From the History of Physics

THE EARLY PERIOD IN THE HISTORY OF THE EXTERNAL PHOTOELECTRIC EFFECT AND THE SIGNIFICANCE OF STOLETOV’S WORKS

(On the Sixtieth Anniversary of the Death of A. G. Stoletov)

P. G. Borzyak

Preface

In our Soviet society the name of A. G. Stoletov, an outstanding scientist and progressive public figure of his time, enjoys great popularity. Stoletov the scientist acquired worldwide renown, in particular, through his investigations of the external photoelectric effect. However, more than six decades have passed since that time, and far from everyone has had occasion to read either the works of Stoletov himself or those of his contemporaries. It is true that in our contemporary literature there are accounts[^49] of the principal results of Stoletov’s photoelectric investigations, but they lack a comparison of these results with what was done by other authors of that period, which is also necessary for forming an objective idea of Stoletov’s role. Meanwhile, in the widespread physics literature one encounters highly contradictory judgments on this latter question.

Often in foreign monographic literature, including works translated into Russian and now widely circulated,[^50] there is a glaring disregard for Stoletov’s contribution to this field of science. The opinion is expressed that, in the field of the external photoelectric effect, alongside Hertz’s discovery, supposedly no less important—and perhaps even more important and fundamental—discoveries were made by Hallwachs, which gives grounds even for calling the phenomenon of interest to us the Hallwachs effect. The absence, along with this, even of any mention of Stoletov is meant to lead to the conclusion that his role was allegedly secondary. Conversely, some of our authors[^66] not only assign Stoletov first place, but also ascribe to him

the discovery of the photoelectric effect itself. Engineers, trusting this scholar, disseminate the latter version in technical and popular-science literature. These circumstances must put our students and young scholars at a loss. Clearly, the need has arisen to set forth Stoletov’s principal results against the background of a brief review of works from the initial period in the history of the external photoelectric effect, so that the reader may judge Stoletov’s role not merely on the basis of trust in the assertions of one author or another, but on the basis of a critical comparison of the factual material. At the same time, since the authors who put Hallwachs’s name in first place have in mind his first two works, these should be discussed in especially great detail. The appended list of literature on the external photoelectric effect, compiled, as far as possible, in chronological order for 1887–1890, i.e., beginning with the time of the discovery of the phenomenon and ending with the last year of A. G. Stoletov’s activity in this field, should also help to create a general picture of the state of the question in that period.

All this is all the more appropriate to do because we are on the eve of the sixtieth anniversary of Stoletov’s death.

INTRODUCTION

The phenomenon of the external photoelectric effect was first observed in Paris in 1885 by a pupil of the French academician Mascart, Nodon. In his experiments, a metal plate connected to an electrometer and placed in a metal box was exposed to direct sunlight through a small window in the box. The electrometer thereby indicated that the plate was charged with electricity of positive sign. However, either because it was impossible to understand what was being observed, or because of uncertainty as to the correctness of the experiments leading to so surprising a phenomenon, the author and his teacher did not venture to publish them. From Nodon the Paris Academy of Sciences received a letter in a sealed envelope, registered on June 29, 1885. The contents of this letter, with a description of the above-mentioned experiments, became known only on August 5, 1889, when it was opened, at the author’s request, at a meeting of the Academy during a public communication on the results of his observations of the influence of meteorological factors on the phenomenon[^34]. But by that time about three dozen articles devoted to the study of the external photoelectric effect had already been published. Therefore, either excessive caution, or uncertainty, as well as insufficient promptness and the absence of proper initiative in developing the investigation, cost the author the loss of priority in the discovery of what subsequently proved to be such an important phenomenon. This honor deservedly fell to the well-known German physicist Hertz, whose work[^1] marked the beginning of the broad study of the external photoelectric effect.

However, Hertz himself, having come upon the phenomenon in connection with investigations of electromagnetic oscillations and vibrators, which soon led to the discovery of electromagnetic waves, after the publication of his article\(^1\) never again engaged in further study of the photoelectric effect. This was taken up by other authors, as a result of which the beginning was laid of a new branch of physical science with its remarkable practical applications. Among these authors of the first period in the history of the external photoelectric effect, the greatest credit belongs to our compatriot A. G. Stoletov.

In May 1887 Hertz’s article was sent to press, and in February 1888 Stoletov began his work. In the same year, beginning in April, three of his publications appeared one after another\(^5, 13, 14\). In 1889 Stoletov wrote a large summary article, including new materials, for “ZhRFKhO”\(^33\) and one communication for the Reports of the Paris Academy of Sciences\(^26\). On June 17, 1890, the last work of this cycle was submitted for publication\(^44\). Thus, the entire work lasted about two years, and one cannot but be astonished at how much was accomplished in such a short period by one person occupied chiefly with teaching. What is striking here is not the number and volume of his publications, but the large number of firmly established fundamental results contained in a comparatively small number of his papers.

1. THE INFLUENCE OF LIGHT ON A SPARK DISCHARGE

Hertz\(^1\) discovered that the passage of a spark current in one spark gap facilitates the passage of current in another, neighboring one. Investigating this action, Hertz, by means of careful and numerous experiments, found that the carrier of this action propagates and is reflected according to the laws of the propagation and reflection of light. Its penetration through various media and its refrangibility in different media proved to be the same as for ultraviolet rays. A number of other experiments likewise led to the conclusion specifically of the light action of the spark on another spark gap. At the same time, the unipolarity of this action was established: illumination of the cathode produced a considerable effect, while illumination of the anode produced an insignificant one. Hertz could not rule out the incidence on the cathode of light reflected and scattered by the anode, which is why he admitted that he had not reliably established whether the action of light was connected exclusively or only predominantly with the cathode.

Thus, by means of numerous, purposefully arranged and painstaking experiments, Hertz clarified the basic conditions for the occurrence of the phenomenon—it is caused by the action of ultraviolet light on the cathode (at least predominantly). The conclusion about the action specifically of ultraviolet light was verified and confirmed by replacing the spark causing the phenomenon with other kinds of light sources rich in ultraviolet rays, among which

especially effective was the electric arc. With it all the preceding experiments on studying the propagation, reflection, and refraction of the acting agent and its absorption in various media were repeated, also leading unequivocally to its identification with ultraviolet light. After this the electric arc, as the most convenient source at that time of short-wave light, came into wide use also in the experiments of subsequent authors.

Hertz acknowledged that, from the methodological side, the conditions of his experiment were complicated, and he pointed out the necessity, for further study of the phenomenon, of simplifying them, eliminating first of all the inductors from the experiments.

Wiedemann and Ebert[^2] continued Hertz’s experiments, improving the experimental technique. Hertz judged the action of light by the increase in the distance between the electrodes at which the passage of the spark current was still observed at a given voltage. The authors mentioned, having checked Hertz’s method, went over to a new one, and in original experiments judged the action of light by the change, at a given voltage, in the frequency of spark discharges. Feeding an improved discharger from an electrostatic machine and including a telephone in its circuit, they listened to the sound produced in it. When the cathode was illuminated the sound acquired the character of a musical tone and, moreover, the pitch of the tone reproduced by the telephone increased in comparison with the tones heard in the dark. This testified to an increase in the frequency of the discharges. The same was also confirmed by experiments with a rotating mirror, which gave a time sweep of the processes of passage of the spark discharges.

Having eliminated the incidence on the cathode of reflected and scattered light, the authors refined Hertz’s result, clearly showing that illumination of the anode produces no effect and that the phenomenon arises only when the cathode is illuminated. The authors imagined that the action of light promoted the occurrence of cathode rays, a correct understanding of the nature of which, of course, did not yet exist at that time. In some cases the authors discovered the action not only of ultraviolet, but also of extremely short-wave visible light.

Without dwelling on other results of this work, let us proceed to a more detailed account of Hallwachs’s investigations.

2. THE INFLUENCE OF LIGHT ON A CHARGED CONDUCTOR

Following Hertz’s indications, Hallwachs already in his first work[^3] did in fact substantially simplify the experiment and made it very clear, passing from spark to silent discharges. With the aid of an electroscope he observed the change in the potential of a singly charged metallic disk, placed in front of a grounded metallic screen with an opening for admitting light to the disk. Under the influence of illumination by the light of an electric arc, negatively

charged disk quickly discharged. The description of the experimental setup constitutes the content of § 1 of his article. The headings of the subsequent paragraphs formulate the results obtained. We shall therefore reproduce them.

“§ 2. The action of the electric arc proceeds predominantly from ultraviolet light.”

This was established by repeating the same experiments that had been carried out by Hertz.

“§ 3. The phenomenon is caused by the action of ultraviolet light on the surface of a charged body.”

Here it was proved that the phenomenon does not occur from the action of light, say, on the gas surrounding the conductor, but precisely from its action on the conductor itself. This followed from the fact that light passing in the immediate vicinity along the surface of a flat electrode exerted no discharging action on the negative electrode. As we have seen, the previous authors had already arrived at the same result on the basis of “sounding” the spark gap with a beam of light. The phenomenon of acceleration of discharges occurred as soon as the light touched the surface of the cathode.

“§ 4. Transfer of electricity.”

The experiment described here was a continuation of the first. In the first experiment, the loss of charge from a negatively charged conductor was observed. Here it was shown that the departing charge does not disappear without trace, but passes to the surrounding bodies, which can also be detected with the aid of electroscopes. For the experiment, two metallic plates connected to electroscopes were taken. A negative charge was imparted to the first plate; the second, being insulated, was at earth potential. The plates were set at a certain angle so that their surfaces facing one another could be illuminated from the front. Upon illumination, the leaves of the electroscope connected to the negatively charged plate fell, while the leaves of the second diverged. It was plainly visible that the charge left the first plate and appeared on the second.

This exhausts the factual content of Hallwachs’s first work. Its result was formulated in one of the author’s subsequent articles^25 as follows: “… some time ago I showed that, upon illumination with suitable ultraviolet light of a negatively charged clean metallic plate, the negative electricity is dispersed, following the electrostatic forces of the field.” True, the latter was not directly shown. The question of the paths of motion of charges between electrodes was later the subject of special investigations by Righi^12,15,45,48. We shall return later to the concluding § 5 of Hallwachs’s work, where factual material is no longer contained.

Thus Hallwachs repeated the results of the previous authors, but with a different method. The difference consisted in the fact that

Galvaks observed the “discharging action” of light on a conductor charged once, say, with the aid of an ebonite rod, whereas, for example, Wiedemann and Ebert observed it on a conductor continuously charged by an electrostatic machine. The difference also consisted in Galvaks’ transition from spark discharges to silent ones. This simplified the experiment and made it a very popular, widespread demonstration experiment, but, in essence, it contained nothing fundamentally new in the scientific respect. Although, it must be supposed, the simplicity and persuasiveness of this Galvaks experiment helped attract attention to photoelectric phenomena.

One might think that we are perhaps now not in a position to assess objectively the situation at a time so remote from us. But if we turn to contemporaries, we shall encounter the same understanding of the matter. Stoletov himself[^5][^35] wrote about his repetition of the experiments of Hertz, Wiedemann and Ebert, and Galvaks. The results of the latter, in his understanding, did not stand out in any fundamental respect in comparison with the preceding ones. In one of the completely impartial abstracts[^51] of Galvaks’ article we read: “The influence of light on an electric discharge has recently been investigated by many physicists, so that one may hope that a complete explanation of this interesting phenomenon will soon be given. Immediately after Hertz’s communication, the author (i.e., Galvaks) investigated this phenomenon, simplifying the experiment as far as possible.” And then follows an exposition of the content of the work. Here we do not find even a hint of discovering anything fundamentally new in the paper being reviewed. Finally, the German authors themselves[^22], one of whom—Lenard—was the most eminent physicist who subsequently discovered the photoelectron, wrote in the introductory part of their article: “Hertz’s discovery of the action of ultraviolet light on the magnitude of the discharge spark gap prompted a series of investigations in which this action was studied more closely. It turned out that light acts not on the gas between the electrodes and not on the anode, but only on the cathode (E. Wiedemann and Ebert); that, consequently, it causes a transfer of negative electricity into the air... It was found that under the action of ultraviolet light, negative electricity also, at a lower voltage, is discharged into the air (Galvaks) and passes to the surrounding conductors. Thus from two metals separated by air, one of which is illuminated, one may make a ‘photoelement’ (Stoletov).” Here, too, Galvaks’ investigation is considered in one series begun by Hertz.

Galvaks himself regarded this differently. In the introductory part of his first article[^3] he wrote not about a continuation of Hertz’s investigations and not about an investigation of the phenomenon discovered by Hertz, but that he “investigated a related phenomenon occurring under simpler conditions.” In § 2 of this article it is written about “the connection of our phenomenon” with Hertz’s phenomenon. But perhaps this is the result of tem-

of misunderstanding? Let us turn to his most detailed review of the external photoelectric effect, written as early as 1913 and published in 1916.^52 Having described his first experiment, on p. 249 he states, with some pretension (but, as we shall see, without sufficient grounds): “This phenomenon provided the desired action of a simple kind, which in this new field made it possible, by means of simple measurements, to penetrate into the entire course of the process, and for that reason it served as the basis for a very large number of works (cf. § 1).” And § 1 speaks of the entire body of work on photoelectron emission up to 1914. Thus the author unambiguously ascribes to himself the role of founder of research on the external photoelectric effect. Therefore, on p. 263 we again read of the “Hertz phenomenon” and of the “phenomenon discovered by the author.” In doing so, Hertz’s own modesty is also invoked. In the second footnote on p. 251 we read: “With regard to the comparison of the works of Hertz and of the author mentioned here, reference should be made to Hertz’s work (Investigations on the Propagation of Electric Force, Leipzig, Barth, 1892, p. 280, Nachtrag 11), where he mentions that he did not succeed in finding such conditions under which the phenomenon of spark discharge, so little understood, would be replaced by a simpler action, and continues: ‘Only Mr. Hallwachs succeeded in this, Wied. Ann., 33, p. 301, 1888.’ Mr. Warburg later (Berl. Ber., 1896, p. 229) was the first to apply the name Hallwachs effect, in which he has been followed by many authors up to the present time.” And in justification of this, on pp. 281–283, he devotes an entire paragraph to proof that “Investigations of the Hertz phenomenon show that the action of light on the spark discharge belongs to a lesser degree to the strictly photoelectric field, and rather to the field of the mechanics of the spark discharge. From this it is also clear why attempts to penetrate into the photoelectric field by means of the action of light on spark formation did not yield real results...” All this, of course, cannot be regarded as serious. Hertz’s merit lies precisely in the fact that, amid this complex mechanics of spark formation, he was able to notice and then investigate a new phenomenon. But, in addition, we shall see that Hallwachs’s own work in no way brought us closer to penetrating into the essence of the phenomenon.

In contrast to all this, Stoletov, who made an incomparable contribution to this field, called Hertz “the very investigator to whom belongs also the first discovery of the facts that I have called ‘actino-electric’” (^35, p. 205). (To explain the name “actino-electric,” in his second communication^13 Stoletov made a footnote: “I use this expression to designate electrical phenomena caused by radiations.”)

But perhaps it is not so much the experiments themselves and their results as the interpretation of the latter that gives Hallwachs the right to priority? Let us reproduce in full the concluding § 5 of his work^3:

“It seems to me that, since our phenomenon is conditioned by a process on the surface of the plates and since the action on positive and negative electricity is different, the most probable explanation for the time being is the supposition that perhaps, in some way, a separation of electricities takes place on the surface. The investigations undertaken in this direction, although they do give a result confirming this supposition, are nevertheless still few in number, and the conditions of the experiment have not yet been varied sufficiently to permit final conclusions. I therefore postpone their publication.”

It should be noted that behind the words “separation of electricities” there was no definite concept of the mechanism of the phenomenon connected with the structure of matter. By that time neither electrons nor, still less, the structure of the atom had yet been discovered. It was therefore unclear why the separation of electricities leads to the departure of negative charge from an electrode that is at a negative potential and is illuminated, while no analogous phenomenon occurs with positive electricity. True, in his very first article Hallwachs himself wrote that, with more exact investigations, one might allegedly detect the action of light also on a positively charged metal. Perhaps this error at first gave him grounds to speak of a “separation of electricities.” But in any case, as we see, in connection with the new phenomenon the author expresses himself very irresolutely and not very clearly. Stoletov calls this “indefinitely expressed remarks” (35, footnote 1 on p. 200). But Hallwachs promises to publish additionally the experimental results relating to this question. We shall therefore have, departing from the chronological sequence of exposition, to turn to his next work, in which his second principal experiment is described.

3. CHARGING OF A NEUTRAL CONDUCTOR UNDER THE INFLUENCE OF ILLUMINATION

In Hallwachs’s subsequent investigations,^11 a zinc plate connected to an electrometer was suspended on an insulator inside an iron box and illuminated through a window transmitting ultraviolet light. Before the beginning of the experiment the plate was connected to earth. In order that no negative charge should thereby be induced on it, the material of the box was chosen so that its contact potential with respect to the plate was negative. After the plate was disconnected from earth and then illuminated by the light of a magnesium lamp, its charging to a small positive potential was observed. Thus it was established that under the action of ultraviolet light there occurs not only the loss of negative charge by negatively charged conductors, but also a certain positive charging of uncharged conductors.

…conductors. The transition to such an experiment was a logical continuation of the line of passing from high potentials to low ones; several other investigators, besides Galvacs, arrived independently at the experimental conclusion concerning the action of light also on uncharged conductors. Among them, Righi should be mentioned first of all.

Righi’s first paper[^4] appeared after Galvacs’s first paper. As the author wrote, in seeking an explanation of the phenomena described by Hertz, Wiedemann and Ebert, and Galvacs, he set himself the aim of studying the influence of light on contact-electric phenomena between metals. For this purpose he took two metals: one in the form of a solid disk, the other in the form of a mesh, arranged parallel and close to each other. Connecting the solid disk with one pair of quadrants of an electrometer, and the mesh with another pair of quadrants and to earth, he observed, when this system was illuminated by the light of an electric arc through the mesh, a deflection of the electrometer. Such a system, consisting of a metallic mesh and a disk subjected to illumination, Righi was the first to call a “photoelement.” He also assembled a battery of four such photoelements connected in series, but did not think to draw a current from it, as from an individual element, and confined himself to electrostatic observations. “This kind of photoelectric battery,” wrote the author in the cited paper, “shows the well-known electrostatic phenomena obtained in a battery with an open circuit…” (See also [^18], C. R., 2°). The author believed that ultraviolet illumination equalizes the potentials of neighboring metallic bodies and that, therefore, the deflection of the electrometer is a measure of the contact difference of potentials of the given metals. For experiments carried out under pure conditions, such a conclusion, as we know, is incorrect.

Removing the mesh and leaving only one plate connected to the electrometer, Righi also observed its positive charging under illumination by ultraviolet light. On this occasion he wrote: “Apparently, in such a case the bodies surrounding the plate replace the metallic mesh; therefore, negative electricity probably passes to these bodies and from there to earth.” This was the first experiment in which photoelectric positive charging of a metallic body not negatively charged from outside was observed. In this paper, however, it was still assumed that, under the influence of the surrounding bodies, a negative charge is induced on the plate under investigation and that, ultimately, the light acts on a negatively charged conductor. But then Righi also carried out an experiment very similar to Galvacs’s experiment. A copper plate, likewise connected to an electrometer, was introduced on an insulator into a copper box. For greater certainty that there was no induced negative charge on the plate, it was slightly charged positively. When it was illuminated by ultraviolet light, it became charged to

of a large positive potential. This experiment was carried out independently of Galvaks’s second experiment, although the publication of the latter appeared earlier. In connection with the appearance of Righi’s review article,^18 Galvaks, in identical letters to the editorial offices of two journals,^41 came out in defense of his priority in the discovery of “photoelectric excitation.” In a reply letter Righi^42 wrote: “As Mr. Galvaks has testified, I was the first to establish the fact that an illuminated conductor acquires a positive potential. The correct explanation of this circumstance was first given by him on May 5, 1888, in a communication to the Göttingen Academy, and after that by me in a communication of July 1, 1888, to the Lincean Academy. At that time nothing was known to me about his publication, and that he had anticipated me I first learned from issue 8 of Wied. Ann. for 1888 and the July issue for 1888 of Phil. Mag. I hastened to insert a quotation from his work when reviewing the proofs of my note of July 1...”.

It must be supposed, of course, that the majority even of non-German physicists learned of Galvaks’s second work from the same sources as Righi. Therefore the experiments of Bichat and Blondlot,^16 published at the very beginning of July, without reference to Galvaks, should likewise be regarded as independent. For their experiments the authors took a carefully cleaned metal plate and grid, which, in order to exclude the influence of the contact difference of potentials, they cut from one and the same sheet of brass. When illuminated by ultraviolet light, the plate, the authors wrote, “acquires a positive charge, i.e., loses negative electricity.” Moreover, when a negative voltage of two volts was applied to the grid and forced convection of air (blowing) was used to remove charges from the illuminated plate, even then both its positive charging and a photocurrent of the corresponding direction (previously discovered by Stoletov) were observed. The authors concluded that “all the facts described here can be explained on the assumption that the combined action of light and blowing acts not only on the apparent charge of the surface of the plate, due in its origin to preliminary electrization, but also on the half of the double layer situated in the air, which gives rise to a jump of potential between the air and the metal. It is sufficient to suppose that the metal is positive relative to the air.”

And, finally, Stoletov too arrived at the establishment of the same fact on the basis of a different, original method. Continuing his initial investigations of the influence of contact potentials on photoelectric currents,^5 he carried out experiments^35 according to the scheme shown in Fig. 1. When the metallic disk \(D\) was illuminated for some time, the plates of the capacitor \(K\) were charged to a certain potential depending on the time of illumination. When the plates of the capacitor were connected through the galvanometer \(T\), there was observed

rejection, the magnitude of which served as a measure of the charging of the capacitor. Under sufficiently prolonged illumination, a deflection of the galvanometer was observed corresponding to a potential difference on the plates of the capacitor greater than the contact difference between \(D\) and \(C\). “One has to admit,” wrote Stoletov, “that the continuous disk not only freed itself from negative charge (equalized its potential with that of the grid), but also became charged positively... It remains to conclude that the very action of the rays charges the conductor with positive electricity. The same is also found on armatures of homogeneous ones (e.g., Ag\(^*\) and Ag).”

This charging by rays was not investigated by me in greater detail, although I repeated some experiments of this kind with a quadrant electrometer. It had been the subject of experiments by Hallwachs, published before I had clarified the matter for myself, and then of the experiments of Bichat and Blondlot, as well as Righi” (35, p. 200).

Thus, three more authors, besides Hallwachs, arrived by independent paths at the establishment of one and the same fact. But the legal priority remained with Hallwachs, who had published his result before all the others. Now, however, the question arises as to whose conclusions from the discovered fact provided the more correct guidance for subsequent researchers in this matter. We have cited the conclusion of Bichat and Blondlot. Righi arrived at the erroneous conviction “that the positive charge produced by illumination is limited by the density of electricity acquired by the plate; as soon as this density reaches a certain value, constant for a given metal, the action is suspended.” (Quoted from C. R.\(^{18}\), l.)

Fig. 1.

Whereas, for example, Bichat and Blondlot did not perceive any difference between the phenomena of discharge of negatively electrified bodies and the positive charging of unelectrified ones, we find in Hallwachs an entirely different approach to this question. In the description of his experiments\(^{11}\) we read the following:

“The box was made of rusty iron so that its contact potential relative to the plate would be negative. Only in this case can the increase of potential shown by the electrometer be explained unambiguously. For if the plate were negative relative to the box, then the rise in potential would also occur as a result of the removal by illumination of negative electricity. But if the plate were positive relative to the box enclosing it entirely and for this reason had a positive charge, then the increase of potential could be explained only by the fact that illumination excites positive electricity on the plate.”

Here Galvaks speaks as if of a new phenomenon of excitation of positive electricity, and not due to the departure of negative electricity. It must be said that this does not create a clearer understanding of the phenomenon than does the concluding paragraph of Galvaks’s first article. The introduction by him of a new term to designate the supposedly newly observed phenomenon, “photoelectric excitation,” did nothing to bring clarity to the question. That Galvaks himself did not possess it, we have just seen. This is evident again from the concluding paragraph of his second article:

“Because the investigations described here had to be interrupted, I have allowed myself for the time being to report the principal results; whether the electrical excitation in electric light, caused by illumination, is in direct connection with the loss of electricity by an electrically charged body when it is illuminated, I cannot yet say anything about this, since up to now I have had no possibility of setting up experiments in this direction”[^11].

Thus, in the first two and at the same time fundamental works of Galvaks two definite assertions are expressed: 1) concerning the loss of charge under the influence of illumination by negatively charged bodies, which was not original, and 2) concerning the excitation by light in a metallic body of positive electricity, and not due to the departure of negative electricity, which was unclear and, as we now know, incorrect. True, from what Galvaks said in the second article it remained unclear whether the experiment described in it was the one promised in the first work but did not justify the hopes placed on it, or whether it was a different experiment whose results it is unknown how to reconcile with the first. But it is known that no third work for clarifying the question posed was carried out by Galvaks. He himself later regarded the work cited here as his greatest merit precisely because of the discovery, as we see, of the incomprehensible “photoelectric excitation.”

Stoletov’s opinion concerning the discovered phenomenon of charging was expressed as follows: “I think that here we are dealing not with a particularly new fact—that this actino-electric charging with positive electricity should be explained as the result of the actino-electric discharge of negatively charged bodies; the named French authors also hint at this.” (The reference is to Bichat and Blondlot[^35], p. 200.) Further, immediately here, in the conclusions of Stoletov’s fundamental investigations, we read:

“1. Rays of the voltaic arc, falling on the surface of a negatively charged body, carry away the charge. Depending on whether the charge is replenished and how rapidly, this removal of charge may be accompanied by a noticeable fall of potential or not.”

In this point the results of observations by means of various methods—those of Hertz, Wiedemann and Ebert, Galvaks, and Stoletov—are generalized.

“2. This action of the rays is strictly unipolar; positive charge is not carried away by the rays.

1. In all probability, the apparent charging of neutral bodies by the rays is explained by the same cause.”

Returning to this question in the concluding § 18 of his article[^35], Stoletov wrote: “In the preceding I spoke of the discharging of bodies by rays; but there are experiments proving that rays charge a neutral body. It seems to me that the first of these facts should be recognized as fundamental, while the second should be explained as a consequence of the first... One may develop the thought, briefly and insufficiently expressed by Bichat and Blondlot, and explain to oneself to some extent both the process of discharging negatively electrified metals and the process of charging neutral or positive ones—if only we accept as a fact that rays of a certain kind tend to carry away negative charges.”

These statements, indeed, as is now clear to us, could have provided the correct orientation in the study of the new phenomenon. But they emphasize still more strongly the obvious fact that, if Galvaks’s first work gave no grounds for ascribing to him the discovery of the phenomenon, then the statements concerning the fact observed in the second work can by no means lay claim to any guiding principles in the further development of the given question. On the contrary, they could rather have disoriented subsequent investigators. The latter can also be supported by certain further investigations of Galvaks himself.

Under the influence of the work of Bichat and Blondlot[^16] on the observation of the combined action of light and blowing upon a metallic plate, Galvaks, with some modification, repeated their experiments and, already in 1890, published the article “A Lecture Demonstration of Photoelectric Excitation”[^40]. In this article he wrote: “G. G. Bichat and Blondlot later found a way to intensify the excitation by directing also a stream of air onto the illuminated plate, whereby they obtained an increase of the potential up to 30 volts. This method of intensifying the excitation at the same time advances the view of a connection between photoelectric excitation and photoelectric discharge, reducing it to the notion that discharge should be regarded as a consequence of excitation, according to which the latter (excitation) can be intensified by the electrostatic forces of the negative charges imparted to the plate, just as by a stream of air. This connection made understandable the experiments of Righi, who proved the proportionality between the illumination produced, the potential, and the electrostatic force manifested at the surface.”

But here Righi, in his letter[^42], probably having in mind his own works[^15], [^18], [^21], caught Galvaks in an erroneous rendering of his conclusions: “What I proved was precisely the independence of the attainable potentials from the electrostatic force on the surface.

In reality I found that, for a conductor of definite nature (in air under definite conditions of temperature and pressure, illuminated by acting rays of a given intensity), the charge excited on it by illumination ceases to grow when the electric density on its surface (and, consequently, also the electrostatic force) reaches a certain constant value, whereas the potential to which the conductor is at the same time brought may attain any value, depending on the shape, position, etc., of the conductor.” (Incidentally, this does not agree with Righi’s conclusion in his first paper, where he proved that light equalizes the potentials of conductors.)

Thus Hallwachs came to the conclusion that the primary phenomenon, in his opinion, can be intensified by the action of a field or of a stream of air, and he even recommended a demonstration experiment with the charging of a conductor up to 100 volts.^40 Righi, on the other hand, came to the conclusion that the primary phenomenon consists in imparting to the surface of a conductor a definite density of electric charge. From the point of view of what we now know about the photoeffect, both the one and the other are incorrect. Hallwachs, priding himself on having supposedly found the primary phenomenon, complicated the experiment by applying blowing, which led to an interpretation of the results that did not help penetrate into the nature of the true primary phenomenon, but led away from an understanding of this nature. From this one should rather conclude not the foundational character of Hallwachs’s works, but the conceptual untenability of his conclusions.

4. PHOTOELECTRIC CURRENT. THE PHOTOELEMENT WITH EXTERNAL PHOTOEFFECT

Hallwachs’s works could not be foundational also for the reason that his research method proved unfruitful. It was convenient for demonstrating the phenomenon, but unsuitable for carrying out quantitative measurements and for the further development of research. And indeed, looking back now on the road traversed, we may note that it not only did not provide the means “by means of simple measurements to penetrate into the entire course of the process,” but in the development of the field of photoelectric phenomena it played no noticeable role and lived on chiefly in school experiments. Even data on the maximum energies of photoelectrons were obtained mainly either from volt-ampere characteristics of photoelectron currents, or (considerably more rarely) from magnetic analysis.

The truly broad methodological basis for the development of investigations of the new phenomenon, and of all subsequent progress in this field, was the idea of characterizing the new phenomena by means of the results of measuring photoelectric currents, put forward by Stoletov.^5,35 For his investigations Stoletov chose a system of electrodes consisting of a solid metallic disk and a metallic-

...metal mesh, i.e., exactly the same as was used by Righi. However, the two authors arrived at this quite independently. It must be remembered that Righi reported his first paper^4 on March 4, when it had already gone to press in a little-circulated organ (Rend. Linc.). Stoletov’s paper^5 was reported at the Moscow Society of Naturalists on March 23. On March 25 it was sent for presentation to the Paris Academy of Sciences to Academician Mascart, who communicated it on April 16. This is also the date of its receipt for publication in the “Reports” of that Academy (see^13, footnote 4 on p. 1593). Therefore, at the time of publication of his paper Stoletov could have known nothing about Righi’s work, something the latter also acknowledged^7.

But although the instruments of Righi and Stoletov were identical, the methods of using them were different, as is clear from Fig. 2. Righi, like Hallwachs, contented himself with electrostatic experiments, measuring the potentials and their changes in the bodies under investigation (Fig. 2, a). Stoletov included a galvanometer in the circuit of the disk and mesh of his, as he called it, “mesh capacitor,” and measured the photoelectric current that proved observable (Fig. 2, b). This circumstance turned out to be more important than the author himself could, naturally, have supposed. The advantage of the new method consisted not simply in the fact that, as is still written, the need to work with high voltages disappeared and precise electrical measuring instruments were employed as technical means, making it possible to pass from a qualitative to a quantitative study of the phenomenon. The electrometer is no less, and even more, precise an instrument than the galvanometer. The significance of the new method was more fundamental and consisted in the fact that it became possible to measure directly the quantity of emitted electrons; in the fact that a characteristic was chosen for study which better reflected the essence of the phenomenon and, at the same time, was better suited to quantitative measurement, being also less subject to the influence of extraneous factors. It was precisely this that made it possible to establish the regularities of the phenomenon even before the discovery of its nature, which, as we now know well, could not have been done before the discovery of the photoelectron and of the quantum nature of light. Such experimentally established quantitative regularities usually also serve as the basis for constructing a correct theory that reveals the nature of the phenomenon. This aim, as is known,

Fig. 2.

served also as the first law of Stoletov in the field of photoelectricity, as did his other results, and Stoletov’s phenomenon in the field of gas discharge. Moreover, the measurement scheme according to Fig. 2, b, and not Fig. 2, a, is also the scheme for using photocells at the present time. In fact, in photocells with the external photoeffect, it is always the photocurrent that is used for applications, and the photo-e.m.f. produced by light is never used.

It is interesting to note that Stoletov’s report on the observation of photoelectric currents called forth, at the very next meeting of the Paris Academy, remarks by the well-known French scientist, Academician Edm. Becquerel[^6]. The latter wrote: “I wish to note in this connection that these phenomena seem to me analogous to those which I observed and analyzed in 1853, but by another method; by that time I had shown that heated gases can conduct electric currents generated by couples with very small electromotive force (de couples à très faible force électromotrice) and that these phenomena depend on the nature of the gas, its density, and also on the relative dimensions of the electrodes.” Reality showed the inconsistency of this remark, and Stoletov did not even have to reply to it specially, which is why it remained little known.

In many cases researchers even after Stoletov continued to use the electrometric method, until their own experience proved the impossibility of obtaining quantitative characteristics of the phenomenon without Stoletov’s method. Thus, for example, the well-known German researchers Elster and Geitel, who had long used exclusively electrometric methodology, when in 1892 they faced the necessity of quantitatively comparing the sensitivities of cathodes made of different alkali metals in visible light[^53], were forced to resort to Stoletov’s method. In the following year, when the question arose of the spectral sensitivities of photocathodes, they wrote: “After, now, by changing the experimental scheme, we replaced the electrometric method of measurement used exclusively by us earlier with the galvanometric one[^53], it seemed possible to us to carry out with some accuracy a comparison of the color sensitivity of the alkali metals: sodium, potassium, rubidium”[^54]. From that time on the monopoly of Stoletov’s method came into force.

5. FIRST REGULARITIES. SATURATION PHOTOCURRENT

A distinctive feature of Stoletov’s very first work, in comparison with others, was the clearly expressed striving in it to obtain not only qualitative but also quantitative results. He established that, other conditions being equal, the photocurrent is proportional to the illuminated surface of the cathode. This was the first quantitative characteristic of the new phenomenon. The result obtained indi-

He showed that each element of the cathode surface participates in the creation of the photoelectric current independently of the others, including neighboring regions. In the absence of an understanding not only of the nature of the phenomenon, but also of the very nature of electricity, the establishment of this fact was very important.

In his very first work Stoletov established a quantitative relation between the photocurrent, the voltage applied between the electrodes, and the distance between them, and studied the features of the volt-ampere characteristic of the photocurrent in air. The development of these investigations with the use of more precise methods both for the case of atmospheric pressure ^26 and at various rarefactions ^14,44 led Stoletov to very important results with consequences of great theoretical and practical significance. Stoletov experimentally established ^14 that the dependence of the magnitude of the current on the gas pressure in the photoelement is a function with a maximum. The pressure corresponding to the maximum current he called critical. Quantitative measurements at different distances between the electrodes and in different fields led him to establish a simple dependence: the ratio of the field strength to the critical pressure is a constant quantity ^35,44. This phenomenon, subsequently called the Stoletov phenomenon, served as one of the experimental foundations for the substantiation of theories of non-self-sustained gas discharge. But with respect to photoelements, what is essential for us is that in these investigations, at low pressures, the possibility was established of increasing the photocurrent severalfold (by 4–6 times) by changing the gas pressure. Thus the foundations were laid—as Hallwachs also testifies (see ^52, p. 435)—for gas-discharge amplification of photoelectric currents in photoelements, which subsequently led to the creation of gas-filled photoelements.

These same experiments first led to the establishment of the fact that even at the greatest attainable rarefaction of air the photocurrent does not fall to zero together with the rarefaction ^14, and later ^44 also to the important discovery of the saturation current. Of course, at that time, as is now entirely clear to us, it was impossible to understand the meaning of the existence of a saturation current in photoelements, but Stoletov appreciated the significance of this fact for the future clarification of the physics of phenomena in photoelements: “Law (4)

\[ \left(\frac{P_m l}{E}=\text{const.}\right), \]

which determines the critical pressure, apparently proves that air takes a direct part in the actino-electric convection; it is difficult to imagine that such a simple relation could be realized if the matter stood otherwise. On the other hand, the existence of a definite and final limit, toward which the current tends as \(P\) tends to zero, suggests the assumption that there are other causes that contribute to this convection. ...” On this occasion Hallwachs wrote: “Stoletov correctly realized that here the matter is ...”

goes on about two different causes: the formation of a favoring pressure forces one to draw the conclusion of a gaseous phenomenon, whereas the existence, at high rarefactions, of a limiting value independent of the potential forces one to draw the conclusion of a primary action due to another cause—let us say, “photoelectric”\({}^{52}\), p. 273).

It is important to note here that, externally, very similar to Stoletov’s photoelectric experiments in rarefied gases were those of Righi\({}^{23,45}\). And it was precisely here that the inadequacy of the electrometric method, which Righi persistently used all the time, showed itself especially vividly. Whereas Stoletov obtained such outstanding results in investigating photocurrents, Righi, in measuring electrode potentials and striving by means of these measurements to obtain information about the electric densities on the electrodes and to connect with them the “coefficient of photoelectric scattering,” arrived at the same time\({}^{45}\) at the incorrect conclusion that “the photo-scattering of negative electricity and the formation of a positive charge on the illuminated conductor are the essence of phenomena of different kinds.” (Quoted from Beibl. Ann. d. Phys.).

Studying the dependence of the photocurrent on the voltage, Stoletov came to the conclusion that in doing so it was also necessary to take into account the contact potential difference between the solid disk and the mesh. And this led him to the conclusion that it was possible to obtain a photocurrent even without an auxiliary battery, solely at the expense of the contact potential difference. In his very first paper (\({}^{9}\), p. 1151) we read:

“Indeed, with a zinc mesh and a copper silver-plated solid disk I obtained a current. Here we are dealing with an element in which the illuminated air replaces the liquid, and which operates as long as the illumination continues, with the current being maintained at the expense of the energy of the radiation.” Thus, if Righi spoke of a photoelement as an analogue of a galvanic cell with an open external circuit, then Stoletov, by an independent route, arrived at a photoelement that gives a current like a galvanic cell. In this case, the energy of light is used to obtain the current. Consequently, the first real photoelement using the external photoeffect was obtained by Stoletov. Therefore, for example, Lenard and Wolf\({}^{22}\), in the quotation given above, connect the obtaining of the photoelement not with Righi’s name, but precisely with the name of Stoletov, who, as we have seen, was the first to obtain a practical photoelement operating with an external voltage source. Stoletov’s circuit, shown in Fig. 2, б, is now recognized by the present authors as the “modern circuit for measuring photocurrents” (see, for example,\({}^{35}\), p. 5).

Studying the current–voltage characteristic of the photocurrent in the region of small voltages, Stoletov established proportionality between voltage and current. This enabled him, in a quite original manner, to determine the contact potential difference

between metallic electrodes^5,35. For this it was necessary to measure the photocurrents \(i_1\) and \(i_2\) for two small applied voltages \(V_1\) and \(V_2\). Since the effective potential difference is \(V_1+K\) and \(V_2+K\), then \(\dfrac{V_2+K}{V_1+K}=\dfrac{i_2}{i_1}\), whence \(K=\dfrac{i_2V_1-i_1V_2}{i_2-i_1}\). Once again it should be emphasized that if the assertion accepted in Righi—that in his measurements the deflection of the electrometer served as a measure of the contact potential difference between the electrodes—was incorrect, then Stoletov’s method, based on studying the form of the regularity \(i=f(V)\), is in principle correct, and for clean vacuum conditions as well. It is precisely by such a method, in principle, that in photoelectric studies the contact potential difference between electrodes is determined in modern work too.

6. THE BEGINNING OF THE PRACTICAL APPLICATION OF PHOTOELEMENTS AND THE DISCOVERY OF THE FIRST LAW OF THE PHOTOEFFECT

In noting the conclusions obtained by Stoletov, we must remember that he obtained the quantitative data necessary for them using such an inconstant light source as an electric arc. Since experimenters of that time had no other, more suitable source of ultraviolet light at their disposal, Stoletov had to seek methods of combating its inconstancy. From his observations he came to the conclusion that “there is hardly any other way of keeping such vigilant watch over the constancy of electric light (or, more accurately, over the intensity of a known category of radiation) as these actino-electric observations” (^35, pp. 172–173). And this led him to the idea, in his experiments with a specially constructed evacuated photoelement, of using an auxiliary control photoelement to obtain comparable data. “Since the intensity of the voltaic arc varied from time to time, in front of this same lamp a control condenser (a disk and a mesh in air) was installed. The battery and the galvanometer were connected alternately to the new instrument and to the control condenser, and the observations were related to the readings of the latter” (^14, p. 92). This was the first practical application of a photoelement.

But with such a method of control, errors connected with rapid changes in the operating regime of the electric arc are possible. Therefore, in experiments to refine the dependence of the photocurrent on the distance between the electrodes and on the voltage, the method was improved and this source of error was eliminated^26. The investigated and the control “condensers” were in one and the same light cone. To make this possible, the control condenser—the photoelement, placed closer to the arc—was made in the form of a cross and allowed part of the light to pass to the instrument under study. To both

two galvanometers of the same period were connected to the photoelements, whose deflections, when the light was uncovered, were read simultaneously by two observers. This improvement in turn led Stoletov to a most important discovery. In his communication26 he wrote:

“Under these conditions both deflections remain strictly proportional when the electric arc is varied without changing the other conditions...

This proportionality of the actions in two different instruments is quite remarkable in itself: it is evident that, under otherwise equal conditions, the actino-electric current is proportional to the intensity of the active rays.”

This was the first formulation of what later came to be called the first law of the photoelectric effect.

Thus, Stoletov not only gave the circuit of the photoelement, but he himself was the first to point out the possibility of its practical use—for application in photometric purposes; he was the first to use it for photoelectric control; he himself discovered the first law of the external photoelectric effect, which is the scientific basis of photoelectric photometry. Therefore Stoletov should rightly be considered the founder of photoelectric photometry and photoelectric control.

The establishment of the first law of the photoelectric effect was of great fundamental significance. After all, this was done still in the period of the dominance of the wave theory of light, when it was unknown why the photoelectric effect is caused only by light of sufficiently short wavelengths. This ambiguity remained even after the discovery of photoelectrons, when it could be assumed that their removal is connected with the amplitude of the electric-field vector of light, as was already known, electromagnetic waves. Stoletov’s law showed precisely that the phenomenon is not determined by this amplitude itself, since the photocurrent is proportional not to it but to its square. On the other hand, if, from a more general point of view, one were to suppose that the removal of an electron (or simply of an electric charge, when the electron was not yet known) requires a definite energy, i.e., from the point of view of the wave theory, a definite intensity of light, then one would have expected the existence of an intensity threshold below which the photoelectric effect would not be observed.

Thus, in order to penetrate into the mechanism of the phenomenon it was extremely important to extend the investigations of the validity of Stoletov’s law to as large a range of illuminations as possible, which was indeed done, as is known, later in the works of subsequent authors. Already from this one example it is clear to what extent Stoletov’s investigations, unlike those of Hallwachs, formed the basis of the logical development of subsequent investigations and truly led along the path toward revealing the mechanism of the phenomenon.

7. ON THE CONNECTION BETWEEN OPTICAL ABSORPTION AND THE PHOTOEFFECT

Stoletov’s conclusion concerning the character of the interaction of light rays with the cathode, which is the immediate cause of the external photoeffect, had a profound physical meaning. The very first experiments, as we have seen, led to the conclusion that the phenomenon discovered by Hertz consists in the action of light on the cathode; but precisely what the character of the interaction of light with the cathode was, what more specific process lay at the basis of the phenomenon, was unclear. Moreover, some investigators[^19] tried to connect the phenomenon not with the electrodes themselves, but with the surrounding or adsorbed gases. Here, one may say, Stoletov’s physical intuition also played no small role.

Above we have already cited a quotation from Stoletov’s first work, in which it was said that the current was maintained by the light energy incident on the photocathode. But one cannot speak of such a transformation of light energy into electrical energy without assuming the absorption of light. “That in all these phenomena an absorption of rays must take place seemed obvious to me even before I was able to prove it directly. It was precisely in this connection that I had to say in my preceding communication that the current of my actino-electric element zinc—air—silver ‘is maintained at the expense of the energy of radiation’,” wrote Stoletov in his second communication.[^13]

It should be noted that Bichat and Blondlot,[^9] who repeated Stoletov’s experiments with the replacement of the metallic cathode by an aqueous one, also arrived at the conclusion that there is a connection between the absorption of light by the cathode and the photoeffect. In doing so they convinced themselves of the absence of any measurable effect. They saw the reason for this in the experimentally demonstrated transparency of water to the rays that produce the action on a metallic cathode. To test his considerations, which had also found support in the work of Bichat and Blondlot, Stoletov extended his investigations to other liquids and solutions.[^13] In doing so he took a further important step and concluded that in the photoeffect the role is played not simply by the absorption of light by the entire thickness of the cathode, but by the “opacity of the thin surface layer, which is the site of the origin of electricity.” Dwelling on this question in his major article,[^35] Stoletov wrote: “From the very beginning of my investigations I suspected that its sensitivity to actino-electric action stands in direct connection with the absorption of rays by one or another plate…

Rays which illuminate the air layer without touching the surface of a (negatively) charged body do not produce an action (Hallwachs); the rays must fall upon it. Moreover, the rays must be absorbed by the negatively charged surface. Obviously, it is important

In this case, absorption in the thinnest upper layer of the electrode, in the layer where, so to speak, the electric charge resides. A substance that is not completely transparent to actinic rays in the form of a sufficiently thick layer may appear absolutely transparent from the standpoint of such “surface absorption” (pp. 166–167). Translated into modern language, this is a question of the absorption of light in the layer from which the photoelectrons emerge. Later this brought to the fore a problem that remains topical even today: that of determining the thickness of such layers. It is interesting to note that gradually the origin of the idea began to be smoothed over in memory, and its discovery to be associated with the names of much later authors. The works in question are those of Ives and his collaborators, dating already from the thirties of our century. Thus, for example, one may read on this matter: “According to Ives’s idea, the photocurrent must be proportional not to the amount of light absorbed throughout the entire thickness of the photocathode, but to the density of the light energy in the region of that layer of the metal from which the photoelectrons emerge” (56, pp. 198–199). In an even more recent book65, on p. 38, it is written that the necessity of correctly estimating the intensity of light in the emitting layer supposedly follows from Ives’s work. It would seem that long before these works such an idea should have become trivial, if only on the basis of the firmly established I and II laws of the photoeffect. And so it in fact was. It was precisely for this reason that Gudden and Pohl63 said that the selective photoeffect is a means of discovering, by electrical means, the absorption spectrum of adsorbed atoms. Consequently, Stoletov’s idea of the connection between the photoeffect and the absorption of light in the emitting layer was already accepted in 1925 as self-evident. That this is not the only case, we shall cite, by way of example, another quotation from Bazilevich’s article64 of the same time: “Assuming: 1) that absorption of radiant energy is a necessary condition for the emission of photoelectrons (A. Stoletov, Actino-electric phenomena, St. Petersburg, 1889; Journ. de Phys. 9, p. 468, 1890; Righi, Beibl. zu d. Ann. d. Phys. 12, 1888) et al.); 2) that photoelectrons are emitted also from internal layers (Rubens u. Ladenburg, Verh. d. D. Phys. Ges. 9, 749; 1907) and 3) that the photoelectric effect is proportional to the amount of absorbed energy (A. Stoletov, loc. cit.; Lenard, Wied. Ber., 108 (IIa), 1649, 1899; Ann. d. Phys. (4) 1, 359, 1910; Richtmeyer, Phys. Rev. 29*, pp. 71, 404, 1909), we shall find....” In reality, the merit of Ives’s school was not the advancement of this old, Stoletovian idea, but the devising of ways of confirming it in the case of the selective photoeffect.

In order to test his idea of the connection between the photoeffect and the absorption of light, Stoletov carried out, as has already been mentioned, numerous

) In the abstracts of Righi’s works in Beibl. Ann. d. Phys.* for 1888 we find no statements on this question.

experiments with various substances. Especially interesting in this respect are the experiments both with aqueous solutions of various dyes and with dried films of them. “But liquids, densely colored, optically opaque even in thin layers, usually proved to be opaque also in the actino-electric respect and, correspondingly, showed considerable sensitivity.” This sensitivity of some dyes considerably exceeded the sensitivity of all metallic electrodes then known. In addition to the conclusion about the connection between photoelectron emission and the absorption of light, Stoletov also drew another essential conclusion: “In my first experiments I was satisfied with the result that there were found liquids which behave like metals, that the liquid state in itself is not an obstacle to actino-electric sensitivity” (^{35}, pp. 168–169). Indeed, this is so. But in subsequent investigations it became clear that liquids are not photoactive at \(\lambda > 1750\) Å. The sensitivity of aqueous solutions of organic dyes at large wavelengths proved to be connected with the formation on the surface of extremely thin solid films^{57}. It is not uninteresting to recall that the same idea of a connection between optical absorption and cathode sensitivity formed the basis of experiments on the sensitization of photocathodes of alkali metals with the same dyes more than 40 years later^{58}.

Following Stoletov, experiments in this direction were continued by Hallwachs^{25}, who came to the conclusion that a really significant photoelectric action is connected with strong optical absorption, but that the converse conclusion is not always justified. Nevertheless, he agrees that Stoletov’s proposition may be accepted as a working hypothesis in further investigations.

Having perceived that the essence of the photoelectric process is based on the absorption of light, Stoletov raised a major physical problem, which could not be solved so quickly. This problem of the connection between optical absorption and photoelectron emission, i.e. the character of the interaction of light with matter in connection with the photoeffect, did indeed become one of the cardinal ones in all subsequent times. It is enough to recall at least the importance which it acquired in the period of investigation of the selective photoeffect. It remains topical even in our own day (recall, for example, the exciton photoeffect).

8. THE PHENOMENON OF FATIGUE. INERTIA

Carrying out comparative studies of the sensitivities of various bodies, in particular metals, and of the influence on this sensitivity of the state of the surface (^{35}, p. 170), Stoletov drew attention to a phenomenon known to us, but not always and not even now clear, to which he gave the name “fatigue” (see also^{26}, p. 170). And other

researchers, in particular also Galvaks, noted the influence on the sensitivity of metal plates of the state of the illuminated surface and of changes in this state over time. But it was precisely Stoletov’s method of observing the photocurrent that made it possible not only to detect clearly, but also to follow the course of the fatigue process during the illumination of the electrode itself. Fatigue was especially strongly expressed in freshly cleaned metals and gradually decreased as the surface aged. From this Stoletov drew a practical lesson for carrying out quantitative experiments: “Therefore, in experiments requiring constancy of the effect over a certain time, I prefer not to use a disk that has only just been cleaned, but to clean it several hours beforehand, or still better on the eve of the experiment.” It is interesting that in our time, when producing nonfatiguing antimony–cesium photocells, Khlebnikov[^59] used a similar method: namely, under strong light he brought the photocells to limiting fatigue, after which they remained stable. A similar procedure is recommended when using photomultipliers ([^65], p. 165).

Finally, Stoletov raised the question—very important both for the physical characterization of the phenomenon and for practice—of the inertia of photoelectric currents, and gave the first solution to it. Already in his first paper[^5] he noted that all changes in the ray illuminating the cathode are instantaneously reflected in the photocurrent. Borgman[^20] decided, with the aid of a telephone connected into the circuit of a photocell, to test to what extent the photocurrent is in fact without inertia. If the phenomenon is inertia-free, the author reasoned, then, illuminating the cathode with intermittent light, I shall hear in the telephone a sound of the corresponding pitch. In the experiment he did not hear the expected tone, and only heard clicks in the telephone when the circuit of the photocell was closed and opened. From this he concluded that the photoelectric current is not modulated in accordance with the modulating light, and inferred a very great inertia of the phenomenon. Stoletov, having analyzed the conditions of the experiment, concluded that the cause of such a result could have been the insufficient sensitivity of the alternating-current indicator—the telephone. The clicks on closing and opening the circuit could have been associated with considerably stronger capacitive currents. Therefore Stoletov developed a very ingenious method, based on the idea of a phase shift between the optical modulation and the times of selection of the photocurrent measured by a sensitive galvanometer. This principle underlies also numerous modern installations for measuring the inertia of photoelectric phenomena, which became especially widespread and developed in connection with the study of elementary processes in photoelectric phenomena in semiconductors. From careful experiments, taking account of experimental errors, Stoletov obtained results from which it followed that the lag of the current relative to the illumination does not exceed 0.001 second. This remarkable

...a result new for that time, proving the practically complete inertialessness of photocurrents, was subsequently refined by other authors60, 61, especially on the basis of already new experimental techniques62, 67.

9. OTHER INVESTIGATIONS OF THIS PERIOD

Without dwelling on the remaining, in our opinion less fundamental, results of Stoletov’s investigations, let us see what else of note was done in the field of the external photoelectric effect during this short period of time. It must be said that not all of the works not yet cited in the list given at the end deserve attention. The results of some of them were subsequently not confirmed8, 19, 22. In some10, 37, 38 there were clearly methodological errors, and completely distorted results were obtained (the action of light also on positively charged bodies, and sometimes even greater than on negatively charged ones).

Hallwachs counted among fundamental results52, pp. 258, 284, his establishment of the presence of an external photoelectric effect in semitransparent silver films when they were illuminated from the rear—from the side of the transparent substrate31. However, if one takes into account the previously established connection of the photoelectric effect with optical absorption in the near-surface layer, then this experiment is only an illustration of that proposition.

In attempts to diversify the experimental conditions, even in this period a beginning was made in the use of polarized light15 and of a magnetic field47, 48, but as yet without substantial results.

The authors of the period under consideration gradually accumulated data on the list of those materials that are photoelectrically sensitive. The first observations were made with metals. Bichat and Blondlot, as well as Stoletov, initiated investigations of liquids. Righi12, 18, 23 showed that the external photoelectric effect is also observed for dielectrics (ebonite, sulfur). In the last work he used, for detecting charges on the surface of the irradiated insulator, an ingenious method of chladenography.

Attempts were made to establish a connection between the photoelectric activity of metals and their contact potentials (their position in the Volta series). At the present time it is clear that reliable data in this respect could be obtained only under good vacuum conditions, which at that time were completely unattainable. It is therefore not surprising that, for example, Stoletov could not establish any systematic connection between these properties35. Righi, carrying out investigations by means of an electrometric method under more complicated conditions, obtained contradictory results in different works. Thus, at first18 he came to the conclusion that the intensity of the photoelectric discharge of negatively charged bodies decreases along the Volta series from electropositive to electronegative...

metals, while the change in the degree of photoelectric charging of neutral bodies follows the Volta series in the opposite direction. Probably already under the influence of this result, and also under the influence of Hallwachs, who in his first works pointed out the existence of different phenomena—photoelectric discharge and photoelectric excitation—Righi began to form the conviction he later expressed,^45 that these were supposedly “phenomena of different kinds.” However, later^23 he came to the conclusion that the series of photoelectric activities of metals does not correspond to the Volta series of voltages, and still later^24 established, in general, a consistent behavior of metals with respect to photoelectric discharge and charging.

Elster and Geitel paid special attention to the circumstance that the electropositive metals possess the greatest photoelectric sensitivity. Thus, for the first authors, zinc was an especially popular object of investigation, owing to its high sensitivity at that time. Magnesium and aluminum, in Elster and Geitel’s work,^29 proved to be still more sensitive. This increased photosensitivity to undecomposed light turned out to be connected with a broader spectral interval of light actinic with respect to the given metals. Thus, the authors mentioned showed that zinc, magnesium, and aluminum are sensitive not only to the far ultraviolet rays, in which the light of artificial sources is rich, but also to the scattered sunlight of the blue sky. Similar results for zinc had earlier been obtained by Goor,^19 but Elster and Geitel were somewhat surprised by the circumstance that he found sensitivity to daylight even for copper and brass, something they could not in any way confirm. In the above-mentioned communication by Nodon^34 it is not indicated with which particular metal he made his observations. Stoletov^35 and Righi^15, ^32, ^4 emphasized that they could not detect any traces of the action of sunlight, and explained this by the fact that absorption in the atmosphere cuts off the actinic rays of the solar spectrum. However, in Stoletov’s work it is not indicated with which particular metal he performed the experiment in sunlight. It was probably silver.

The results obtained by Elster and Geitel with zinc, aluminum, and magnesium led them to the idea of subjecting the most electropositive metals—the alkali metals—to tests. In this way they came to experiments with amalgamated sodium,^30, ^46 and subsequently, already after the period under consideration, to the creation of photocathodes from alkali metals, which was an extraordinarily important stage in the evolution of photocathodes.

Much attention was devoted to elucidating the mechanism by which negative charge is transferred from the illuminated body. Is the light beam itself a special kind of conductor through which the charge is removed? To answer this question, so curious for us, Bichat^17 carried out an experiment with a hollow metallic cylinder. Illu-

emitting actinic light onto the outer surface of the cylinder, he observed a considerable acceleration of the loss of the negative charge imparted to the cylinder. Illumination through a small aperture of the inner part of the cylinder did not produce any noticeable effect. “If,” wrote the author, “the beam of light were a conductor, it would cause discharge in the one case and in the other to the same extent.” The negative result of Bichat’s experiment followed already from the proportionality, established by Stoletov, between the magnitude of the photocurrent and the area of the illuminated surface of the cathode.

The independently conducted experiments of Righi[^12][^18] and of Bichat[^17] with light metallic vanes on vertical suspensions (“an electric Ségner wheel”) showed that, when a negatively charged vane wheel is illuminated, the suspended system is set in motion, which can be registered with the aid of a mirror. From this both authors drew the conclusion that electrical convection is brought about by air particles. For Stoletov, such a conclusion was a natural consequence of the results of his investigation of photocurrents at different pressures. Righi, however, evidently attaching special importance to this question, returned to it again and again.[^12][^18][^21][^23][^24][^45][^48] In this connection it is very interesting to quote from § 18 of Stoletov’s major article,[^35] where he wrote: “But however that may be, the idea of convection of charges in one way or another—i.e., either by particles of the gas or by particles of dust from the cathode—is not yet a complete explanation of the whole phenomenon, as some apparently tend to think. This idea explains to us how the removal of charge through the gas continues; it explains why the phenomenon can be observed only in a gaseous medium; but we still do not fully understand why and how the process begins. Why do certain particles separate from the surface of the electrode, why is the action unipolar, why is it stimulated only by rays of a known category and is closely connected with the absorption of these rays by the surface of the cathode? These points constitute the chief nerve of the riddle.”

From the quotation given it is clear how distinctly Stoletov perceived the fundamental problems in this field, whereas other investigators often saw the main thing in what was incidental. A vivid example of this is a series of Righi’s works on the study of the paths of motion of charged particles carrying out electrical convection,[^12][^18][^23][^24][^45] or on the study of the electric density on the electrodes under investigation, supposedly determining the effect.[^15][^18][^21][^24] If we turn to the bibliography, we shall see what energetic activity in the field of the photoeffect Righi developed during this period. He often devised ingenious and subtle experiments for that time, such as, for example, the elegant experiment using the method of klydonography to determine the velocities, in an electric field, of charged air particles,[^18] and an interesting experiment to demonstrate the motion of charged particles along the lines of force of the electric field.[^23] But

although all this may, to some extent, have prepared the future results of Elster and Geitel and Lenard, in itself it did not lead to any classical results relating to the problem of the photoelectric effect—results so fundamental and so numerous that they could in some measure be compared with the results of Stoletov.

10. CONCLUSIONS

Summing up the investigations during 1887–1899, Galvaks52 wrote: “From a review of the work of the first period it follows that almost all the principal advances were achieved in the first three years after the discovery of this field,” and then he systematized these most important results as follows (p. 284):

1. General results. 2. Data concerning electrical conductivity in gases. 3. Data concerning photoelectric excitation. 4. Subsequent data (i.e., data after 1889).

Group of questions 4 will not concern us in the present case. If one takes into account the author’s bias, understandable from what has preceded, in matters of group 3, and the obvious dominant importance, even in Galvaks’s own view, of Stoletov’s results in group of questions 2, it is of interest to cite in full Galvaks’s formulation of the general results:

1. General results

“The action of light takes place on bodies, not on the surrounding gas. A necessary condition is the absorption of light. The quantities of electricity discharged are proportional to the intensity of the light. Photoelectricity is produced by light arriving also from behind. (This refers to a semitransparent metallic film.) There is no loss of time in the appearance of the phenomenon, nor any aftereffect. A very large number of different bodies prove to be photoelectrically active.”

Of the six points listed here, the clarification of the second, third, and fifth is due to Stoletov. If we take into account that in the appearance of the sixth he made his contribution alongside other authors, and that the fourth point is a consequence of the second, then, despite the fact that the principled and enormously important practical significance of Stoletov’s method is not noted here, his preeminent role is evident, to say nothing of the very discovery of the phenomenon.

Thus we arrive at the following résumé. After the discovery of the external photoelectric effect by Hertz, Wiedemann and Ebert clearly showed that the phenomenon is characteristic only of the cathode and is not observed at the anode. Galvaks carried out experiments to detect the photoelectric effect on negatively charged bodies in a very simple, attractive

...form, and was also the first to show that light leads not only to the discharge of negatively charged bodies, but also to a certain small positive charging of neutral ones, although his conclusions from the facts confused the question. However, the electrometric method of investigation proposed by Hallwachs and Righi and adopted by other investigators (the observation of potentials and of their changes) proved unfruitful and incapable of serving as a basis for further progress in this field. Such a basis was provided by Stoletov’s method of observing photoelectric currents. Thanks to it, Stoletov became the founder of quantitative investigations of the phenomena, and his “grid condenser” became the prototype of the modern photocell, although the very name “photocell” was first applied by Righi. Stoletov himself was the first to apply the method of photoelectric control of the intensity of light, thanks to which he obtained results that led to the establishment of the first law of the photoelectric effect, which in turn became the theoretical basis of photoelectric photometry. At the same time, Stoletov devised the first method of combating the phenomenon he called the “fatigue” of photocathodes. He posed the problem of the inertia of the external photoelectric effect and gave its first solution with the accuracy permitted by the technical means of the time (down to 0.001 sec). He also raised the fundamental problem, still relevant today in the field of the external photoelectric effect, of the connection between the photoelectric effect and the absorption of light in the emitting layer. Studying the current-voltage characteristics of photocurrents, he laid the foundations of the photoelectric method for determining the contact potential difference, and, extending these investigations to rarefied gases, discovered the existence of saturation current and also laid the foundations of gas-discharge amplification of photocurrents—not to mention the outstanding results that were obtained for the problem of the gas discharge itself and are known as the “Stoletov phenomenon.”

A definite contribution to this period was made by the work of Bichat and Blondlot, Righi, Elster and Geitel. It was very important to clarify the circumstance that high photosensitivity is associated with the electropositive properties of the metal. In connection with Righi’s work on the photoelectric effect, results were also obtained which, though not directly related to the photoelectric effect, were of substantial significance for other areas of electronics.

However, not one of the authors of that period can compete with Stoletov in obtaining such a number of outstanding results of both fundamental and practical significance in the field of the external photoelectric effect—results which in themselves constituted a sufficiently broad basis for the further development of research. We must acknowledge that, if the discovery of the phenomenon belongs to Hertz, then in laying the foundations for the study and application of the external photoelectric effect the first place, unquestionably, belongs to Stoletov.

LITERATURE

I. Literature on the external photoelectric effect for 1887–1900

1. H. Hertz, Ueber einen Einfluss des ultravioletten Lichtes auf die elektrische Entladung, Wied. Ann. 31, 983 (1887).

2. E. Wiedemann und H. Ebert, Ueber den Einfluss des Lichtes auf die elektrischen Entladungen, Wied. Ann. 33, 241 (1888). (See also the brief account in Phil. Mag. (5) 25, 162 (1888), borrowed from Sitzber. d. phys.-med. Inst. d. Erlangen.)

3. W. Hallwachs, Ueber den Einfluss des Lichtes auf elektrisch geladene Körper, Wied. Ann. 33, 301 (1888). (Abstract in ЖРФХО 20, phys. sec., div. 2, issue 5, 41 (1888).)

4. A. Righi, Di alcuni nuovi fenomeni elettrici provocati delle radiazioni, I Nota, Rend. Linc. 4, 1 Sem., 185 (1888), and also: Nuovo Cim. 23, 61 (1888); Phil. Mag. 25, 314 (1888); Journ. de Phys. 7, 153 (1888). (Abstract in Beibl. Ann. d. Phys. 12, 286 (1888).)

5. A. Stoletow, Sur une sorte de courants électriques provoqués par les rayons ultraviolets, C. R. 106, 1149 (1888). (The communication was reprinted in Phil. Mag. 26, 317 (1888); abstract in Beibl. Ann. d. Phys. 12, 605 (1888).)

6. Observations de M. Edm. Becquerel à propos d’une Note de M. A. Stoletow, présentée à l’Académie dans sa dernière séance; C. R. 106, 1213 (1888).

7. A. Righi, Sur les phénomènes électriques produits par les rayons ultraviolets, C. R. 106, 1349 (1888).

8. I. I. Borgman, On the influence of light on electrical discharge, ЖРФХО 20, phys. sec., div. 1, issue 4, 111 (summary of a report); Phil. Mag. 26, 272 (1888).

9. E. Bichat et R. Blondlot, Action des radiations ultraviolettes sur le passage de l’électricité à faible tension au travers de l’air, C. R. 106, 1349 (1888).

10. F. Narr, Ueber die Wirkung des Lichtes auf statische Ladungen, Wied. Ann. 34, 712 (1888). (Abstract in ЖРФХО 20, phys. sec., div. 2, 64 (1888).)

11. W. Hallwachs, Ueber die Elektrisierung von Metallplatten durch Bestrahlung mit elektrischem Licht, Wied. Ann. 34, 731 (1888).

12. A. Righi, Di alcuni nuovi fenomeni elettrici, provocati delle radiazioni, Rend. Linc. 4, 1 Sem., II Nota, 498 (1888); III Nota, 578 (1888); IV Nota, 691 (1888). (Abstract in Beibl. Ann. d. Phys. 12, 721 (1888).)

13. A. Stoletow, Sur les courants actino-électriques au travers de l’air, C. R. 106, 1593 (1888). (Abstract in Beibl. Ann. d. Phys. 12, 723 (1888).)

14. A. Stoletow, Suite des recherches actino-électriques, C. R. 107, 91 (1888); Collected Works, vol. I, Gostekhizdat, M.–L., 1939, p. 267; Selected Works, Gostekhizdat, M.–L., 1950, p. 246. (Abstract in Beibl. Ann. d. Phys. 12, 723 (1888).)

15. A. Righi, Di alcuni nuovi fenomeni elettrici, provocati delle radiazioni, V Nota, Rend. Linc. 4, 2 Sem., 16 (1888); VI Nota, Rend. Linc. 4, 2 Sem., 66 (1888). (Abstract in Beibl. Ann. d. Phys. 13, 198 (1889).)

INITIAL PERIOD OF THE HISTORY OF THE EXTERNAL PHOTOEFFECT

1. E. Bichat et R. Blondlot, Action combinée de l’insufflation et de l’illumination sur les couches électriques qui revêtent les corps conducteurs, C. R. 107, 29 (1888).

2. E. Bichat, Sur les phénomènes actino-électriques, C. R. 107, 557 (1888).

3. A. Righi, Sui fenomeni elettrici provocati dalle radiazioni, I Memoria, Mem. di Bologna 9, 369 (1888). (Summary article on 12 and 15, brief summaries in C. R. 107, 559 (1888).)

4. M. Ueber, den Einfluss des ultravioletten Lichtes auf negativ elektrisch geladene Konduktoren, Wien. Ber., Abt. II 97, 719 (1888). (Abstract in ЖРФХО 21, phys. sect., div. 2, 11 (1889). See also: Rep. d. Phys. 25, 91 (1889); Zentralbl. f. Elektrot. 11, 314; 340 (1889).)

5. I. Borgman, On the investigation of actino-electric phenomena, Preliminary communication, ЖРФХО 21, 23 (1889); C. R. 108, 733 (1889).

6. A. Righi, Sulla cariche elettriche generate dalle radiazioni, Rend. Linc. 5, 1 Sem., 331 (1889). (Abstract in Beibl. Ann. d. Phys. 13, 566 (1889).)

7. P. Lenard und M. Wolf, Zerstäuben der Körper durch ultraviolettes Licht, Wied. Ann. 37, 443 (1889).

8. A. Righi, Sui fenomeni elettrici provocati delle radiazioni, II Memoria, Atti Ist. Veneto 7 (1889), 25 pp.; Nuovo Cim. 25, 193 (1889). (Abstract in Beibl. Ann. d. Phys. 13, 567 (1889).)

9. A. Righi, Sui fenomeni elettrici provocati delle radiazioni, III Memoria, Atti Ist. Veneto 7 (1889), 54 pp.; Nuovo Cim. 26, 135; 217 (1889); 27, 33 (1890). (Abstract in Beibl. Ann. d. Phys. 13, 976 (1889).)

10. W. Hallwachs, Ueber den Zusammenhang des Elektrizitätsverlustes durch Beleuchtung mit der Lichtabsorption, Wied. Ann. 37, 666 (1889); Gött. Nachr., 325 (1889).

11. A. Stoletow, Sur les phénomènes actino-électriques, C. R. 108, 1241 (1889). (Abstract in Beibl. Ann. d. Phys. 13, 902 (1889).)

12. A. Stoletow, Lettera alla direzione del giornale «Il Nuovo Cimento», N. Cim. 26, 58 (1889).

13. A. Righi, Osservazioni alla precedente lettera, N. Cim. 26, 59 (1889).

14. J. Elster und H. Geitel, Notiz über die Zerstreuung der negativen Elektrizität durch das Sonnen- und Tageslicht, Wied. Ann. 38, 40 (1889).

15. J. Elster und H. Geitel, Ueber die Entladung negativ elektrischer Körper durch das Sonnen- und Tageslicht, Wied. Ann. 38, 497 (1889).

16. W. Hallwachs, Lichtelektrische Versuche, Tagebl. Naturforscherversamml. zu Heidelberg, 214 (1889).

17. A. Righi, Ueber die durch Strahlung hervorgerufenen elektrischen Erscheinungen, Rep. d. Phys. 25, 380 (1889). (Abstract in Beibl. Ann. d. Phys. 14, 68 (1890).)

18. A. Righi, Sulla misura delle forze elettromotrici di contatto dei metalli in vari gas, per mezzo delle radiazioni ultraviolette, Rend. Linc. 5, 1 Sem., 860 (1889). (Abstract in ЖРФХО 22, phys. sect., div. 2, 43 (1890) and in Beibl. Ann. d. Phys. 14, 69 (1890).)

19. A. Nodon, Étude sur les phénomènes électriques produits par les radiations solaires, C. R. 109, 219 (1889).

1. A. Stoletov, Actino-electric investigations, ZhRFKhO 21, phys. sect., sec. 1, 159 (1889); Collected Works, vol. I, Gostekhizdat, Moscow—Leningrad, 1939, p. 217; Selected Works, Gostekhizdat, Moscow—Leningrad, 1950, p. 246. (There is a translation into German in Physikalische Revue I, 723 (1892). In Russian it was issued as a separate brochure in St. Petersburg (1889); in part see La lumière électrique 34, No. 47–52, 576 (1889). Abstract in Beibl. Ann. d. Phys. 14, 322 (1890).)

2. J. Elster und H. Geitel, Elektrische Beobachtungen auf dem hohen Sonnblick, Wien. Ber. 99, 1008 (1890).

3. E. Branly, Déperdition des deux électricités dans l’éclairement par des radiations très réfrangibles, C. R. 110, 751 (1890). (Abstract in Beibl. Ann. d. Phys. 14, 539 (1890).)

4. E. Branly, Courants photoélectriques entre les deux plateaux d’un condensateur, C. R. 110, 898 (1890). (Abstract in Beibl. Ann. d. Phys. 14, 541 (1890).)

5. W. Hallwachs, Untersuchung einer Fehlerquelle bei der lichtelektrischen Erregung, Wied. Ann. 40, 332 (1890).

6. W. Hallwachs, Vorlesungsversuch zum nachweis der lichtelektrischen Erregung, Wied. Ann. 40, 343 (1890).

7. W. Hallwachs, Bemerkungen zu einem Prioritätsanspruch des Herrn Righi, Wied. Ann. 40, 338 (1890); N. Cim. 28, 59 (1890).

8. A. Righi, Erwiderung auf die Bemerkung des Herrn W. Hallwachs, Wied. Ann. 41, 505 (1890); N. Cim. 28, 62 (1890).

9. J. Elster, Ueber die Entladung negativer Elektrizität durch Licht im magnetischen Felde, Naturw. Verein. Braunschweig 24, IV (1890).

10. A. Stoletow, Sur les courants actino-électriques dans l’air raréfié, Journ. de Phys. 9, 468 (1890); Collected Works, vol. I, Gostekhizdat, Moscow—Leningrad, 1939, p. 270; Selected Works, Gostekhizdat, Moscow—Leningrad, 1950, p. 249; Bull. Soc. fr. de Phys., 202–207 (1890). (Abstract in Beibl. Ann. d. Phys. 15, 233 (1891).)

11. A. Righi, Sulla traiettorie percorse nella convenzione fotoelettrica, e su alcuni nuovi fenomeni elettrici nell’aria rarefatta, Rend. Linc. 6, 2 Sem., 81 (1890); Sulla convenzione fotoelettrica e sul altri fenomeni elettrici nell’aria rarefata, IV Memoria, Mem. di Bologna 10, 85 (1890). (Extract—N. Cim. 30, 197 (1891).) (Abstract in Beibl. Ann. d. Phys. 14, 1167 (1890).)

12. J. Elster und H. Geitel, Ueber die Verwendung des Natriumamalgams zu lichtelektrischen Versuchen, Wied. Ann. 41, 161 (1890).

13. J. Elster und H. Geitel, Ueber den hemmenden Einfluss des Magnetismus auf lichtelektrische Entladungen in verdünnten Gasen, Wied. Ann. 41, 166 (1890).

14. A. Righi, Sulla convezione elettrica, Rend. Linc. 6, 1 Sem., 151 (1889). (Abstracts in ZhRFKhO 22, issue 7, phys. sect., sec. 1, 147 (1890) and in Beibl. Ann. d. Phys. 14, 663 (1890).)

II

1. N. A. Kaptsov, Uchenye zapiski Moskovskogo universiteta 52, 71, 1940; N. S. Khlebnikov, UFN 22, 384 (1939).

2. Zh. G. De-Bur, Electron emission and adsorption phenomena, ONTI, Moscow—Leningrad, 1936, p. 22; G. Simon and I. R. Suhrmann, Photophenomena and their application, ONTI, Moscow—Leningrad, 1936, p. 11; A. L. Hughes and L. A. DuBridge, Photoelectric phenomena, ONTI, Moscow—Leningrad, 1936, pp. 6–7.

3. ZhRFKhO 20, phys. sect., sec. 2, 41 (1888).

1. W. Hallwachs, Die Lichtelektrizität, Handb. d. Radiol., IIIb, pp. 245—618, Leipzig, Akad. Verlagsges. MBH (1916).

2. J. Elster und H. Geitel, Wied. Ann. 46, 287 (1892).

3. J. Elster und H. Geitel, Wied. Ann. 52, 433 (1894).

4. V. K. Zworykin and E. G. Ramberg, Photoelectricity and its Applications, J. Wiley a. Sons, New York, 1949, p. 5.

5. L. N. Dobret͡sov, Electron and Ion Emission, Gostekhizdat, Moscow—Leningrad, 1952, pp. 198—199.

6. O. Rhode, Ann. d. Phys. 19, 935 (1906).

7. A. R. Olpin, Phys. Rev. 36, 251 (1930).

8. N. S. Khlebnikov, Phys. Zapiski 9, 171, 1941; ZhTF 15, 99 (1945).

9. J. Elster und H. Geitel, Phys. Zs. 13, 468 (1912).

10. E. Marx und K. Lichtenecker, Ann. d. Phys. 41, 124 (1913).

11. E. O. Lawrence and J. W. Beams, Phys. Rev. 29, 903 (1927); Phys. Rev. 32, 478 (1928).

12. B. Gudden und R. Pohl, Zeits. f. Phys. 34, 245 (1925).

13. V. V. Bazilevich, ZhRFKhO, Phys. Section 56, issue 4, 421 (1925).

14. N. O. Chechik, S. M. Fainshtein and T. I. Lifshits, Electron Multipliers, Gostekhizdat, Moscow, 1954.

15. A. M. Butov and E. G. Shvidkovskii, UFN 48, 151 (1952); S. E. Frish and A. V. Timoreva, Course of General Physics, vol. III, p. 11, Gostekhizdat, Moscow (1953).

16. A. T. Forrester, R. A. Gundmundsen and Ph. O. Johnson, Phys. Rev. 99, 1691 (1955).