as well as
typical programming techniques therefor, it is nevertheless plain that
appellants' invention, as defined in apparatus claim 10, was not obvious
under 35 USC 103 because one not having knowledge of appellants' discovery
simply would not know what to program the computer to do.  See Ex parte King,
146 USPQ 590 (Pat. Off. Bd. App. 1964).

We do not perceive of any "mental steps" issue in regard to apparatus claim
10.  It is quite clear that claim 10, in typical means-plus-function language
as expressly permitted by the third paragraph of 35 USC 112, does not
encompass the human being as the "means" or any part thereof.  Cf. Brown v.
Davis, 116 U.S. 234 (1886); Republic Iron & Steel Co. v. Youngstown Sheet
& Tube Co., 272 F. 386 (6th Cir. 1921); Permutit Co. v. Village of Poynette,
61 F.Supp. 305 (W.D. Wisc. 1945) aff'd per curiam 158 F.2d 799, (7th Cir.
1947); Wilcox v. Danner, 19 CCPA 802, 53 F.2d 711 (1931); Mabon v. Sherman,
34 CCPA 991, 161 F.2d 255 (1947). The pencil, paper, and ruler - referred
to by the board in regard to 35 USC 102 - do not  anticipate the
claimed "means" since the former additionally require human manipulation.

Accordingly, the board's decision as to claim 10 is REVERSED.


As to claims 1, 6-9, 12, and 17-21, the board's decision is AFFIRMED; as to
claim 10, the board's decision is REVERSED.

WORLEY, Chief Judge, concurring.

The court is indebted to counsel for the respective parties and amici curiae
for the able and earnest fashion in which they have, in both argument and
briefs, presented their understandably diverse positions.

One of the more interesting points raised is how far Congress intended to go
in conferring patentable status on mental steps as they are interwined in
computer programs generally.  It is questionable whether prior decisions
denying patentability of purely mental steps, or the statute, read singly or
together, can support a broad rule either sanctioning or prohibiting the
patentability of such steps in relation to computer programs.  Where the line
will be drawn can only be determined on a case by case basis in building, as
best we can, sound and intelligible precedent.  While I agree here with the
results reached by my colleagues I do not necessarily subscribe to all
that is said.

Congress, of course, had no way of knowing in 1952 what lay in the test tube
then or what would become a reality tomorrow.  But it devised a statute - a
model of legislative craftsmanship and foresight - broad enough to anticipate
and nourish the technological explosion we have witnessed.  Our patent
system is a delicate balance of interests; it protects the fruits of the
extraordinary efforts it demands of inventors compatibly with the public
interest.  It has largely  fostered the favorable climate resulting in
the tremendous strides this country has made in reaching the pinnacle in the
world-wide competition in the arts and sciences - from atomic energy to
antihistamines, computers to catalysts, lasers to lunar landings. {1'}

So it is with no little surprise and concern that one learns of proposed
patent "reforms." I can appreciate, from my own Congressional experience,
the difficulty of enacting legislation acceptable to all concerned.
However, after having dealt with the patent statutes, particularly the 1952
Patent Act, for nearly twenty years as a member of this court, I have grave
misgivings concerning the desirability or need for any substantive change in
a system that has worked so well in following the constitutional mandate
"to promote the Progress of Science and useful Arts."


{1} In addition to the principal parties' oral arguments and briefs, Bell
Telephone Laboratories, Incorporated (Bell Labs), the real party in interest
in Patent Appeal No. 8216, filed a brief amicus curiae.

{2} Consisting of Chief Judge Worley and Judges Rich, Smith, Almond, and
Kirkpatrick, Senior District Judge, Eastern District of Pennsylvania,
sitting by designation.

{3} Reversed, with opinion authored by the late Judge Smith and joined by
Judges Rich and Almond.  Judge Kirkpatrick concurred in the result, and Chief
Judge Worley did not participate in the merits of the decision.  In re Prater,
56 CCPA 1360, 415 F.2d 1378 (1968). The decision was handed down the day
Judge Smith died.

{4} Consisting of Chief Judge Worley and Judges Rich, Almond, Baldwin, and
Kirkpatrick, Judge Rich dissenting with an opinion in which Judge Almond 
joined. In re Prater, 56 CCPA 1376, 415 F.2d 1378 (1969).

{5} In addition to new briefs of the principal parties and Bell Labs,
International Business Machines Corporation (IBM) and Honeywell Inc. 
(Honeywell) filed briefs amicus curiae, and Applied Data Research, Inc.
(Applied Data) and the Association of Independent Software Companies
(Software Association) filed a brief amici curiae.

{6} The principal parties argued the appeal.  All of the amici curiae,
with the exception of Honeywell, presented brief oral arguments.

{7} Consisting of Chief Judge Worley and Judges Rich, Almond, Baldwin, and
McGuire, Senior District Judge, District of Columbia, sitting by designation.

{8} Serial No. 155,236, filed November 20, 1961, for "Reduction of Data from
Spectral Analysis," allegedly a continuation-in-part of serial No. 49,921,
filed August 16, 1960, now abandoned.  The real party in interest appears to
be Mobil Oil Corporation, formerly Socony Mobil Oil Company, Inc.

{9} 35 USC 101, 102, 103, and 112 are the relevant sections of the statute.

{10} Claims 1, 6-9, 12, and 17-21 are directed to the "method," or "process."
See 35 USC 100(b) and 35 USC 101.

{11} Claim 10 is directed to the "apparatus," or its statutory counterpart "
machine," the terms being synonymous and used interchangeably here. 
See 35 USC 101.

{12} The set of equations is of the general form:

   Y(1)@(11)x(1)+a(12)x(2)+a(13)x(3) ... a(1m)x(m)

   Y(2)@(21)x(1)+a(22)x(2)+a(23)x(3) ... a(2m)x(m)




   Y(n)@(n.1)x(1)+a(n.2)x(2)+a(n.3)x(3) ... a(nm)x(m),


   Y(1), Y(2) ... Y(n)=peak heights of peaks 1, 2 ... n, respectively, of
   the spectrogram,

   x(1), x(2), x(3) .. x(m)=concentrations of components 1, 2, 3 ... m,
   respectively, present in the mixture,

   a(11), a(12), a(13) ... a(nm)=constant coefficients representing the
   contribution that each component 1, 2, 3

   ... m, respectively, makes to the height of each peak

   1, 2 ... n, and

   n and m are both integers, n being larger than m.

{13} "For example, [states the specification,] in a spectrogram obtained from
the analysis of a 10-component mixture in which there are 20 peaks from which
to choose, there are 184,756 possible subsets of 10 equations available."

{14} Of course, if the spectrographic data were error-free, or exactly 100%
accurate, then each subset of equations would yield the same concentration
of values.  However, any electrical measuring instrument, including the
spectrograph, has some degree of error in it and cannot produce error-free
data.  Thus, a variety of concentration values may be produced by the
different subsets of equations.

{15} "Error amplification," which occurs in the data analysis, is not to be
confused with the measurement error discussed in n. 14, supra, or the
magnitude of the latter.  So it is that the mere use of the peaks having
the most accurate heights per se does not ensure the most accurate
concentration values.

{16} So stated the board after it quoted from the examiner's Answer that:

   Appellants have discovered that those sets of simultaneous equations
   having the largest determinants are among those sets which can be
   solved with greatest accuracy.  * * *

Indeed, the solicitor has not controverted the novelty or non-obviousness of
appellants' discovery.

{17} As disclosed in the specification, the machine includes a
battery-energized motor which drives the armatures of ganged rotary switch
sections through a mechanical linkage.  The motor is stopped when the solenoid
of a relay is energized by the amplified output of a photocell which scans an
oscilloscope screen and is responsive to a maximum signal appearing thereon.
The signals produced on the oscilloscope screen correspond to the magnitude
of the determinants of the several subsets of equations, and are brought
together on the screen within the view of the photocell.  Movement of the
photocell downwardly across the face of the oscilloscope is accomplished
by the aforementioned motor via a gear box; and, when the photocell
photo-electrically senses the upper end of a determinant trace, its
electrical response is amplified and it energizes the relay to stop the
motor.  At this point, the machine has identified the "determinant of
greatest magnitude." Stopping of the motor simultaneously stops operation
of the motor-driven rotary switches, and the values of the circuits which
are established at that point are displayed on meters.  The meter readings
reveal the component concentrations.

The various mathematical coefficients to be dealt with in utilizing
appellants' method are represented in the disclosed machine by various
voltages in various circuits.  In this manner, the coefficients of the
relevant mathematical equations appear as electromechanical elements in
appellants' machine.

The specification sets forth in detail the electrical and mechanical
components of this machine and their respective functions to relate the
constant coefficients inherent in the present invention to a step in the
method and to a means in the machine which directly performs the functions
which underlie the present invention. The specification describes the
operation of the machine in relation to the mathematical content of the
invention and points out that, as the motor rotates the armatures of the
ganged rotary switch banks through pre-set positions, the values of the
determinants for the subsets represented by the positions can be

{18} The specification states that "the flexibility of a digital computer
lends itself to the solution of problems of the type above described and
in most instances represents an instrumentality preferred for the carrying
out of the method of the present invention." Nevertheless, no program for
a general-purpose digital computer, either in block diagram or machine
language form, has been disclosed.

{19} Directing himself at rehearing to appellants' allegation that they have
disclosed a statutory invention, the solicitor orally stated, "Nobody
questions that.  The examiner conceded it, and the board has conceded [it].
But the question is, what do the claims define?"

{20} The solicitor, in his petition for rehearing, argues that "the grant
of a patent containing process claims of such breadth as to confer upon a
patentee the right to exclude others from thinking in a certain manner"
would run afoul of the First, Ninth, and Tenth Amendments to the Constitution.
He urges that Article I, Section 8 must be construed in the light of the other
constitutionally assured rights and that freedom of mind or thought may not
be abridged by the patent laws.  However, amicus Bell Labs points out "that a
claim may verbally read on that which does not infringe without being itself
invalid for that reason alone." Bell Labs notes, first, that the breadth of
a claim may be judicially restricted so as to be no broader than the actual
invention when the claim is read in connection with the specification,
suggesting that no court would hold purely mental activity to be an
infringement even though the claim may "read on" such mental activity;
second, that no court would construe purely mental activity to be structure
or acts equivalent to the present machine and its function as described in 
the specification, 35 USC 112, third paragraph; and, third, that purely
mental activity would not be infringement since there would be no harm to
the patentee, an essential element for any tort.  In Bell Labs' view,
 freedom of mind or thought would not necessarily be abridged by the
grant of a patent with claims broad enough to read on mental processes.
As will be seen, we find it unnecessary to further discuss this aspect
of the case.

{21} In their Objections to the Grant of the Petition for Rehearing,
appellants stated that this court had not held "that a disclosure of thought
processes by way of equations and the like in the absence of any machine,
system or apparatus making use of them would be patentable." Appellants
go on:

   [The] mere fact that a claim absent the disclosure is broad enough to
   read upon a pencil and paper execution is not per se a reason why that
   claim should be rejected.  This does not say that a monopoly is in that
   way granted upon a pencil and paper execution.  * * * Infringement is
   never determined by the language of the claim divorced from the

In the same paper, appellants stated that they "have not urged * * * that
any claim will be infringed by one sitting at a desk and making mental
calculations.There is no such thing as mental infringement." In fact,
appellants argue that "Thought Is Still Unpatentable."

Appellants' reply to IBM's amicus brief states that they "have contended all
along that the meaning to be deduced from appellants' specification, as well
as the dictionary, suggests that the method claims do not cover purely
mental operations."

{22}  "Purely mental steps" are considered to be steps which may only be
performed in, or with the aid of, the human mind.  This is quite in contrast 
to "purely physical steps" which may only be performed by physical means,
machinery, or apparatus. Purely mental steps (e.g., "believing") are quite
different from purely physical steps (e.g., "heating") in many respects, not
the least of which is that the former are much less susceptible to specific
definition or delineation.  Between the purely mental and purely physical ends
of the spectrum there lies an infinite variety of steps that may be either
machine-implemented or performed in, or with the aid of, the human mind (e.g.,
"comparing" and "determining").  In ascertaining whether a particular step
is "mental" or "physical," each case must be decided on its own facts,
considering all of the surrounding circumstances, to determine which end
of the spectrum that step is nearer.  It may well be that the step of
"comparing" may be "mental" in one process, yet "physical" in another.
Disclosure of apparatus for performing the process without human
intervention may make out a prima facie case that the disclosed process
is not mental and is, therefore, statutory.  See Kayton, Patent
Protectability of Software: Background and Current Law, in The Law of
Software 1968 Proceedings B-25 (1968).  Here, of course, the patentability
of the disclosed process is not in question; but that is not to say that
the claims delineate a patentable process.

{23} At the time Abrams and Yuan were decided, the Patent Act of 1952, under
which we decide this case, was still in the future.  Couched in terms of
the present statute, it may well be that today purely mental steps are
unpatentable because purely mental steps may, at present, be too vague
and indefinite to comply with 35 USC 112. See n. 22, supra.  Whether or
not a sequence of purely mental steps comes within the bounds of "process"
as used in 35 USC 100 and 101 is, we feel, an issue which has never
been squarely decided.

{24} "That a process may be patentable, irrespective of the particular form of
the instrumentalities used, cannot be disputed." Cochrane v. Deener, 94 U.S.
780, 787. The word "art" appeared in the statute then in force and the word
"process" did not.  See R.S. 4886.

{25} Further expansion of the law relating to process patentability followed
shortly after in Tilghman v. Proctor, 102 U.S. 707 (1880) and The Telephone
Cases, 126 U.S. 1 (1887). See also Smith v. Snow, 294 U.S. 1 (1935) and
Waxham v. Smith, 294 U.S. 20 (1935), and the analyses thereof in
Judge Smith's previous opinion, supra.

{26} See n. 21, supra.

{27} See Kayton, Patent Protectability of Software: Background and Current
Law, in The Law of Software 1968 Proceedings B-25 (1968).

{28} See n. 21, supra, and text associated therewith.

{29} No reason is now apparent to us why, based on the Constitution,
statute, or case law, apparatus and process claims broad enough to
encompass the operation of a programmed general-purpose digital computer
are necessarily unpatentable. In one sense, a general-purpose digital
computer may be regarded as but a storeroom of parts and/or electrical
components.  But once a program has been introduced, the general-purpose 
digital computer becomes a special-purpose digital computer (i.e., a
specific electrical circuit with or without electro-mechanical components)
which, along with the process by which it operates, may be patented
subject, of course, to the requirements of novelty, utility, and
non-obviousness.  Based on the present law, we see no other reasonable

{30} By construing a claim as covering only patentable subject matter, courts
are able, in appropriate cases, to hold claims valid in order to protect the
inventive concept or the inventor's contribution to the art.  The patentee
at that time usually may not amend the claims to obtain protection
commensurate with his actual contribution to the art.

{31} This thought, in the public interest, is deemed to be paramount to an
applicant's interest, since the applicant is not foreclosed from obtaining
the proper coverage by express claim language.

{32} See the second paragraph under the heading "THE INVENTION", supra.

{1'}  Reminiscent of an old timer's lament that he was just beginning to
grasp the principle radio.