Marconi Wireless Telegraph Co. of America v. United States

GreeN, Judge,

delivered the opinion of the court:

The act of June 25, 1910, 36 Stat. 851, as amended by the act of July 1, 1918, 40 Stat. 705, confers upon the owner of a valid patent the right to bring suit in the -Court of Claims against the United States to recover compensation for any infringement of the patent or patents by the Government. During the period involved in the suit the plaintiff was the owner of certain patents which are the basis of the suit herein, namely: Marconi Reissue, #11913; Lodge, #609154; Marconi, #763772; Fleming, #803684.

Each of these patents relates to' electrical communication between different points without connecting wires, which is-generally known as radio communication. It is alleged by plaintiff that all of these patents have in one or more respects been infringed by the defendant.

On behalf of defendant it is alleged:

(1) That plaintiff has released the defendant;

(2) That plaintiff has assigned its claims herein;

(3) That the patents are invalid;

(4) That even if none of the above defenses are sustained, plaintiff cannot recover for the reason that the evidence fails to show that any of the patents have been infringed.

The questions that relate to plaintiff’s right to maintain the suit will be taken up in their order. All questions relating to the validity and alleged infringement of the patents will be taken np in the order in which the patents were issued, each patent being considered separately.

It is contended on behalf of defendant that under the provisions of the contract of sale made by plaintiff to defendant of wireless and ship stations, as set out in Finding Y, the suit should be dismissed as to all patents in suit except the. Fleming patent #803684, it being conceded that the apparatus sold included the combinations set forth in the other-patents. The finding shows that the contract released the defendant from the payment of any further sum “on account of its (plaintiff’s) patent rights involved in any of the apparatus hereby transferred.” We think it obvious that the release applied only to the property sold and “transferred” by the contract which is not involved in the suit.

November 20, 1919, plaintiff sold to the Radio Corporation its factory, letters patent, including the letters patent here in suit, and certain land transmitting stations, reserving to itself:

“A claim against the United States Government arising from unlicensed use by and for the Government of the apparatus covered by the patents of the Marconi Company.”

The agreement further provided that all amounts received in cash by the Marconi Company after paying expenses should “forthwith be paid to the Radio Corporation” which should issue therefor preferred stock, and it is argued that plaintiff’s claim has been assigned and therefore no suit could be brought thereon. We do not think this follows. It is true that the ultimate effect of the agreement was to enable the Radio Corporation to become the owner of all that might be recovered in this suit but it did not become the owner thereof until plaintiff had prosecuted the suit to judgment and assigned any judgment in its favor to the Radio Corporation. Up to that time the Radio Corporation merely had a contingent chose in action. It is suggested that after conveying the title to the patent and as to infringements reserving only the right to bring action against the defendant, plaintiff had lost the right to bring suit against the defendant but the general rule is that the right to sue for infringement rests with the one who was the owner of the patent at the time the infringement occurred. Some question might possibly arise as to the validity of the assignment' to the Radio Corporation and with respect to the right of that company to sue for other infringements, but this is the most that can be inferred from the cases cited. We think it clear that in its transactions with the Radio Corporation plaintiff has done nothing to invalidate its right to sue for any infringement committed by the defendant prior to the time of the sale and this is all that is claimed.

THE MARCONI REISSUE PATENT #11913

Guglielmo Marconi, an Italian scientist, is sometimes called the father of wireless telegraphy but he. was not the first to discover that electrical communications could be made without the use of connecting wire. The theory upon which this art is based was first disclosed by Professor Maxwell of Cambridge as far back as 1865. Maxwell, however, did not propose any apparatus for developing its use, if indeed he had any idea that it might be made useful, and no practical demonstrations of his theory were made until Heinrich Hertz, between 1879 and 1890, discussed the Max-wellian theory and described apparatus for carrying it into effect in an article published September 21, 1888.. The experiments which Hertz made public substantiated and amplified the theory of Maxwell. Hertz showed that circuits through which a current of high frequency electricity passed developed electrical disturbances in the form of waves which had many of the properties of light waves and that this radiant energy set up what might be called ether waves somewhat like sound waves which induced currents in other circuits not connected by wire therewith. His apparatus consisted of a transmitter so called because it transmitted the electric wave and a receiver so named because it! was put into electric vibration by electric waves from the transmitter although it was not connected with the transmitter. The transmitter included a battery and an induction coil connected with an oscillator or radiator circuit. There was a gap in the circuit where two rods which were connected to the secondary terminals ended in polished metal balls in close proximity, all so arranged that when the current in the primary coil was interrupted there was a discharge of sparks across the spark gap which resulted in the radiation of the wave. To prove the existence of the wave and its transmission Hertz used a turn or loop of wire either in the form of a rectangle or a ring, the two ends terminating in metallic knobs separated by a minute gap. When the ring or loop was held near an active oscillator, electric impulses were set up which revealed themselves by minute sparks at the gap between the balls. Thus Hertz proved the existence of the waves and the fact that they could be transmitted, but he never succeeded in producing waves which were detectable except at a short distance. His experiments created great interest in the. scientific world and other scientists took up his discoveries where he left off and carried them still further.

Nearly six years later, Sir Oliver Lodge, in 1894, wrote a series of articles in the “London Electrician” describing the original and later work of Hertz and explaining it by means of diagrams.

A number of distinguished scientists continued to experiment and develop theories with reference to the transmission of electric waves without wires and methods for accomplishing such a transmission. Among these were Popoff, Tesla, Crookes, Dolbear, and Edison. Some of these men took out patents as did Lodge, although his patent was later than the patent of Marconi, and it becomes necessary to consider the effect of their discoveries and disclosures upon the patent' in suit.

Marconi applied for his first patent in 1896 and it was reissued under date of June 4, 1901. The reissue patent #11918 relates to a “complete system or mechanism capable of artificially producing Hertz oscillations and forming the same into and propagating them as definite signals and capable of receiving and reproducing, telegraphically, such definite signals; * *

The subject-matter of the reissue patent was extensively used until displaced by later improvements. The plan of this system is shown by figs. 10 and 11 .of the patent, reproduced in Finding IX. The same finding also shows the explains tbe diagrams and tbe manner in which the apparatus worked. We do not think it necessary to repeat this explanation in the opinion.

The only claim of the reissue patent which is involved in this suit reads as follows:

“3. The combination, in an apparatus for communicating electrical signals of a spark producer at the transmitting station, an earth connection to one end of the spark producer, an insulated conductor connected to the other end, an imperfect electrical contact at the receiving station, an earth connection to one end of the contact, an insulated conductor connected to the other end, and a circuit through the contact, substantially as and for the purpose described.”

Figure 10 illustrated in Finding IX is a diagram of the apparatus used in the combination.

Finding X, by reference to fig. 10, locates the essential parts of the combination of the transmitting station.

The original patent having been considered 10 be defective because some of the claims were thought to have covered apparatus employed by Prof. Popoff, changes were made in the claims with reference to the tube j in the receiving station (see fig. 11). The original patent No. 586193 stated:

“The tube j may be replaced by other forms of imperfect electrical contacts, but this is not desirable.”

In the reissue patent the meaning of this quotation was changed by omitting the words, “but this is not desirable.” No satisfactory explanation of the reasons for this change in the scope of the description of the invention has been given.

It is urged on behalf of defendant that there was nothing new in any of the apparatus described in the patent and that the combination thereof involved no'inventive skill and presented no new discovery, but was' one which could be perceived by a person skilled in the art and consequently no valid patent could rest thereon. Counsel for plaintiff vigorously dispute these contentions and call attention to the fact that new results followed from its use of the Morse code for signalling with Hertz waves to a greater distance than was previously thought possible. It should be noted in this connection that the aj>-pears to be the first one that went into practical and general commercial use for the purpose of communication through •electric signals transmitted without wires.

When we consider the evidence it shows that there was nothing new in any of the individual elements of the apparatus described in claim 3 and while counsel for plaintiff in argument endeavor to show an infringement based on some one or more of these features we are clear that the combination alone was patentable and think a reading of the claim shows that nothing more was intended. If we are correct in this, it follows that if the defendant in the apparatus which it used omitted some element of the combination of the patent and introduced other 'and different elements instead having a different and useful effect, there would be no infringement of the patent.

Tesla, in 1893, disclosed a practical apparatus for generating and utilizing Hertzian waves. His sending apparatus is described in what is called the Martin book, introduced in evidence, and showed that he used a powerful alternating current generator inductively coupled to a lateral charging circuit containing a spark gap and a condenser, whereby oscillations were generated and this charging circuit was inductively coupled to a distributing circuit.

Popoff, in 1896, prior to the reissue of the Marconi patent, described a receiving apparatus for detecting Hertzian electric rays. The transmission station had an earth connection and he employed a coherer substantially identical with that alleged to be disclosed in the Marconi reissue patent. It was this ldnd of a coherer that furnished the imperfect electrical contact referred to in the claim in suit. Tesla, Popoff, Crookes, and others fully appreciated the fact that their inventions and disclosures would in the future form part of a system for the transmission of wireless messages.

In the case of Marconi Wireless Tel. Co. v. DeForest Wireless Tel. Co. (138 Fed. .657, 671), Judge Townsend, as we think, correctly disposes of any claim for originality based upon the separate features of the Marconi reissue patent. He said:

“If now we examine the patent in suit in the light of this discussion, we shall find that every element of the claims in suit is taken from the prior art. The signaling and receiving instruments are common to the various apparatus from Dolbear, in 1882, to Popoff in 1895; the Morse key is specifically referred to by Dolbear, Edison, and Kitsee, and is suggested by Lodge; the spark-gap, invented by Hertz for producing the Hertz oscillations, is found in Lodge and Popoff; the 1891 Branley coherer or ‘imperfect electrical contact’, comprising a tube with filings and its operation by means of a tapper, are elaborately explained in Lodge’s lecture and utilized by Popoff in his experiments; and ‘choking-coils’ are illustrated, and described by Popoff. The ‘insulated conductors’, described and shown as comprising elevated plates suspended on wires, were shown in Dolbear, Edison, and Kitsee.”

Judge Townsend held in effect that the combination was patentable, but we shall not discuss the question of whether his conclusion in this respect is correct. In the case before him, a different kind of detector was used from any involved in the instant case. It must be conceded, however, that if his reasoning should be followed it would require us to hold that the apparatus used by defendant, which comprised a crystal detector with certain forms of transmitters and receivers, constituted an infringement of the Marconi reissue patent. In this conclusion we do not agree, but hold with Judge Veeder’s decision in Marconi Wireless Tel. Co. v. National Electric Sig. Co. (213 Fed. 815), a case in which the issues so far as the Marconi reissue patent is concerned were exactly the same as those in the case at bar.

A determination of the question of infringement requires a comparison of the apparatus described in the patent with that used by the defendant. The particular features of the combination set out in plaintiff’s patent which are alleged to be infringed are the imperfect electrical contact and the transmitter. The imperfect electrical contact shown in the patent in suit was a device known as a coherer, which was a detector of the electrical waves. The coherer or detector used by Marconi was a tube containing metal filings which was placed in the circuit of the receiving station. When a weak high-frequency current was passed through this circuit, these metal filings tended to cohere, and from this feature the name coherer was derived. The patent contemplated such a degree of imperfection in the contact of the particles in the coherer that the local-battery circuit which included it would be broken on tapping the coherer. Means were provided for tapping the coherer and thus the current would be broken by the tapping and again restored by the high frequency effect.

None of the receiving sets used by the defendant, however, used a coherer similar to that described in plaintiff’s patent. On the contrary, a device altogether different and functioning in a different manner was used. In the apparatus of defendant there was no tube containing metal filings, nor was there any imperfect electrical contact. The device used is commonly known as a crystal detector. In the crystal detector the connection is made by a fine wire which completes the circuit by being brought into contact with a crystal also forming a part of the circuit. This crystal acts as a rectifier in the manner of a one-way valve, allowing much larger portions of the high-frequency waves to flow in one direction than in the other. This constituted a great improvement on Marconi’s invention, as was shown by the fact that in practice it rapidly displaced the coherer. Not only was it a better detector, but it also enabled speech or music to be heard and it is still used in some of the simpler forms of the modern “radio.” It is urged on behalf of plaintiff that these crystal detectors utilize or constitute “imperfect electrical contacts” within the meaning of the claim and as a matter of fact.

There is nothing in the claim or the specifications of the patent to indicate that the words “imperfect electrical contact” are used otherwise than in their ordinary signification. The meaning of the words is so plain that it cannot be aided by definition. The crystal detector does not utilize an imperfect electrical contact in the detection of Hertzian waves but requires a good small contact of such a degree of perfection that it is necessary when the contact becomes imperfect to restore the perfect contact. It is argued on behalf of plaintiff that because the crystal detector has a higher resistance than the terminals between which it is interposed it constitutes an imperfect electrical contact. On this point Judge Yeeder says: ...

“From an electrical point of view the defendant’s' detectors are not imperfect electrical contacts; as the term is used by Marconi, because the electrical condition is permanent from wave to wave of electrically received impulses. Moreover, the defendant’s devices are rectifiers, which means that in any complete cj^cle or complete wave of alternating current the positive half wave passes through the device with less obstruction or greater power than the negative half wave. This rectifying action is a definite property, which is the same from wave to wave. The defendants detectors do not operate in the same manner as the Marconi coherer.”

In a detailed statement, Judge Yeeder pointed out that the coherer is a make-and-break device, whereas the crystal device is not at all of that nature but one which changes the received alternating signalling oscillations into direct current, the energy received and detected being that of the transmitted energy rather than that of a local battery.

It is also contended that the patent is infringed by the transmitters which defendant used. It is argued in effect on behalf of plaintiff that the connection with the earth described in claim 3 of the Marconi patent refers to any connection with a secondary circuit. But plaintiff cannot have both a general claim for any connection and a specific claim for a connection made in a particular and definite way. If the claim is general it would be anticipated by prior disclosures of ground connections in wireless signalling apparatus but Marconi could not have an all inclusive right to every kind of earth connection. The patent in suit describes and claims a specific arrangement of grounded antenna. It requires one end of the spark producer and one end of the coherer to be directly connected with the earth and the other end of the spark producer and the other end of the coherer to be connected with an insulated elevated conductor.

The effect of these direct electrical connections is obviously in the case of the receiver to cause all the high-frequency current induced in the antenna circuit to pass or flow directly through the coherer and in the case of the transmitter to cause all of the energy created by the spark discharge to circulate in the antenna circuit. The transmitting circuits and receiving circuits used by defendant were quite different in their arrangement and function. All included a transformer connection between the antenna circuit and the closed circuit; the latter circuit in the transmitter included the spark producer or gap and in the receiver included the crystal detector. The transformers were of two types: viz, those with (a) separate primary and secondary windings; (b) an auto-transformer type of winding.

In all instances in which a transformer having separate primary and secondary windings was utilized there was no direct electrical connection between the ground and the antenna and the spark producer or gap in the transmitter, or between the ground and the antenna and the detector in the receiver. With this type of transformer no physical metallic connection exists. The coupling is merely inductive in its character. This arrangement is shown in the diagram set out in Finding XXII illustrating the apparatus of the Wireless Specialty Apparatus Company, In the transmitting and receiving circuits utilizing the auto-transíonner type of coupling there was a direct electrical communication between the primary and secondary winding in the coupling of the closed circuit to the open circuit, but as shown in Finding XXI this is merely incidental to the construction of the auto-transformer, the true function of which is to elec-tro-magnetically couple the open and closed circuits together, and such electrical connection is not for the purpose of connecting the receiving antenna and the ground directly to the detector so as to pass all of the high frequency currents generated in the antenna through the detector as was done by the apparatus described in the Marconi patent. In fact this connection was merely incidental and not for the purpose of connecting the antenna and the earth directly to the spark producer or gap. In other words it functioned differently than did the apparatus of the reissue patent.

We do not overlook the fact that in the cases cited by Judge Yeeder no physical contact was shown to exist between the apparatus and the ground, whereas in the case before us in part of the apparatus there was such a physical contact, although it can be said that it was made in a differ-exit manner and operated differently. But this fact does not affect the primary question here.

As stated above, plaintiff’s patent is on the combination. In Loom Company v. Higgins, 105 U. S. 580, it is said: “If a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention.” This quotation may sustain plaintiff’s claim that the combination was patentable but it also shows that the patent thereon was not infringed.

At this point, keeping in mind that it has already been shown that all the elements of the reissue patent are old and that none of its individual features considered separately and alone were patentable, it becomes quite clear that a new combination made by another party presenting a new arrangement, producing new and beneficial results does not infringe a former combination by using some of the elements thereof.

In conclusion we hold that the reissue patent was not infringed :

(1) Because the apparatus used by defendant included in its combination a new element, namely, the crystal detector;

(2) Because defendant’s apparatus was differently arranged in that the spark gap and detector were not in the antenna circuit where they were located under plaintiff’s patent;

(3) Because the changes made in the combination by defendant had a new and beneficial result.

THE LODGE PATENT, #009154

The findings show that the British patent to Oliver Joseph Lodge, #11575, was filed May 10, 1897, over seven months before the United States application for the same invention, the patent in suit No. 609154, was filed. The British patent, however, was not sealed until October 25, 1898, and the United States patent to Lodge was issued August 16, 1898. The actual certificate of the British patent therefore was not issued until after the American patent had been granted. By section 3 of the act of 1897, section 4887 of the Revised Statutes was amended to read as follows:

“No person otherwise entitled thereto shall be debarred from receiving a patent for his invention or discovery, nor shall any patent be declared invalid, by reason of its having been first patented or caused to be patented by the inventor or his legal representatives or assigns in a foreign country, unless the application for said foreign patent was filed more than seven months prior to the filing of the application in this country, in which case no patent shall be granted in this country.”

Under the provisions of this statute the invalidity of the patent will not be shown merely by the fact that the application for the patent in the United States was not made until more than seven months after the application for the foreign patent was filed. It must also appear that the invention was first patented in a foreign country. In the case of the Lodge patent, if we take the dates when the patents were actually issued, we find that for some reason the date of the letters patent issued in England is subsequent to that of the one issued in the United States. We do not think, however, that this fact, when we consider the English statute and the obvious intention of the provisions of the law in the United States, is determinative of the question now before us. The complete specification for the British patent was accepted August 10, 1898, and the British patent was sealed October 25, 1898. Under the British law acceptance gavfr “like privileges and rights as if a patent for .the invention had been sealed.” In Davies v. Townsend, 13 R. P. C. 265, the defendant was accused of having sold goods marked with the word “patented” when no patent had been granted. The suit was dismissed because his application had been accepted and he had the same rights as if a patent had been sealed on that day.

In Smith v. Goodyear Dental Vulcanite Co., 93 U. S. 486, 498, the Supreme Oourt held with respect to an English patent: “Though the provisional specification was filed March 14, 1855, the completed specification was not until the 11th of September following. It was, therefore, on the last-mentioned date that the- invention was patented.”

The last clause of the statement above quoted obviously has no general application, but refers merely to the case then before the Court.

In the Smith v. Goodyear case, supra, the date when the British patent was granted to Goodyear was not important, except upon the question of the priority of the invention. It did not affect the validity of the patent which Goodyear took out in the United States. The question which the court was deciding was a very different one from the one involved in the case at bar in which we are required to determine what is meant by the words “first patented * * * in a foreign country” as used in the statute. In Great Britain, while the rights of a patentee date back to the time when the completed specifications are filed, his rights do not become fixed and settled until the patent is “sealed.” The specifications may be sufficient and approved but if the patent is not “sealed” he receives no certificate that a patent has been granted and no rights accrue to him as a patentee. We think, therefore, that the word “patented” as used in the statute refers to the time when plaintiff’s right to a patent is fixed and determined and this in Great Britain would be the date when it is “sealed.” In the case at bar the Lodge patent was not sealed in Great .Britain until after it had been granted in the United States. We conclude therefore that Lodge’s patent taken out in the United States cannot be held invalid by reason of the subsequent issuance of the British patent on the same invention.

It is urged that there was such laches in enforcing the claim for infringement under the Lodge patent as to prevent any recovery for that reason even if the patent be held valid.

Title to the Lodge patent was acquired by plaintiff on March 19, 1912, a little less than three and a half years before it expired. Mr. Marconi admitted in his testimony that his company had previously been openly infringing the patent since the time of its grant. The evidence also shows that the loading coil which is one of the basic features of the Lodge patent had been used openly by the Navy since 1903, by the Army since 1908, and by the Stone Wireless Telegraphy Syndicate in 1902. In fact, the Stone Company openly used the induction coil in its antenna for several years, but was never notified of infringement or sued under the Lodge patent. Suit could have been brought thereon as early as 1903 against the manufacturers of the apparatus manufactured by the Stone Wireless Telegraph System and other contractors who made it for the Government. But this was all prior to the time that the plaintiff acquired the Lodge patent. About that time or shortly after it acquired the patent, as stated in Finding XII, the United States was orally notified with reference to the four patents in suit including the Lodge patent in the winter or spring of 1912 at a White House conference at which the President and four members of the Cabinet were present and written notice of the two Marconi patents, and the Lodge patent, and their infringement was given by registered mail to both the Navy and War Departments on March 7, 1913, and March 8, 1913, respectively.

The facts recited above show clearly that prior to the time the plaintiff acquired the Lodge patent there had been such laches in asserting any rights thereunder that any person interested might well believe that the patent had been abandoned. The contention of the defendant is that the laches on the part of the owners of the Lodge patent prior to the time when plaintiff acquired the ownership thereof' not only prevents a recovery for any use of the patented device prior to that time but in effect invalidates the patent and prevents any recovery for infringement thereof after notice was served upon defendant that it would be held liable for any infringement thereof. Several decisions are cited in support of this contention. All of them hold that the party to whom a patent has been transferred must bear the burden of the acts of his predecessors in ownership and where they have been guilty of laches the effect thereof carries on to the new owner. Consequently, where the apparatus alleged to infringe the Lodge patent had been purchased prior to the time when the notice was given by plaintiff to defendant, as shown above, there can be no recovery on account of such use. This rule we think will apply whether use was made of the patented device before or after the purchase of the apparatus used.

But as to apparatus purchased or any use of the patented device after notice of the plaintiff of intent to enforce its rights under the patent, we think the rule is different. None of the cases cited go so far as to hold that, after notice has been given that the owner of the patent will insist upon his rights, previous laches prevents any recovery for subsequent infringement arising from the use of apparatus made or purchased after the notice had been given. On the contrary, it was held in the case of Simpson v. Newport News Shipbuilding & Dry Dock Co., 18 Fed. (2d) 318, 321, where the notice was given prior to manufacture and use by the defendant of the patented device, that previous laches on the part of the plaintiff with respect to other parties would not prevent a recovery. A recovery was denied only because the final conclusion was that the patent was invalid. In the case of Marconi v. National Co., supra, which involved the same plaintiff and the same patent (Lodge patent), the plaintiff was not permitted to recover on account of use of the patented device prior to the suit but the patent was sustained and an injunction granted against further infringement. In this case the opinion does not show that the defendant received any notice with reference to the infringement prior to the time when the suit was begun. Richardson v. Osborne & Co., 82 Fed. 95, is an extreme case. The suit was not begun until two months before the patent had expired. The opinion shows that the defendant had been using the patented device for nearly fourteen years without being disturbed or molested and was carrying on a very large business with every reason to believe that the plaintiff would not make any claim under his patent. If the plaintiff had been permitted to maintain his suit, it is manifest that the defendant would have sustained a large loss upon investments made in good faith and the defense of laches was sustained. The opinion in this case cites a large number of cases bearing on the effect of laches generally. Only a few of these cases involve patents. The principles which relate to the effect of laches apply to all suits where laches in commencing a suit is shown and are not peculiar to patent cases. The rule of law which is applied is based upon the principle that it is manifestly inequitable to permit a plaintiff to recover upon an act of the defendant for which the plaintiff himself is largely to cases, while recovery is often denied on account of laches, the patent is not held invalid.

In the case at bar, we see nothing inequitable in holding the defendant liable where the apparatus alleged to infringe was made or purchased after the defendant was served with notice of plaintiff’s claim, but where the apparatus was purchased or made prior to" the service of such notice our conclusion is that there can be no liability.

Mere use of infringing apparatus after the notice was given, we think, is not sufficient to make the defendant liable. Where defendant purchased or made the apparatus before notice and had thus invested its money or gone to expense in good faith, having reason to believe that no claim would be made by plaintiff, we think the plaintiff would be estopped from complaining of the use of the apparatus after the notice.

VALIDITY OF THE LODGE PATENT

Before the Lodge patent was taken out it was known that with alternating currents of given frequency a circuit will carry its strongest current under a given pressure or voltage when there is a condition of resonance, or, in other words, when the natural period of the circuit corresponds with the frequency of the-current. In the alternating-current circuit its natural period depends on the extent of its capacity and inductance, or self-inductance as it is sometimes called. It was also known that a helical coil of Avire had principally inductance and an electric condenser had essentially capacity, and that variation in the amount of either the inductance or capacity changed the period of the circuit. The most important feature of the Lodge patent was the use of a self-inductance coil between a pair of capacity areas in the oscillating circuit of either or both a sending or receiving set for Hertzian wave telegraphy. As shown in fig. 9 of the patent drawings set out on page 691, Lodge made use of an adjustable coil whereby the extent of its self-inductance could be varied. The purpose of such a coil was to control the oscillations excited in the system and constitute the system a resonator or absorber of definite frequency or pitch, which acted, cumulatively with a distant radiator of corresponding-period also capable of acting cumulatively. By this method, as he stated in claim 5, “signalling may be effected between any two or more correspondingly attuned stations without disturbing other differently attuned stations.” (See claims 1, 2, and 5 of the Lodge patent as set out in Finding XXVIII.)

It is quite evident from the claims made by Lodge, as set out in Finding XXVIII and his diagrams of apparatus which accompanied the patent and which are set out on pages 690 and 691, that Lodge contemplated varying either the inductance or capacity or both in such a manner as to attain syntony or resonance and prolong the electrical oscillations excited in the system. An important feature of the apparatus for this purpose was the adjustable coil. The defendant, to show that Lodge offered no new invention when he proposed to tune the circuits in syntony, cites Marconi v. National Co., supra, in which Judge Veeder said: “Tuning the two circuits of a transformer was common knowledge, and was practised by Ducretet, Braun, Lodge, Tesla, and Pupin, and was described and used by Fessenden prior to 1900”; and also the decision of the British court in a case where the owners of the Lodge patent brought suit for prolongation of the Lodge British patent reported in Reports of British Patent Cases, vol. 28, page 365. In this case the court considered the third, fourth, and fifth claims of the Lodge patent with reference to tuning the transmitters and receivers and said:

“With regard to tuning, in the sense of varying the natural frequency of a circuit, it had long been well known that this could be done by varying either the induction or the capacity, or both the capacity and the induction of the circuit. For tuning his syntonic jars, Lodge had provided means for varying the induction. Under these circumstances, again, I cannot think that tuning by means of induction coils switched in or out as desired, or by means of adjustable induction coils can be considered in itself a great invention.”

And the court refused to grant extension of these claims. Apparently the British judge considered that everyone skilled in the art would know that the effect of an induction coil could be varied by varying the number of separate turns of wire used and for this reason it ivas no invention to use an adjustable coil in the circuit for the purpose of tuning it. All this may be conceded and yet we think it does not render the Lodge patent invalid and ineffective as to infringement. Lodge’s invention, as we view it, consisted of a new combination in which features which may be said to be old when considered apart and alone, were so associated as to produce new and valuable effects.

We do not think it is necessary to consider in detail the patents granted to Pupin, Hutin, Tesla, and Stone, referred to in the findings, although they were granted prior to the time the patent to Lodge was issued. These patents show that the laws of closed resonant circuits in general were known prior to the Lodge invention, but there is nothing in these patents which shows knowledge of the desirability of tuning an open oscillating circuit for Hertzian-wave signalling. Lodge took features that were old and combined them in such a way as to produce new and useful results which had never been accomplished before. While the laws regarding the relationship of capacity and inductance to achieve the tuning of an alternating current circuit were known prior to Lodge’s invention, no satisfactory evidence has been j)roduced to show that the elevated wire of a Hertzian-wave transmitter or receiver and the ground connection were known to constitute an electric condenser in the open oscillating circuit prior to the Lodge invention. Lodge appears to have been the first to recognize that the elevated wire and ground connection constituted a condenser and was the first to realize the desirability of applying the known laws of resonance for closed circuits to a Marconi open oscillating circuit. The Marconi patent #586193 (the original patent which was reissued as the first patent in suit) contemplated the use of plates of such length that the receiver would be “electrically tuned with the electric oscillations transmitted”, but there is no evidence that this method would work satisfactorily, and Lodge asserted at the time his patent was reconsidered that the method recited in the Marconi patent would not work at all. In any event, the evidence shows that if this method ever was put in use it was entirely superseded by the method which Lodge used. Lodge’s invention increased the range for signalling with Hertz waves and it went into wide and almost universal use, superseding other methods.

We think on the whole that while none of the features of the Lodge patent may be said to be entirely new, Lodge had taken these features and in a new combination had so used the most useful function of each as to obtain better results than had ever before been reached. We conclude, therefore, that his patent was not void for want of novelty,

Tesla’s patent #645576, the application for which was filed before that of Lodge, although the patent itself was issued later, is mentioned as anticipating the features of the Lodge patent. But the system of Lodge made use of a different combination from anything disclosed by Tesla and was operated differently.

It is said that the decision in Marconi v. National Co., supra, which sustained the Lodge patent, is based on erroneous assumptions, but we do not need to agree with all that Judge Veeder said in order to concur with his conclusion, which we think was well founded.

In the case of Marconi v. Kilbourne & Clark Mfg. Co., 239 Fed. 328, the court treated the validity of the Lodge patent as being settled and held it was infringed. This case is chiefly useful as showing the improvements and special features of the combination which Lodge used and the manner in which it operated.

The Lodge patent was also held valid in the case of Marconi v. DeForest Radio Tel. & Tel. Co., 225 Fed. 65, affirmed 225 Fed. 373. If we are correct in holding the Lodge patent valid, then the defendant has infringed this patent by the use of the apparatus of the Wireless Specialty Company and other apparatus as more particularly set out in Finding XXXIY, and the defendant would be liable for such infringement unless it purchased or acquired the apparatus prior to the time it received notice from the plaintiff that it proposed to enforce its rights under the Lodge patent.

THE MAECONI TUNING- PATENT #763772

This patent is sometimes refered to as the “four circuit tuning patent” but is commonly called the Marconi tuning patent. The patent relates to means for providing a stronger transmitted signal in wireless telegraphy and means for rendering the receiver more selective than had resulted in the previous practice. Both the receiver and the transmitter had an open antenna circuit and a closed oscillation circuit which were tuned to the same frequency or time periods, and all four also synchronized. This resulted in having a somewhat stronger signal radiated by the transmitter and a more sensitive reception by the receiver. In this respect it was successful and went into wide use in the United States and elsewhere. It is claimed that by means of this invention there was a substantial increase in the distance of reliable radio transmission and reception and also an increase in the selectivity of radio apparatus which in turn made possible an increase in the number of transmitting stations which could be used at the same time without interfering with each other. Apparatus embodying the invention of #763772 was employed in the first transatlantic radio signalling.

A number of claims under this patent are involved in the suit. These claims are set out in Finding XL which also contains an explanation of some of the more important claims in connection with the diagram contained in Finding XXXVIII. Some of these claims are directed to the transmitter, some to the receiver, and some to a communication system involving both the transmitter and the receiver. But so far as the validity of the patent is concerned it will not be necessary to consider these claims separately.

Two defenses are made against this patent, the first being that the patent is invalid and the second that there has been no infringement in any event. The alleged invalidity of the patent is based upon prior patents and disclosures.

Marconi had a great deal of difficulty in obtaining this patent. The original application for the patent was filed November 10, 1900. Before it was reached for action and on July 1, 1901, the entire original specification and all of the claims were canceled and a new specification and set of claims were substituted. The examiners who first considered the case rejected the application. During more than three years afterwards new applications and petitions for revival were filed and rejected by reason of the prior art set forth in the Braum British patent: Lodge #609154; Marconi #627650; and Tesla #645576. During the prosecution in the Patent Office of this patent, the examiner rejected all claims made therein, except 15, basing his rejection on the Tesla patent #649621 and particularly said in his letter that claims 4, 13, 14, and 16, which specify a variable inductance and means for adjusting the inductance, were not patentable over Tesla because Lodge had shown that this was old, and the examiner also stated that claims 3 and 5, which specify an adjustable condenser, were not patentable over Tesla.

After this holding of the examiner, nothing was done by Marconi until October 6, 1903, when a petition was filed asking that the application be revived. The examiner recommended that the petition be denied, citing as additional prior art a paper by Pupin and the Stone patent #714756 which had been issued on December 2,1902, and on Décember 3, 1903, the Commissioner denied the petition to revive. • At the time of filing the petition for revival on October 6, 1903, the applicant presented an amendment containing the present claims 1, 2, 3, 6, 8, 9, 10, 13, 14, 15, 16, 17, 18, and 20. Elaborate arguments and affidavits were submitted and the Commissioner submitted a long report in effect denying the application. Another petition for revival was presented on February 19, 1904, and granted by the Commissioner. The case was acted upon by a new examiner, and all the claims formerly rejected by the other examiner except four were immediately allowed in accordance with a brief letter in which it was stated, among other things, that all claims which included the element of variable inductance were considered allowable over the Stone patent, and that claims 4, 5, 6, and 7 would be allowed if amended by inserting the words “a variable inductance and” after the word “including” in line 3 of each claim.

Patent #763772 lias several times been before the courts.

In the case of Marconi v. National Co., supra, this patent with other patents was in issue and was held valid and infringed as to claims 1-3, 6, 8, 10-13,16-20. The device used by the defendant in that case had loose coupling, signif3dng a plain open gap. It is not clear whether the court considered the rotary and quenched gap apparatus involved in the instant case to be also^ infringements. Preliminary injunctions were granted in the case of Atlantic Communication Co. (not reported) and the DeForest Radio Tel. db Tel. Co., 225 Fed. 65. In the case of Marconi Wireless Telegraph Co. v. Kilbourne & Clark Mfg. Co., 239 Fed. 328, the validity of the patent was not determined, but it was held that quenched gap transmitters did not infringe. In another case with the same plaintiff against the United Wireless Telegraph Co. in the Southern District of New York, claims 1 to 20 were held valid and infringed but the case has not been reported.

The British patent #7777, of 1900, corresponding to the patent in suit, was sustained, as was also the corresponding French patent.

It will not be necessary to consider the claims of the patent in detail as the matter in controversy between the parties is confined to a narrow issue.

The fundamental principle of the patent is embodied in an apparatus utilizing a transmitting station of two circuits, an open antenna circuit and a closed oscillation circuit, which are coupled together and both of which are tuned to the same or substantially the same frequency or wave length; and the use of a receiving station of two similar circuits Avhich are likewise tuned to the same or substantially the same frequency or wave length with each other and the transmitter.

All of the wireless sets of defendant made use of apparatus constructed in accordance with the principle which formed the basis of the patent. We will, therefore, next consider the question of the validity of the patent. In so doing it is important to observe that when used with reference to electrical circuits the words “tune” and “synchronize” have the same meaning and that when circuits are “tuned” or “synchronized” with each other they are said to be in “resonance.”

Finding XLIII shows that the examiner who at first refused to allow the patent based his rejection on the patents of Tesla and of Lodge. Tesla obtained two patents prior to the Marconi tuning patent — #645576 and #649621. In these patents Tesla made use of an open antenna circuit and a closed oscillation circuit in both the transmitter and receiver, all of which were tuned together or synchronized. Tesla did not refer to the matter of tuning by adjustable inductance but one of the examiners who rejected Marconi’s application stated that it was fair to assume Tesla intended to use either that method, or a condenser, or both, to accomplish this purpose. This was necessarily so for these methods were known at the time and it could be accomplished in no other manner.

Judge Veeder in the National Electric Sig. Co. case, supra, says that Tesla’s patent #454622 of 1891 contains no suggestion of tuning and seems to lay great stress upon that fact. This patent is not in evidence in the instant case and is not before us. Tesla’s patent #645576 of date March 20, 1900, the application for which was filed before the Lodge pi a tent, was upon a four circuit system and referred to the synchronization of the circuits no less than six times therein. In one place he says “the primary and secondary circuits in the transmitting apparatus being carefully synchronized, * * and that “the results were particularly satisfactory when the primary coil or system A' with its secondary O', was carefully adjusted, so as to vibrate in synchronism with the transmitting coil or system AG A

Finding LXII includes a diagram of Tesla’s system as set out in his piatent. The specifications show that Tesla had the two circuits in the transmitting apparatus synchronized or tuned together and the two coils in the receiving appsaratus tuned together and made to “vibrate in synchro-nism with the transmitting coil.” Further on Tesla specifies “the feature of a transmitting and receiving coil or conductor, both connected to the ground and to an elevated terminal and adjusted so as to vibrate in synchronism.”

In three other places Tesla refers to the circuits of his system being “synchronized.” Just what method he proposed to use for synchronizing the circuits is not stated, but Lodge had shown how it could be done by the system of adjustable inductance and as we have specially found (see Finding LXII), anyone skilled in the art was capable of applying the Lodge method of tuning to the Tesla system. The last Patent Office examiner in his letter of March 31, 1904, giving reasons why the patent should be granted, after referring to the method used by Stone to synchronize the circuits, said:

“Applicant discloses another means whereby the same result may be effected, viz. by including a ‘variable inductance’ in the elevated conductor. All claims which include this element are considered allowable over the Stone patent,

* sjs % # ❖

“Claims 4, 5, 6, and 7 will be allowed if amended by inserting the words a variable inductance and after the word ‘including’ in line 3 of each claim.”

From this statement it is evident that the only basis the examiner could find for allowing the claims was the feature of variable inductance in a specific location as a means of bringing all the circuits into resonance. The examiner does not seem to have given very careful consideration to the matter at hand for he overlooked the fact that claims 10 and 12 as well as claims 4, 5, 6, and 7 contain no reference to variable inductance. He should have known that Lodge had shown and patented the use of a variable self-inductance coil to obtain a condition of resonance or syntony between circuits, and it cannot be contended that the feature of variable inductance was new in itself. Marconi used it for the purpose of doing what Tesla had done without specifying the method. After Tesla had filed his application in #645,576, Lodge showed how it could be applied and advantageously used in the application for his (Lodge’s) patent. Tesla filed his application for #645,576 in September 1897. Lodge filed his application February 1, 1898. Tesla therefore did not have the knowledge of the manner in which Lodge made use of the principle of tuning by adjustable inductance, and as Lodge’s patent was issued in 1898 and Tesla’s #645,576 not granted until 1900, it may be tbat neither knew of the work of the other. But after these patents were issued it was only necessary to put them together to have the Marconi system. We think the examiners who first rejected the Marconi application were right and that the last examiner who granted the patent was wrong.

Judge Veeder in the Nat. Electric Sig. Co. case, supra, calls attention to a new theory propounded by Tesla in his divisional patent #649,621 and seems to consider that because Tesla’s theory was different from that of Marconi, Tesla’s system does not anticipate that of Marconi. But we are not here concerned with theories. What we have to deal with are Tesla’s specifications, the apparatus which he described, the manner in which he proposed to use it, and the effects which it would accomplish. He may be and probably was entirely wrong as to some of his theories but there is no doubt about these essential matters. Claim 10 of the Marconi patent recites:

“A system of wireless telegraphy, in which the transmitting-station and the receiving-station each contains an oscillation transformer, one circuit of which is an open circuit and the other a closed circuit, the two circuits at each station being in electrical resonance with each other and in electrical resonance with the circuits at the other station, substantially as described.”

The specifications of this claim are repeated in a different form and with different words in claim 3 of Tesla’s #649621. If anything is lacking it is supplied elsewhere in other claims made in Tesla’s patent. In this connection it should be said that some importance has been attached to the fact that Marconi used a persistent oscillator, but Tesla also describes a persistent oscillator. Claim 10 is sometimes referred to as the “broad claim” of the Marconi patent. It is quite plain that it was anticipated by Tesla whatever his theories may have been, and so far as the other claims of the Marconi tuning patent are concerned we think we have already shown good reasons why they should be rejected.

It is sometimes said that Tesla’s purpose was only to transmit electrical energy, but the electrical waves transmitted by any wireless system are merely one form of electrical energy. Moreover, the specification of Tesla’s patent recited that the method of energy transmission would bo useful when it was desired to transmit intelligible messages great distances. In view of this statement it would be within the knowledge of those skilled in the art to interrupt the continuous generation of high-frequency energy in the transmitting system by a telegraph key and substitute for the current receiving instrumentalities disclosed in connection with the receiving system a radio signal detector device.

If we are correct in what has been stated above, the Marconi tuning patent is invalid, but it should be noted that other defenses are strongly urged by the defendant based upon the Stone patent and certain disclosures made in letters written by Stone. As stated above, Stone’s application was filed prior to that of Marconi and his patent was also granted before that of Marconi, although it was not allowed until after an amendment to the specifications had been made and this amendment was filed after Marconi’s application.

Stone, in the original application for his patent, discloses a four-circuit system in which the transmitting and receiving open circuits are each associated with a closed circuit and the two closed circuits are tuned in resonance with each other. As we have before stated, all of the different features of the Marconi patent considered separately and individually were old and Marconi could obtain a valid patent only by proposing a new combination of these features which produced new results and this combination also must not be one which would naturally suggest itself to one skilled in the art. Here again in connection with Stone’s patent we conclude that Marconi’s patent does not come within this rule. Tesla had shown the advantage of all four circuits being syntonized; Lodge had taken the two-circuit system and tuned or synchronized the open circuits in the same manner afterwards used by Marconi; Stone had disclosed in his original application a four-circuit system in which the closed circuits were tuned together. A consideration of these three systems would naturally suggest to one skilled in the art the tuning oí all four circuits together by the use of the adjustable self-inductance method in the manner proposed by Lodge, and Stone put this suggestion into practice when he filed the amendment to his specifications. Marconi used the suggestion earlier in the application for his patent, but under the circumstances as we think neither Stone nor Marconi was entitled to credit for it.

A quite unusual situation is presented by the conflicting patents of Marconi and Stone. As before stated, Marconi applied for the patent which subsequently became #763772 November 10, 1900. Stone filed the application for his patent on February 8, 1900, and on April 8, 1902, Stone filed an amendment to his specifications which proposed a four-circuit system with all the circuits synchronized. His patent was allowed on December 2, 1902, while the Marconi patent #763772 was not granted until June 28, 1904. In 1899, Stone wrote two letters to a friend of his by the name of Baker in which he disclosed a four-circuit system with all the circuits synchronized, but, as we have seen, he did not actually put such a system in practice until he filed the amendment to his specifications on April 8, 1902.

The Stone patents and disclosures were not before the British court, or Judge Veedor in the National Electric Sig. Co. case, supra, but were before the court in the Kilbourne & Clark Co. case, supra, decided by Judge Neterer. In the argument for plaintiff it is said: “Judge Neterer overruled this defense.” The opinion of Judge Neterer shows that he considered the Stone letters and that they disclosed a four-tuned circuit system. It was unnecessary for him to pass on the validity of the Marconi tuning patent and he did not do so for the reason that he found there was no infringement. If any inference is to be drawn from, his opinion it is that it was proper to take into consideration the Stone letters.

It is contended with much force on behalf of the defendant that the letters written by Stone which are referred to in Finding LV anticipated the Marconi patent even if it otherwise be held to be valid, but it is not necessary that we should decide' whether a private letter such as Stone wrote to his friend would constitute a disclosure which would affect the Marconi patent, for without considering these letters we think patent #763772 was invalid for the reasons stated above, except as to claim 16, which relates to a matter not covered by the other claims.

Some argument is made in favor of the Marconi patent, based upon the fact that Marconi made use of a persistent oscillator, but this was used in the earlier Marconi patent # 586193. The use of the persistent oscillator is, we think, important only when the question of infringement by the use of the quenched-gap apparatus is considered.

The alleged infringing apparatus may be classified in three groups: (a) The plain open-gap apparatus; (5) that having a rotary gap; (c) the larger group using a quenched gap.

If we are correct in holding that the Marconi patent #763772 is invalid, there would, of course, be no infringement in any event, but if it should be held that the patent was valid then it becomes necessary to consider whether any or all of the apparatus used by defendant infringed thereon. We think it clear that the plain open-gap apparatus would be an infringement if the patent was valid, and we are likewise of the opinion that this would be true as to the apparatus using the rotary gap which, as we think, was merely a variation of the plain open gap and apparently used for the purpose of cooling the terminals used in making the gap across which the spark passed. In this connection it should be observed that Marconi did not confine his patent to any particular kind of spark producer and evidently intended that any known kind might be used. The rotary gap might in some respects have been superior to the plain open gap, but it did not function in any different manner.

The form of the device which constituted the quenched-gap spark producer and its action as such are set forth in Finding XLYIII. The evidence is conflicting as to whether in its effect upon other parts of the apparatus it functioned differently from the plain open gap. In any event, it would seem that it was only another device for a spark producer and that Marconi in an earlier patent had shown a somewhat similar contrivance. It also appears that when it was used it was necessary to follow the same method of tuning as before. On the other hand, this device apparently was more efficient than the plain open gap for in a comparatively short time it largely displaced the latter system.

In view of the fact that we have found the Marconi patent #763772 invalid except as to claim 16, we do not find it necessary to determine whether apparatus making use of the quenched gap constituted an infringement.

Claim 16 is set out in Finding XL. We shall not consider it at length as we have no doubts with reference to its validity. It relates to a different subject matter from any of the other claims and is directed to a combination of elements which is new and useful.

The claim refers to a receiving station in which the principal features are the combination of an oscillation transformer with an open circuit connected with one coil of the transformer and including an oscillation-receiving conductor at one end and capacity at the other end, with an adjustable condenser in a shunt connected with the open circuit and around the transformer coil, a wave-responsive device electrically connected with the other coil of the oscillation transformer, and means for adjusting the two transformer circuits in electrical resonance with each other. These features are shown in the diagram of the receiving station found in fig. 2 of Finding XXXVTII, where the means for adjusting the two transformer circuits in electrical resonance with each other are indicated. It may be said, however, that the purpose of the adjustable condenser in shunt with the primary winding of the transformer in the open circuit of the receiver is not described in this patent; but the, specifications of the patent include tables showing the preferred adjustments in connection with the induction coil as numbered therein, and we think these tables would indicate to one skilled in the art that the use of the condenser connected in shunt with the primary winding of the transformer of the receiver was for the purpose of enabling the electrical periodicity or tuning of the open circuit of the receiver to be altered and adjusted. So far as disclosed by the evidence, prior to the Marconi patent, no such apparatus had ever been used or devised. Being new, if it was useful, it was patentable. Its usefulness, we think, is shown by the fact that the receiving apparatus of the Kilbourne & Clark Company and that made by the Telefunken Company used by the Government embodied this device. Claim 16, therefore, is valid and was infringed.

THE FLEMING PATENT. #803684

(Granted November 7, 1904)

This patent is entitled an “Instrument For Converting Alternating Electric Currents Into Continuous Currents”, and the nature of the invention patented is well described in the patent itself (page 1, line 11) as follows:

“This invention relates to certain new and useful devices for converting alternating electric currents, and especially high-frequency alternating electric currents or electric oscillations, into continuous electric currents for the purpose of making them detectable by and measurable with ordinary direct-current instruments, such as a ‘mirror-galvanometer’ of the usual type or any ordinary direct-current ammeter.”

Fleming’s patent was granted during a period when the development of electric communications without connecting wires was extremely active. Out of the work of electrical scientists during this period grew some of the most marvelous discoveries of the properties of electricity, and great progress was made in the art of reproducing the human voice and musical tones by means of wireless apparatus in the form of what we now call radio communications. Fleming’s patent in some respects was new but the defendant contends that it was new only in the use of a discovery that had been made long before Fleming applied -for his patent. The apparatus used by Fleming is not very complicated. A detailed description of it with drawings is found in finding LXIY, but we think it will be easier to understand its fundamental principles if we go back to a prior patent.

Edison was granted patent #307031 on October 21, 1884. In Edison’s apparatus he made use of an ordinary incandescent lamp which, as everyone knows, consists of a glass vacuum tube containing a filament through which an electric current is passed bringing it to a very high degree of heat. Fig. 4 of finding LXX shows the tube marked A and ■within the loop made by the filament a plate marked T) which is connected with a wire which passes down and out of the tube and forms another circuit (not shown in the figure) by being connected with the lamp circuit. After Edison’s patent came out scientists experimenting with his tube observed that there was a unidirectional current flow in the vacuous space between the heated electrode of the lamp filament and the plate electrode, and that if an alternating current were applied to a circuit, including the plate and filament, rectification would take place and the resultant current fiow would be unidirectional. This change from an alternating current to a direct current flowing only in one way has been referred to as rectifying the current and the apparatus producing this result has been termed a rectifier. Why it passed only in one direction has never been very satisfactorily explained but the fact that it did so pass attracted the attention of scientists the world over. Fleming himself disclosed this fact in a lecture before the Koval Society in 1890. Clark Howell in a paper read before a meeting of the American Institute of Electrical Engineers in 1897 described and illustrated the “Edison effect” and mentioned the fact that if an alternating current is used to render the filament incandescent the additional current thus produced is unidirectional, and at the same meeting Dr. Kennelly said that so far as he knew this paper pointed out for the first time that an alternating current passing through an incandescent lamp giving the “Edison effect” is capable of producing in a branch circuit through a third wire in a lamp continuous or at least unidirectional currents. These proceedings were published to the world.

Edison did not mention this feature in his patent and probably was not aware of it at the time he took it out, but if an alternating current was impressed upon the hot and cold electrodes of his apparatus it would function so as to rectify the current and change it from one that was alternating to one that was continuous in the additional closed circuit.

The descriptive matter taken from the Fleming patent as originally filed and set out above shows that one of its principal features is the device for converting alternating currents into direct currents whereby his apparatus acted as a rectifier. The apparatus of the Fleming patent used for rectifying the current is shown by fig. 1 in Finding LXIV, and if we examine it with respect to the vacuum tube shown therein with its heated electrode (the filament of the lamp) and its cold electrode connected with an outer closed circuit, we find that it is substantially the same as the corresponding part in the apparatus, of Edison. Edison used for his cold electrode a plate placed within the filament loop while Fleming used for his cold electrode a cylinder outside of the filament loop, but this is an immaterial variation.

It is contended on behalf of defendant that the Fleming patent involves nothing more at most than a new use which is not patentable. There seems to be much force in this contention. Fleming made no substantial change in the structure of the Edison vacuum tube with its hot and cold electrodes. His first claim as originally made when taken in connection with his specifications is merely a restatement in different words of claim 5 of the Edison patent. It is true Fleming does not refer in claim 1 t® an incandescent electric lamp as Edison does, but his explanations and specifications show he intended that an incandescent lamp should be used, also that there should be two circuits and that the terminals and connections thereof should be substantially the same as those specified in the Edison patent. In the use of the respective patents, however, there was an important difference. Edison believed that the current flow through his tube was proportionate to the degree of incandescence or candle power of the filament and connected his tube in circuit with a galvanometer or relay' for indicating or controlling conditions in the filament circuit. While his apparatus was capable of rectifying an alternating current he did not make use of an alternating current. On the other hand, Fleming used an alternating current which was rectified by the Edison tube, and he also made use of the rectified current in connection with additional apparatus in such a way that it operated a detector in transmitting wireless messages. In this respect his apparatus was new and useful.

As a general rule a patent cannot be based upon a new use of an old device, but to this rule there are exceptions. Judge Maver in Marconi Wireless Tel. Co. v. DeForest Radio T. & T. Co., 236 Fed. 942 (affirmed 243 Fed. 660), held in substance that the use which Fleming made of the rectified current and the manner in which he applied it constituted invention. He accordingly upheld the Fleming patent, and his decision was sustained by the Court of Appeals. We need not, however, pass upon the question of whether the Fleming patent merely sets forth a new use of an old device, as there is another and more decisive objection to its validity.

It must be conceded that the original application for the Fleming patent contained specifications of inventions that were not new, and upon which the applicant was not entitled to a patent. Making a claim known to be void ordinarily vitiates the entire patent. The statutes, however, provide that under certain conditions the invalidity may be avoided by a disclaimer. In the case of Ensten v. Simon, Ascher & Co., 282 U. S. 446, 452, the Supreme Court said with reference to these statutes:

“Construed together they ‘enact that where a patentee claims materially more than that which he was the first to invent, his patent is void, unless he has preserved the right to disclaim the sux’plus; and that he may fail to preserve that right, by unreasonable neglect or delay to enter a disclaimer in the Patent Office.’ Walker on Patents, 6th ed., Sec. 254.”

We have already shown that the patent recited that the invention related to new devices for converting alternating-electric currents into continuous electric currents, and Fleming stated further in his patent:

“I have discovered that if txvo conductors are enclosed in a vessel in which a good vacuum is made, one being heated to a high temperature, the space between the hot and cold conductors possesses a unilateral electric conductivity, and negative electricity can pass from the hot conductor to the cold conductor but not in the reverse direction.”

But this was not a new invention. As stated above, Fleming himself some fifteen years before (1890), in a lecture, showed that the current flowed only in one direction when an alternating current was used in connection with the lamp of an Edison apparatus, and discussed the flow of current from the hot negative terminal of the lamp to the cold electrode of the plate, observing that negative electricity flowed from the plate into an additional circuit connected with the lamp. In 1897 similar statements made by Howell and Kennelly were published. In stating this feature in his patent specifications as new and patentable, Fleming must have known that he Avas making a claim that could not be sustained. There could have been no inadvertence in making this statement, and this matter alone would seem to be sufficient to render the Fleming patent invalid. True, he entered a disclaimer later, but the statutes which give relief where invention has been wrongfully claimed do so only on the condition that the inventor has not unreasonably neglected .or delayed to enter the disclaimer. In the case at bar the disclaimer was not entered until eleven years after the patent had been taken out and long after other inventors had discovered other and different methods of making use of the Edison device without claiming to have invented it. We think that if ever there was a case of unreasonable delay in making the disclaimer it is presented in what is now before us and that the Fleming patent must be held invalid for that reason.

The defendant also raises the question as to whether any proper part of the patent was disclaimed, as seemed to be required in Corn Products Refining Co. v. Penick & Ford, 63 Fed. (2d) 26, and also in Grasselli Chemical Co. v. National Aniline Co., 26 Fed. (2d) 305, but we need not determine this question as we think the patent is void in any event for failure to make disclaimer within a reasonable time.

Even if the Fleming patent be held valid, it is still contended on behalf of defendant that there was in fact no infringement. The determination of this issue makes it necessary to examine more in detail the claims and specifications of the patent and the apparatus described therein.

The invention of the Fleming patent #803684 relates to a rectifier that is suitable for use as a part of a detector Avith signalling to permit the use of an ordinary direct-current galvanometer to indicate'the presence of the signals. The diagram figures in Finding LXIV reproduce the drawings of the Fleming patent, and it will be necessary to turn to this finding to understand the drawings and the description thereof as given in the finding. This description will not be repeated here. In general, however, it may be said that the Fleming device made use of a receiver consisting- in part of a vacuum bulb within which was an aluminum cylinder suspended by platinum wires from the glass bulb. Within the cylinder was a carbon filament supplied with an electric current somewhat in the same manner as in an incandescent lamp, with the result that the filament is heated to a very high temperature and the current flows or was intended to flow only in one direction between the filament and the cold electrode. The radio-frequency impulses received in the antenna are induced in the closed circuit of a transformer and rectified by the vacuum tube, with the result that half of the radio-frequency waves are lost and half pass through the vacuum tube and are effective to actuate a direct-current galvanometer. Fig. 1 of the diagrams set out -in Finding LXIV shows what is called the Fleming valve because it cuts off part of the current. As this valve tends to reduce the current flow, a method is shown in fig. 2 of reducing the loss of current, and fig. 3 shows how a number of these so-called “one-way valves” may be used to increase the possible current flow.

The claims in suit under the Fleming patent are numbered 1 and 37, and are set out in Finding LXV.

Oh November 17, 1915, the assignee, plaintiff herein, filed a disclaimer limiting the application of certain claims including claim 1, except when used in connection with high-frequency alternating electric currents or electric oscillations, and also eliminating words which showed that the apparatus described in the patent was intended to produce low-frequency currents or to make use thereof. The word “detector” as used in Fleming’s specification plainly referred to a device adapted to make feeble high-frequency impulses measurable with an ordinary direct-current mirror-galvanometer, or, in both the rectifier tube and the direct-current indicator of the rectified current. The term as used in the Fleming patent, however, did not include all apparatus associated with the detector tube, and did not include any amplifying tube or tubes.

The so-called Fleming valve, as shown in the description, had only two electrodes, and did not go into any extended commercial use, only about 700 Fleming valves having been shipped by plaintiff to anyone between 1910 and 1915. While in some respects an improvement on the crystal detectors, it did not entirely displace them. There is no evidence that the United States used any sets with two electrode tubes. Instead it used the three-electrode tubes which were much superior. The Navy used what is known as the DeForest audion which had a three-electrode tube, and will be described later on to show the difference between apparatus using the two-electrode tubes, and those using the three-electrode tubes.

In Marconi v. DeForest, supra, claims 1 and 37 were held valid and infringed by an audion detector having three electrodes. In another case with the same title, 261 Fed. 393, the decree rendered in the former suit was extended and Judge Mayer held the two-electrode tube to possess the same inherent ability to generate radio waves that is possessed by the three-electrode audion, and in effect that although Fleming did not know what could be accomplished with a three-electrode tube, and did not contemplate using it, he was entitled to the benefits of the device. With this conclusion we are unable to agree for many reasons. The basis of our conclusion is what we consider to be the fundamental difference in the manner in which the two-electrode tube and the three-electrode tube function.

The three-electrode tube was invented by DeForest in 1906 and patented by him in 1908. It is so called because it interposed between the heated electrode of the filament and the cold electrode of the plate or cylinder used by Fleming a third electrode in the form of a spiral of wire called the “grid”, which was connected with the closed input cir-between the two. patents, it will be necessary to consider just what features were covered by the Fleming patent after the disclaimer had been filed, what effects were accomplished by it, and what results attained as compared with the three-electrode apparatus used by the Government.

The Fleming patent did not include the discovery that an alternating current could be rectified by the use of what we might call the Edison tube so that only a direct current would pass through the outside circuit, nor could his patent include the discovery that the negative electrons emitted by the heated filament were of very small mass, or that the electron stream could be controlled by the potential of the cold electrode which occurs in rectifying an alternating current. In general, it may be said that the Fleming patent did not include the discovery that the Edison two-electrode valve was a suitable rectifier to change the alternating current into a direct current, nor did it include all uses for the electron emissions from the heated electrode.

Claim 1 of the Fleming patent recites:

“1. The combination of a vacuous vessel, two conductors', adjacent to but not touching each other in the vessel, means, for heating one of the conductors, and a circuit outside the vessel connecting the two conductors.”

These features which form the foundation of the Fleming-patent are each and all, as we have shown above, old in the-sense that they were well known to the prior art and the-patent would therefore be invalid if it were not for the disclaimer which puts these features to a new use which is. made more definite by claim 37 and the disclaimer. Claim 37 applies the apparatus to “electrical oscillations of high frequency” and as a detector uses the combination stated in claim 1. The disclaimer disclaimed the combination of elements in the claims “except as the same are used in connection with high-frequency alternating electric currents or electric oscillations of the order employed in Hertzian wave transmission”, and also omitted the words referring to low-frequency currents appearing elsewhere in the patent application, thereby showing plainly that it was not intended the: patent should apply to or low frequency.

The invention described in the patent was broad enough to show to one skilled in the art the desirability of using the Edison two-electrode vacuum tube as a rectifier for radio frequencies of high oscillations. By the term “detector” Fleming’s specification meant to one skilled in the art a device adapted to make feeble high-frequency impulses measurable with an ordinary direct-current mirror-galvanometer, and the term “detector” includes both the rectifier tube and the direct current indicator of the rectified current, but not any amplifying tube or tubes which were not mentioned in the Fleming patent and not discovered until after it had been taken out. Apparently Fleming had not thought of their use in connection with the apparatus described in his patent.

The defect in the Fleming apparatus arose from the fact that instead of amplifying the flow of the current (alternating current so rectified as to become direct current) it was reduced. It was sought to overcome this reduction by using a number of the one-way valves as shown by fig. 3 of the patent, while the three-electrode tube used by the Government tended to amplify the current and for this reason was much superior to anything that could be effected by the Fleming apparatus. The evidence shows that in the trial before Judge Mayer the two-electrode tube was made to amplify by the ionization of the residual gas in the tube, but there is nothing in the patent to indicate that anything of this kind was contemplated nor is there any satisfactory evidence to show that a Fleming valve was ever commercially operated in this manner by anyone. On the other hand, the evidence tends to show that this method was so unstable and uncertain it could not be used commercially.

1 The evidence also shows that the two-electrode tube can be made to amplify by means of a field outside the tube controlling the electron flow in a specially shaped tube. The effect obtained was similar to the control exerted by the grid of the three-electrode tube, but there is nothing inherent in the Fleming' valve nor in the patent descriptions and specifications that would suggest the use of such a tube. On the other hand, De Forest discovered that the three-electrode tube had a different effect upon the current from that produced when a two-electrode tube was used, and most important of all, that when the three-electrode tube was used the effects of the current flowing in the input circuit could be greatly amplified. The invention of the three-electrode tube and its amplifying use had a greater effect on the art than the two-electrode tube and the apparatus as described in the Fleming patent. The Fleming tube did not possess any inherent ability to generate radio waves, and it could not be made to amplify them, except by circuit connections never appreciated by the inventor of the patent in suit. The defendant never operated any two-electrode tubes by any of the methods above mentioned.

. The alleged infringements are all based on the use or sales of three-electrode tubes and apparatus for using them. The .general construction of the apparatus for.using three-electrode tubes is shown in the diagram of the De Forest audion set out in finding LXXV. This apparatus was used by the Navy and contains the principal and most important features .of the apparatus which made use of the three-electrode tube. The same finding describes at length in connection with the diagram the special features of the apparatus and its purpose. This descriptive matter will not be repeated except to show the difference between the De Forest invention and that of the Fleming patent. One of the marked differences- was that the Fleming patent'contemplated detection by rectification of the full radio or high-frequency oscillations impressed upon an indicator adapted to register the strength of the rectified one-way pulsating current, while the three-electrode tube in the diagram set out in finding LXXV and marked “Exhibit 83” detects by substantially suppressing the high-frequency oscillations and amplifying: principally the low or audio-frequency oscillations. In this connection it will be remembered that the plaintiff filed a disclaimer with .relation to the patent now under consideration which in effect withdrew its claims except as they related to high-frequency currents or oscillations. In contrast, the DeForest audion made use of the low-frequency oscillations. Part of the reason for this was because the audion was intended to be used in connection with a telephone, and the low-frequency oscillations are better adapted for use with that instrument.

The De Forest audion introduces the new feature of the so-called “grid” (marked “G” in the diagram), illustrated in Finding LXXY, which makes a third conductor or electrode :in the outer tube marked “T” which contains three conductors adjacent to but not touching each other and means for heating one of the conductors.

The purpose of the apparatus making use of the third electrode was to effect a control of the electron stream and to furnish a means of amplifying the radio signals. This third .electrode is connected to an input circuit and is separate from the output circuit, which contains an additional source of energy, with the result that a radio signal impressed on the input (grid) circuit is reproduced in amplified form in the output circuit; also the grid circuit constitutes a means for impressing upon the plate circuit principally the audio' (low) oscillations received and not the radio or high-frequency oscillations contemplated by claim 37 of ■the patent. These high-frequency oscillations are substantially supressed by the grid condenser of the three-electrode tube, while the Fleming patent, as before shown, contemplated detection by rectification of the radio or high-frequency oscillations as impressed upon an indicator. The great advantage of the apparatus used by the defendant was that it could be so used as to amplify almost indefinitely the received signal and thus enable communications to be made over far longer distances and with much greater accuracy than could be accomplished when the two-electrode apparatus was used with its reducing effect.

Counsel for plaintiff cite the case of Marconi v. De Forest, supra, as adjudicating the validity of the Fleming patent and holding that it is infringed by the De Forest audion. <jn behalf of defendant it is urged that the opinion rendered therein shows that both sides were suing upon two-electrode patents but as the patents involved in that suit have not been offered in evidence this cannot be definitelv determined. It does appear, however, from all of the evidence in the cases between the plaintiff and the De Forest Company that both sides represented to the court and agreed that the inherent qualities of the two-electrode tube and the three-electrode tube are the same. This being admitted by counsel, the decision, of course, was based upon that conclusion. If this conclusion was not correct, and we think it was not, the decision last cited has no application to the instant case.' It should be observed, however, that the case involved ten patents, of which certain ones belonging to the plaintiff were held valid and infringed and as we understand the opinion it would seem that certain of the De Forest patents were also held valid. But we have nothing before us from which we can determine what was covered by these patents, respectively, and consequently cannot say how far Judge Mayer’s decision went. The decision recites that the plaintiff (Marconi Co.) confessed judgment on two of the De Forest patents, and it is said that “De Forest in his three-electrode audion has undoubtedly made a contribution of great value to the art and by the confession of judgment in respect thereof defendant company may enjoy the just results of this contribution”, and a type of the De Forest audion was used by defendant in the case now before us. But it is not necessary for us to determine whether any of the De Forest patents were valid. What we have to determine in this case is whether the De Forest audion and other three-electrode apparatus used by defendant as.,shown by the findings constituted an infringement. One other matter which should be considered on this point, as we view it, is that a comparison of the apparatus of the De Forest audion, as shown in the diagram in finding LXXY, with the Edison apparatus, as shown in fig. 4 in finding LXX, definitely establishes that the De Forest audion, with the exception of the added third electrode in the form of a grid and the additional circuit attached to the grid, used absolutely nothing but what had been pictured by Edison long before Fleming thought of it; and that the additional apparatus used by De Forest was something that Fleming had never imagined and wa.s not suggested in any way by his patent. What De Forest accomplished thereby could never be ejected by the use of the Fleming tube, except by the use of some special additional apparatus never contemplated by Fleming and never put into commercial or practical use.

The defendant not only introduced new features in the apparatus which it used, but the apparatus functioned altogether differently from that of Fleming’s and produced different and improved results.

Dr. Van der Bijl, a noted radio expert, was quoted with approval by Dr. Miller, a witness in the case, as follows: “In fact, the insertion of the grid into the valve resulted in a device of tremendous potentialities, one that can justly be placed in the same category with such fundamental devices as the steam engine, the dynamo, and the telephone.” And also, “It is hardly necessary to say that the insertion of the grid has made the audion a device of immense practical importance and enabled it to perform functions that would otherwise have been impossible.”

Plaintiff’s expert, Weagant, practically admits that these quotations state the fact correctly, but says that “the two-electrode Fleming valve is the foundation on which this important .structure has been reared and, as has been demonstrated in my recent tests, possesses inherently all of the characteristic capabilities of the three-electrode tube.” This statement of Weagant’s is the foundation of plaintiff’s case. Let us see how it agrees with the facts shown in evidence and the reasonable deductions that may be drawn therefrom.

We have already, shown how different was the function of the three-electrode tube from the two-electrode tube, but Weagant says that the two-electrode tube can be made to function in the same manner “by means of an electrostatic field control of the electron stream.” He does not explain how, but it was shown elsewhere in the testimony that it was by means of an electrostatic field being brought close to a specially shaped bulb of the Fleming valve but outside of it, the effect of which was to a greater or less extent to deflect the electron stream from the heated filament. In the DeForest apparatus the “grid” was placed inside of the bulb and not outside and, as we have seen, functioned in a very different manner. But perhaps the most complete and fatal objection to this claim for the Fleming patent is that while it was shown by tests at the trial that the Fleming tube could be made to amplify in this manner, there was nothing whatever said in the Fleming patent about such an amplification and not the remotest suggestion of setting up an outside field to produce an amplifying or oscillating effect. Apparently no one ever thought of anything of the kind until the three-electrode tube had gone into successful use — in fact, into universal use ■ so far as long distances had to. be overcome. After experiments with the three-electrode tube had shown how superior it was not only to the Fleming two-electrode tube but even to the crystal detectors, some of which were much more effective than the Fleming apparatus, after a disclaimer had been filed on the Fleming patent, and after suits had been begun for its infringement; .in.short, after all this had taken place, someone found that some of the same results could be produced with the Fleming valve in the manner described above and demonstrations were made on the trial of cases affecting the De Forest patent. When it comes down to the case before us; we find that years have elapsed since that discovery was made and yet there is not a syllable of evidence to show that this method, which is now said to be inherent in the Fleming patent and by which the same results could be accomplished as by the three-electrode tube, has ever been commercially used, nor is there any evidence that it ever was used in a practical way, although there can be no doubt that the plaintiff would have produced testimony on that point if it could be obtained. On the other hand, although De Forest was probably not conscious of the full limits to which his invention might be extended, it is a matter of common knowledge that the three-electrode tube came into general use for long distance purposes, superseding even the best of the crystal detectors. The G-overn-' ment at all times and especially in war times had need of the best which could be obtained and it never used any of the two-electrode tubes. All the tubes which it used were of ■ three electrodes. In fact, only about 700 of the Fleming valve two-electrode tube sets were ever put out, so completely did the three-electrode tube surpass them.

The Circuit Court of Appeals in Marconi v. De Forest, supra, says that in the audion the high-frequency oscillations and their electrical result get into the indicator or battery circuit just as in the Fleming valve. This, we think, is quite incorrect. The audion excluded the high-frequency oscillations but what is more important it used two circuits in transforming the alternating current and functioned altogether differently. It is said that a detector must act on alternating current arid that this makes the defendant (De Forest) an infringer although he invented an improved detector. If by the word “detector” is meant the tube containing the heated filament, it is true that an alternating current is used both by Fleming and by De Forest in his audion. But this does not make the audion an infringement, for De Forest used it in a different combination, with different apparatus and for a- different purpose. It is true Fleming discovered that Edison’s apparatus could be used to transform the alternating current into a direct current, as shown by his article before referred to, but all this was several years before his patent was issued and could not be patented. The audion made use of Edison’s apparatus and the discovery that the alternating current was thereby rectified, but beyond this it made use of a new device which was probably the greatest improvemént -that has ever been effected and upon which all long distance wireless communications still depend. .

'It is contended that when DeForest introduced the screen grid he merely divided' the current which entered the tube and passed in.and out of the heated filament and that there was no invention in so- dividing it. We are not impressed with this argument. Ever since the possibility of wireless communication in some manner entered the brain of Hertz, scientists have been experimenting to ascertain what could be done with original current to extend the distance in which its effects could be felt or made manifest in a distant apparatus. In so doing, they have strengthened the current' by parallel coils and condensers; they have varied it with spark gaps, .with greater or less intervening space; they have trans-. formed it from an alternating current to a direct current,* and finally, by use of an apparatus, in which the screen grid is essential, they have succeeded in amplifying- it many thousand times. Today we can hear around the world not merely signals but the voice of man reproduced from the same current that originated in the transmitter, strengthened, transformed, and amplified. If this is not discovery or invention or both, it is difficult to understand what would be.

The other apparatus used by the defendant and alleged to infringe the Fleming patent is referred to in Findings LXXYI, LXXIX, and LXXX. Facts are set forth in these findings which, in our opinion, furnish sufficient reason for holding that such apparatus did not constitute an infringement and we think no useful purpose would be accomplished by repeating the statements contained in the findings.

Our conclusion on the whole is that there has been no infringement of. the Fleming patent.

• Whalet, Judge; Williams, Judge; LittletoN, Judge; and Booth, Chief Justice, concur.