WOODBURY, Circuit Judge.

The Leeds Northrup Company, hereinafter for convenience referred to as the plaintiff, a Pennsylvania corporation, filed a petition for a declaratory judgment against the Doble Engineering Company, hereinafter referred to as the defendant, a Massachusetts corporation, praying for an adjudication that patent No. 1,945,263, issued on January 30, 1934, to Frank C. Doble and assigned by him to the defendant, is invalid, or, if valid, that it is not infringed by certain apparatus manufactured by the plaintiff. The defendant answered and filed a counterclaim alleging validity and infringement. The case was referred to a master who found Claim 15 of the patent valid but not infringed. Neither party questioned the master's findings of fact and the district court adopted them as its own. The plaintiff made no objection whatever to the master's report but the defendant objected to it on the ground that the master erred in ruling as a matter of law that the plaintiff's apparatus did not infringe. After hearing, the district court concluded that this objection of the defendant must be overruled and the report affirmed, and judgment in that court was entered accordingly. Thereupon the defendant took this appeal.

The facts are exceedingly complicated and involved, but fortunately they are not in dispute and, also fortunately, they do not have to be stated in complete detail in order to understand the issues raised by this appeal. It will suffice to explain the patented apparatus and its setting in the electrical art to which it appertains in general terms only.

The Doble patent is for an "Apparatus for Testing Insulating Values". Describing it the master says: "The Doble testing system (disregarding for the moment protection against charging currents) includes a voltage source, the grounded specimen of insulation to be tested, conductors or leads between the voltage source and the specimen to be tested, and a set of measuring instruments between the voltage source and one side of the specimen." For clarity we venture to describe the apparatus in more detail. Basically it consists of a source of electrical energy (alternating current), typically of 110 volts; a transformer to step this current up to any desired voltage, usually up to approximately 10,000 volts; a lead wire or conductor from the high tension terminal of the secondary coil of the transformer to one side of the insulator to be tested, in practice this is usually to the conductor which passes through the center of a bushing on an oil circuit breaker; another lead wire from the other side of the insulator under test, which side in practice is usually if not always grounded, to electrical measuring instruments; and another lead wire or conductor from these instruments to the low tension terminal of the secondary coil of the transformer.

It appears that this general system for testing insulators was worked out some forty or fifty years ago by Steinmetz and that it worked satisfactorily when the insulator under test was not grounded, and when a stationary apparatus was used in a laboratory where there were no external electrical interferences from high tension lines, generators etc. in the neighborhood. It was not satisfactory, however, when embodied in a portable apparatus and used to test installed insulators only temporarily disconnected from service. The reason for the unsatisfactory operation of the old system in the field is that insulators when in place for use are almost invariably connected to ground; that when in use they are usually located in close proximity to high tension transmission lines; that moving a portable apparatus around for different tests causes its parts to be in different relationship to the surface of the earth for each test made, and that the condition of the earth's surface as wet or dry varies from place to place and from day to day. It seems that these factors cause stray electric currents which could not be eliminated or compensated for, to pass through the measuring instruments of the apparatus and vitiate their readings.

These stray currents are of two kinds; those which arise from the flow of electricity through the apparatus itself, and those which are caused by the different positions of the apparatus with respect to the surface of the earth and the wet or dry condition thereof, and by the presence in the neighborhood of the testing apparatus of high tension transmission lines and similar electrical installations. The former are called "charging currents", but we shall refer to them hereafter by the more specific name of "internal charging currents". The latter are variously called "extraneous electrostatic disturbances", "external electric stresses" or "induced voltages from extraneous electrostatic fields". For simplicity we shall refer to them by the possibly technically inaccurate but descriptive and convenient name of "external charging currents".

It appears that in the Steinmetz apparatus which we have described internal charging currents could be prevented from passing through its measuring instruments so that accurate readings were obtainable but that it was only possible to do this when the insulator under test was not grounded. There does not appear to have been any known way of avoiding in it the disturbing effects of external charging currents. Thus by the old system the only known way to test insulators was to disconnect them and take them bodily to a laboratory where a stationary apparatus could be used, where external conditions could be controlled and where the insulator being tested could be disconnected from ground.

This technique for testing was unsatisfactory and Doble directed his attention to the problem of supplanting it by a better one. He solved the problem by devising means for protecting the old apparatus from the effects of both internal and external charging currents so that as a result it could be used to test insulators under service conditions at the place of installation, as in the yard of an electrical station, that is, in the field. The disadvantages of the old system, the need for improving upon it, and the problems which confronted Doble in supplanting it are stated in greater detail in the first paragraphs of the patent and in the excerpt from the master's report quoted in the margin as footnotes two and three respectively.

It is admitted that the apparatus which Doble devised rendered the old technique for testing obsolete and, for the purposes of this case at least, it appears to be conceded by the plaintiff that Doble's contribution to the art in so doing is entitled to rank as an invention. Concerning it the master found: "The Doble tester made in accordance with this patent has had wide use, and witnesses for both parties agreed that it has been an important contribution to the art. By its use many deteriorated bushings have been located and removed or repaired in cases where they would otherwise have failed in service, with resulting great trouble and expense. The tester of the patent is simple to operate and gives readings from which reliable results can be consistently obtained." The master found that it soon enjoyed considerable commercial success.

Doble solved the problem of protection against charging currents by devising a system of what are called electrical shields; an inner shield to cope with internal charging currents and an outer shield to cope with those which originate outside of the apparatus itself. By means of these shields, which need not be described, he succeeded in keeping practically all charging currents of both kinds out of the measuring instruments of his apparatus so that with it insulators could be accurately tested while still installed.

We turn now to the plaintiff's apparatus. After the advent of the Doble tester the plaintiff began to manufacture two forms of apparatus, called respectively its 1935 and its 1936 sets, both of which served the same purpose as the Doble apparatus. That is to say, either of them could be successfully used to test the insulating value of dielectrics in the field. There can be no doubt that the defendant felt the plaintiff's competition.

The master found that in both of the plaintiff's sets a shield which "corresponds in substance" to the defendant's inner shield was used to eliminate the disturbing effects of internal charging currents. But he found that the plaintiff in each of its sets succeeded in eliminating the effects of external charging currents by devices or techniques which were wholly dissimilar to the defendant's outer shielding system. External charging currents occasioned by the proximity of high tension lines, etc., were averaged out of the plaintiff's 1935 set by using a reversing switch, and disturbances of the same kind were avoided in the 1936 set by energizing it from a source of alternating current of a different frequency from that ordinarily used commercially. External charging currents arising from varying the positions of the apparatus and from the variable moisture content of the surface of the earth were eliminated from each set by means quite unlike those used by Doble. Thus the master found, and the defendant concedes, that the means used by the plaintiff in both its 1935 and 1936 sets for protecting the measuring instruments thereof against external charging currents were not "the same means as Doble's" nor were they "the equivalent thereof in the sense of the patent law", although they accomplished "the same ultimate result".

In spite of the master's finding of essential equivalence between the plaintiffs and the defendant's inner shielding systems he did not find infringement because he ruled that in Claim 15 of his patent Doble claimed not only his inner shielding system which the plaintiff used but also his outer shielding system which admittedly the plaintiff did not use. The defendant takes exception to this ruling, saying that, properly construed, Claim 15 of the patent covers only Doble's inner shielding system so that the plaintiff must be found to infringe because it used an inner shielding system which the master said "corresponds in substance" to Doble's. The solution of the controversy, then, hinges upon the interpretation of Claim 15 which reads:

"An apparatus for measuring in the field electrical qualities of insulation interposed between a grounded structure and a high tension terminal, comprising a source of alternating voltage adapted to be connected in circuit with said high tension terminal and said grounded structure, measuring means interposed in circuit between said voltage source and one side of the insulation for taking measurements determinative of energy loss and current, and means for shielding said voltage source and its connection with said high tension terminal so as to divert their charging currents away from the measuring means."

Although Doble's outer shielding system is not specifically mentioned in the claim in suit the master concluded that because any testing apparatus for use in the field must have some means for eliminating external charging currents his outer shielding system must be included in the claim by necessary inference. He reasoned: "As has been stated, one of the conditions for the use of such measuring apparatus 'in the field' is that it must have some provision for eliminating the effect of electrostatic disturbances which are almost always encountered in the field. This is a vital part of the invention. Claim 15 is not in terms limited to any specific means for accomplishing that end. The literal language of the claim would be satisfied by the provision of any such means and the plaintiff admittedly provides means for accomplishing the same result but in a way which is not Doble's way and which is not an equivalent of Doble's way in the sense of the patent law. To hold that the plaintiff's equipment is within the claim would in effect give the defendant a patent for a result, so far as the limitation 'in the field' is concerned, which is not permissible. The Doble patent is entitled to cover the, Doble structure and its equivalents, but is not entitled to cover all means for accomplishing the same result." In consequence he ruled that "the words 'in the field' in claim 15 restrict the claim to Doble's outer shield or to equivalents thereof and that since neither the plaintiff's 1935 set nor its 1936 set uses such shield or its equivalent, the plaintiff has not infringed."

The defendant, in support of its assertion that this ruling is erroneous and that Claim 15 covers only the inner shield, it being "an ordinary sub-combination claim directed not to the whole system shown in the patent, but to a subsidiary combination of elements which, on the master's findings, was new with Doble", says: (1) that since the phrase "in the field" occurs in the introductory clause of the claim it does not limit the claim but is merely a preliminary statement specifying "the type of machine in which the claimed subsidiary combination of elements works an improvement over the prior art", (Williams Mfg. Co. v. United Shoe Machinery Corp., 316 U.S. 364, 369, 62 S.Ct. 1179, 1182, 86 L.Ed. 1537) and (2) that since Doble's outer shield is specifically covered by other claims of the patent which are not here in suit, the master's ruling violated "the established rule of claim differentiation", that is, the rule that "If some of the claims are limited to details, those which remain are prima facie not to be fettered by insisting that they contain, as necessary elements, the particulars which are specifically covered elsewhere". Bresnahan v. Tripp Giant Leveller Co., 1 Cir., 102 F. 899, 900.

The plaintiff, in addition to asserting that the master's construction of the claim is correct and the defendant's construction of it is incorrect, contends that construed as the defendant would have it, the claim would be invalid. In its brief the plaintiff says: "The external shielding system, which is absolutely necessary to an apparatus capable of realizing the object stated in the preamble of the specification (use 'in the field') and overcoming the difficulties therein set forth, is a brilliant invention * * *," but "The internal shielding system, which merely diverts the intra-set effects or internal charging current from the meters (irrespective of the place of use of the test set) is the mere carrying forward of an old idea by substantially the same means, in fact, by old means, to get slightly better results."

The foregoing states the essence of the conflicting arguments of the parties.

It appears to be firmly established that the question of the meaning of a patent is one for the court as distinguished from one for the jury and that a patent is subject to the same general rules of construction as any other written instrument. 2 Walker on Patents, Deller's Ed., §§ 245 and 240. But in addition to the general rules of construction applicable to all written instruments, the courts over the years have formulated a great number of minor rules or canons of construction applicable to patents only. id. ch. XI. Possibly the courts, as non-expert tribunals, felt inadequate to definitely and finally decide the question of the meaning of such a highly technical document as a patent, many of which are extremely difficult if not impossible to understand by one not versed in the arts to which they appertain, and hoped by formulating these minor rules to provide definite guides to meaning for themselves and others to follow in the future. If this was the hope it seems to us that it has not been realized but that on the contrary these numerous overlapping and sometimes conflicting canons of construction and the exceptions thereto shed only an illusion of light upon, and so only add confusion to, the exceedingly difficult question of the meaning of a patent.

To illustrate: A patent covering what is called a pioneer invention, that is, one covering a wholly novel device, or one of such novelty and importance as to mark a long step forward in the progress of the art to which it appertains, is to be construed liberally but one which covers a device which marks only a minor step forward in a crowded art is to be narrowly construed. Then there is another rule to the effect that a patent covering an invention or discovery of great utility which has enjoyed practical success is to be liberally construed, while a mere paper patent, that is, one covering something which has never gone into practical commercial use, is to be narrowly construed. Considered separately these rules do not seem unreasonable, — they may even be helpful in some cases, — but considered together they breed confusion because we are aware of no rule indicating the proper attitude for a court to take when called upon to construe a pioneer patent which for some reason or another, maybe because of lack of funds or lack of aggressiveness, or possibly even sheer obstinacy on the part of the patentee, has remained a paper one. Another illustration is found in the case at bar. As we have already stated the defendant here, citing cases, relies upon the rule that words in the introductory clause of a claim import no structural elements into it, so that, it says, the words "in the field" merely specify the type of machine in which the elements actually claimed work as improvement over the prior art. In answer to this argument the plaintiff, although it admits that "In some cases the introductory clause of a claim is a mere title and is not a limitation", says that there are other cases, and it claims that the case at bar is one of them, in which an introductory clause in a claim "cannot be ignored and is a limitation", so that, it says, Claim 15 here in suit, although it specifically claims only the inner shield, must be limited by construction to claim that shield only when used in conjunction with an outer one. It cites cases in which introductory clauses have been held to have such a limiting effect. Neither party, however, cites any case, and we have found none, in which any indication is given as to how a rational choice is to be made between these divergent lines of authority. Of course we are at liberty to choose between them without stating any reasons for our choice, but this is not a candid and straightforward technique of decision.

It seems to us that what we have said indicates how unsatisfactory it is to solve the problem of the meaning of a patent by resorting to minor canons of construction which, instead of providing reasons for a result provide only pegs upon which a result already reached in some other way may conveniently be hung. In consequence we prefer to decide the question of the meaning of a patent, not by heavy reliance upon subsidiary canons of construction, but rather by resorting to broad general principles applicable to the construction of all similar written instruments.

The courts have said (for authorities see 2 Walker on Patents, Deller's Ed., §§ 240, 245) that letters patent are contracts. This seems to us too broad a statement. If patents are contracts at all, surely they are contracts of a peculiar sort. But the fact remains that patents, like contracts, are bilateral instruments, and this common feature makes the rules for the construction of contracts applicable to them. Thus as we see it our problem is to determine first what a patentee intended to claim as his invention or discovery and second upon what invention or discovery the patent office intended to grant a temporary monopoly. To make this determination we turn to the words of the patent, viewing them as objectively as we would view the words of any ordinary contract.

The rule that the claim or claims of a patent measure the scope of the invention is too well established to call for the citation of authorities and equally well established is the rule that claims are to be construed with reference to the specification and in the light of the drawings and of the prior state of the same and analogous arts. That is, a claim in a patent, like a clause in a contract, is to be construed in connection with the other terms of the instrument of which it forms a part and the whole instrument is to be interpreted with reference to the circumstances surrounding its inception. So viewed it seems to us that in the claim in suit the patentee intended to claim only his internal shielding system.

From the paragraphs of the specification quoted in footnote two above it is clear that what the patentee had primarily, although not exclusively, in mind was an apparatus for testing grounded insulators or dielectrics "in the field" and that he well knew that any such apparatus to be useful must have means for preventing both internal and external charging currents from vitiating the readings of the sensitive electrical measuring instruments required to measure the minute currents of electricity passed through the insulator by the apparatus. The means he provided, and which undoubtedly worked, were the inner and outer shielding systems which we have mentioned. Obviously then, he must have intended to claim both systems, and looking at all of the claims of his patent it seems to us that this is in fact what he did.

In many of the claims of the patent (it contains nineteen), he claims his outer shield, sometimes in connection with his inner one and sometimes not, and in several claims beside the one here in suit he claims only his inner shield. This strongly indicates that Doble intended to claim his shields when used together and also to claim each shield when used alone. And the surrounding circumstances do not contradict this indication. From the record it appears to be conceivable that there might be occasions when it would be desirable to test ungrounded insulators in the field, and in this event the testing apparatus would only have to be shielded against external charging currents since internal ones could be dealt with by means other than an internal shield. Reasons also exist for Doble to have wished to claim his internal shield when used without his outer one. The quotation from the patent in footnote two above shows that Doble had in mind the possible use of his apparatus in a laboratory, a place where external charging currents are not ordinarily encountered, but where, as the record shows, there may be occasion to test grounded insulators. Thus in an apparatus designed for use in a laboratory there would be no need for an external shield but there would be a need for an internal one.

From the words of the claim in suit, in which no specific mention is made of the external shield, read in connection with the other claims of the patent and viewed against the background of the electrical art, we conclude that Doble in Claim 15 intended to cover only his inner shield, that is, to claim it as a separable part, or sub-combination of parts, of his entire apparatus. And this was a permissible technique for him to have adopted under the rule of Deering v. Winona Harvester Works, 155 U.S. 286, 302, 15 S.Ct. 118, 124, 39 L.Ed. 153, in which the Supreme Court said: "Admitting that additional elements are necessary to render the device operative, it does not necessarily follow that the omission of these elements invalidates the claim, or that the precise elements described in the patent as rendering it operative must be read into the claim." See also 2 Walker on Patents, Deller's Ed., § 166 (p. 789) where it is stated "A part or a combination may be claimed separately, though it cannot do useful work separately from the residue of the machine or apparatus of which it constitutes a part", and cases in point are cited. The foregoing disposes of the master's conclusion, that the claim construed in this way is invalid for the reason that it is for a result.

The question now arises whether or not the Commissioner of Patents intended to grant a temporary monopoly to Doble on his internal shielding system standing alone. The plaintiff contends that the proceedings in the patent office show that the commissioner had no such intention and that he indicated as much to the patentee. That is, the plaintiff argues that the defendant is estopped by the file wrapper from contending that Claim 15 covers only the internal shielding system.

The file wrapper in this case is voluminous and much of the language used therein is highly technical. In all honesty we cannot pretend that we fully understand every word of it. It clearly appears, however, that claim 15 of the patent is identical with application claim 37; that this claim was offered by way of amendment after the rejection of earlier application claims 13, 15 and 17, and that these last claims were offered as amendments after the rejection of original application claims 5, 6, 7 and 8.

In the application as it was filed in the first place Doble mentioned testing insulators in the field as the primary use for his apparatus, but he also said that it "may be used in the laboratory", and in that application he made twelve claims, in four of which, the 5th, 6th, 7th, and 8th, he claimed his internal shielding system without specifically mentioning his outer one. The phrase "in the field" does not appear in any of these original claims but in three of them, the 2d 4th, and 12th, he claimed "a portable field apparatus". All of these original claims were rejected as unpatentable over an earlier patent to one Hadaway, the examiner in charge saying in part: "Although no shielding means is shown (in Hadaway) it obviously would not involve invention to provide shields for the apparatus and connecting wires in view of the patent to Strauss."

In response to this action of the patent office Doble offered some amendments to his specification, cancelled all of his original claims, and offered twelve new ones which he numbered 13 to 24. Of these claims only 13, 15, and 17, specifically covered the internal shielding system standing alone, the others, except the last which covered a feature not involved in this litigation, specifically claimed both systems. Neither the phrase "a portable apparatus" nor the phrase "in the field" appears in any of these claims. But in "remarks" filed with these amendments Doble not only stressed the importance and novelty of his invention but he also said that it "has accomplished for the first time the valuable result of enabling measurements of great delicacy and accuracy to be performed in the inspection of dielectrics, at the places where disturbing electrical stresses of great magnitude are encountered", and by this he must have meant "in the field" because that is the only place where such stresses are likely to be found. He then went on to say: "This accomplishment is due to the system of shields which not only protect the instruments against external electrical stresses and disturbing influences, but, what is still more important, confine the currents to their prescribed paths in the apparatus itself and eliminate the influence of the high tension current and connecting cable from the measuring instruments. The shielding system for the latter purpose, and its combination with the shielding means against external influences are wholly new. Applicant understands that the examiner recognizes these elements of novelty and concedes patentable invention therein." In his "remarks" the applicant then went on to differentiate his shields from the one shown in the patent to Strauss by saying that Strauss' shield is at the most only an external one, as well as that it differed in other respects, and then added that he had "created an apparatus capable of being brought into the field, and there used to make measurements of extreme delicacy under conditions where the disturbing factors otherwise would make the readings erroneous and entirely unreliable".

The patent office, however, rejected claims 13, 15, and 17, as well as some of the others which were offered in amendment to those originally filed, on an Electric Journal paper in view of a patent to one Shackelton and a British patent. With respect to the paper the examiner in charge said: "Although no shielding means is disclosed there is no invention broadly in providing shields for the instruments, source, and connecting wires in view of the common use of shields with measuring apparatus as disclosed by either Shackelton or the British patent." He also rejected Doble's claims covering both shielding systems on the two above patents saying that each of them "shows measuring apparatus using two sets of shields. One set of shields is grounded and the other set is connected to one of the terminals of the secondary of the supply transformer".

In response to this action of the patent office Doble offered amendments to claims 13 and 15, and some others, and cancelled the rest including claim 17. In his "remarks" filed with these proffered amendments Doble differentiated his apparatus from that of Shackelton and from the one described in the British patent on the ground that both of them showed only laboratory apparatus and then said:

"Applicant's shielding means are so connected as not only to protect the measuring instruments from external electrostatic fields and currents, but also to prevent them from being affected by charging currents and losses in the high tension conductors of the apparatus. The inner shielding system provides a closed circuit path for losses, and charging currents of the high tension conductors, preventing them from being indicated on the instruments, while the outer shield eliminates the effect of induced voltages in high electrostatic fields from passing through the instruments * * *. The important utility of the invention resulting from these characteristics is that the apparatus is usable in the field to measure insulating parts which are so mounted as to be in effect grounded at one side or one electrode, and could not possibly be shielded, as is done in making laboratory tests, and as shown by the Shackelton patent. Applicant is the first to provide an apparatus capable of being taken to the place where oil circuit breakers and the like are installed for use, and there testing the insulation accurately and in a brief time, without necessitating dismantling of the switch, with the expense, delay and interruption of service incident thereto."

But, in spite of the amendments offered and the arguments made in support of them the patent office again rejected claims 13 and 15, the examiner reiterating his position that they did not disclose invention over Shackelton, saying:

"While it is true that the Shackelton patent is not adapted to test the insulation of circuit breakers, the casing of which are necessarily grounded, applicant is informed that these claims are not so limited. The broad idea of shielding the source of supply and the high tension lead is clearly shown in the Shackelton patent."

He also gave as an additional reason for his action that they failed to "patentably distinguish" over an apparatus described in the article published in the Electric Journal, and a German patent. In response to this the patentee cancelled claims 13 and 15, and substituted others in their stead one of which, as application claim 37, became without change patent claim 15. Commenting on this action Doble remarked:

"The examiner's comment on claims 13, 14, and 15 recognizes the difference between applicant's accomplishment and what is described by the Electric Journal and the Shackelton patent. The new claims contain the distinctions over the art, for lack of which the examiner rejected the former claims; wherefore applicant expects that the new claims will be recognized as patentable. Likewise they are distinguished from the German patent, which is concerned with the measurement of conductors, cables, condensers, coils and the like, in the laboratory, and not with the measurement of insulation, and particularly that of oil circuit breakers, in the field. In this latter patent, as well as in Shackelton, the entire apparatus is enclosed in a shield and it is not adapted to measure a capacitor of which one terminal is grounded and the other is exposed to high potential."

"Not only do the inner and outer shields have their individual functions, but the two shields together perform a function and [in] combination which is different from the function of either alone; which is to maintain a substantially constant capacity effect between guard and ground."

The patent office acted favorably upon these amendments and the patent in suit was issued.

Analyzing the foregoing it seems to us that the patentee is not estopped by the file wrapper from contending that the claim in suit should be construed as covering only his internal shielding system. In the first place it seems to us evident from the quotations from the applicant's "remarks" that, although he stressed the value of the use of his shields in combination, he did not, as the plaintiff argues, give up or abandon, in the course of his negotiations with the patent office, his claim to his inner shielding system standing alone. On the contrary it definitely appears that he stressed the importance of that system in preventing internal charging currents from affecting the measuring instruments when the apparatus was used to test grounded insulation and, since insulators as installed are almost always if not invariably grounded, he thereby stressed the vital importance of that system to any apparatus for use "in the field". We do not see that he singled out either one of his systems as more important than the other for the primary purpose for which he designed his apparatus. This conclusion renders it unnecessary for us to pass upon the question disputed at the bar of whether or not "remarks" of this sort provide an adequate basis for the application of the doctrine of file wrapper estoppel.

We turn now to the various actions of the patent office and to the communications addressed by it to the applicant.

It appears that the patent office declined to issue a patent on Doble's original application on the ground that the shields disclosed and claimed did not constitute an invention in view of the prior art. Doble then, instead of appealing from that action as he might have done, argued that his apparatus, due to its shields, was useful in testing insulators as installed under service conditions, a use to which the unshielded, or inadequately shielded apparatus of the prior art could not be put; offered amendments to bring out this feature of his alleged invention, and argued that devising his shields called for an exercise of the faculty of invention. But the patent office was not convinced and maintained its original stand. Doble then tried again, by further arguments and amendments of the same sort, met with no better success, and tried still again, further emphasizing the use of his apparatus "in the field" and the importance of his shields, both separately and collectively, to attain that end. This time he succeeded and the patent office acted favorably on his application as amended.

It seems to us from the file wrapper that the patent office must have understood Doble's position, so that, in granting the patent, it must have finally concluded that the prior art disclosed only apparatus capable of use in a laboratory while Doble showed one so shielded as to be capable of use "in the field", and that to devise such an apparatus, in view of the nature of the step taken and of the prior art, involved an act of invention. Thus we conclude that the emergence of the phrase "in the field" during the course of the negotiations between the applicant and the patent office was intended by the latter to operate only to limit the claim to an apparatus capable of use where dielectrics are installed for service, whether the apparatus should be actually used there or in a laboratory. But since both shielding systems are equally important separately as well as important collectively we do not think the phrase was intended by the patent office to be construed to limit the claim only to the inner shielding system when used in conjunction with the outer one.

Further support for this conclusion is found in the nature of the several claims of the patent as it was finally issued. As we have already noted some of its claims cover both shielding systems together and others either one or the other of the systems separately. If it had been the intention of the patent office to permit Doble to have a monopoly only on his shielding systems when used together, it seems reasonable to suppose that when it carefully examined the application and required many amendments to correct inaccurate or indefinite language in the claims, it would have required that each claim distinctly and definitely specify both systems in conjunction. Having allowed the claim as it did, we think the patent office intended to permit Doble to have a patent on his shields separately as well as collectively, and that in the claim in suit it gave him a temporary monopoly on his inner shield as a sub-combination of parts, or a separable part, of his complete apparatus, which, as we have already pointed out, he was at liberty to have under the statute as it has been repeatedly construed by the Supreme Court.

Only one further point with respect to this issue remains to be considered. The plaintiff contends, and the district court concluded, that the defendant had admitted that the phrase "in the field" limited the claim in suit to the inner shield only when used in conjunction with the outer one. We find no support for this conclusion. The defendant agreed at the trial that the phrase "imports into the claim certain limitations", but we find nothing in the record to indicate the scope of the limitations agreed upon. Under these circumstances we cannot say that the defendant by agreement intended to admit anything more than that the phrase limits the claim to an apparatus capable of use in the field as distinguished from one useful only in the laboratory.

This brings us to the question of the validity of the claim in suit as we have construed it.

The master clearly and definitely found the claim valid as he construed it and the defendant contends that in addition he also found it valid as we have construed it. The plaintiff, on the other hand, denies that the master made any such additional finding, but says that we should ourselves, on the record before us, pass upon the question of its validity as covering the internal shielding system standing alone. An elaborate argument to the effect that so construed it is invalid as not embodying a patentable invention is contained in the plaintiff's brief.

A careful examination of the record discloses to our satisfaction that the master did not in fact definitely pass upon the validity of the claim except as he had construed it. There are to be sure indications pointing to the conclusion that he would have found it valid even if he had agreed with us as to how it should be construed, but we do not think that valuable property rights should be definitely determined by us on the basis of mere inferences drawn from statements in the report concerning other issues.

Neither do we propose to pass upon the validity of the claim ourselves. The record as transferred to us is not complete and even if it were our conclusion would not be different. The questions of fact necessarily raised by any consideration of the question of validity are exceedingly complicated and involved and the answer to be given to them depends in large part upon the credibility of expert and other witnesses. Under these circumstances it seems to us that the question of the validity of the claim as now construed is definitely one of fact and as such should be passed upon by the master and the court below in advance of consideration by us.

The judgment of the District Court is reversed and the case is remanded to that court for further proceedings in conformity with this opinion; no costs in this court to either party.