Appellant brought an action against appellee for a declaratory judgment establishing that appellee's Letters Patent No. 2,191,939 were invalid and not infringed by appellant.1 Appellee filed a counterclaim charging infringement of all 12 claims of the patent except claims 5 and 7 and demanding an injunction and an accounting for profits and damages. On trial the lower court found that all claims of the patent were valid and infringed except claims 5 and 7, which were found to be not infringed. A declaratory judgment was entered accordingly and appellant was enjoined from further infringement. This appeal followed.

Appellee's patent relates to an improved process for making halftone printing plates for reproduction of wash drawings in printed matter. A wash drawing is an original art work prepared by an artist on drawing paper using black water color paint diluted with water to produce various tones of gray when applied to the paper with a brush. Those portions of the work where paint has been applied are called tone areas. The parts where the white paper has been left untouched are called the highlight areas. The finished wash drawing is called 'copy.'

The halftone method of reproducing wash drawings is old in the photoengraving art. In this process a halftone negative is first made by photographing the copy with a halftone screen placed in the camera a short distance in front of the negative. This screen is like an optically perfect window screen mounted in glass. The reflected light from the wash drawing passes through the lens of the camera and then through the screen openings. The screen blocks a certain amount of this light from reaching the negative. Where the light passes through the openings in the screen it exposes the negative so as to record black dots of opaque deposit thereon. Where the light reflected from the drawing is of relatively great intensity, the light which passes through an opening in the screen will tend to 'spread out' when it strikes the negative, making a comparatively large black dot. Thus, the lighter tones of the wash drawing will be represented upon the negative by large black dots. These dots merge, forming small white dots between them. The black dots will be smaller on those portions of the negative corresponding to the darker tone sections of the copy, since less light is reflected from these areas. On a positive print it will be just the reverse; the lighter tones will be represented by black dots on a white background and the darker tones will show the screen pattern, with small white dots. It is this dot and screen pattern which creates the illusion of the various tones of gray in the reproduced drawing, with the relatively small and widely-spaced dots representing the lighter tones and the black screen pattern, with small white dots, giving the impression of darker tones.

After the negative is obtained it is developed and a positive print made on a zinc or other metal plate. This plate is then treated with acid which dissolves all of the surface of the plate except the resistant black portions. When ink is applied to the plate, only the raised black parts will be reproduced on the printing paper.

It is impossible to obtain a reproduction of a wash drawing with a clear white in the highlights by the halftone process alone. The obstruction by the halftone screen of light reflected from the highlights of the copy invariably results in small black dots appearing in the corresponding areas of the reproduction, giving the highlights a gray cast which is unattractive. In the prior art this dot pattern was eliminated by physically routing out the raised dots on the printing plates with a hand tool or machine, or by painting out the dots on the negative. Another method was masking the tone areas of the copy with a substance which made these areas relative non-reflective of light, and then timing the halftone exposure with reference to these areas of weak actinic value so that the unmasked highlight areas would be overexposed and the dot pattern in the highlights obliterated. All of these methods required skilled labor and were expensive, time consuming and imperfect. Still another method involved making the usual halftone exposure, then displacing the screen and making a second exposure. This had the effect of exposing and thus eliminating the dots in the highlights, but it involved complicated lens and screen manipulation to avoid over-exposure of the tone areas, and the lighter tones were often lost in the process.

We come now to a consideration of appellee's patented process. This patent was issued to Walter S. Marx on January 27, 1940 and will hereafter be referred to as the Marx Patent. While there are 12 claims in the patent, they differ only in the inclusion or omission of steps in the process and in matters of detail and emphasis. A typical claim, covering the essentials of the entire process, is printed in the margin.2

The process consists of three steps, as follows:

Step 1. The wash drawing is made by an artist on white paper or an equivalent reflector of ultra-violet light with any of the customarily used black water-color pigments diluted in what appellee calls a 'fluorographic solvent.' This solvent consists of an ultra-violet absorbent substance, preferably fluorescent, dissolved in water. The description specifies quinine bisulphate as the preferred ultra-violet absorbent, though other substances might also be used. Appellee now uses a different absorbent in its process.

Step 2. The usual halftone exposure of the copy is then made, with the halftone screen in place. This step results in an ordinary halftone negative, which is left in the camera for Step 3.

Step 3. A second exposure is made with the halftone screen removed or displaced, using only ultra-violet light. Ultraviolet light is light having wave-lengths of between approximately 1500 and 4000 angstrom units3 and it is invisible to the naked eye. The light sources preferred are the ordinary white flame carbon arc or a high intensity mercury vapor argon-filled arc lamp, both of which are rich in ultra-violet light. An ultra-violet transmission filter, which filters out all visible light and transmits only ultra-violet light, must be interposed between the light source and the negative. The filter in practice is either in the form of a sleeve-like covering over the lamp itself or is placed inside the camera immediately behind the lens. If the filter is placed over the lamp the exposure must be made in total darkness so that the negative will be exposed only by ultra-violet light. The preferred procedure is to place the filter behind the lens of the camera, in which case the copy may be illuminated by the ordinary rays of the lamp, consisting of both visible and ultra-violet light, and only the reflected ultra-violet light will reach the negative. There is no functional difference between placing the filter in one place or the other. In this exposure the reflected ultra-violet light from the highlights of the copy exposes and thus eliminates the dot pattern in the corresponding areas of the negative. The tone areas of the copy, however, have been made ultra-violet light absorbent by the fluorographic solvent. No ultra-violet light is reflected from these areas and the corresponding sections of the negative are therefore not affected in any way.

Steps 2 and 3 may be reversed in the process. The process is complete when the two exposures have been made.

Since we are of the opinion that the Marx Patent is not infringed by appellant's process, we shall proceed immediately to discuss that question. Appellee urges that the trial court's finding of infringement was a finding of fact, not to be disturbed unless clearly erroneous.4 However, it is well settled that where, as here, there is no dispute as to the evidentiary facts, and the record and exhibits enable us to clearly comprehend the nature both of the process patented and the alleged infringing process, the question of infringement resolves itself into one of law, depending upon a comparison between the two processes and the correct application thereto of the rule of equivalency.5 The testimony in this case was largely expository and descriptive of the elements and operation of the two processes and was not disputed.

The alleged infringing process utilizes the phenomenon of fluorescence to eliminate the dot-pattern in the highlights of a wash drawing.6 This process, like the process of the Marx Patent, may be broken down into three steps.

Step 1. A wash drawing is made by an artist on 'Kemart Illustration Board.' This paper is treated in the process of manufacture with a special fluorescent material which causes the paper to fluoresce brilliantly with visible light when irradiated with ultra-violet light. The ordinary black pigment is used in making the drawing but a liquid called 'Kemart Neutralizer' is used as a diluent for the pigment in place of water. This diluent contains quinine bisulphate, among other substances, and its function is to destroy the fluorescent quality of the Kemart Illustration Board when applied by the artist in making the drawing.

Step 2. The prepared copy is then photographed for the standard halftone exposure, and the negative is left in the camera for Step 3.

Step 3. A second exposure is then made with the halftone screen removed or displaced, using light radiated from 'Kemart Purple Camera' lights. These are white lamps, rich in ultra-violet light, with sleeve-like ultra-violet transmission filters fitting over them. The type of filters used are well known in the photographic arts and there was evidence that the same type of filter is used in the process of the Marx Patent. The filters transmit light which is 90-95% ultra-violet; the small amount of visible light transmitted is of no photographic consequence. The ultra-violet light activates the fluorescent material in the highlights of the copy where the Kemart paper is exposed. The highlights are thus caused to radiate visible fluorescent light with wave lengths of 4100 to 4800 angstrom units. In the tone areas the fluorescent quality of the paper has been destroyed by the application of Kemart Neutralizer, and therefore no light is emitted from these areas. No ultraviolet light is reflected from either the tone areas or highlights of the copy; the copy absorbs ultra-violet light in all areas. To make sure that only the visible fluorescent light reaches the negative, a Kodak Wratten 2A filter may be used just behind the lens. This filter blocks all ultra-violet light and permits only visible light to pass through to the negative. The result is that the dots in the highlights of the negative are exposed to visible fluorescent light and obliterated while the tone areas are not affected.7

On these facts the court below found that the Kemart Process includes all of the steps of, and is described by, every claim of the Marx Patent except claims 5 and 7 thereof; that the Kemart Process proceeds in substantially the same manner as does the method of the Marx Patent to accomplish the same result; and that the dot eliminating visible fluorescent light which in the Kemart Process moves from the highlight areas of the drawing into the camera 'is the full equivalent of ultra-violet light reflected unmodified from the highlight areas of the drawing for the same purpose.'

We cannot agree. In so holding we assume, without deciding, that the process of the Marx Patent is a meritorious invention, entitled to a relatively broad range of equivalents.8

Appellee contends that because the Kemart Process involves use of a type of copy which absorbs ultra-violet light in its tone areas, a standard halftone exposure, and a second exposure using ultra-violet light, the process falls within the literal terms of the claims of the Marx Patent and resort to the doctrine of equivalents is unnecessary. We believe there are sufficient differences in the Kemart Process to avoid literal infringement. At any rate, the fact that the claims of the Marx Patent are broad enough to cover appellant's process does not establish infringement. The claims are to be read in connection with the specifications, and a patentee's broadest claim can be no broader than his actual invention.9

The test of identity or equivalence of two processes is not the apparatus or materials used but whether they involve identical or equivalent steps.10 Steps are equivalent if they work in substantially the same way to accomplish the same results.11

The similarities in the two processes are few and superficial. It is true that Kemart Neutralizer when applied on ordinary drawing paper absorbs ultra-violet light and may be used in place of fluorographic solvent in the Marx Process with fairly satisfactory results.12 But the fact that the neutralizer is an ultra-violet light absorbent is unimportant in the Kemart Process, for its sole function in that process is to destroy the fluorescent quality of the Kemart paper. The neutralizer could be an excellent reflector of ultra-violet light without affecting the efficacy of the Kemart Process, since all reflected ultra-violet light would be blocked from reaching the negative by placing a Wratten 2A filter at the lens of the camera.13

Neither are the two processes equivalent by virtue of the fact that both use ultra-violet light. Again, the function of the light in the Kemart Process is very different. In the Marx Process the ultra-violet light is reflected from the highlights of the copy to the negative so as to efface the dot pattern therefrom. In the Kemart Process the purpose of ultra-violet light is to excite fluorescence in the highlights of the copy. It is essential that ultra-violet light only be radiated to the copy. At the copy the ultra-violet light is absorbed both in the highlights and in the tone areas. Once absorbed, the ultra-violet light as such ceases to exist. Its function has been performed. When the fluorescent material in the highlights of the copy absorbs the ultra-violet light a change takes place in the structure of the atoms of this material. The changed structure is unstable and in returning to its former state energy is released. This energy consists of radiations of visible light of longer wave lengths than the ultra-violet. It is this visible light which exposes the negative and eliminates the dots in the highlights.

Thus, while the Marx Process involves use of reflective photography, the Kemart Process uses what is known as 'direct light' photography. In the Marx Process the same ultra-violet light which is radiated to the copy is 'cast back' from the copy to expose the negative. In the Kemart Process the photographically effective light is emitted from the copy itself; the dot-eliminating exposures involves the same principle as that involved in photographing the headlights of an oncoming automobile.

Comparing the two processes, we find that the Kemart Process involves use of a different kind of copy, a different kind of dot-eliminating light, the wholly different phenomenon of fluorescence, and a different kind of photography. These differences go to every feature of the Marx Process alleged to be novel. They do not appear to be those 'slight, well known variations'14 or those 'unimportant and insubstantial changes and substitutions'15 in a patented process which would characterize a piracy of the substance of the invention.

We recognize that equivalence, in the patent law, is 'not the prisoner of a formula and is not an absolute to be considered in a vacuum', and 'things for most purposes different may sometimes be equivalents.'16 The test is whether the steps in the alleged infringing process were known equivalents in the particular art here concerned when Marx obtained his patent.17 'An important factor is whether persons reasonably skilled in the art would have known of the interchangeability of an ingredient not contained in the patent with one that was.'18

Certainly it cannot be said that, at the time of the issuance of the Marx Patent, ultra-violet reflective and fluorescent photography were known equivalents in the photoengraving art. If the use of ultra-violet reflective photography was itself novel in the art, then fluorescent photography could not have been a 'known equivalent.'

Appellee contends, however, that the two methods were recognized photographic equivalents. Assuming that this would sustain a finding of equivalence of the two processes here in question, there is no evidence to support the contention. Appellee points to no particular applications of the two methods in which they could be said to operate similarly with similar results. The testimony relied upon establishes only that both methods of photography were 'known' prior to the issuance of the patent. Appellee relies upon a work entitled 'Ultra-Violet Photography' published by the General Electric Company in 1936. This paper shows only that the two methods were known and that both involve the use of ultra-violet light. In fact, in the introduction to this work it is stated, in referring to the two methods: 'There are two distinctly different methods of using ultra-violet radiation for making photographs.' (Emphasis ours.) The paper proceeds to point out the very substantial differences in the methods. The applications described are the deciphering of altered documents and the examination of old paintings by photographic means. Far from stating that the two methods are similar or interchangeable in these applications, it is said that the question whether one method or the other will be useful in a given situation can be determined only by experiment. This is not an assertion of equivalence, even in this context. Still less does it lend itself to the inference that the two methods were 'known equivalents' in all of their conceivable applications, including the one here involved.

All of the claims of the Marx Patent state that the dot-eliminating photographic exposure is made 'by,' 'with,' or 'in the presence of' 'ultra-violet light only.' The discussion between court and counsel below indicates that the court considered the use of 'ultra-violet light only' in illuminating the copy as being the essence of the Marx Process, and since Kemart uses the same light at this phase of the photographic process, the differences in the way the light is used are superficial and do not go to the heart of the invention. It appears also to have been the lower court's view that since the desired results could be achieved according to Marx's teaching by using only ultra-violet light, the use of fluorescence was merely an addition without a difference. Apparently these views underlie the finding of the court that the visible light moving from the Kemart copy is the 'full equivalent' of ultra-violet light reflected from the copy in the Marx Process. Since the term 'ultra-violet light only' in the Marx claims is obviously not so unlimited as to comprehend a process using wholly visible light, this finding of the court can be rationalized only on the ground that the kind of light moving from the copy to the negative is unimportant, so long as ultra-violet light is used to illuminate the copy.

We cannot agree with this construction of the term 'ultra-violet light only' in the claims of the Marx Patent. The claims are ambiguous in this regard; it is not clear whether the term 'ultra-violet light only' refers specifically to the light illuminating the copy or to the light striking the negative. It is likely that the term was intended to refer indifferently to either or both.19 The same ultra-violet light is involved at both points; the ultra-violet light is radiated from the light source to the copy and is reflected back to the negative. But if it is doubtful whether the term refers to the 'ultra-violet light only' at the point where it strikes the copy or at the point where it strikes the negative, we think it is plain that it was meant to describe the light which is photographically effective in the dot-eliminating exposure, i.e., that light which ultimately strikes the negative. The specification teaches that the ultra-violet transmission filter may be placed at any point between the light source and the negative. If the filter is placed over the light source, it is both 'ultra-violet light only' which illuminates the copy and 'ultra-violet light only' which strikes the negative. If, as in Marx's preferred procedure, the filter is placed in the camera, it is ultra-violet light plus visible light which illuminates the copy, but it is 'ultra-violet light only' which exposes the negative. In any event it must be 'ultra-violet light only' which exposes the negative and the claims clearly so indicate. The use of 'ultra-violet light only' as claimed is, by appellee's own assertions, the heart of the alleged invention of Marx. The problem is whether these claims can be said to include a process using visible light only to expose the negative in the dot-eliminating exposure. We think they cannot.20

It is apparent in the record that it is the use of ultra-violet light in both processes which chiefly influenced the determination of the court below. The problem may be made clearer by assuming a variation in the facts. For instance, infra-red rays excite fluorescence in certain substances. If infra-red rays were used in the Kemart Process, we think there would be little doubt that the process would not infringe. Yet from a functional standpoint the ultra-violet light of the Kemart Process and the infra-red rays of our hypothetical case are the same, since both would be used to excite fluorescence in the copy. A finding of infringement here would be tantamount to a finding that two dissimilar machines, which operate in an entirely different way, are equivalents because the same kind of energy used to operate one is used to power the switch of the other. The ultimate results, i.e., the finished negatives, of the two processes are alike, but similarity of result is not sufficient to show infringement.21 It might be said that both processes rely upon the broad principle that by proper use of a particular kind of light in connection with a particular type of copy the dots in the highlights of a halftone negative may be photographically eliminated without affecting the tone areas. But Marx was given a patent for a process; he did not, and could not, patent a principle.22

There is still another reason why the judgment cannot be affirmed. At the close of the trial appellee consented to a finding and declaratory judgment that claims 5 and 7 of the Marx Patent were not infringed by appellant. Appellee consented to this finding and judgment for the sole reason that in these claims it is expressly stated that ultra-violet light is reflected from the highlight portions of the Marx Copy in the dot-eliminating exposure. In the Kemart Process this reflection does not occur.

It is important to note that the question of infringement of claims 5 and 7 as well as all other claims of the patent were put in issue by petition and answer in appellant's action for a declaratory judgment. It was stipulated by the parties prior to trial that an actual controversy existed with respect to the allegations of appellant in his petition. In the discussion in the court below prior to the granting of appellant's motion for judgment of non-infringement of claims 5 and 7, appellee admitted that the question of infringement of these claims was in issue. Appellee, it is true, did not charge infringement of these claims in its counterclaim, but they were not thereby removed as issues in the case.23 The judgment of non-infringement of these claims was final and binding on appellee. The effect of the judgment and of appellee's consent thereto cannot be avoided by the assertion that it was unnecessary to pass upon infringement of these claims in view of the other and more broadly stated claims. Appellee is bound by the admissions inherent in his consent to the finding and judgment in question.

To support his contention that this finding and judgment is consistent with a finding of infringement of the other claims, appellee invokes the rule that in interpreting a series of claims, a limitation not present in one must not be implied where the same limitation appears in later claims.24 The reason of the rule is that each claim is in theory a separate patent, so that two claims should not be so construed as to make them identical.25

But this rule of construction is subordinate to the controlling rule that a patentee's broadest claim can be no broader than his actual invention, no matter how it may be expressed or what other claim his patent may contain.26 If, therefore, the Marx Patent is limited to a process involving reflection of ultra-violet light in its dot-eliminating exposure, expressly or by necessary implication, appellee has conceded away his entire case on infringement.

If the claims of the patent are not so limited, then we cannot fathom what those claims mean. The claims are to be read in the light of the specifications.27 The specifications expressly state, not once but twice, that ultra-violet light is reflected from the highlights of the copy in the dot-eliminating exposure. They also teach that the paper on which the wash drawing is made must be a reflector of ultra-violet light. All of the claims are clear that the tone areas must be absorbent of ultra-violet light. If reflection is not necessarily involved in the process, no reason appears for using reflective and absorbent materials in the copy. It is conceded that the light source must be rich in ultra-violet light, and it is undisputed that it is ultra-violet light which moves from the copy to expose the negative. Appellee informs us of no means, and the Marx Patent teaches none, whereby this could take place unless it be by reflection of the light from the copy. The court below found as a fact that the fluorescent light of the Kemart Process is the 'full equivalent of ultra-violet light reflected unmodified from the highlight areas of the drawing for the same purpose.' (Emphasis ours.) If the Marx Patent is not restricted to use of ultra-violet reflective photography, then this finding was irrelevant. And if it is not so restricted, the vigorous argument of appellee that fluorescent and ultra-violet reflective photography are equivalents is similarly superfluous. In his application for a patent Marx attached no particular significance to the word 'reflected' in claims 5 and 7 of the patent as issued. Claim 5 was stated to be inserted to 'introduce the movement of the screen between the impressions,' claim 7 to specify that 'the exposure to ultra-violet light is admitted in sufficient quantities to remove the screen pattern from the highlight sections.' There is not a single indication in the claims, the specifications, the file wrapper or the record that the dot-eliminating exposure in the Marx Patent is or could possibly be effected by any means other than the reflection of ultra-violet light from the highlights of the copy.28 Appellee's concession of non-infringement of claims 5 and 7 left nothing more to decide since those claims merely define the limits of his invention. The effect is the same as if appellee had consented to a finding and declaratory judgment that none of the claims of the Marx Patent was infringed by appellant.

We believe it unnecessary to discuss the question of validity. In oral argument counsel for appellant stated that appellant had no interest in the question of validity of the patent if it should be held uninfringed.29 It has been said by the Supreme Court that it is the 'better practice' always to pass upon the question of validity.30 This language has been generally understood as a mere cautionary admonition to the courts to exercise their discretion whether to pass upon the question of validity, and to strike down clearly invalid patents where it would be in the public interest to do so, even though a finding of non-infringement would dispose of the case.31 A court may not, however, find a patent uninfringed and at the same time hold it valid, since a holding of validity would be a decision of a hypothetical case.32 Thus, if we should go fully into the question of validity of the Marx Patent here, an issue which is as 'fugitive, impalpable, wayward, and vague a phantom as exists in the whole paraphernalia of legal concepts',33 and should conclude that the patent is valid, we should not be able to so hold. Suffice it to say that the patent in suit is not clearly invalid.34 A brief study of this patent against the background of the prior art leads to quite the contrary conclusion, but a decision on this question should await a case requiring it.

That part of the judgment holding claims 5 and 7 of the Marx Patent not infringed is affirmed. That part of the judgment holding the remainder of the claims of the Marx Patent infringed is reversed. So much of the judgment as holds the Marx Patent valid is ordered stricken. The cause is remanded to the District Court with directions to enter judgment holding all claims of the Marx Patent not infringed by appellant and to order, adjudge and decree that appellee take nothing by its counterclaim and that appellant recover costs of appellee.