Marconi Wireless Telegraph Co. of America v. De Forest Radio Telephone & Telegraph Co.

236 F. 942 | S.D.N.Y. | 1916

MAYER, District Judge.

Whatever differences may exist between men of science in respect of the theories by which they account for the movement and action of the unseen forces about which so much has been testified and argued in this case, the solution of the points of the controversy, with a single exception, is not difficult. This, because courts, in an art of this kind, place their decisions upon things demonstrable, and cannot speculate as to theories in regard to which there is not a common agreement among recognized authorities.

In endeavoring to resist plaintiff’s attack, defendants have proceeded on the theory that, beginning with his parent patent, No. 979,275, antedating Fleming, De Forest gradually developed his first conception until finally it found practical exemplification in the two so-called three-electrode “Audion” devices as to which plaintiff has confessed judgment. In line with this plan of defense, defendants have elaborately built up an unsteady theoretical structure, and upon this have superimposed an observatory from which they can see in the mind’s eye only that which they call “Audion” action. Therefore, in these circumstances, it is desirable, in order to avoid confusion, to consider first the patents in issue, and then the question of infringement ; ’ for, when their true value is assigned to the patents, the controversy as to infringement will be better understood. The patents deal with those instrumentalities which, in the art, are aptly named detectors.

“The purpose of the detector,” as explained in simple language by Waterman, plaintiff’s expert, “is to enable some indicating instrument to respond to and thus reveal the presence of the high'-frequeney oscillatory currents which are the result in the receiving system of the transmission of the wireless waves. These wireless waves are of the same nature as light, but are of greater wave length'. We have sense organs for perceiving ether waves of the length known as light, also sense organs for perceiving that range of wave lengths known as radiant heat; but we have no means of detecting ether waves of those lengths which are employed in wireless telegraphy, and it is therefore necessary that their presence should be detected and indicated to us through some means that we can perceive. A wireless telegraph transmitter is thus a sort of lighthouse, which emits light of an invisible nature, and the receiver must furnish the eye to detect th'e waves which are emitted. * * * If we go back for the moment to the figure of the lighthouse, we see that, if we think of light sent out from it, then the receiving antenna casts a shadow. The energy which it receives corresponds to that shadow, just as, when a material object casts a shadow of light, it absorbs the energy of the light which it intercepts, so that receiving antenna absorbs the minute amount of energy which it intercepts in the moving wave. On account of their excessive frequency and minute energy, the oscillatory charges which are set up in the receiving antenna are not, generally speaking, able to affect directly any known measuring or indicating apparatus. They therefore must make their presence felt indirectly, by producing some local effect which will permit of a signaling or indicating apparatus to be operated in accordance with a sufficiently definite code so that intelligible signals may be sent. Hence a detector, as the term is used in wireless telegraphy, is a means of causing the oscillations to produce or vary a local current, in accordance with variations of the waves produced at th'e sending station, and of such a character that an indicating instrument can respond to them.”

See, also, Pierce’s Principles of Wireless Telegraphy (1910) page 142.

As the practical radio art developed, there was a constant effort to improve the detector in three directions; First, and most important, in sensitiveness to received signals; secondly, in reliability; and, thirdly, in ease of manipulation ’ by the receiving operator. There were many types of detectors prior to Fleming, the most useful of which were known as tire coherer, the microphone, the magnetic, the electrolytic, and the crystal. Some detectors, such as that of Hertz and the hot wire barretter of Fessenden, were never of any commercial utility, and may be disregarded.

. The coherer was in standard use for a fairly long period. It consisted of a glass tube with metal filings, and its operation was caused by the cohering of the filings, due to high-frequency oscillations, thus transforming a practically nonconducting device into a conductor and permitting a local battery current. The coherer lacked sensitiveness to feeble waves, and required to be shaken or otherwise moved to restore the coñtact to its sensitive condition after the receipt of a signal. See Pierce, supra, p. 143 et seq.

The microphone consisted of a loose contact of two terminals, preferably dissimilar in character, such as carbon and steel, and operated by reason of a change of contact resistance effected by the incoming oscillations. This device was used both with and without a local battery. Its failure to attain any large commercial use was due to the delicacy and difficulty of adjustment.

Of the magnetic detectors, that of Marconi was widely used and displaced the coherer. This magnetic detector consists, in substance, of a moving- band of soft iron passing in front of two magnets, which magnetize the iron. A coil is so connected to the antenna that the oscillations demagnetize the iron band and ate thus detected. Although still useful, because of simplicity of operation and indifference to static discharges, this type lacked the keen sensitiveness which has become so important to- the increasing usefulness of the art.

The electrolytic consisted of a cell containing an electrolyte (usually 20 per cent, nitric acid, but, in any event, any electrolytically conductive liquid, such as common salt solution, dilute sulphuric acid, or caustic soda) and having two immersed electrodes. One form was the Shoemaker cell, where the two elements were dissimilar, and another ivas where the elements were of the same material, such as fine platinum wire. In the first form a local battery was not used, while in the second it generally was. Though this detector was highly sensitive, it was extremely difficult of adjustment, especially on shipboard, and the fine wire was liable to be burned out by strong signals or static discharges.

The crystal detector, invented by Bose, opened up a new line of experiment and investigation, to which, among others, Gen. Dun-woody, of the United States Army (retired), defendants’ expert Pick-ard, and Pierce, of Harvard, la.ter (and after the Fleming date) made valuable contributions. Detectors of this class consist of a self-restoring high resistance between solid bodies, one of which is usually crystalline in character, such as carborundum and molybdenite. The operation depends upon the phenomenon that, when a contact is made with certain crystals, current will flow more easily in one direction than another.

The crystal detector, particularly because of ruggedness of material, is still in extensive use; but, as is generally accepted and was fully demonstrated in the courtroom, it is somewhat unsatisfactory by the reason of the necessity of taking time to feel around, as it were, sometimes for a sensitive point, and sometimes for the best point on the crystal, and the liability that such a point may be destroyed, or its sensitiveness impaired, by a strong incoming signal, by static, or by the local sending station.

These criticisms or defects of one kind or another in the detectors prior to Fleming — or since, for that matter — will be fully appreciated when it is realized that efficiency in this art consists in attaining accuracy and quickness in reception of signals, as well as distance, whether the radio message is across the ocean, from one merchant to another, from a vessel in distress, calling' for help from land or sea, or from a naval officer to the ships under his command.

[1] With the state of the art as briefly outlined, supra, John Am-brose Fleming disclosed the incandescent lamp detector. While the United States patent application was filed April 19, 1905, the effective date is that of the British specification, filed November 16, 1904. Fleming is a British scientist of the highest standing, and, as appears from his patent and his papers read before learned societies, is and long has been recognized as a man or major accomplishments, with the ability to make clear what he intends to convey.

Stripped of technical phraseology, what Fleming did was to take the well-known Edison hot and cold electrode incandescent electric lamp and use it for a detector of radio signals. No one had disclosed, nor even intimated, the possibility of this use of a device then long known in another art. Cohering filings, magnets, electrolytes, and sensitive crystals, at that time, failed to. give any hint of the utility in this art of the Edison lamp. What led Fleming to his result was his adherence to the theory of the “rectified” alternating currents. In his patent specification he put his proposition thus:

“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 ‘mirror galvanometer’ of the usual type or any ordinary direct-current ammeter. Such instruments as the latter are not affected by alternating electric currents, either of high or low frequency, which can only be measured and detected by instruments called ‘alternating current’ instruments, of special design. It is, however, of great practical -importance to be able to detect feeble electric oscillations, such as are employed in Hertzian wave telegraphy by an ordinary movable coil or movable needle mirror galvanometer. This can be done if the alternating current can be ‘rectified’ ; that is, either suppressing all the constituent electric currents in one direction and preserving the others, or else by changing the direction of one of the sets of currents which compose the alternating current, so that the whole movement of electricity is in one direction. * * * I have discovered that, if two conductors are inclosed 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. As the- hot conductor should be heated to a very high temperature — say near to the melting point of platinum (1700° centigrade) — it should be of carbon, preferably in the form of a filament, such as is used in any ordinary incandescent electric lamp. The cold conductor may be of many materials; but I prefer a bright metal, such as platinum or aluminum, or else carbon. The two conductors/ are inclosed in a glass bulb similar to that of an incandescent lamp, and I generally heat the carbon filament to a high state of incandescence by a continuous electric current, the electrical connection to the filament and the cold conductor being made by platinum wires, sealed air-tight through¹ the glass.”

He clearly described the necessity for a high degree of vacuum and a highly incandescent filament. Fleming patent, page 1, line 96, and page 2, line 5.

[4] In his lecture before the Royal Society, read February 9, 1905 (Proceedings of the Royal Society, vol. 74, particularly pages 477, 481 — 485; also see Waterman’s testimony, page 1083 et seq.), he described the mode of operation of his device, making equally clear what he had set forth in his patent specification, and, further, illustrated his views by a sheet of “characteristic curves.” The “characteristic curve” is a curve plotted between voltage applied to the detector and the current through the detector, resulting from the application of this voltage. It is obtained by connecting a circuit containing a battery and a galvanometer or ammeter and a resistance whereby the potential of the battery may be varied to the hot and cold elements of the detector. Connected across the detector is a voltmeter to measure the applied local battery voltage. I agree with plaintiff that this sheet disclosed to one skilled in the art everything necessary to obtain a complete knowledge of the operation of the device. It showed that, as the incandescence of the filament increases, this detector device becomes more sensitive, and logically, therefore, the device should be operated at a high degree of incandescence, obtainable by whatever were known means therefor. Why the device thus operates successfully to detect signals is not as yet surely understood, but that it does so operate is an unescapable fact.

Fleming called the operation “rectification,” and held that, substantially speaking, the current will flow through the lamp in one direction only; i. e., from the cold cylinder to the hot filament. In his 1905 lecture he said:

“Perfect rectifying power, liowever, does not exist. There is not an infinite resistance to movement of negative electricity from the metal cylinder to the hot filament through the vacuum, although this resistance is immensely greater than that which opposes the movement of negative .electricity in the opposite direction. * * * Returning, then, to the vacuum valve, we may note that the curves in Pig. 3 show that the vacuum space possesses a maximum conductivity corresponding to a potential difference of about 20 volts between the electrodes, for the particular valve used. The interpretation of this fact may, perhaps, be as follows: In the incandescent carbon there is a continual production of electrons* or negative ions by atomic dissociation. Corresponding to every temperature there is a, certain electronic tension or percentage of free electrons. If the carbon is made the negative electrode in a high vacuum, these negative ions are expelled from it; but they cannot be expelled at a greater rate than they are produced. Therefore there is a maximum value for the outgoing current, and a maximum value for the ratio of current to electromotive force; that is, for the conductivity.”

Whether right or wrong in his theory, the result of Fleming’s invention was to give the art a new, valuable, and easily obtainable detector, which has gone into important commercial use. This Fleming detector is highly sensitive, quickly adjusted by an operator of even inferior skill, and only momentarily disturbed by static or strong signals. The thoroughness and earnestness of this litigation is its most significant testimonial. Nothing in the prior art urged by defendants in negation of invention calls for extended discussion. The Tesla patent (No. 645,576) and the Fessenden patents (Nos. 706,742, 706,743, and 706,744) were far removed from the incandescent lamp and were commercially useless; and nothing could be learned for this purpose from the Valbreuze and Zehnder tubes.

Rectifiers of low frequency oscillations, such as those of Wehnelt and Cooper-Hewitt, taught nothing. These are rectifiers for commercial power frequencies, and it was not common knowledge, as of Fleming’s date, that rectifiers of low-frequency oscillations would rectify radio waves; nor is it a fact that all rectifiers of low frequencies are likewise rectifiers of radio high frequencies. Further, it was not common knowledge, as of Fleming’s date, that a rectifier of radio oscillations would act as a detector. For instance, Pickard first attributed the action of crystal detectors to thermo-electric effects; but, when Pierce published his investigations in 1907, Pickard amended many of his patent applications to conform with Pierce’s theory of rectification. See Pierce supra, page 162, and testimony of Pickard.

In the absence of a well-accepted theory of operation which needed merely some physical embodiment, and in the absence, also, in the art of the physical device itself, at a time when men of great skill were constantly endeavoring to bring forth an advance in this branch of the art, the contribution of Fleming was clearly invention, and is entitled to liberal interpretation and consideration — unless impeded by De Forest.

[2] This brings us to the parent patent of De Forest, No. 979,275, to which, on the evidence, the effective date of November 4, 1904, must be accorded. Plaintiff is well justified in calling this and the divisional applications the Bunsen burner patents. Nowhere is there a suggestion of an incandescent electrode. On the contrary, in the specification and the drawings it is entirely apparent that De Forest pointed out only what the layman understands as heating gas. This De Forest "stated in language which sounds impressive. He said:

“I have discovered that, if two bodies adapted for use as electrodes or conductive members be electrically separated partially or wholly, after the manner common in analogous devices, the separation between them may be neutralized sufficiently to enable them to act as a detector of electrical oscillations, if the intervening or surrounding gaseous medium be put into a condition of molecular activity, such, for instance, as would be caused by heating it in any manner, as by radiation, conduction, or by the combustion of gases in the space which surrounds the poles. Such condition or molecular activity causes what would otherwise be a nonsensitive device to become sensitive to the reception of electrical influences. I am thus enabled to employ as such sensitive member devices which would otherwise be of no value, or to make those devices now used more sensitive to the electrical waves. This principle is embodied in the apparatus illustrated in the various figures shown.”

Translated into plain English, this meant;

“I will try to make the gas conductive between two electrodes by heating it to the dissociating point.”

It was attempted to read incandescence into the specification, or rather to infer much that later knowledge has taught; but incandescence had long been a word of art, and Fleming had no trouble in using it, either in his specification or his Royal Society paper. Why not De Forest? Merely because the incandescent lamp detector was the farthest from his thoughts.

[3] True, gas is a generic term of wide meaning, as is clear from the very beginning of J. J. Thomson’s notable “Conduction of Electricity through Gases”; but, when the language of a patent is to be interpreted, the document must be construed as a whole, just like any ordinary contract, and words cannot be isolated from their context to give them a more comprehensive meaning than was originally intended. What defendants have attempted is to establish that De Forest described in these patents ionization by impact as distinguished from dissociation by flame, and thus forestalled Fleming, on the hypothesis that De Forest was the first to realize the value and effect of electronic emission. No better confirmation of the negligible character— in this connection — and, perhaps, obscurity of the disclosure can be found than the testimony of Pickard, in answer to the court’s questions (Q. 925 et seq.).

An elaborate discussion, at this juncture, of electronic action might be interesting, for the subject is really fascinating; but it is unnecessary, for the simple reason that the patent discloses merely the heating of the gas, without any direction from which the most learned scientist of that day could have gleaned any further information.

In the divisional patent, 867,876, the expression is used, “This gas may be air, or the electrodes may be inclosed”; but bow this device works is still to be explained, for the experiment at High Bridge with the Ncrnst lamp was not in accordance with the disclosure of the patents. This burner detector oí De Forest lias never been commercially used, and thus has not made any impression on the art. A mere inspection of the device in operation will show that this flickering flame is impractical. The most that can be said is that it may contain the germ of an idea which, in this rapidly progressing art, may hereafter be utilized in some way. While, therefore, it is- not necessary to declare this patent and its three subsidiaries invalid, they may be eliminated from this case for all practical purposes.

Before considering the patent, No». 824,637, and its division, No. 846,070, filed originally January 18, 1906, it must be remembered that De Forest in December, 1905, knew of Fleming’s Royal Society paper of March, 1905, as appears from a reference to that effect in his application for a certain patent not here in issue (No. 823,402), where he used the expressions “exhausted vessel” and “heated to incandescence.” Further, on December 21, 1905, he instructed his solicitor to “look out for Fleming’s recent patent.”

The point is that what in effect defendants urge inter alia is that De Forest’s idea of employing a local battery, which has come to be known as “Battery B,” in any event, imparts invention to his patents, and its use by plaintiff amounts to infringement, or, if that contention be not sustained, then finally defendants do not infringe. With his knowledge of Fleming’s theory, it should have been very easy to describe the incandescent lamp detector plus battery B, but, in 824,637 De Forest now had in mind a receptacle inclosing a gaseous conductive medium. He said:

“With these objects in view, my invention comprises a receptacle inclosing a sensitive gaseous conducting medium, the conductivity of which does not necessarily depend upon the heat of combustion, although such conductivity may be increased by heating said gaseous medium, and which in some cases requires practically no heating at all, a wave-intercepting means associated with said gaseous conducting medium, whereby the feeble electrical currents or oscillations resulting from the energy absorbed from electromagnetic signal waves may be impressed upon said gaseous conducting medium to alter its conductivity, and a signal-indicating device operatively connected with' said gaseous conducting medium, whereby alterations in the conductivity of the latter may be made manifest.”

The only possible reference to a vacuum is at page 1, lines 101— 105, as follows:

“In all embodiments of the present invention the electrodes are inclosed and are surrounded by a suitable gas, and they may be inclosed in a receptacle which may be partially exhausted.”

The only reference to incandescence is in one compound word at page 2, line 4, as follows:

“In Mg. 1 two filaments, C, which may be ordinary incandescent-lamp carbon filaments, are sealed into the receptacle, B. * * * ” ²

These “mays” at best are meager disclosures, but that these patents dealt only with the heated gas idea is clear from De Forest’s correspondence with his solicitor in December, 1905 and January, 1906, from his ordering incandescent lamps from one McCandless in the same December and January, with his thereafter change of phraseology and tone (see No. 841,386, and his January 20th letter to his solicitor, “Keep it dark, but the new receiver is the best yet”), but most convincingly from the'patent itself.

According to Pickard’s theory of ionization by impact, there must be a source of electrons; but in this patent no electron producing nor impelling means are shown. On the contrary, Figures 4, 5, and 6 show two cold electrodes, and, referring to Figure 4, De Forest specifically dispenses with both heated electrodes (page 2, line 72), an inconsistency with defendants’ theory which cannot be reconciled (XQ. 1147 et seq.). These patents (Nos. 824,637 and 836,070) were never of any commercial utility and at best can be sustained only within the limits of their precise disclosures.

The so-called selective per se patent, No. 841,386, is so utterly useless that it might well be declared invalid; but it will suffice in this suit to> construe it as limited to a structure selective per se and irrespective of any circuit connections. It follows, of course, that plaintiff does not infringe any of defendants’ patents and that the counterclaim will be dismissed.

We now come to what I think is the only substantial question in the case- the infringement claimed against defendants. The Fleming patent was originally framed in rather broad language, so that it might have been construed as applying to other than radio uses, in addition to its use in the radio art. By disclaimer, filed in the Patent Office November 17, 1915, plaintiff disclaimed the combination of claims 1 to 6, inclusive, and claims 10 to 15, inclusive, except as the same are used in the radio art, and to certain correlated words in the specification.

The claims selected to sue upon were Nos. 1 and 37, because typical. They read:

“1. Tlio 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.”

“37. At a receiving station in a system of wireless telegraphy employing electrical oscillations of high frequency a detector comprising a vacuous vessel, two conductors adjacent to, but not touching, each! other in the vessel, means for heating one of the conductors, a circuit outside of the vessel connecting the two conductors, means for detecting a continuous current in the circuit, and means for impressing upon the circuit the received oscillations.”³

Claim 1, as limited by the disclaimer, is a broad claim for the incandescent lamp as a radio detector., Claim 37, in respect of which disclaimer was unnecessary, covers the detail applicable to a radio system; i. e., a local circuit containing means for detecting a continuous (direct) current, such as a telephone or galvanometer, and means of impressing high-frequency oscillations on the detector, such as the secondary of the oscillation transformer.

Fleming’s theory, as has already been stated, was that of rectification; while defendants account for the action of their “Audion” on the theory that it is a telephone relay, or, in other words, that its products arc. alternating currents of “audio” frequency and of the local energy, and not of the “input” energy. As a result of these differences, the effect and relation of the iocal battery was one of the sharply contested points in controversy.

It was satisfactorily proved that, for some reason not yet understood, incandescent lamps possess idiosyncrasies of operation, as demonstrated by a batch of a dozen lamps of identical dimensions, made of identical stock, pumped at the same time for a vacuum, and sealed at the same time. Farrand’s testimony; Waterman, 1244 and 1820 et seq. Of these, some worked best at the negative end (i. e., without a battery), some with a small amount of battery, some with a battery equal to the battery for lighting the filament, and some with a battery in addition to that used for lighting the filament.

While, with care and time, lamps could be selected which would work best without a local battery, such a course would obviously be foolish commercially, and unnecessary, when a simple and well-known means could, Jse employed to utilize all the lamps, whatever their idiosyncrasies. This means was a local batteryj and a potentiometer, whereby a varying local potential may be applied to the lamps. The potentiometer is a resistance connected across the lighting battery of the detector, so that any fraction of the lighting battery may be tapped' off and applied to the local circuit. The local battery is used to bring the lamp detector to the sensitive point of. its characteristic curve, and the potentiometer is the simple and effective device which, varying the local battery, accomplishes this task.- Nearly all prior art detectors were used in this way — the coherer of Marconi and Lodge ; the microphone of Hughes and Branley; the electrolytic of Fessen-den, Vreeland, and others; and the crystal of Bose. Plaintiff’s Exhibits 77 and 82.

The use of the local battery to locate the sensitive point on the characteristic curve was Well known and accepted as” of Fleming’s time, and, as appears by his 1905 lecture, was fully understood' by him. See, also, particularly, Vreeland patent, No. 780,842, Plaintiff’s Exhibit 82. Plaintiff is undoubtedly entitled to use the Fleming detector with a well-known instrumentality, and therefore to employ the variable local battery, for practically all the prior art detectors required local batteries to locate the operating points. Plaintiff is likewise entitled to use the Fleming device in the ordinary detector circuits of the prior art. The circuits of the Marconi patent, No. 627,650, are the specific circuits which, plaintiff has used, and the modern operative Fleming device has simply been substituted for the coherer in old and familiar circuits (Q. 145, 146, et seq.).

Defendants’ alleged infringing device is the so-called P. N. Type Audio-n De Forest Detector. Plaintiff’s Exhibits 11 and 12 are drawings reducing to the simplest form the P. N. circuits and comparing Fleming with De Forest.

Waterman’s description (in part) states in simple language plaintiffs view of this P. N. device as follows:

“The defendant’s apparatus is an incandescent lamp detector of high-frequency oscillations for wireless telegraph purposes, and consists of an incandescent lamp with the nsual high vacuum, a filament, and a cold element, which in the particular construction here shown is divided into two portions, one a plato and the other a wire gridlike structure placed between the plate and the filament, both having leads brought out through the glass for exterior connection. * * * ”

In the diagram of the Fleming patent there is missing the grid found in the De Forest structure.

“In other words, the two functions which are essential in the incandescent lamp detector, namely, impressing on the space the received oscillation and the detecting of what happens as a. result, are performed by a single circuit, In which are located the oscillation transformer and the Indicator. In tills De Forest circuit, as the devices are arranged, * * * two circuits are used, and the cold element is correspondingly divided up. These are two well-known, standard, equivalent circuits. You may use either device in either circuit.”

Of all the explanations of the action of the De Forest, that of Armstrong in the Electrical World (December 12, 1914, Plaintiff’s Exhibit 45) seems most convincing, and that article, for purposes of brevity, may be regarded as being read into this opinion. Sec, also, Dr. Austin’s Bulletin of the Bureau of Standards of the Department of Commerce and Babor. In reading this literature, it must be remembered that both sides agree that the De Forest two-element arid three-element bulbs operate on the same principle.

As Armstrong was on the stand acid subject to cross-examination, his article is to be treated, not merely as a scientist’s essay, but as equivalent to testimony. Its details cannot be satisfactorily abstracted, but the result is that the weight of the evidence points to the conclusion that defendants’ device is one of unilateral conductivity, or, in other words, a rectifier permitting current to flow in one direction only, viz. from plate to filament and from grid to filament.

The two circuits used are, as Waterman said at the beginning of the case, a division of Fleming’s single circuit, the grid-element division being the branch for impressing the oscillations on the detector, and the plate-filament division being the branch for indicating the signal; and it is established with reasonable certainty that defendants’ device, in order to operate, must have a heated electrode connected to the negative terminal of the local battery. See, also, Fleming’s Technics article, Plaintiff’s Exhibit 22 and Waterman’s experiment at High Bridge, testimony, 1566 et seq. and 1481a et seq. See, also, transcript of oral argument of Mr. Hoguet at pages 30-34, expressing briefly the views which I accept as to the action of defendants’ device and the controverted points as to the galvanometer and telephone.

In Exhibit 123, plaintiff has compiled some 2Q articles to show the identity of the Fleming and De Forest detectors. Of course (except Armstrong’s article, because he testified), these articles do not prove plaintiff’s proposition; but they do show the point of view and the opinions entertained by many scientific authors. Against all these we find Pickard standing alone, except in so far as the interesting theoretical exposition of Dr. Davis supports him.

Pickard has developed the theory that the local battery changes the mode of operation of the incandescent lamp from rectifier to relay; but, while many experiments were -made and much testimony was given, this theory is still in the realm of speculation, and certainly has not been satisfactorily demonstrated. Indeed, it was necessarily stated by Waterman, and admitted by Pickard, that the ultimate mechanism of the subject-matter is not known, and that physicists are compelled to change their theories from time to time in the light of later investigations.

In order to reconcile the explanation of the action of the De Forest grid detector with the language of De Forest’s earlier patents, so as to work out the idea that De Forest’s two and three electrode detectors were simply the logical development of an original thought, Pickard advanced the theory that the action of the De Forest grid was by ionization by impact, and therefore that it was necessary to have a local battery to impel electrons at a high speed on their journey of succeeding collisions. But this theory is shattered, or at least impaired, by the tests, which showed that, when ionization by impact occurred, the detector showed a blue glow and stopped operation. If anything was shown in this regard, it was rather, as plaintiff contends, that the device operates in spite of, and not because of, ionization by impact.

Within the limits of an opinion it is, of course, impossible to analyze at length a mass of experiments, tests, ánd theses, and an infinity of detail necessarily involved in the testimony of experts, in an art of this kind; but, if plaintiff’s theory that its own device and that of defendants operate on the same principle has not been proved (and I think it has as far as such proof is yet possible), at least defendants’ theory has not been satisfactorily demonstrated, and, finally, the physical facts all support plaintiff’s claims. Here, as is so often the case in lawsuits, resort is had to the story of events and the outcroppings of human nature.

De Forest had long been proceeding on a theory different from that of Fleming. Having read Fleming’s article, he began to experiment with the incandescent lamp. He probably doubted its efficacy at first, but within a very short space of time — perhaps a week, perhaps a month — he changed his mind, and, discovering that Fleming was right, wrote his solicitor, after he had filed his application for No. 824,637, that the "new receiver is the best yet.” Thereafter he used the language of the incandescent lamp, and in an address on October 20,' 1906, before the American Institute of Electrical Engineers, really described fundamentally the Fleming lamp detector, although using phraseology which has since become Audion vocabulary. Thus the physical ocular fact is that in the alleged infringing P. N. device, the Fleming detector, and not the Bunsen burner, is used, and the broad claim No. 1 of the Fleming patent is infringed, precisely the same as if a patented crystal has been placed in some old or new type of circuit with a local battery — such, for instance, as the Weagant and Armstrong circuits.

In respect of claim 37, defendants’ device does not escape because the circuit outside the vessel is divided into two branches, nor because Fleming’s detector of a “continuous current” was a galvanometer and De Forest’s is a telephone long well known in the art. 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; but, on the other hand, Fleming’s invention was likewise a contribution of value, and is to be treated liberally, and not defeated, either by unconfirmed theory or by association in apparatus, where later developments have taught how other useful adjuncts can be employed.

Claims 1 and 37 of plaintiff’s patent are valid and infringed by defendant company; defendants’ counterclaim will be dismissed, and, as there is no evidence against De Forest individually, the bill as to him will be dismissed.

1 Noto. — Eccles in his “Handbook of Wireless Telegraphy and Telephony” defines “electron” as “the ‘atom’ of negative electricity, the smallest quantity of electricity knowto to take part in electrical phenomena.”

2 Note. — Italics mine,

3 Note. — In construing claim 37, it must be remembered that “continuous current’* is used in its English sense of 1905 of a direct current, -whether intermittent, varied, or not. See, also, page 2, line 109, of Fleming patent. By a recent convention, continuous current now means a direct current of unvarying value.