05& 
 
 
 C\J 
 
 LO 
 
 V 
 
LIBRARY 
 
 OF THE 
 
 UNIVERSITY OF CALIFORNIA. 
 
 OIFT OF 
 
 Class 
 
I 
 
THE DISCHARGE OF 
 
 ELECTRICITY IN GASES 
 
 BY 
 
 NICHOLAS M. WILHELMY, S. M. 
 
 A Dissertation Presented to the Faculty of Philosophy 
 
 of the 
 
 Catholic University of America 
 
 For the Degree of Master of Philosophy. 
 
 Washington. 
 1905. 
 
PREFACE 
 
 THE discoveries made during the last fifty years concerning 
 the discharge of electricity through gases and the determination 
 of the close relationship between the phenomena of this dis- 
 charge and the recently studied properties of the radioactive 
 substances have opened new fields for experiment and specula- 
 tion. 
 
 The prediction of Faraday, that the study of the electrical 
 discharge in gases might eventually lead us to a more thorough 
 understanding of the nature of matter and electricity, is being 
 gradually realized. Thus the existence in the chemical atom 
 of much smaller particles, which propably are the same in all 
 substances may now be looked upon as an established fact. 
 The structural character of electricity also is, at the present 
 day, more than a mere hypothesis. 
 
 Moreover this discharge has already proved of more than 
 purely scientific interest. The mere mention of the mercury- 
 arc lamp and of the mercury-arc alternating-current rectifier 
 will be enough to show the practical use now made of some of 
 the properties discovered The mercury-arc rectifier, which has 
 been put on the market quite recently, bids fair to take a very 
 prominent place among rectifying devices on account of its effi- 
 ciency and convenience in handling. 
 
 To Professor Daniel W. Shea I desire to acknowledge my in- 
 debtedness and express my sincerest gratitude for his indispen- 
 sable assistance and able direction in the study of the bewilder- 
 ing amount of literature on this subject. I also gratefully ac- 
 knowledge my obligations to Professor Jno. J. Griffin for the 
 instruction received in the department of Chemistry. 
 
4 CHAPTER i. FIRST PERIOD. 
 
 gradual changes that take place in the electric discharge as the 
 pressure of the air decreases. At normal pressure, the electric 
 strain may be released in two different ways : if the distance 
 between the electrodes is not too great for the available poten- 
 tial, the spark discharge occurs; in this discharge itself the 
 spark proper and the discharge through the surrounding elec- 
 trified air have to be distinguished; if the distance between the 
 electrodes is increased the sparking becomes less frequent and 
 is accompanied by a brush discharge, till the latter finally be- 
 comes the only noticeable phenomenon. 
 
 If the electrodes are inclosed in a vessel from which the air 
 may be exhausted, the phenomena are the same as those des- 
 cribed above as long as the pressure is not reduced below a def- 
 inite value. After this there is a gradual change in the char- 
 acter of the discharge. The brush disappears and a luminous 
 "Stream Discharge" reaching from one pole to the other sets in. 
 If the pressure is lowered still more, a bluish glow first appears 
 at the kathode and a pinkish light then begins to surround the 
 anode. This light gradually extends towards the kathode 
 though it never reaches the negative light. This last mentioned 
 phenomenon was styled by Faraday the "Dark Discharge" 
 owing to the dark space that always separates the two lights. 
 Here is his own description of this interesting discovery: (1) 
 "I will now notice a very remarkable circumstance in the lumi- 
 nous discharge accompanied by negative glow, which may, per- 
 haps,be correctly traced hereafter into discharges of much higher 
 intensity. Two brass rods, 0.3 of an inch in diameter, entering a 
 glass globe on opposite sides, had their ends brought into con- 
 tact, and the air about them very much rarefied. A 
 discharge of electricity from the machine was then made 
 through them, and whilst that was continued the 
 ends were separated from each other. At the moment of sepa- 
 ration a continuous glow came over the end of the negative rod, 
 
 (1) 1. c. p. 490, No. 1544. 
 
CHAPTER i, FIRST PERIOD. 5 
 
 the positive termination remaining quite dark. As the distance 
 was increased, a purple stream or haze appeared on the end of 
 the positive rod, and proceeded directly outwards towards the 
 negative rod; elongating as the interval was enlarged, but never 
 joining the negative glow, there being always a short dark spa- 
 ce between, This space of about one tenth or one twentieth of 
 an inch, was apparently invariable in its extent and position 
 relatively to the negative rod; nor did the negative glow vary. 
 Whether the negative end were inductric or inducteous, the 
 same effect was produced. It was strange to see the positive 
 purple haze diminish or lengthen as the ends were separated, 
 and yet this dark space and the negative glow remain un- 
 altered." 
 
 Faraday had never carried his vacuum very high. The idea 
 of studying the discharge at the lowest obtainable pressure na- 
 turally suggested itself. Masson (1) attempted to pass an 
 electric discharge through a barometric vacuum but failed to 
 obtain any results, whence he concluded to the non-conductivity 
 of an absolute vacuum. In a later paper (2) he brought further 
 proof for this opinion. The first one to remark the resolution 
 of the positive light into layers was W. R. Grove (3) who had 
 also noticed the dark space observed by Faraday and had dis- 
 covered, in addition, that under certain circumstances the latter 
 may disappear. Quet (4) obtained the same striation of the 
 positive light and described anew the different phenomena of the 
 discharge, showing that it was disruptive and that it was pos- 
 sible to effect a complete disappearance of the positive light for 
 a certain distance of the electrodes. 
 
 B. VALVE-TUBE AND FUNNEL TUBE. 
 
 The study of the elctric discharge in rarefied gases was given 
 
 (1) C. R. 7, p. 671. Pogg. Ann. 46 pp. 487. 39. 
 
 (2) Ann. de Oh. et de Phya. 15, p. 486. 1868. 
 
 (3) Phil. Trans. 1852. Pogg. Ann. 93. p. 492. 1854. 
 
 (4) Pogg. Ann. Ergzb. 4. p. 507. 1854. 
 
(5 CHAPTER i. FIRST PERIOD. 
 
 a greater impetus by the advent of M. Gaugain (1) who dis- 
 covered that in a vacuum-tube with two electrodes whose sizes 
 are very unequal, a current is transmitted in one direction much 
 more easily than in the other, He called this tube "oeuf-soupa- 
 pe" or "soupape electrique", from its valve-like action. Accord- 
 ing to his view, the closing current of an induction coil was al- 
 ways allowed to pass with much greater ease than the opening 
 current. He also studied the nature of the layers and concluded 
 that they are a phenomenon of matter. 
 
 Gaugain 's views regarding the real nature of the occurrences 
 in his valve-tube were challenged by P. Riess, who showed, in a 
 series of papers, (2) that the phenomenon was to be traced to 
 the direction of the current and not to the fact that it was either 
 an opening or a closing current. He formulated the law that 
 an induced current after traversing the valve-tube, deflects a 
 magnet in the same manner as a current flowing from the large 
 to the small electrode, or in other words, that the resistance of- 
 fered by the tube is greater for currents which make the small 
 electrode their anode than for those in which the anode is the 
 larger surface. The same subject was studied by Feddersen (3) 
 and Knochenhauer, (4) neither of whom found results favoring 
 Gaugain's views. 
 
 Another form of tube resembling the previous one in its 
 effects is what is known as the "Funnel-tube", namely, a 
 vacuum-tube divided into several compartments by funnel-shaped 
 obstructions which leave only a narrow opening between the 
 several parts. These tubes were first made by Geissler in 1858 
 and subsequently perfected by Hoitz. The nature of the effects 
 occurring in them was studied by J. C. Poggendorff. (5) But 
 
 (1) C. U. 40. p. 640, & Pogg. Ann. 95 p. 163. 1855. 
 
 (2) Pogg. Ann. 96. p. 177. 1835. 120, p. 513, 1863. 136, p. 81, 1869. 
 
 (3) Pogg. Ann. 115, p. 336, 1862. 
 
 (4) Pogg. Ann. 126, p. 228, 1866. 129, p. 78, 1866. 
 
 (5) Pogg. Ann. 134, p. 1. 1868. 
 
CHAPTER 1 FIRST PERIOD. 7 
 
 before giving the results of his investigations, it may be well to 
 mention that, in 1859, J. Pluecker (1) had noticed the peculiar 
 behavior of obstructions in a tube, producing what he called 
 "recurring currents". He again touched upon the same subject 
 in 1861, (2) stating that if the current flows from a wider to a 
 narrower part, there is a partial recurring of the current towards 
 the anode. Poggendorff found that these tubes also offer a 
 greatly different resistance with different connections: the re- 
 sistance is always smaller when the points of the funnels are 
 directed towards the anode. At the end of these same points 
 there appeared a dark space from which a cone of light extended 
 in the direction of the next funnel. From these several obser- 
 vations it became clearly evident that the funnel points were 
 behaving somewhat like a kathode. With the aid of this tube 
 Poggendorff detected a peculiar influence exerted by a rise of 
 temperature on the striation of the positive light: namely that 
 if one of the central compartments was heated, the striae of the 
 adjoining compartments increased in number and brilliancy. 
 The explanation of this phenomenon would present no difficulty 
 were it not for the fact that it persisted for weeks, even after 
 the cooling of the tube in ice. 
 
 C. STRIATION. 
 
 During this period a certain amount of attention was also 
 devoted to the conditions giving rise to striation. As we have 
 already noted the pioneer observer of the phenomenon was W. 
 R. Grove. V. S. M. Van der Willigen (3) was the first to pro- 
 duce it by means of the influence-machine both with and with- 
 out a condenser in circuit, (4) thus removing one more of the 
 supposed differences between frictional and induced electricity. 
 
 (1) Pogg. Ann. 107, p. 87. 1859. 
 
 (2) Pogg. Ann. 116, p. 29, 1861. 
 
 (8) Pogg. Ann. 98, p. 494, 1856. 99, p. 175, 1856. 
 (4) Pogg. Ann, 118, p. 511, 1861. 
 
8 CHAPTER r FIRST PERIOD. 
 
 It was learned that striation was brought about by the 
 insertion of a resistance (1) into the outer circuit; which resist- 
 ance might be that of an air-gap, a metallic resistance, a wet 
 twine, etc. Striation was also influenced by the application of 
 heat, by mere touching of the tube and by connecting one of 
 the electrodes with the earth. (2) The distance between the 
 striae was shown to be some function of the diameter of the 
 tube (3) and of the pressure therein. (4) All these observations 
 about striation refer to the positive light, but it must be noted 
 that some striation was also observed in the negative glow. (5) 
 
 D. INFLUENCE OF A MAGNET ON THE ELECTRIC DISCHARGE. 
 
 In 1849, A. De la Rive (6) noticed that if one of the electro- 
 des of a vacuum-tube was magnetised, the electric light rotated 
 around the pole of the magnet. This interesting and suggestive 
 discovery seems to have passed unobserved. Several years 
 later, the fact that a magnet does really exert some influence on 
 the electric discharge was again discovered by J. Pluecker, (7) 
 who found that the general behavior of the electric light under 
 the influence of a magnet is apparently similar to that of a wire 
 carrying a current and free at one end. The behavior of the 
 positive light is different from that of the negative: while the 
 former is brought together into a cone, the latter rotates around 
 a magnetic curve. If the tube is subjected to the influence of 
 a magnet of sufficient strength, the positive light may be bent 
 back towards the anode, causing a negative glow to appear at 
 
 (1) Paalzow, Pogg. Ann. 112, p. 567, 1861. Poggendorff, Pogg. Ann. 134, 
 p. 17, 1868. Holtz, Poog. Ann. 170, p. 555, 1878. 
 
 (2) Poggendorff, Pogg. Ann. 134, p. 43, 1868. P. Riess, Pogg. Ann. 104, p. 
 321, 1858. 
 
 (3) Pluecker, Pogg. Ann. 103, p. 101, 1858. 
 
 (4) Waltenhofen, Pogg. Ann. 126, p. 527, 1865. 
 
 (5) J. Pluecker, Pogg. Ann. 103, p. 92, 1858. 
 
 (6) Pogg. Ann. 104, p. 129, 1858. 
 
 (7) Koelnische Volksztg. July 22, 1857. Pogg. Ann. 103, p. 88, 1858. 
 
CHAPTER i FIRST PERIOD. 9 
 
 the latter. (1) De la Hive (2) again called attention to his 
 early experiments and repeated them. He also ascertained (3) 
 that under the influence of a magnet the layers in the positive 
 light become more brilliant and that the so-called dark space (4) 
 may be made luminous. Generally the magnet was found to 
 increase the resistance in the tube. 
 
 F. SPECTROSCOPIC STUDY OF THE LIGHT IN VACUUM-TUBES. 
 
 The discovery by J. Pluecker (5) in 1858 of the fact that 
 each gas gives a special spectrum when fluorescing under the 
 influence of the electric discharge in a vacuum-tube, created a 
 new branch in spectrum analysis. This discovery, as its author 
 himself foresaw, was to prove of great technical and scientific 
 usefulness. The first scientific conclusion he deducted from 
 it was the non-luminosity of the electric discharge through 
 rarefied gases, the light being entirely due to ponderable mat- 
 ter under the influence of the current. (6) The difference be- 
 tween the spectra of the same gas near the kathode and the 
 anode was described by F. W. Dove. (7) These spectra were 
 studied by Van der Willigen, (8) Waltenhofen (9) and many 
 others, but most of the work is only incidentally connected 
 with this subject. 
 
 G. FACTORS OF THE DISCHARGE. 
 
 PRESSURE AND CURRENT. Gaugain (10) remarked an increase 
 
 (1) Fogg. Ann. 107, p. 77, 1859. 
 
 (2) Pogg. Ann. 104, p. 129, 1858. 
 
 (3) Ann. de Ch. et de Phys. JO, p. 159, 1867. 
 
 (4) Ann. de Ch. et de Phys. 20, p. 103, 1870 
 
 (5) Pogg. Ann. 104, p. 113, 1858. 
 
 (6) Pogg. Ann. 107, p. 497, 1859. 
 
 (7) Pogg. Ann. 104, p. 184, 1858. 
 
 (8) Pogg. Ann. 106, p. 526, 1859. 
 
 (9) Pogg. Ann. 126, p. 535, 1865. 
 
 (10) C. R. 40, p. 640. 
 
10 CHAPTER i. FIRST PERIOD. 
 
 of current coinciding with an increase of vaccum. A. Mor- 
 ren (1 > made a more careful study of this subject and learned 
 that the relation between current-intensity and gas-pressure 
 was not a simple one. The curve which would represent this 
 relation (pressure being represented along the X-axis, and 
 current on the Y-axis) would, with high but decreasing pres- 
 sures, at first rise very slowly till the gas pressure had reached 
 as low as 2-4 mm, according to the nature of the gas; with a 
 further decrease of pressure the curve would rise very rapidly 
 up to a maximum which oocured at about 0.7-lmm, after 
 which it would again quickly decrease. 
 
 DISCHARGE POTENTIAL. In 1834 Harris (2) had shown that 
 the quantity of electricity required to obtain a discharge in 
 rarefied air varied directly as the distance between the elec- 
 trodes for a determined gas pressure and also directly 
 as the gas pressure for a given distance between 
 the electrodes. From this he concluded that the smallest 
 charge could not be retained on a conductor in a suffi- 
 ciently high vacuum. Matteucci (3) accepted the same theory. 
 But in 1839 Masson (4) failed in his attempts to produce a dis- 
 charge through a high barometric vacuum. Q-assiot also 
 obtained vacua which did not allow the discharge of an induc- 
 tion coil, hence the conclusion already drawn by Faraday that 
 a perfect vacuum was a perfect non-conductor seemed to be 
 justified. The relation between the gas pressure and the dis- 
 charge potential was investigated more thoroughly by Walten- 
 hofen (5) who demonstrated that this potential depended not 
 only on the pressure of the gas but -also on the nature and the 
 shape of the electrodes. From his several experiments he con- 
 
 (1) Ann. de Ch. et de PhyB. IV, 4, p. 325, 1865. 
 
 (2) Phil. Trans, p. 243, 1834. 
 
 (3) Ann. de Oh. et de Phys. Ill, 28, p. 385, 1850. 
 
 (4) C. R. 7, p. 671. 
 
 (5) Pogg. Ann. 126, p. 527, 1865. 
 
CHAPTER i. FIRST PERIOD. 11 
 
 eluded that for very high vacua, a high potential is required 
 only to start the discharge and not to maintain it. On the 
 strength of this he advanced the theory that the electric dis- 
 charge could really be propagated through vacua which are 
 considered as perfectly impervious to electricity, thus rejecting 
 the several proofs adduced in support of the hypothesis that a 
 perfect vacuum was a non-conductor. Karl Schultz (1) also 
 observed a minimum of discharge potential for certain gas 
 pressures, above and below which the required potential rises 
 slowly at first but with a gradually increasing velocity. He 
 also investigated the effect of the dimensions of the tube on the 
 discharge potential and found that the latter increases as the 
 cross-section decreases and that it becomes greater as the length 
 of the air column increases for all pressures above 1 mm. 
 Below these pressures the discharge potential does not seem to 
 be influenced by the length of the tube. 
 
 A. De la Rive (2) noticed the different changes of tempera- 
 ture near the kathode and the anode, but this manifestation of 
 energy was studied more closely by Poggendorff. (3) In the 
 Bakerian lecture of 1858, J. P .Grassiot (4) also alluded to the 
 special development of heat at the kathode and attributed it to 
 taa overcoming of a large resistance near this electrode. As a 
 further proof of the presence of this large resistance he brought 
 forward the fact that metal is projected from the surface of the 
 kathode, i. e. the so-called "Zerstaeubung". 
 
 H. EFFECTS OF THE DISCHARGE. 
 
 The strong green or blue fluorescence caused on glass under 
 certain circumstances was observed by J. Pluecker (5) and P. 
 
 (1) Pogg. Ann. 135, p. 249, 1868. 
 
 (2) Ann. de Ch et de Phys. 8, p. 437, 1866. 
 
 (3) Pogg. Ann. 138, p. 642, 1869. 
 
 (4) Pogg. Ann. 119, p. 131, 1863. 
 
 (5) Pogg. Ann. 103, p. 88, 1858. -104, p. 113, 1858. 
 
12 CHAPTER i FIRST PERIOD. 
 
 Riess (1) and attributed by them to the negative light. 
 Pluecker also noticed that whenever the negative light was 
 brought to the walls of the tube it produced this vivid fluores- 
 cence. Another curious and important phenomenon of fluores- 
 cence was described by H. W. Dove, (2) namely, that uranium 
 glass and barium-platinum cyanide fluoresced intensely when 
 placed in close proximity to the tube. Had he examined more 
 deeply into the nature of the phenomenon, it can hardly be 
 doubted that he would have anticipated the discovery subse- 
 quently made by W. Roentgen, because this fluorescence was 
 in all probability an effect of the X-rays. 
 
 In some tubes constructed by Geissler, the gas still retained 
 some fluorescence or "after-glow" after the discharge had been 
 stopped. This peculiar effect was extensively studied during 
 this period. The first to offer an explanation was E. Becque- 
 rel, (3) who assigned the presence of oxygen as the cause of the 
 phenomenon. Riess (4) ascribed it to sulphuric acid gas. H. 
 Wild (5) believed it was produced by the oxidation of sulphur 
 after the discharge. A. Morren (6) attacked Becquerel's 
 opinion and showed that pure oxygen would not give any after- 
 glow; he found that if oxygen contained but slight traces of 
 nitrogen it would fluoresce. E. Sarazin (7) attributed the 
 after glow to chemical causes, and particularly to the formation 
 of ozone. Prom his experiments, he thought himself justified 
 in concluding that no other gases but pure oxygen and oxygen- 
 compounds are able to give this fluorescence. This same view 
 was confirmed by A. De la Rive, (8) in whose laboratory most 
 of Sarazin's experiments had been performed. 
 
 (1) Pogg. Ann. 104, p. 321, 1858. 
 
 (2) Pogg. Ann. 113, p. 511, L861. 
 
 (3) 0. 48, p. 404, 1859. 
 
 (4) Pogg. Ann. 110, p. 523, 1860. 
 
 (5) Pogg. Ann. Ill, p. 621, 1860. 
 
 (6) C. R. 53, p. 794, 1865. 
 
 (7) Ann. de Ch. et de Phys. 17, p. 501, 1869. 19, p. 191, 1870. 
 
 (8) Ann. de Ch. et de Phys. 19, p. 191, 1870. 
 
CHAPTER i FIRST PERIOD, 13 
 
 Two more effects of the discharge brought to light during 
 this period deserve to be mentioned: 
 
 1 ) Its effect on a photographic plate. This was discovered 
 by H. W. Dove in 1861. (1) The fact that he obtained a well 
 defined shadow of a piece of uranium glass shows that he had 
 no ordinary light effect. 
 
 2) The fact that as the discharge continues through a gas, 
 the latter becomes more rarefied. (2) This change of vacuum 
 was attributed to a combination of oxygen with other gases or 
 solids in the tube. / ' 
 
 (1) Pogg. Ann. 113, p. 511, 1861. 
 
 (2) Pluecker, Pogg. Ann. 105, p. 67, 1858. 
 
CHAPTER II. 
 
 SECOND PERIOD. FROM HITTORF TO LENARD. 
 
 The publication in 1869 of W. Hittorf's first paper on the 
 electrical conductivity of gases (1 ) may be rightly looked upon 
 as the beginning of a new period in the study of this subject. 
 Heretofore there had been more or less casting about and a 
 lack of defin iteness ; but henceforth a more systematic study of 
 the discharge of electricity through gases was instituted. 
 Observations were more accurate, new facts were discovered, 
 their theoretical bearing was discussed and a successful attempt 
 was made to analyse some of the more complicated phenomena. 
 All this was due mainly to the efforts of such men as W. 
 Hittorf, E. Goldstein, the two Wiedemanns, Sir W. Crookes, 
 J. J. Thomson and several others. 
 
 I. W. HITTORF. 
 
 Faraday had already described the appearance of the electric 
 discharge in rarefied gases. Although many experimenters in- 
 tervened between him and Hittorf, the latter may be looked 
 upon as continuing the former's work, for he prefaced his first 
 paper by an extensive extract from Faraday's Experimental 
 Researches and then pursued his own investigations on similar 
 lines but under more favorable conditions. These allowed 
 him to get a higher E. M. F. and higher vacua, thanks 
 to the improvements made in the Rhumkorff coil and 
 
 (1) Pogg. Ann. 136, p. 1, 1869. 
 
CHAPTER n. HITTORF 15 
 
 the mercury pumps constructed by Geissler. By the 
 aid of these new appliances, he noted (1) that as the 
 gas pressure became lower than 2mm of mercury, there 
 was a rapid change in the phenomena observed by Fara- 
 day, The glow on the kathode soon extended itself not only 
 over greater portions of this electrode but also throughout the 
 surrounding space driving back all the while the reddish light 
 towards the anode. Both lights became striated with a conca- 
 vity towards the positive pole. But in the mean time the three 
 parts of the discharge noted by Faraday, namely, the positive 
 light, the dark space and the negative glow still remained in 
 the tube. While the number of striae in the positive light was 
 variable, depending on the size of the tube, the quantity of 
 available electricity, the nature and pressure of the gas, 
 etc., the layers in the negative light on the other hand exhibited 
 a remarkable constancy. They were always three in number. 
 Directly on the kathode there appeared a narrow band of light 
 which sometimes was very faint; beyond this there was a still 
 darker but well defined band, which is now called the Hittorf 
 or kathode dark space; at the end of this band begins what is 
 called more particularly the negative light or glow. / > 
 
 It may be noted here that the kathode dark space has been 
 called by some writers the ''dark space" without any further 
 specification; this is misleading as the term was first and still 
 is generally applied to the region between the positive and the 
 negative light. 
 
 At the highest vacua obtainable by Hittorf during this period, 
 the positive light disappeared completely and the negative glow 
 filled the whole tube. A very important characteristic of this 
 negative light is its rectilinear propagation in a direction nor- 
 mal to the surface of the kathode, the position of the anode 
 exercising no influence on its path. This, property was aptly 
 
 (1) Pogg. Ann. 186, p. 6, 1869. 
 
16 CHAPTER n. HITTORF 
 
 illustrated by a tube wherein the kathode pointed away from 
 the anode, in which case the negative light travelled directly to 
 the end of the tube opposite the positive pole. By placing a 
 solid or liquid in the path of this light, a well defined shadow 
 was thrown on the opposite wall of the tube, whence Hittorf 
 concluded that the path of the glow was bounded by any solid 
 which it happened to strike and that there was never any devia- 
 tion from the rectilinear transmission. ( 1) This statement con- 
 cerning the ending of the negative light on striking any solid 
 was considerably modified by later experimenters: by E. Gold- 
 stein (2) and H. Hertz (3) in particular. 
 
 Another property of this negative light which has been very 
 important in the study of the subject is its power to produce 
 fluorescence on substances which may fluoresce under the influ- 
 ence of light. Thus if this negative glow is brought to the 
 walls of the tube there is a vivid fluorescence; the color of the 
 latter is bluish for lead glass and a very bright green for sodium 
 glass. The light may be brought to the walls of the tube in 
 several ways: by giving the kathode such a shape or position 
 as will lessen the distance from the walls to its surface, by 
 evacuating to such a degree as will allow the negative light to 
 fill the whole tube or, finally, by subjecting the tube to the in- 
 fluence of a magnet. Hittorf found that in this last case the 
 negative light behaves like a stream of negatively electrified 
 particles moving from the kathode to the anode. 
 
 Besides noting these general facts he made quantitative 
 experiments to investigate the conditions prevailing in the 
 tube. He first studied the conductivity of the gas and reached 
 the conclusion (4) that the maximum of conductivity does not 
 
 (1) Pogg. Ann. 136, p. 8, 1869. 
 
 (2) Wied. Ann. 11, p. 832, 1880. 
 
 (3) Wied. Ann. 45, p. 28, 1892. 
 
 (4) Pogg. Ann. 136, p. 30, 1869. 
 
CHAPTER n. HITTORF 17 
 
 depend on the gas-pressure alone but also on the dimensions of 
 the tube and the shape of the kathode. The conductivity in- 
 creased until the pressure had been reduced to a certain value 
 beyond which it decreased rapidly. In fact he was able to 
 reach vacua through which the highest E. M. F. then at his 
 command could not send an electric current. The source was a 
 42 cm. coil, which was put at his disposal in Paris. But he 
 soon discovered that the conductivity was not uniform through- 
 out the different parts of the tube. He was able to state that 
 with decreasing gas-pressure, the resistance in the positive 
 layers diminishes, while that in the kathode light and particu- 
 larly that near the kathode increases. The application of heat 
 to the kathode decreases this resistance very sensibly: thus, if 
 this electrode was heated to white heat a few Bunsen cells were 
 enough to maintain a constant current through a highly 
 evacuated tube. (1) 
 
 The negative light itself also greatly increases the conduct- 
 ivity of a gas not only in the direction of its propagation but 
 also in all other directions. (2) 
 
 The current-intensity likewise has an effect on conductivity. 
 Hittorf (8) ascertained that the conductivity of a gas varies 
 proportionately to the current-intensity and, in a later paper, (4) 
 he stated more particularly that in the positive light the con- 
 ductivity increases proportionately to the intensity of the current 
 for a constant gas-pressure. Studying the fall of potential in 
 the tube, he found that the difference of potential between any 
 two cross-sections of the positive light does not depend on the 
 current-intensity. (5) 
 
 The fall of potential at the kathode is always very great. (6) 
 
 (1) Wied. Ann. 21, p. 90, 1884. 
 
 (2) Wied. Ann. 7, p. 553, 1879 
 
 (3) Wied. Ann. 7, p. 622, 1879. 
 
 (4) Wied. Ann. 20, p. 705, 1883. 
 
 (5) Wied. Ann. 20, p. 705, 1883. 
 
 (6) Wied. Ann. 21, p. 90, 1884. 
 
18 CHAPTER n . E. GOLDSTEIN 
 
 This great potential-gradient is accompanied by an excessive 
 heating effect on the kathode, causing the so-called "Zerstaeu- 
 bung" of the latter. Hittorf even performed some experiments 
 to show that this heat is sufficient to account for the rotation of 
 the Crooke's radiometer. 
 
 II. E. GOLDSTEIN. 
 
 Hittorf had generally used the term "Glimmlicht", (which 
 may be rendered by "Negative glow" or "light"') in connection 
 with the main phenomenon which he studied in rarefied gases. 
 By this he meant an action which is propagated from the 
 kathode in the nature of a ray; "strahlenartig". (1) E. Gold- 
 stein calls the same phenomenon the electric ray of the kathode 
 light, "der elektrische Strahl des Kathodenlichtes", (2) and 
 later ,simply <4 the kathode rays," which name passed into general 
 use. He found that Hittorf s statement viz. that the kathode 
 rays ended wherever a solid was put in their path, had to be 
 modified. The kathode rays, under these circumstances, do not 
 merely end and produce fluorescence but they are "differen- 
 tiated", (3) this differentiation taking place both at the subs- 
 tances which did and those which did not fluoresce. 
 
 Allowing kathode rays to impinge on one of these non-fluores- 
 cing substances afforded the most convenient method of study- 
 ing this new kind of rays or this modification of the primary 
 kathode radiation. The path of the new ray is rectilinear and 
 it forms every possible angle with the surface from which it 
 issues. The modified ray ressembles the more refrangible light 
 not only in its rectilinear propagation but also by 
 its power of exciting fluorescence. This was shown 
 by several interesting experiments, for instance: a beam of 
 parallel kathode rays was allowed to enter a tube sufficient in 
 
 (1) Pogg. Ann. 136, 223, 1769. 
 
 (2) Wied. Ann. 11, p. 833, 1880. 
 
 (3) Wied. Ann. 11, p. 833, 1880. 
 
CHAPTER n. E. GOLDSTEIN 19 
 
 length to prevent the rays from reaching its farther end. No 
 fluorescence could in this case be noticed in the tube. If a 
 fluorescing screen was then placed anywhere in the path of the 
 kathode rays, fluorescence immediatly appeared not only on the 
 screen but also on the walls of the tube. It was found that this 
 modification of the kathode rays was not connected with any 
 particular density of the gas or strength of the electrical dis- 
 charge. 
 
 Another important point brought out by the work of Gold- 
 stein is the fact that in the much discussed funnel-tubes or any 
 similar devices, the narrower part which points towards the ano- 
 de acts as a secondary kathode. 1) The rays which it emits 
 as well as all the other phenomena are the same as at an 
 ordinary kathode but "quantitativ gemildert". Goldstein gave 
 these devices many forms for the purpose of a thorough invest- 
 igation and called them secondary kathodes. 
 
 Early in his work he had already deemed himself justified in 
 speaking of rays in connection with the positive light. He 
 pursued the study of these rays by constructing a tube bent 
 several times at right angles, thus preventing the kathobe rays 
 from reaching beyond the first bend. Nevertheless he found 
 fluorescence at each succeeding bend and if a solid was put in 
 the path of the discharge towards the kathode a shadow was 
 produced, while if placed in the path towards the anode it did 
 not effect the fluorescence. From this Goldstein concluded that 
 the rays in the positive light are, like the kathode rays, directed 
 from the kathode to the anode. Like the same rays also they 
 produce fluorescence, are propagated rectilinearly and end at a 
 solid wall whereon they produce a new source of the same kind 
 of rays. A discovery made the following year by E. Wiede- 
 mann ( 2 ) shows that the effects on which Goldstein based these 
 
 (1) Berl. Monatsber. p. 279, 1876. Wied. Ann. 11, p. 836,1880. 
 
 (2) Wied. Ann. 10, p. 236, 1880. 
 
20 CHAPTER n E. GOLDSTEIN 
 
 conclusions may, at least in part, be ascribed to some other 
 cause. Wiedemann was working with a tube on which, oppo- 
 site the kathode and beyond the anode, there was a narrow 
 attachment bent at some distance from the main tube. When 
 the current passed, fluorescence appeared not only at the bend 
 which could be reached by a straight ray coming from the 
 kathode but also at the very end of the bent tube. Goldstein had 
 discovered the deflection of the kathode rays when they pass 
 near a second kathode in the tube. (1) E. Wiedemann explained 
 his results as a new instance of deflection, due to the static 
 electricity gathered at the bend by the striking of part of the 
 kathode ray. Goldstein however proved that this was not a case 
 of deflection but of reflection and thus had the honor of again 
 discovering a new property of the kathode rays. He found that 
 this reflection was diffuse like that of light from a non-polished 
 surface. 
 
 He also made a very extensive and interesting study of the 
 effects produced by different shapes of the kathode on the 
 images which they cast on the opposite walls of the tube. ( 2 ) 
 The most important results of his experiments are the fact that 
 kathode rays may cross one another and that their path is not 
 always rectilinear, being modified either by mutual repulsion or 
 by an action emanating from different parts of the kathode. E. 
 Wiedemann (3) showed later on that the first of these two expla- 
 nations could not be admitted because the apparent repulsion 
 still took place if one of the rays was cut off; thus the cause of 
 the effect had to be traced to the kathode itself. 
 
 In connection with the striated discharge, Goldstein (4) de- 
 monstrated that the distance between the layers is a function 
 of the diameter of the tube and of the gas-pressure. Though 
 
 (1) Berl. Monatsber. p. 285, 1876. -Wied. Beibl. 4, p. d832, 1880. 
 
 (2) Wied. Ann. 15, p. 246, 1882. -15, p. 254, 1882. 
 
 (3) E. Wiedemann & H. Ebert, Wiod. Ann. 4H, p. 158, 1892. 
 
 (4) Wied. Ann. 15, p. 277, 1882. 
 
CHAPTER II.OTHER OBSERVERS 21 
 
 he could not find the exact expression for the latter function, 
 he succeeded in showing that Sir W, Crookes was wrong in 
 supposing it to be where p represents the density of the 
 
 gas, 
 
 In his earlier papers, Goldstein (1) had already noticed that 
 the regular kathode could br> replaced in a tube by any sub- 
 stance perforated with a number of small holes, Somewhat la- 
 ter,!; 2) while working whith a kathode of this nature,he discover- 
 ed near the kathode some rays not deflected by a magnet, From 
 the place of their origin he called them "Canal Rays", 
 
 III. OTHER OBSERVERS. 
 
 Although Hittorf and Goldstein did the most important work 
 during this period, they are far from being the only ones who 
 labored in this extensive and interesting field, There are even 
 some whose work received much greater notice than theirs, Sir 
 William Crookes for instance, This was due partly to the fact 
 that the discoveries of Crookes were of a kind that readily 
 appealed to the popular mind ; whereas the others were too abs- 
 truse to receive general attention. 
 
 The impossibility of giving a detailed analysis of the work of 
 all the experimenters makes it advisable to group their many 
 observations under different headings. This will avoid all 
 unnecessary repetitions and at the same time give a more 
 general view of the subject under consideration. 
 
 The observed facts may be grouped into four classes : 
 
 1. The general phenomena and the properties of the dis- 
 charge. 
 
 2. The factors of the discharge and their mutual relations. 
 Under this heading will be grouped the conductivity of gases, 
 
 (1) Wied. Ann. 11, p. 832, 1880. 
 (2) Berl. Sitzungsber. 37, p. 691, 1886. 
 
22 CHAPTER n GENERAL PHENOMENA AND PROPERTIES 
 the discharge-potential, the fall of potential in the different 
 parts of the tube, the current-intensity, the influence of the gas- 
 pressure and the size, shape and nature of the electrodes. 
 
 3, The external causes that influence the discharge or the 
 accompanying phenomena, such as resistance, heat, a magnetic 
 field, an electric field, ultra-violet light, etc. 
 
 4. The effects of the discharge and of the different rays pro- 
 duced in a vacuum tube, such as mechanical, chemical and 
 optical effects, secondary radiation, etc. 
 
 1, GENERAL PHENOMENA AND PROPERTIES. 
 
 The general appearances in the tube and the various charac- 
 ters capable of being assumed by the discharge again claimed a 
 certain amount of attention. They were studied under slightly 
 altered circumstances but the results failed to reveal anything 
 new, hence it will be sufficient to make but a mere reference to 
 them, (1) 
 
 During this period the question as to whether the electrical 
 discharge was continuous or intermittent in character was 
 extensively considered. G. Wiedemann and R. Ruehlmann, (2 ) 
 while studying the light in the tube by the use of a heliometer, 
 reached the conclusion that the discharge was always intermit- 
 tent. The same view was held by E. Wiedemann, (3) E. Gold- 
 stein, (4) Warren de la Rue and H. Mueller, (5) and E. Fer- 
 net. (6) But Hittorf (7) and Hertz (8) showed rather definite- 
 
 (1) O. Lehmann. Wied. Ann. 11, p. 686, 1880. 22, p. 305, 1884. Warren de 
 la Hue and H. Mueller. Proc. Roy. Soc. 35, p. 292, 1883. Crookes. Chem. 
 News. 39, p. 155, 1879.-63, 1891. C. Chree, Proc. Phil. Soc. Cam. 7,p.222,1891. 
 
 (2) Pogg. Ann. 145, p. 235, 1872. 
 
 (3) Wied. Ann. 10, p. 245, 1880. 
 
 (4) Wied. Ann. 12, p. 101, 1881. 
 
 (5) Ann. de Ch. et de Phys. 24, p. 461, 1881. Phil. Trans. 169, p. 225 
 1878. C. R. 86, p. 1072, 1878. 
 
 (6) C. R. 90, p. 680, 1880. 
 
 (7) Wied. Ann. 7, p. 553, 1879. 
 
 (8) Wied. Ann. 19, p. 782, 1883. 
 
CHAPTER n. FACTORS OF THE DISCHARGE 23 
 
 ly that a battery discharge through gases was not to be looked 
 upon as discontinuous any more than if it occurred through an 
 entirely metallic circuit. 
 
 J. T. Bottomley (1 ) described the property of a vacuum tube 
 to act like a condenser when held in one hand or surrounded by 
 a conductor. 
 
 The discharge in the tube is propagated with a finite velocity. 
 This was already studied by Wheatstone. J. J. Thomson (2) 
 also investigated this velocity and determined its value for the 
 propagation of the positive light towards the kathode. This 
 value was found to be about half that of the velocity of light, 
 viz. 1.6 X 10 10 cm/sec. 
 
 Most of the other observations refer to the kathode rays and 
 their effects, mainly to fluorescence. Crookes (3) described the 
 apparent mutual deflectibility of two kathode rays and the 
 peculiar phenomenon known as the tiring of the glass, a pro- 
 perty which prevents the glass from fluorescing with the same 
 intensity after having been exposed for some time to the 
 influence of the kathode rays. This tiring is very persistent. 
 
 Spottiswoode and Moulton (4) obtained peculiar shadow 
 effects which they explained as interference of the kathode 
 rays. 
 
 2. FACTORS OF THE DISCHARGE. 
 
 The resistance of a gas was shown by G. Wiedemann (5) to 
 be independent of the cross-section of the column of gas. War- 
 ren de la Rue and Hugo Mueller (6) found that there is no 
 
 (1) Nat. 23, p. 218, 1880. 
 
 (2) Proc. Roy. Soc. 49, p. 84, 1891. 
 
 (3) Chem. News, 39, p. 155, 1879. 
 
 (4) Wied. Beibl. 8, p. 73, 1884. 
 
 (5) Pogg. Ann. 158, p. 53, 1876. 
 
 (6) C. R. 86, p. 1072, 1878. 
 
24 CHAPTER n FACTORS OF THE DISCHARGE 
 
 relation between the E. M. F. and the current-intens'ty in rare- 
 fied gases. Hittorf (1) also reached the same result and, with 
 regard to the positive light in particular, stated that the 
 difference of potential between any two cross-sections does not 
 depend on the intensity of the current, He drew the conclusion 
 that the conductivity in the positive light increased proportio- 
 nately to the current-intensity. 
 
 The question concerning the conductivity of a complete va- 
 cuum was again discussed during this period. J. Puluj (2) 
 thought that a vacuum would be a conductor and that the 
 resistance of high vacua is due to the electrostatic charge on 
 the walls of the tube. According to Edlund" also, there is no 
 resistance in a vacuum. But to this, K. Krayewitsch (3) 
 objected that in extreme vacua the discharge-potential again 
 varies with the distance between the electrodes and that highly 
 evacuated funnel-tubes still show a great difference of conduc- 
 tivity with different connections. 
 
 The distribution of resistance in the tube is far from being 
 uniform. It is found mostly near the electrodes and mainly on 
 the kathode, as shown by Hittorf, (4) Warren de la Rue and 
 Hugo Mueller (5) and E. Goldstein. (6) The last named author 
 stated very clearly that the real resistance in a highly evacuated 
 tube is on or in close proximity to the kathode surface. This 
 resistance can be reduced in several ways: one is by using a 
 kathode which can be heated to white heat; another ingenious 
 method adopted by Goldstein was to use an inverted U tube 
 on whose electrodes some cadmium was placed. If the exhaus- 
 tion was too high, the discharge again took place as soon as a 
 
 (1) Wied. Ann. 7, p. 573, 1879. 20, p. 729, 1883. 
 
 (2) Sitzungsber. der Wien. Akad. der Wiss, 2te Abt. 85, p. 871, 1882. 
 
 (3) Rep. der. Phys. 19, p. 118, 1883. 
 
 (4) Wied. Ann. Jubelb. p. 430, 1874. 
 
 (5) Proc. Hoy. Soc. 35, p. 292, 1883. 
 
 (6) Wied. Ann. 24, p. 79, 1885. 
 
CHAPTER n FACTORS OF THE DISCHARGE 25 
 
 little of the cadmium had been evaporated by heating, although 
 cooling devices, whose efficiency was tested by the spectroscope, 
 were used to confine the vapor to the immediate neighborhood 
 of the kathode. 
 
 This great resistance at the kathode means a rapid fall of 
 potential in the same region: under this aspect it was also 
 studied by A, Righi. (1) 
 
 The conductivity of gases in general began to receive con- 
 siderable attention during this period. E. Becquerel (2) had 
 found that gases begin to be conductors of electricity when they 
 become red hot. G. Wisdemann (3) extended this discovery 
 by finding that the heating of a tube considerably lowers the 
 amount of electricity necessary for a discharge. 
 
 R. Blondot, (4) characterized the conductivity of hot gases as 
 galvanic. The conductivity of flames also was investigated by 
 W. Hittorf, (5) A. Macfarlane and D. Rintoul, (6) Mascart (7) 
 and Angelo Nob. Emo. ( 8) In the same article the last men- 
 tioned author treated likewise the conductivity of damp air, 
 which subject was also taken up by Macfarlane and Rintoul. (9) 
 
 In many respects the resistance of gases is different from 
 that of solid or liquid conductors. The smallest electro-motive 
 force is sufficient to set up a current in a solid conductor, 
 whereas to obtain a current in gases, the electric tension at the 
 electrodes must first attain a definite value, which varies with 
 the nature, density and temperature of the gas. G. Wiedemann 
 
 (1) Nuovo Cim. 8, p. 93, 1880. 
 
 (2) Ann. Ch. Phys. (3) 39, p. 377, 1853, 
 
 (3) Fogg. Ann. 158, p. 68, 1876. 
 
 (4) C. R. 92, p. 870, 1882. 
 
 (5) Wied. Ann. Jubelb. p. 430, 1874, 
 
 (6) Proc. Ed. Roy. Soc. p. 567, 1882. 
 
 (7) C. R. 95, p. 917, 1882. 
 
 (8) Riv. Sclent. Ind. Firenze, 15, p. 67, 1883. 
 
 (9) Proc. Ed. Roy. Soc. p. 801, 1882. 
 
26 CHAPTER n FACTORS OF THE DICHARGE 
 
 and R. Ruehlmann (1) studied this potential and found that a 
 higher one was required to start the discharge at the positive 
 than at the negative electrode. It was also investigated by 
 Macfarlane acting alone (2) and in connection with R. Simp- 
 son (3) and P. M. Playfair. (4) W. Hittorf held the view that 
 some finite potential was always required to set up a current in 
 a gas. But A. Schuster ( 5) attacked this opinion, maintaining 
 that the smallest electro-motive force is sufficient to bring about 
 this current. The fact, that in reality a very great electro- 
 motive force is required, would be due merely to the resistance 
 at the electrodes. If this could be obviated, the current, it was 
 supposed, would pass readily. A means of doing away with the 
 electrodes was found by J. Moser (6) who succeeded in exciting 
 a tube by the induction produced in the tube itself. Contrary 
 to the expectations of many, high vacua resisted this current as 
 well as the ordinary electrode current. J. J. Thompson ( 7 ) 
 also studied this new way of obtaining a current in a vacuum 
 tube; curiously enough he never succeeded in obtaining any 
 striation by this method. 
 
 The electricity necessary for the current may also be supplied 
 by influence as had already been shown by Pluecker. (8) This 
 was again treated by J. T. Bottomley, (9) apparently without 
 any knowledge of Pluecker 's work. 
 
 E. Wiedemann (10) studied the effect of varying the distance 
 between the electrodes at constant pressure; he succeeded in 
 
 (1) Pogg. Ann. 145, p. 235, 1872. 
 
 (2) Ed. Roy. Soc. Trans. 28, p. 633, 1877. 
 
 (3) Ed. Roy. Soc. Trans. 28, p. 673, 1877. 
 
 (4) Ed. Roy. Soc. Trans. 28, p. 679, 1877. 
 
 (5) Proc. Roy. Soc. 42, p. 371, 1887. 
 
 (6) C. R. 110, p. 397,1890. 
 
 (7) Electrician, 27, p. 139, 1890. Phil. Mag. 32, pp. 321, 445, 1893, 
 
 (8) Pogg. Ann. 107, p. 81, 1859. 
 
 (9) Electrician, 28, p. 463, 1892. 
 
 (10) Wied. Ann. 20, p. 756. 1883. 
 
CHAPTER n EXTERNAL INFLUENCES 27 
 
 arranging his apparatus in such a manner as to allow him to 
 vary the distance between the electrodes continuously. As the 
 anode moves towards the kathode the positive layers do not 
 change their position but disappear one after the other as the 
 anode enters them. When the last layer has disappeared and the 
 anode has entered the dark space, a brush appears on the latter. 
 This brush is bent back when the negative glow is entered. As 
 soon as the kathode dark space is reached, the negative light 
 appears back of the kathode and fluorescence is seen on the 
 glass behind the same. 
 
 3. EXTERNAL INFLUENCES. 
 
 Some new facts were discovered concerning the influence 
 of a magnet on the electric discharge and the kathode rays. 
 J. J. Thomson (1) verified those already known, parti- 
 cularly those dealing with the production of striation in 
 the positive light and found, moreover, that the place at which 
 the negative glow appears on the electrode is changed by a 
 magnet. W. Spottiswoode and J. F. Moulton (2) discovered 
 the interesting fact that under some circumstances, the kathode 
 rays are brought to the walls of the tube, not at one particular 
 spot but at a whole series of them, as was proved by the fluores- 
 cence; in their case the magnet was put close to the kathode. 
 
 Pluecker ( 8) had already described the effects that take place 
 in the tube when the walls are touched by the hands of the 
 operator. When the finger or some other conductor is brought 
 near the tube, the light therein is generally attracted but some- 
 times repelled. Pluecker thought he could trace this difference 
 to the nature of the gas and the special form of the tubes. But 
 this was shown to be wrong by Edm. Reitlinger and Alph. von 
 
 (1) Proc. Camb. Phil. Soc. 5, pt. 6, p. 391, 1886 
 
 (2) Proc. Roy. Soc. Lon. 32, p. 388, 
 
 (3) Pogg. Ann. 104, p. 121, 1858. 
 
23 CHAPTER n- EFFECTS OF THE DISCHARGE 
 
 Urbanitzki (1) who named the double phenomenon ^electro- 
 attraction" and "electro-repulsion" respectively, They found 
 that the effect depended mainly on the pressure of the gas, At 
 high pressures there was electro-attraction, but by gradually 
 lowering the pressure a point was reached where no effect was 
 noticed; at all points below this, electro-repulsion occurred, 
 The effect is influenced, moreover, by the current-strength 
 and by resistance inserted in the circuit. It is of an electric 
 nature and preceded by some other effect due to influence, as 
 the conductor is brought near the tube, but it cannot come 
 from the presence of free electricity since it is not produced 
 by a charged non-conductor. 
 
 Another result brought about by approaching a conductor 
 is the production of a new kathode at the very place where 
 the tube is touched, This was noticed by K, Domalip, (2) 
 W. Spottiswoode and J, F. Moulton, (3) The two last named 
 authors also observed that this new kathode showed a dark 
 space and possessed the other properties of an ordinary 
 kathode. 
 
 4, EFFECTS OF THE DISCHARGE, 
 
 HEATING G-, Wiedemann and E. Kuehlmann (4) learned 
 that the heat produced by a current in rarefied gases varies 
 directly as the current-intensity and not as the square of this 
 same current-intensity, as required by Joule's law for solid 
 conductors. Subsequently G. Wiedemann (5) studied the 
 heating effect of the current more extensively. The heat 
 produced on capillary tubes increases as the gas-pressure in- 
 
 (1) Wiener Anz. p. 100, 1877. -Wied. Ann. 10, p. 574, 1880.-13, p. 670 
 
 1881. 
 
 (2) Sitzungsber. der K. Bochm. Akad. der Wise. July, 2d, 1880. 
 
 (3) Proc. Roy. Soc. 29, p. 21, 1879.^PM1. Trans. 1879, 1880. 
 
 (4) Pogg. Ann. 145, p. 237, 1872. 
 
 (5) Pogg. Ann. 158, pp. 55, 252, 1876. 
 
CHAPTER n EFFECTS or THE DISCHARGE 29 
 
 creases, but somewhat more slowly. The heating on any 
 cross-section of a long or small capillary tube is almost the 
 same between large limits of the respective lengths of the tube. 
 Tubes of different inner but same outer diameter receive 
 nearly equal quantities of heat, but the temperature is not 
 equal in the several parts of the same tube:-that in the dark 
 space being notably lower than that of the brighter parts. 
 From his studies of the heat effects in rarefied gases, E. 
 Wiedemann (1) reaohed the following conclusions: The 
 spectrum given by the light in a vacuum tube is not necessa- 
 rily due to high temperature; a spectrum may be obtained 
 even at temperatures below 100 0. If these statements are 
 referred to the average temperature of the gas at any place in 
 the column of light, they are, no doubt, true. But from this 
 it would be wrong to conclude that a light-emitting atom or 
 other small particle of gas does not exceed these temperatures. 
 The low values given by E. Wiedemann contrast rather strongly 
 with those given later by Paalzow and Neessen, (2) who found 
 that the temperature in rarefied gases ranged between 10 000 
 and 100 000 under the influence of the electric discharge. Wie- 
 demann also found that the passing from the band spectrum to 
 the line spectrum requires 128 300 calories per gram for hydro- 
 gen, and in connection with this that the heat of dissociation 
 of the hydrogen molecule is 126 000 Gr. D. units of heat. The 
 study of the relative temperature of the different parts of the 
 
 (1) Wied. Ann. 5, p. 500, 1878. 6, p. 298, 1979.-10, p. 202, 1880. 20, p. 
 
 756, 1883. 
 
 (2) Verb, der Ges. D. Naturf and Aerzte, 63 Vers. zu Bremen, p. 51, 1890. 
 
30 CHAPTER n EFFECTS OF THE DISCHARGE 
 tube gave the general results represented by the following cur- 
 
 ve: 
 
 a 
 
 s 
 
 a/ 
 H 
 
 o 
 
 Kathode 
 
 Anode 
 
 Immediately near the kathode the temperature is very high; 
 but this part of the curve could not be traced very accurately. 
 The maximum always falls well within the kathode light and 
 consequently its position depends on the extension of this light. 
 
 D. Groldhammer (1) learned that the total heat developed 
 under the influence of the electric discharge depends in a great 
 measure on the current-intensity but not on the gas-pressure 
 whereas the relative temperature of the electrodes depends on 
 the gas-pressure. As this pressure is increased, the tem- 
 perature at the kathode at first becomes equal to and then lar- 
 ger than that at the anode. This heating of the electrodes was 
 also studied by A. Naccari and G. Guglielmo. (2) 
 
 The heat developed by the striking of kathode rays may be- 
 come intense enough to melt glass and platinum. (3) 
 
 The kathode rays produced in high vacua give rise to a vivid 
 
 (1) Journ. der Russ. Phys. Chem. Ges. 6, p. 325,1884. 
 
 (2) Nuovo Cim. 17, p. 1, 1835. Atti della R. Ace. di Torino, 20, p. 263, 
 1885. 
 
 (3) W. Crookes, Chem. News, 39, p. 155, 1879 H. A. Cunningham, Nat. 
 19, p. 458, 1879. 
 
CHAPTER n EFFECTS OF THE DISCHARGE 31 
 fluorescence on the glass of the tube and in many substances. 
 This was explained by Crookes ( 1 ) as due to the impinging of 
 radiant matter; by Goldstein (2) as a result of the ultra-violet 
 light produced by the kathode rays ; by Puluj (3) as coming from 
 an action of the ether carried along by the moving particles on 
 the ether on the surface of the fluorescing bodies. 
 
 The spectrum given by these fluorescing substances is gen- 
 erally -continuous but Crookes (4) found that some of the rare 
 earths give a band spectrum. On this observation he founded 
 a new branch of spectrum-analysis. 
 
 In some cases, the continued impinging of the kathode rays on 
 a fluorescing substance causes a distinct change of color; in 
 others different spectra may be given by the same substance. 
 Crookes claimed for these bodies a different fluorescence in dif- 
 ferent vacua. But E. Wiedemann (5) showed that all the facts 
 could be easily explained by a change in the physical or chemi- 
 cal condition of the body; for instance, by the loss of water of 
 crystalization which would naturally increase with the vacuum. 
 
 Crookes and Hittorf obtained the condition of vivid fluores- 
 cence only at very high vacua and in, the opinion of the former 
 especially, these high vacua where matter became radiant were 
 absolutely required. In connection with this it may be in- 
 teresting to note that D. Goldhammer (6) began to obtain the 
 same conditions at pressures ranging from 1.2 to 0.9 mm of 
 mercury. 
 
 A peculiar case of fluorescence was found in the so-called 
 after-glow of some gases, which had already been extensively 
 
 (1) Chem. News, 33, p. 153, 1379. 
 
 (2) Wien. Ber. 80, 1879. 
 
 (3) Wien. Ber. 81, April 15, 1880. 
 
 (4) Chem. News, 47, p. 261, 1883. Radiant Matter Spectroscopy. Phil. 
 Trans. 1888-1885. 
 
 (5) Wied. Ann. 9, p. 157, 1880. 
 
 (6) Jour, der Russ. Phys. Chem. Ges. 6, p. 325, 1884. 
 
32 CHAPTER n EFFECTS OF THE DISCHARGE 
 
 studied in the previous period. Since then some attempts had 
 been made to prove that this effect was due to secondary dis- 
 charges from the walls of the tube; but E. Warburg (1) showed 
 that the older explanation, which attributed it to a chemical 
 action, was still more satisfactory. 
 
 MECHANICAL EFFECTS. A.de la Rive (2) succeeded in rotating 
 a pair of vanes mounted in a vacuum tube by directing the dis- 
 charge on them with the aid of a magnet. Six years later Sir 
 W. Orookes ( 3 ) illustrated the same property of the kathode 
 rays by several striking experiments. But it must be remarked 
 here that, although the so called bombardment may contribute 
 to the rotation, it is not the only cause of it. Thus W.Hittorf (4) 
 showed that the heat produced on the radiometer- vanes is of itself 
 sufficient to cause rapid rotation. Moreover the phenomenon is 
 still complicated by an electrostastic action between the char- 
 ged walls of the tube and the radiometer itself. Another ar- 
 rangement of these experiments, made both by Crookes and 
 Puluj, (5) is to make the wheel of the radiometer the kathode 
 of the discharge by covering every alternate side with a fluores- 
 cing screen so that the kathode rays will be produced only on 
 the opposite sides. But this is still more complicated than the 
 previous one since the heat produced on the walls of the tube 
 by the impinging kathode rays is again a new cause in- 
 fluencing the rotation. 
 
 CHEMICAL EFFECTS. Many chemical compounds iindergo chan- 
 ges when subjected to the influence of an electric discharge 
 through gases ; but it is hard to determine how far these changes 
 are due to purely electrical causes since there cannot be any 
 doubt that the intense heat caused by the electric discharge 
 
 (1) Arch, de Gen. 12, p. 504, 1884. 
 
 (2) Ann. de Ch. et de Phys. 29, p. 207, 1873. 
 
 (3) Phil. Trans, p. 152, 1879. 
 
 (4) Wied. Ann. 21, p. 125, 1884. 
 
 (5) Radiant Electrode Matter. Phya. Soc. Heprint of Memoirs, p. 275. 
 
CHAPTER n EFFECTS OF THE DISCHARGE 33 
 
 and particularly by the kathode rays is a great factor in che- 
 mical change in general. 
 
 In a very great number of gases the more complex molecules 
 are broken up into simpler ones and quite frequently into the 
 simplest possible. But conversely, if a mixture of different 
 gases is exposed to the discharge, chemical combination may 
 take place between them. These several changes are not con- 
 fined to gases, but may also be effected in solids with which 
 the gases come into contact: thus a chemical action has 
 sometimes been noticed on the constituents of the glass-walls 
 and the electrodes of a vacuum tube. 
 
 A third action of a chemical nature is seen in the frequent 
 changing of oxygen to ozone. Phosphorus is similarly affected. 
 
 Moreover there is evidence of a transfer of gases from one 
 electrode towards the other; this often assumes an electrolytic 
 character and may cause a different degree of vacuum in the 
 different parts of the tube. 
 
CHAPTER III 
 
 THIRD PERIOD 
 
 The interest daring this period centers mainly around the ka- 
 thode rays and the several causes that bring about the conduc- 
 tivity f gases. The nature of the former was largely discussed, 
 their properties were studied in a masterly way by Ph. Lenard, 
 their effects and particularly secondary radiation and Roentgen 
 rays opened new fields for investigation. The stupendous 
 amount of the literature on this general subject during this pe- 
 riod will not allow of more than a passing reference to most of 
 the work. Physicists all over the world have investigated the 
 laws governing the discharge of electricity through gases; they 
 have tried to build up a theory which would account for all 
 known facts and have investigated many new phenomena. 
 
 This chapter will be devoted mainly to the historical review 
 of the question of conductivity and to the study of kathode 
 rays. In the latter part, the work of Lenard will receive special 
 attention on account of its importance. 
 
 I. CONDUCTIVITY OF GASES. IONIZATION. 
 A. INFLUENCE OF ULTRA-VIOLET LIGHT ON GASES. 
 In studying the resonance-phenomena between very rapid 
 electric oscillations, H. Hertz ( 1 ) accidentally discovered that 
 
 (1) Wied. Ann. 31, p. 983, 1887. 
 
CHAPTER in. CONDUCTIVITY OF GASES 35 
 
 one spark greatly facilitated the formation of another. He 
 traced the effect to ultra-violet light and expressed the desire 
 that this subject should be studied under simpler conditions. 
 This wish received the response that such an important disco- 
 very deserved. Eilhard Wiedemann and H. Ebert (1) were the 
 first to publish the results of their investigations. Ultra-violet 
 light may facilitate the discharge in the proportion of 2:1. The 
 greatest effect is noticed when the kathode is illuminated; the 
 illumination of the anode gives much smaller results and when 
 the action of the ultra-violet light is confined to the intervening 
 gas by carefully excluding reflection, the effect does not take 
 place. The active light lies mainly beyond the visible spectrum 
 but for carbon dioxide the maximum effect was obtained with 
 the wave-lengths between the Gr and K lines. The part of the 
 arc which proved most effective is that on or in the immediate 
 vicinity of the positive carbon. The effect was explained as a 
 sympathetic action between the ultra-violet light-waves and the 
 very short wave-lenghts which, in their theory, are the kathode 
 rays. Experimenting with different metals and liquids they 
 found that, when the kathode is a substance which readily ab- 
 sorbs ultra-violet light, the effect is greatest: thus among the 
 ordinary metals the most effective was found to be platinum. 
 Pure water produced little effect, whereas solutions showed a 
 result in proportion with their absorptive power. Aniline dyes, 
 which readily absorb ultra-violet light, produce very strong 
 effects. 
 
 Hallwachs (2) studied the question from another standpoint 
 and thus discovered what has been known ever since as the 
 "Hallwachs Effect". A negatively charged body is rapidly and 
 in some cases almost instantaneously discharged under the iii- 
 
 (1) Wied. Ann. 33, p. 241, 1888. 35, p. 209, 1888. 
 
 (2) Wied. Aun. 33, p 301, 1888. 34, p. 731, 1888. -37, p. 666, 1889. 
 
36 CHAPTER m CONDUCTIVITY OF GASES 
 
 fluence of ultra-violet light. Hallwachs has definitely esta- 
 blished that this effect is due to ultra-violet light and that it 
 occurs at the surface of the charged body. If a 
 positively charged body is put into the path of the 
 light, the loss of electricity is scarcely observable but if the 
 body is not electrified, it will acquire a small positive charge. 
 This is owing to the negative electricity liberated under the 
 influence of the ultra-violet light. Whilst working with electri- 
 fied liquids, the same experimenter confirmed E. Wiedemann 
 and H. Ebert's conclusion that there is a direct relation bet- 
 ween the absorptive power of a body for ultra-violet light and 
 the rate of discharge. 
 
 Other important work was done in this line by A. Righi, (1) 
 Stoletow (2) JElster and G-eitel, (3) Lenard, (4) Hoor, (5) and 
 many other physicists. (6) The principal facts brought out 
 
 (1) C. R. 106, p. 1349, 1888. 107, p. 559, 1888. Jour, de Phys. 7, p. 153, 
 1888. Rend, della R. Ace. dei Lincei, 4, p. 16, 1888.^-6, p. 185, 1888. 
 Exner Rep. 25, p. 185, 380, 1889. Atti del R. 1st. Yen. 7, 1889. Mem. Bol. 
 9. 1888. 10, p. 85, 1890. 
 
 (2) C. R. 106, p. 1149, 1888. 106, p. 1593, 1888. 107, p. 91, 1888. 108, p. 
 1241, 1889. Jour, der Rusa. Phys. Chem. Ges. 21, p. 159, 1889. Jour, de 
 Phys. 9, p. 468, 1890. Bull. Soc. Fr. de Phys. p. 202, 1890. Phys. Rev. 1, 
 p, 721, 1892. 
 
 (3) Wien. Ber. 97, p. 1175, 1888. Sitzungsb. der Wien. Ak. Math. 
 Naturw. Cl. 99, p. 1008, 1891. Wied. Ann. 38, p. 40, and 497, 1889. 39, p. 
 321, 1890. 41, p. 161, 1890. 42, p. 564, 1891. 43, p. 225, 1891. 44, p, 722, 
 1891. 46, p. 281, 1892. 48, p. 625, 1893. 52, p. 432, 1894. 55, p. 684, 1895.- 
 57, p. 24 and 401, 1896. 62, p. 599, 1897. 
 
 (4) Lenard and Wolf, Wied. Ann. 37, p. 443, 1889,-^Lenard, Wieii. Ber. 
 108, p. 1648, 1889. 
 
 (5) Wien. Ber. 97 p. 719, 1888. Exner Rep. 25, p. 105, 1889. 
 
 (6) Bichat and Blondlot, C. R. 106, p. 1349, 1888. Bichat, C. R. 107, p. 
 557. 1888. Arrhenius, Wied. Ann. 33, p. 638, 1888. Borgraann, C. R. 108, 
 p. 733, 1889. Branly, C. R. 110, p. 751 and 898, 1890. 116, 741, 1893. 
 114, p. 68, 1892. -120, p. 829, 1895. Lum. El. 41, p. 143, 1891. Jour, de 
 Phys. 2, p. 300, 1893. Precht, Wied. Ann. 49, p. 150. 1893. Cantor, Wien. 
 Ber. 102, p. 1188, 1893, Warburg, Wied. Ann. 59. p. 1, 1896. - Drude 
 Ann. 5, p. 811, 1901. Simon, Wien. Ber. 104, p. 565, 1895. Jaumann, Wied- 
 Ann. 62, p. 396, 1897. G. C. Schmidt, Wied. Ann. 62, p. 407, 1897. 64, p. 
 
CHAPTER in CONDUCTIVITY OF GASES 37 
 
 by this study are the following: The effect may be produced 
 by any source of light that will furnish ultra-violet rays. Those 
 rays that cause the discharge are always absorbed but the ab- 
 sorption is not necessarily accompanied by the discharge. The 
 effect of polarized light depends to a great extent on the posi- 
 tion of the plane of polarization, showing two maxima and two 
 minima. The minima occur when the plane of polarization 
 coincides with that of incidence for the ray of light. The max- 
 ima occur for positions which differ from the previous ones by 
 an angle of 90 degrees. 
 
 The loss of electricity is also greatly influenced by the physi- 
 cal condition of the surface experimented on, being greater for 
 highly polished surfaces. The surfaces themselves are attacked 
 and sensibly roughened by the action of ultra-violet light. This 
 phenomenon led to the almost general belief that the discharge 
 was brought about by the so-'caDed "Zerstaeubung" of the elec- 
 trode. Ph. Lenard (1) did not look upon this as probable and 
 began a series of experiments to discover the true nature of this 
 action of the shorter wave-lengths of light. He first construc- 
 ted a tube with a quartz window that allowed the ultra-violet 
 light to strike one of the electrodes. This electrode was made of 
 sodium amalgam and could be charged to any desired potential. 
 The opposite electrode was of platinum. After the platinum 
 electrode had been charged negatively by casting a beam of 
 ultra-violet light on the sodium amalgam, it was taken out and 
 tested for sodium in the spectroscope. Although 3X10 7 mg. of 
 
 708, 1898. Henry, Proc. Cambr. Phil. Soc. 9, p. 401, 1898. Buisson, C. R. 
 127, p. 224, 1898. 130, p. 1298, 1900. Zeleny, Phil, Mag. 45, p. 272, 1899 
 Rutherford, Proc. Cambr. Phil. Soc. 9, p. 401, 1898. Schweidler, Wien 
 Ber. 107, p. 881, 1898. 108, p. 273, 1899. J. J. Thomson, Phil. Mag. 48, 
 p. 547, 1899. Merritt and Stewart, Phys. Rev. 11, p. 230, 1900. Guthe, 
 Drude Ann. 5, p. 818, 1901. Kreusler, Drude Ann. 6, p. 398, 1901. 6, p. 
 412, 1901. E. Ladenburg, Drude Ann. 12, p. 558, 1903. 
 (1) Drude Ann. 2, p. 359, 1900. 
 
38 CHAPTER in CONDUCTIVITY OF GASES 
 
 sodium may easily be detected in this way, not the slightest 
 trace of it was noticed on the platinum. Another proof against 
 the Zerstaeubuiig theory clearly appears from the fact that if 
 the kathode is charged anywhere from 1 to 45.000 volts, the cou- 
 lombs discharged per second represent a constant quantity of 
 about 22.5X10' 10 . By inserting a screen with a hole in the cen- 
 tre between the kathode and the antikathode, he still obtained 
 negative electrification on the latter from the beam of rays that 
 could reach it. Applying a magnet to this beam, he saw he 
 could deflect it in such a manner as to cause electrification solely 
 at the place to which it had been deflected. From this he con- 
 cluded that ultra-violet light produces kathode rays on negati- 
 vely electrified bodies which it discharges. The velocity of 
 these kathode rays is smaller than that of those produced in an 
 ordinary vacuum tube and changes with the potential to which 
 the kathode is charged. The following table gives some of the 
 values : 
 
 Charge of the kathode in volts. Velocity. 
 
 607 0.12X10 10 cm per second- 
 
 4 380 0.32X10 10 " " 
 
 12600 0.54X101 
 
 The maximum velocity of the rays obtained by this method 
 was 10 8 cm per second. 
 
 Ultra-violet light has also a direct effect on gases, serving to 
 make them conductors by producing in them what Lenard has 
 styled "Electric Carriers". This process is more generally 
 known as "lonization" and was studied by Arrhenius, ( 1 ) Bran- 
 ly (2) and especially by Ph. Lenard. 
 
 Lenard (3) was the first to find that the very short wave- 
 lengths between 0.00014 mm and 0.00019 mm produce fogging, 
 
 (1) Wied. Ann. 33, p. 638, 1888. 
 
 (2) C. R. 110, pp. 751, 898, 1890. 120, p. 829, 1895. 
 
 (3) Drude Ann. 1, p. 486, 1900.,- 3, p. 298, 1900. 
 
CHAPTER HIINFLUENCE OF HEAT ON CONDUCTIVITY 39 
 form ozone and make the gas through which they pass a good 
 conductor. The best source of these rays is incandescent hydro- 
 gen though they are also contained in an aluminium spark gap, 
 the arc-light and the rays of the sun. A positively charged 
 body placed near the path of these rays is rapidly discharged. 
 This discharge is accelerated if the air acted on by ultra violet 
 light is directed towards the electrified body, Under the influ- 
 ence of this light both positive and negative carriers are pro- 
 duced. The initial velocity of the latter is much greater than 
 that of the former and according to Lenard's estimate ranges 
 from 10 7 to 10 8 cm per second. 
 
 B. INFLUENCE OF HEAT ON CONDUCTIVITY. 
 
 The increase of conductivity by heat, which had been known 
 and investigated during the previous period, still continued to 
 receive a considerable amount of attention. (1) In connection 
 with this subject it was learned that all gases may be classified 
 under two general heads, viz.: 1. Those that conduct with 
 difficulty even at the highest temperatures, as air, nitrogen* 
 carbon dioxide, ammonia, and the vapors of sulphuric acid, of 
 tin and of mercury. 2. Those that readily allow the passing 
 of a current, such as the halogens, hydriodic acid, hydrobromic 
 acid, hydrochloric acid, sodium chloride and potassiun chloride. 
 The cause of this diversity of behavior is probably the decomposi- 
 tion of the second class of gases and vapors into electric carriers. 
 Those of the first class are simply broken up into less compli- 
 cated molecules which take part in the transfer of electricity 
 only by convection. 
 
 (1) J. J. Thomson, Phil. Mag. 29, p. 359, 1890. 29, p. 441, 1890. Branly, 
 C. R. 114, p. 831, 1892. -114, p. 1531, 1892. Braun, Zeitschr. fuer Phys. 
 Chem. 13, p. 155, 1894. Pringsheim, Wied. Ann. 55, p. 507, 1895. Merrit 
 and Stewart, Phys. Rev. 7, p. 129, 1899. J. Stark, Wied. Ann. 68, pp. 931 
 and 942, 1889. Die Elektrizitaet in Gasen, Leipzig, 1902. Arrhenius. Wied. 
 Ann. 42, p. 18, 1891. 
 
40 CHAPTER in. INFL. OF KATHODE RAYS ON CONDUCTIVITY 
 
 Lenard (1 ) calculated the velocity of the positive carriers 
 or ions in a flame ( 2) and found that for lithium it was 108 cm 
 per second for 1 volt/cm. His method did not show any evi- 
 dence of negative carriers probably on account of their high 
 velocity. This was computed by M. Mareau (3) and shown to 
 be greater than 1200 cm/sec, per volt/cm. 
 
 C. INFLUENCE OF KATHODE RAYS ON CONDUCTIVITY. 
 
 The ionization of a gas under the influence of kathode rays 
 was studied chiefly by Ph. Lenard, (4) Des Coudres, (5) E. 
 Wiedemann and Gr, C. Schmidt, (6) Arnold, (7) McClennan, (8) 
 Townsend (9) and Durac. (10) 
 
 The absorption of the kathode rays is, according to Lenard, 
 proportional to the mass-lengths of a gas. Its effect is an ioni- 
 zation which remains in the gas for some time. McClennan 
 proved this ionization to be the same in all gases at the same 
 density. From this Lenard concluded that there is a relation 
 between the absorption of the ^kathode rays and the conductivi- 
 ty resulting therefrom. Durac found that the number of pairs 
 of ions produced in 1 cm of gas at 1 mm pressure is 0.43 for one 
 kathode quantity. This is about one fiftieth of the value that 
 had been previously given by Townsend. 
 
 (1) Drud Ann. 9, p. 642, 1902. 
 
 (2) cfr. also K. Wesendouck, Wied. Ann. 66, p. 121, 189*8. J. A. McClelland 
 Phil. Mag. 46, p. 29, 1898. H. A. Wilson, Phil. Trans. Roy. Soc. Lond. 192, 
 p. 499, 1899. 
 
 (3) C. R. 134, p. 1575, 1902. 
 
 (4) Wied. Ann. 56, p. 255, 1895. 63, p. 253, 1897. Drude Ann. 12, p. 449 
 1903. 
 
 (5) Wied. Ann. 62. p. 143, 1897. 
 
 (6) Wied. Ann. 62, p. 468, 1897. 66, p. 330, 1898. 
 
 (7) Wied. Ann. 61, p. 327, 1897. 
 
 (8) Zeitschr. f. Phys. Chem. 37, p. 513, 1901. 
 
 (9) Phil. Mag. 1, p. 198, 1901. 3, p. 557, 1902. 5, p. 389, 1903. 
 
 (10) Phil. Mag. 4, p. 29, 1902. 
 
CHAPTER in INJ?L. OF ROENTGEN RAYS ON CONDUCTIVITY 41 
 
 This ionization of the gas by kathode rays has furnished a 
 con sistent explanation of many of the phenomena that occur in 
 vacuum tubes, mainly with regard to the striated and the non- 
 striated positive light, the different dark and luminous layers 
 of the discharge near the kathode, etc. These different pheno- 
 mena will be discussed in the subsequent chapter; thus a mere 
 reference to the theoretical work done along this line will be 
 sufficient at this place. (1) 
 
 D. INFLUENCE OF ROENTGEN RAYS ON CONDUCTIVITY. 
 
 The ionization of a gas may be brought about directly by 
 Roentgen rays and indirectly by the so-called secondary radia- 
 tion which they produce. 
 
 The direct ionization by Roentgen rays was one of the first 
 properties of these rays observed by their discoverer. (2) 
 This property was so evident that Roentgen believed himself 
 justified in questioning Lenard's work on ionization by kathode 
 rays. The degree of ionization depends greatly on the intensity 
 of the rays and is proportional to the pressure of the gas. 
 
 The secondary rays produced at the surface of any body on 
 which Roentgen rays impinge, are generally better ionizing 
 agents than the primary rays themselves on account of their 
 higher coefficient of absorption. Ever since the day of the dis- 
 covery of these rays they have been largely used as a primary 
 or secondary source of ionization and studied as such. (3) 
 
 (1) J. J. Thomson, Phil. Mag. 50, p. 278, 1900. 1. p. 368, 1901. Stark, 
 Phys. Zts. 2, p. 664, 1901. Drude Ann. 3, p. 237, 1900. 5, p. 110, 1901. 7, 
 p. 426, 1902. Townsend, Nat. p. 340, 1900. Phil. Mag. 1, p. 198, 1901. 
 Townsend and Kirkby, Phil. Mag. 1, p. 630, 1901. 
 
 (2) Wied. Ann. 64, p. 12, 1898. Sitzb. Wuerzb. Phys. Med. Ges. 1895. 
 Beitrag. 
 
 (3) Roentgen, Wled. Ann. pp. 12, 18, 1898. Righi, C. R. 122, pp. 376, 
 601, 1896. -Rend, della R. Ace. dei Lincei, 5, p. 342, 1896. Mem. Bol. 5, p. 
 723,1896. J.J.Thomson, Proc. Roy. Soc. 09, p. 274, 1896. Proc. Cambr. 
 
42 CHAP, in INF. OF THE BECQUEREL RAYS ON CONDUCTIVITY 
 E. INFLUENCE OF THE BECQUEREL RAYS ON CONDUCTIVITY. 
 
 The so-called Becquerel rays, (1) which are a complexus of 
 very different rays, are emitted by the radioactive substances. 
 From such substances as are naturally radioactive, bodies have 
 been separated which exhibit this phenomenon in a marked 
 manner. These are actinium (separated by Debierne), polo- 
 nium and radium ( separated by the Curies ) . Of these the most 
 active is radium whose rays are divided into the , /5 and Y rays 
 which are distinguished by their penetrating power and their 
 deflectibility. The a. rays are the least and the Y the most 
 penetrating. The Y rays are not deflected by a magnet; the 
 /? rays are easily deflected and the a rays only with difficulty 
 and in a direction opposite to that of the /? rays. The Y ra y s 
 closely resemble those discovered by Roentgen; the /? rays 
 are negatively electrified particles traveling with great velocity 
 and hence may be looked upon as kathode rays. The rays 
 carry a positive charge. On bodies where the Becquerel rays 
 strike they produce a secondary radiation. (2) The primary 
 
 Phil. Soc. 10, p. 10, 1898. Id. and McClelland, Proc. Cambr. Phil. Soc. 
 9, p. 129, 1896. Id. and Rutherford, Phil. Mag. 42, p. 392, 1896.. Ruther- 
 ford, Phil. Mag. 43, p. 241, 1897. Id. and McClung, Proc. Roy. Soc. 67, p. 
 245, 1900. Perrin, C. R. 122, pp. 186, 716, 1896. Jour, de Phys. 5, p. 350, 
 1896. 6, p. 425, 1897. Ann. de Ch. et de Phys. 11, p. 496, 1897. Winkel- 
 mann, Wied. Ann. 66, p. 1, 1898. Sagnac, C. R. 125, pp. 168, 
 
 230, 942, 1897. 126, pp. 36, 467, 521, 887, 1898. 127, p. 46, 1898. 128, pp. 
 300, 546, 1899. Jour, de Phys. 8, p. 65, 1899. Id. and P. Curie, C. R. 130, 
 p. 1013, 1900. Jour, de Phys. 1, p. 13, 1902. Townsend, Proc. Cainbr- 
 Phil. Soc. 10, p. 217, 1900. Dorn, Arch. Neerl. 5, p. 595, 1900. J. A. Cunn- 
 ingham, Proc. Cambr. Phil. Soc. 11, p. 431, 1902. 
 
 (1) H. Becquerel, C. R. 122, 1896. 128 and 129, 1899. 130 and 131, 1900. 
 G. C. Schmidt, Wied. Ann. 65, p. 141, 1898. P. and S. Curie, C. R. 1898, 1899, 
 1900, 1901 and 1902. Id. and Bemont, C. R. 127, p. 1215, 1898. Rutherford, 
 Phil. Mag. 47, p. 109, 1899. Debierne, C. R. 129, p. 593, 1899. 130, p. 906, 
 1900. Giesel, Wied. Ann. 69, p. 91, 1899. Phys. Zts. 1, p. 16, 1899. Ber. 
 Chem. Ges. 33, p. 1665, 1900. 
 
 (2) H. Becquerel, C. R. 128, p. 771, 1899. 129, p. 716, J899. 132, pp. 371, 
 734, 1286, 1901. 
 
CHAPTER in CHEMICAL SOURCES OF IONIZATION 43 
 as well as the secondary rays have many interesting properties, 
 but here we are only concerned with their power to produce 
 ionization in gases. This ionization is always an effect of ab- 
 sorption and is different for the different kinds of rays. It 
 usually follows the same general laws as the other rays here- 
 tofore studied. ( 1 ) 
 
 F. CHEMICAL SOURCES OF IONIZATION. 
 
 Chemical reactions are very often accompanied by ionization 
 in the gases which enter into combination. In combustion we 
 may have the double process of direct ionization of the burn- 
 ing gases and their electrification by a solid introduced into 
 the flame. (2) 
 
 The discharge of an electrified body by moist air which had 
 been in contact with phosphorus is also in all likelihood due 
 to ionization resulting from a chemical action. (3) 
 
 Chemical decomposition by electrolysis or otherwise may 
 also give rise to ionization. Thus Townsend (4) has proved 
 that the rapid evolution of hydrogen from sulphuric acid and 
 iron causes the gas to assume a positive charge, leaving the so- 
 lution oppositely electrified. He likewise found positive charges 
 
 (1) H. Becquerel, C. R. 122, pp. 559, 689, 762, 1086, 1896. 123, pp. 856, 
 1896. 124, pp. 438, 800, 1897. G. C. Schmidt, Wied. Ann. 65, p. 141, 1898. 
 Rutherford, Phil. Mag. 47, p. 109, 1399. Strutt, Nat. 61, p. 539, 1900. Proc. 
 Roy. Soc. 68, p. 126, 1901. Elster and Geitel, Wied. Ann. 69, p. 673, 1899. 
 M. Cantor, Drude Ann. 9, p. 452, 1903. McClelland, Phil. Mag. 8, p. 67, 1904. 
 p. 230, Febr. 1905. 
 
 (2) Svante Arrhenius, Wied. Ann. 42, p. 18, 1891. Smithels, Dawson and 
 Wilson, Proc. Roy. Soc. 64, p. 142, 1899. H. A. Wilson, Phil. Trans. 192, p. 
 499, 1899. McClelland, Phil. Mag. 46, p. 29, 1899. Warburg, Drude Ann. 
 2, p. 304, 1900. Marx, Drude Ann. 2, p. 768, 1900. 
 
 (3) Bidwell, Nat. 55, p. 6, 1897. Barus, Phys. Rev. 10, p. 257, 1900. 
 Amer. Jour, of Sc. 11, p. 237 and 310, 1901. Phil. Mag. 1, p. 572, 1901 2, p. 
 40, 1901. 
 
 (4) Proc. Carabr. Phil. Soc. p. 244, 315, 1897. Phil. Mag. 45, p. 13-% 1898- 
 
44 CHAPTER in IONIZATION BY ZERSTAEUBUNG. 
 on chlorine and oxygen after their liberation. The rapid elec- 
 trolytic dissociation of sulphuric acid leaves hydrogen with a po- 
 sitive charge whereas oxygen does not show any electrification. 
 This difference is easily explained by the fact that the libera- 
 tion of oxygen is not a primary but a secondary electrolytic re- 
 action. The following equations show one of the ways in which 
 sulphuric acid may be electrolyzed and explain the different be- 
 havior of the two gases that are evolved. 
 
 H 2 SO 4 == H 2 -|- SO 4 
 
 2 H^O -|- 2 S0 4 = 2H 2 SO 4 -|- O 2 
 
 G. IONIZATION OF LIQUIDS BY ''ZEBSTAEUBUNG". 
 
 The anomalous electrification of air near water-falls had been 
 known for a long time. Lenard ( 1 ) showed that when water 
 or mercury is allowed to drop on a metallic plate, the liquid be- 
 comes positively electrified and the surrounding air shows ne- 
 gative electrification. Lord Kelvin (2) obtained a similar result 
 by allowing air to bubble through water. The Zerstaeubung 
 of pure water in air seems to produce only the negative carriers 
 or ions in the latter, while the slightest trace of a solvent in the 
 water causes both kinds of carriers to appear. The velocity of 
 the positive ions produced by pure water is given by Karl Kaeh- 
 ler (3) as 4. 17 cm/second per volt/cm. 
 
 This subject has received a good deal of attention quite 
 lately. (4) 
 
 (1) Wied. Ann. 46, p. 584, 1892. 
 
 (2) Lord Kelvin, McClean and Gait, Brit. Ass. Rep. 1894. Proc. Roy, 
 Soc. Lond. 57, p. 335 and 436, 1895. 
 
 (3) Drude Ann. 12, p. 1119, 1903. 
 
 (4) Elster and Geitel, Wied. Ann. 47, p. 496, 1892. Wesendonck, Wied. 
 Ann. 47, p. 529, 1892.^51, p. 353, 1894. J. J. Thomson, Phil. Mag. 37, p. 
 
CHAP, in PHENOMENA CONNECTED WITH THE DISCHARGE. 45 
 II. PHENOMENA CONNECTED WITH THE DISCHARGE, 
 
 During this period some new and important facts were dis- 
 covered such as the production of Roentgen rays from kathode 
 rays which was described by Professor Roentgen of Wuerzburg 
 in 1895. The most important work however consists in a closer 
 study of previously known phenomena. Thus, Lenard published 
 his papers on kathode rays which were honored and deservedly 
 so by many scientific associations. For among all the impor- 
 tant work done in the busy physical laboratories of our days, 
 that of Lenard stands out as a model of accurate investigation 
 and of scientific thought. 
 
 The other important kind of rays, named by Goldstein 
 "canal rays", also received their share of attention, mainly from 
 their discoverer and Willy Wien. 
 
 A. KATHODE RAYS. 
 
 We have already seen how a discovery made by H. Hertz led 
 to the extensive study of the effects of ultra- violet Irght on 
 gases. It likewise was another discovery of his that enabled 
 Lenard to bring the kathode rays out of the kathode tube 
 and thus study them under a great variety of conditions 
 unobtainable in the discharge tube. 
 
 Hertz (1) noticed that the walls of a vacuum tube still 
 fiuoresced if a thin sheet of gold leaf or aluminium foil were 
 interposed between them and the kathode. The fluorescing 
 spot thus obtained changed its position under the influence of 
 a magnet and showed all the other characteristics of the 
 
 341, 1894. 4, p. 352, 1902. Usener, Ztschr. fuer Phys. Chem. 18, p. 191, 
 1895. F. Himstedt, Ber. der Naturf. Ges. Freib. im Breisgau, April, 1903. 
 Pacini, Atti della R. Ace. del Lincei. 13, pp. 559 and 617, 1904. A. Schmaus 
 Drude Ann. 9, p. 224, 1903. 
 (1) Wied. Ann. 45, p. 28, 1892. 
 
46 CHAPTER in PATH OF THE KATHODE RAYS. 
 fluorescence produced by kathode rays. Thus it seemed to be 
 proved that kathode rays could traverse a thin sheet of metal. 
 
 Lenard fully appreciated the importance of this discovery 
 and set to work to utilize it. He finally succeeded in construct- 
 ing a tube with an aluminium window thin enough to allow 
 the kathode rays to go through, yet strong enough to stand the 
 atmospheric pressure and to keep a perfect vacuum. The alu- 
 minium which he used was 0,003 mm thick. 
 
 His first studies of the rays thus obtained were published 
 under the title:- "Ueber Kathodenstrahlen in Gasen von 
 atmosphaerischem Druck und im aeussersten Vacuum." (1) 
 The fact that up to then kathode rays could only be obtained 
 and studied between pressures which scarcely exceeded the 
 range of 1 mm is sufficient to suggest the full significance of 
 this title. The work of Lenard is very extensive and prac- 
 tically covers the whole subject, so that an analysis of it to- 
 gether with references to and an occasional addition from that 
 of other experimenters will be sufficient to give a complete 
 view of the subject. 
 
 1. PATH OF THE KATHODE RAYS. The path of the kathode 
 rays is generally rectilinear but it may be modified by interven- 
 ing obstacles and by electric and magnetic fields . 
 
 Thus the kathode ray issues from the aluminium window not 
 only normally but at every angle from to 90 degrees, and in 
 gases it suffers diffusion at all higher pressures. (2) This dif- 
 fusion increases with increasing pressure of the gas and is al- 
 ways greater for slow rays than for more rapid ones. This 
 general law holds for solids as well as for gases. The latter ab- 
 
 (1) Ber. der Berl. Akad. p. 3, 1893. Wied. Ann. 51, p. 225, 1894. 
 
 (2) Lenurd, Wied. Ann. 51, p, 225, 1894. See also: Goldstein, Wied. Ann. 
 51, p. 622, 1894. 67, p. 84, 1899. W. Kaufmann, Wied. Ann. 69, p. 95, 1899. 
 Stark, Phys. Ztschr, 2, p. 233, 1900; McLennan, Ztschr, f uer Phys. Chein. 
 p. 513, 1901. Seitz, Drude Ann. 6, p. 1, 1901. 
 
CHAPTER m PATH OF THE KATHODE RAYS. 47 
 sorb the rays and thus shorten their path. This absorption is 
 the same for equal mass-lengths of all substances. The extent 
 of a certain kind of rays was found to be 2 cm in air of 760 mm 
 pressure and 10 cm for hydrogen at the same pressure. For 
 all gases at the lowest obtainable pressure, (0.000 009 mm or 
 about 1/85 000 000 th of an atmorphere) this path was found to 
 be limited only by the end of the observation tube which was 
 150 cm long. 
 
 The absorption is not always the same but varies greatly with 
 the conditions in the kathode tube; thus showing that kathode 
 rays of different velocities must exist. 
 
 PATH OF RAYS IN A MAGNETIC FIELD.(!) A bundle of kathode 
 rays traveling through a magnetic field in a direction perpen- 
 dicular to the lines of force is deflected. This deflection is not 
 uniform for the whole beam of rays as appears from the fact 
 that the rays are separated into a so-called magnetic spectrum, 
 some being but slightly and others considerably deflected. This 
 again shows that the kathode rays are not all alike; their velo- 
 cities vary and probably may assume any values from to a 
 certain maximum which, according to Lenard, lies between 
 0.67 X10 10 and 0.81 X10 10 cm per second, i. e. less than 1/3 of the 
 velocity of light. The deflection of the kathode Tays is largely 
 dependent on the gas-pressure in the kathode tube but is in no 
 wise affected by the nature or the pressure of the gas in the ob- 
 servation tube. 
 
 (1) Lenard, Wieel. Ann. 51. p. 225. 1894. 52, p. 23, 1894. 56, p. 255, 1895. 
 Birkeland, C. R. 123, p. 492, 1896. 128, p. 228, 1898. Deslandres, C. R. 125, 
 p. 373, 1897. 126, pp. 997, 1323, 1897. 127, p. 1210, 1898. Wiechert, Wied. 
 Ann. 69, p. 739, 1899. J. J. Thomson, Phil. Mag. 44, p. 293, 1897. 48, p. 
 547, 1899. Proc. Cambr. Phil. Soc. 9, p. 248, 1895. 10, p. 49, 1900. Brauu. 
 Wied. Ann. 60, p. 552, 1897. 65, p. 368, 1898. W. Kaufmann, Wied. Ann. 
 61, p. 544, 1897. 65, p. 431, 1898. E. Wiedemann and Welmelt, Wied. Ann. 
 64, p. 606, 1898. W. Wien, Wied. Ann. 65, p. 440, 1898. Strutt, Phil. Mag. 48, 
 p. 478, 1809. II. A. Wilson, Proc. Cambr. Phil. Soc. 11, p. 179, 1901. 
 
48 CHAPTER in EFFECTS OF THE KATHODE KAYS. 
 
 PATH OF RAYS IN AN ELEOTEICAL FIELD. ( 1 ) If the path of the 
 kathode rays is perpendicular to the lines of force, it becomes 
 curved towards the positive plate. This deflection is directly 
 proportional to the fall of potential in the field and inversely to 
 the square of the ray's velocity. This law supplies one of the 
 most convenient methods of measuring the velocity of kathode 
 rays. 
 
 If the electric field is parallel to the direction of transmission, 
 it exerts a force tending to change the velocity of the kathode 
 rays. This change is positive if the kathode ray is directed 
 oppositely to the lines of force and negative if it travels along 
 them; or in other words, the kathode rays are attracted by and 
 accelerated in the direction of a positive plate, whereas they 
 are retarded if they are propagated in the direction of a negati- 
 ve plate. The most convenient way of showing this property 
 of the kathode rays is to allow them to pass through an opening 
 in the centre of two condenser-plates and then to observe the 
 change in their magnetic deflectibility. The velocity of katho- 
 de rays may be accelerated in an electric field to fully 1/3 that 
 of light. 
 
 2. EFFECTS OF THE KATHODE RAYS. One of the most notice- 
 able effects of the kathode rays is the fluorescence (2) they 
 produce in many substances. In addition to the facts already 
 known with regard to this property Lenard showed that the 
 bluish light observed in vacuum tubes is nothing bat the fluo- 
 
 (1) Lenard, Wied. Ann. 64, p. 279, 1898. 65, p 504, 1898. Goldstein, 
 Wied. Ann. 48, p. 787, 1893. Ver. der Phys. Ges. 2, p. 142, 1900. 3, p. 192, 
 1901. E. Wiedemann and H. Ebert, Wied. Ann. 46, p. 158, 1891. E. Wied. 
 and G. C. Schmidt, Wied. Ann. 60, p. 510, 1897. E. Wiedemann, Wied. 
 Ann. 63, p. 246, 1897. - 67, p. 714, 1899. Id. and Wehnelt, Fort, der Phys. 
 2, p. 811, 1898. J. J. Thomson, Phil. Mag. 44, p. 293, 1897. 48, p. 547, 
 1898. E. Schneider, Inaug. Diss. Erlangen, 1903. 
 
 (2) Wied. Ann. 51, p. 225, 1894. 
 
CHAP, in KATHODE RAYS FROM ULTRA-VIOLET LIGHT. 49 
 rescence of the gas under the influence, of the kathode rays 
 which in themselves are completely invisible. These rays also 
 produce fog-nuclei in gases traversed by them, (1) and charge 
 negatively all bodies within their path. This is the case even 
 when the vacuum is so great as to offer too much resistance for 
 the ordinary transfer of electricity. Similarly kathode rays 
 carry their negative charge through dielectrics which, under 
 ordinary circumstances, isolate very well. 
 
 3. KATHODE RAYS FROM ULTRA-VIOLET LIGHT. After having 
 proved that ultra-violet light that is absorbed on a negatively 
 charged surface produces kathode rays thereon, Lenard (2 ) began 
 to study these rays which, on account of their much smaller ve- 
 locity, offer advantages not possessed by the ordinary kathode 
 rays. The study of these phenomena gradually led him to 
 change his views with regard to the kathode rays. At first he 
 had been inclined to consider these rays as waves in the ether 
 but through his experiments with ultra-violet light, he came to 
 the conclusion that the kathode rays are nothing but the path 
 of the free elementary quantities of negative electricity. Under 
 ordinary circumstances this electricity is bound to matter but 
 when the ultra-violet light is absorbed at the surface of a body, 
 the vibrations of the elementary quantities are increased until 
 these quantities finally break away from matter and travel into 
 space because of their own velocity and the action of some ex- 
 ternal force; this force.is generally in the nature of anelectrical 
 field. If this field exerts a retarding force of sufficient intensi- 
 ty, the electrical quantities may be brought back to the charged 
 surface which under these circumstances will show no loss of 
 electrification. The production of negative electrical quantities 
 is always accompanied by the production of an equal number 
 
 (1) Wied. Ann. 64, p. 279, 1898. 
 
 (2) Drude Ann. 2, p. 359, 1900.-3, p. 298, 1900.-8, p. 149, 1902. 12, p. 
 449, 1903. 12, p. 714, 1903. 
 
50 CHAP. m. KATHODE RAYS FROM ULTRA-VIOLET LIGHT. 
 of carriers of positive electricity. The velocity of these is ge- 
 nerally much smaller than that of the negative carriers. When 
 a gas is ionized by kathode rays of a velocity less than that 
 corresponding to -11 volts, there is no evidence of positive 
 carriers on account of their possessing no velocity in this case. 
 The positive carriers are of the same order of magnitude as 
 atoms and molecules arid are most probably the atoms or mole- 
 cules of the gas. 
 
 In connection with this it may be interesting to give Le- 
 nard's views on the different quantities with which we are deal- 
 ing in physics. According to him, four things are always to be 
 distinguished: 
 
 1. The atoms of chemistry. These constitute matter. 
 
 2. Ether, that hypothetical substance which transmits light 
 electricity, etc. 
 
 3. Electrical quantities. They are not material although 
 coming out of matter. The path of these quantities of negative 
 electricity constitutes the kathode rays. 
 
 4. Electrically charged atoms, or carriers of electricity. 
 
 It must be noticed however that many physicists apply the 
 term ion to the quantities under the two last headings and that 
 many do not admit the existence of these "free" electrical quan- 
 tities. According to them, the quantities of negative electricity 
 which constitute the kathode rays are bound to a small fractional - 
 part of the atom. Lenard does not suppose this splitting up 
 of the atom. According to Lord Kelvin and Helmholtz, 
 we must admit the structural nature of the atom bat its 
 splitting up is not concluded from their discussions. 
 
 From a later paper by Lenard may be seen what is to be un- 
 derstood by chemical atoms or matter. The atoms are groups 
 of "dynamides" having in all probability the same extension 
 and inertia for all substances. They are electrical fields of for- 
 ce, their true radius is exceedingly small and the space between 
 the several dynamides is very large in relation to the space 
 
CHAPTER in. REFLECTION OF THE KATHODE RAYS. 51 
 occupied. If the dynamides in a cubic metre of platinum could 
 be crowded together so as not to leave any intervening space, 
 they would not occupy more than one cubic mm. In one of his 
 last papers (1) he says that a pair of elementary electrical 
 quantities in rapid rotation would be the simplest conception of 
 a dynamide, or of part of it, thereby showing that he perhaps no 
 longer regards the splitting up of the atom as an impossibility. 
 All these views are not mere speculation. They are an 
 attempt to account for the absorption of kathode rays of 
 different velocities ( 2 ) and the diverse other facts concerning 
 them. The most important of the facts that led to the pre- 
 ceding views are the following:- The effective radius of the 
 component parts of the atom is variable. For rapid kathode 
 rays the absorption is proportional to the density of the ab- 
 sorbing medium, and is in no wise influenced by either its 
 chemical or its physical constitution. The diffusion and the 
 secondary radiation follow the same law. 
 
 4. REFLECTION OF THE KATHODE RAYS. -The only im- 
 portant property of kathode rays not thoroughly- studied 
 by Lenard is their reflection from solid surfaces. Conse- 
 quently the facts concerning this phenomenon will be 
 taken from the work of other experimenters. (3) 
 
 This reflection depends largely on the velocity of the kathode 
 rays. For very slow rays it is small; with increasing velocity 
 it increases up to a certain maximum, from which it drops 
 
 (1) Drude Ann. 12, p. 714, 1903. 
 
 (2) Lenard, Drude Ann. 12, p. 714, 1903. R. J. Strutt, Nat. 81, p. 539, 
 1900. H. Beequerel, C. R. 130, p. 208, 1900. 130, p. 809, 19.0. P. and S. 
 Curie, C. R. 130, p. 647, 1900. 
 
 (3) Starke, Wied. Ann. 66, p. 49, 1898. Drude Ann. 3. p. 75, 1900- 
 Campbell Swiuton, Phil. Mag. 48, p. 132, 1899. Austin and Starke, 
 Drude Ann. 9, p. 271, 1902. Segny, C. R. 122, p. 134, 1896. Swinton, Froc. 
 Hoy. Soc. 64. p. 377, 1899. Villard, C. R. 127, p. 223, 1898. 130, p. 1010. 
 1900. Seitz, Drude Ann. 6, p. 1. 1901. Stark, Phya. Zeitschr. 3, p. 161, 1902. 
 
52 CHAPTER in. REFLECTION OF THE KATHODE RAYS. 
 again and approaches a nearly constant value for all higher 
 velocities. 
 
 The angle of maximum emanation, i. e. the angle formed 
 by the incident rays and the direction of maximum emanation 
 is different for different substances. 
 
 The phenomenon of reflection is always complicated and 
 consequently difficult to study quantitatively since the reflec- 
 tion is accompanied by a secondary emission which can not 
 well be separated from the reflected rays. 
 
 When the kathode rays undergo reflection as well as diffusion, 
 their velocity is generally decreased and they are less homogen- 
 eous than they were before incidence. (1 ) 
 
 5. CHARGE OF THE KATHODE QUANTITIES: e/m. The value 
 for the charge e, carried by an ion in electrolytes has been de- 
 termined by electrochemists as 1.29X1O 10 electrostatic units. 
 Several authors have also established the charge of an ion in 
 gases by means of the ionization obtained through Roentgen 
 rays or ultra-violet light. The mean of the values that were 
 reached is 6.5X10' 10 . This is at least in the same order of mag- 
 nitude as that obtained from electrolysis and, considering the 
 uncertainty of the methods used, the unit charge of an ion may 
 be looked upon as a universal constant. 
 
 The relation of charge to mass, e/m, in the kathode rays was 
 investigated by many experimenters. Their first results were 
 widely divergent but the latest computations are in closer agree- 
 ment. The most reliable determinations are probably those 
 given below :- 
 
 J. J. Thomson, (1897) 11.43X10 7 C. G. S. Units. 
 Ph. Lenard. (1899) 1.15 
 
 Kaufmann, (1901) 1.86 
 
 Sarke, (1903) 1.85 
 
 (1) E. Gehrke, Drude Ann. 8, p. 81, 1902. 8, p. 480, 1902. 
 
CHAPTER m. ROENTGEN RAYS. 53 
 
 From these results and from the preceding ones about the 
 charge of one of the ions the apparent "mass" of one of the 
 kathode quantities may be obtained if we assume that their char- 
 ge is equal to that of the ions. On this assumption it has been 
 found to be more than 1 000 times smaller than that of a hydro- 
 gen atom. But the most recent work of the best physicists 
 seems to require a much smaller mass for those particles. 
 
 All attempts to determine the relation of charge to mass in 
 the elementary negative quantities of the kathode rays, or in 
 other words, the specific charge of those quantities, have been 
 carried out with the greatest care especially during the past few 
 years. Despite this results so widely different have been obtain- 
 ed that their disagreement could no longer be considered as 
 being between the reasonable limits for errors of observation: 
 thus for instance, W. Wien found 0.3X10 7 and Simon 1.865X 
 10 7 . A closer investigation of the subject revealed the fact that 
 for lower velocities of the kathode rays, this relation was con- 
 stant or nearly so, whereas for greater velocities it varied. The 
 following values given by Kaufmann show these variations: 
 
 V. of Kathode Rays / 
 
 in 10 l cm/sec. I 2 ' 36 2 - 72 
 
 10 7 CMC*! S. U. I ' L31 L17 ' 975 - 77 ' 65 
 
 Since from other considerations we may assume that the 
 electric charge e is a constant, we are compelled to admit that 
 the mass of the kathode quantity varies . This apparent increase 
 in mass becomes especially evident when the velocity of light 
 is being approached by the kathode rays. 
 
 B. ROENTGEN RAYS. 
 
 When kathode rays of sufficient velocity fall upon a metallic 
 or other solid surface, their energy is partially transformed into 
 a much more penetrating form of radiation. 
 
54 CHAPTER in. CANAL RAYS. 
 
 The new rays were first studied by Roentgen (1) who called 
 them "X-Rays" on account of their unknown nature. Gold- 
 stein's "Differentiated Rays" which were studied in an earlier 
 chapter may have been partly Roentgen rays; but they included 
 certainly also reflected kathode rays whereas Goldstein does not 
 make any distinction between several kinds of differentiated 
 rays. So Roentgen is rightly known as the discoverer of the 
 rays which bear his name although he certainly was not the first 
 to observe effects produced by them. 
 
 The main characteristics of these rays are: their behavior 
 in a magnetic field, their penetrating power, their ionizing and 
 fluorescing properties. A magnet has no influence on them 
 and they do not suffer any refraction. Their penetrating power 
 is determined almost entirely by the density ef the body on 
 which they impinge. The vivid fluorescence which they cause 
 and their effect on photographic plates have proved of no small 
 practical usefulness. 
 
 C. CANAL RAYS. (2) 
 
 As the kathode rays are the path of free negative electrical 
 quantities, so the canal rays are the path of positive electrical 
 quantities or positively electrified particles. 
 
 The velocity of the latter is generally much smaller than that 
 of the kathode quantities but it is considerably increased when 
 
 (1) Sitzb. der Wuerzb. Phys.-Med. Ges. 1893. Ueitrag. 
 
 (2) Goldstein, Wied. Ann. 64, p. 38, 1898. Verh. der D. Phys. Ges. 3, p. 
 205. 1901. 4, p. 228, 1902. W. Wien. Verh. der D. Phys. Ges. p. 165, 
 1897. p. 10, 1898. Wied. Ann. 65, p. 441, 1898. Drude Ann. 5, p. 421 ? 
 1901.-8. p. 244, 1902.-9, p. 660, 1902. Phys. Zeitschr. 3, p. 440, 1902.' 
 Schuster, Proc. Roy. Soc. 47, p. 557, 1890. Arnold, Diss. Erlangen, 1897. 
 Wied. Ann. 61, p. 325, 1897. E. Wiedemann and G. C. Schmidt, Wied. 
 Ann. 62, p. 468, 1897. G. C. Schmidt, Drude Ann. 9, p. 703, 1902. ' Vil- 
 lard, C. R. 126, p. 1564, 1898. Wehnelt, Wied. Ann. 67, p. 421, 1899. 
 Ewers, Wied. Ann. 69. p. 167, 1899. Towiisend, Phil. Mag. 6, p. 598, 1903. 
 
CHAPTER. in CANAL RAYS. 55 
 
 the positive particles traverse the rapid fall of potential at the 
 kathode. There they produce the fluorescence which is gene- 
 rally known as the first kathode layer . If the kathode is not 
 continuous they proceed through and beyond the openings as 
 canal rays because of the velocity acquired in the kathode fall. 
 It is in this manner that canal rays are usually produced. 
 
 They are not deflected by a magnet so easily as are the kath- 
 ode rays. This deflection is opposite to that of the latter and 
 separates the canal rays into several groups, showing that the 
 positively electrified particles which constitute them have masses 
 of different orders of magnitude. There may be other kinds 
 of rays in this group of canal rays besides those defined by the 
 order of magnitude of the particles which constitute them. 
 Thus E. Goldstein, (1) basing his division on the color of the 
 fluorescence produced by them and on their different methods 
 of propagation, thinks that what is commonly known as ''canal 
 rays" is a complex of at least five different kinds of rays. 
 
 These conclusions are not astonishing in view of the careful 
 work of Willy Wien, (2) who showed conclusively that the 
 canal rays are very different in different gases and described the 
 difficulties he encountered in his efforts to obtain pure gases 
 in his discharge tubes on account of the occluded gases con- 
 stantly given off by the kathode. 
 
 From his experiments it seems probable that while the canal 
 rays approximately retain their initial velocity, the relation of 
 charge to mass is a variable one for the same particle. The 
 values of this relation vary from to 36 000, the latter being 
 for canal rays in hydrogen at a discharge-potential of 9 000 volts. 
 It also seems probable that the canal rays are produced from 
 the gas in the tube, since in highly exhausted tubes they are 
 
 (1) Verb, der D. Phys. Ges. 4, p. 228, 1902. 
 
 (2) Drude Ann. 8, p. 244, 1902.- 9, p. 660, 1902. 
 
56 CHAPTER in POTENTIAL PHENOMENA. 
 
 entirely absent while kathode rays are still in evidence: these 
 are produced from the negative electrode. 
 
 Canal rays as well as kathode rays are influenced by an elec- 
 trostatic field. When the rays are parallel to the lines of force 
 they are either retarded or accelerated, the former taking place 
 when they travel with the lines of force, the latter when they 
 travel against them. In a field that is perpendicular to their 
 path of propagation they are deflected towards the negative 
 plate. 
 
 The chemical effects of these rays generally consist in the 
 dissociation and splitting up of more complex molecules in the 
 gas which they traverse. When they acquire sufficient velocity 
 the dissociation produced by them may assume an electrical 
 character which is evidenced by secondary radiation and ion- 
 ization. 
 
 III. POTENTIAL PHENOMENA. 
 1. FALL OF POTENTIAL. (1) 
 
 The general behavior of the fall of potential was carefully 
 studied during this period. As might have been expected, 
 the potential gradient was found to be very different in the 
 three main parts of the discharge:- the positive light, the 
 dark space and the negative glow. The maximum fall is 
 always in the dark space near the kathode, the minimum in 
 the negative glow. At the anode there is also a great fall 
 followed by a much smaller value. Within the unstriated posi- 
 
 (1) Graham, Wied. Ann. 64, p. 49, 1898. Skinner, Wied. Ann. 68, p. 752, 
 1899. Phil. Mag. 50, p. 563, 1900. G. C. Schmidt, Drude Ann. 1, p. 625, 
 1900. H. A. Wilson, Phil. Mag. 49, p. 505, 1900. Wehnelt, Drude Ann. 7, 
 p. 237, 1901. J. Stark, Drude Ann. 5, p. 89, 1901. H. Starke, Verh. der 
 D. Phys. Ges. 5, p. 364, 1903. D. Rudge, Proc. Cambr. Phil. Soc. 12. p. 
 155, 1903. N. S. Taylor, Phys. Rev. p. 321, May, 1904. R. S. Willows 
 Phil. Mag. 9, p. 370, 1905. Townsend, Phil. Mag. 9, p. 289, 1905. 
 
CHAP, in POTENTIAL PHENOMENA. 57 
 
 tive light there is no change in the potential gradient but if the 
 tube presents striatioiis, the fall of potential curve is a suc- 
 cession of relative maxima and minima. The former begin at 
 the beginning of the bright layers, i. e. at the bright part 
 nearest the kathode; the latter are found towards the end of 
 the layers, i. e. in the dark part nearest the anode. 
 
 In the dark space the potential gradient rises gradually in 
 the direction from the kathode to the anode, sometimes showing 
 small maxima and minima. 
 
 2. KATHODE FALL. (1) 
 
 The difference of potential between the kathode and the begin- 
 ning of the negative glow is called "Kathode fall." It decreases 
 very rapidly in the kathode dark space but this decrease does 
 not always obey the same laws. As long as the negative glow 
 can extend itself symmetrically over the surface of the electro- 
 de, the kathode fall is normal, but as soon as this negative glow 
 is forcibly restricted to surfaces smaller than those which it 
 would naturally tend to cover, the kathode fall becomes abnor- 
 mal. The normal fall is independent of the current-intensity 
 but varies with the pressure and the nature of the gas. When 
 the negative light is prevented from extending freely, the po- 
 tential fall at the kathode increases with increasing current. 
 The relation between these two quantities is represented by one 
 branch of a parabola. 
 
 The normal kathode fall is not influenced by the pressure 
 of the gas but when the kathode fall becomes abnormal it 
 
 (1) Paalzow and Neessen. Wied. Ann. 56, p. 700, 1895. Capstick, Proc. 
 Roy. Soc. 63, p. 356, 1898. E. Wiedemann, Wied. Ann. 67, p. 714, 1899. 
 Strutt. Proc. Roy. Soc. 65, p. 446, 1900. G. C. Schmidt. Drude Ann. 1, p. 
 625, 1900. Heuse, Drude Ann. 5, p, 670, 1901. Wehnelt, Drude Ann. 7, 
 p. 237, 1902. Skinner, Phil. Ma#. 2, p. 616, 1901. E. Riecke, Drude Ann. 
 4, p. 592, 1901. J. Stark, Phys. Zeitschr. 3, pp. 83, 165, and 274, 1902. 
 Drude Ann. 12, p. 1, 1903. 12, p. 31, 1903. Cunningham, Phil. Mag. 4, 
 p. 684, 1903. 9, p. 193, 1905. 
 
58 CHAPTER in POTENTIAL PHENOMENA. 
 
 increases slowly at first and then rapidly as the vacuum 
 becomes higher. 
 
 Finally it may be noticed that heating the kathode con- 
 siderably decreases the potential gradient, and that a weak 
 transverse magnetic field somewhat lowers the kathode fall, 
 while a strong field makes it rise rapidly. On the contrary if 
 a magnetic field is parallel to the kathode rays, it does not 
 influence the kathode fall. 
 
CHAPTER IV. 
 
 THEORY OF THE DISCHARGE. 
 
 1. VOCABULARY. 
 
 Unfortunately there is still in this branch of physical 
 science a confusing variety of terms without any universally 
 recognized definition. Some terms are used to denote entirely 
 different things, while inversely the same thing often has 
 several appellations. The two words "ion" and "electron" 
 which are certainly very extensively used, may serve as an 
 illustration. 
 
 The term "ion" was first used by Faraday:- "Finally", he 
 says speaking about electrolytes, "I require a term to express 
 those bodies which can pass to the electrodes, and I propose 
 to distinguish such bodies by calling those anions which go to 
 the anode of the decomposing body; and those passing to the 
 kathode, kations; and when I have occasion to speak of these 
 together, I shall call them ions. Thus the chloride of lead is 
 an electrolyte, and electrolyzed evolves the two ions, chlorine 
 and lead, the farmer being the anion, and the latter the 
 kation." Here ion in its original sense means an atom or 
 molecule carrying an electrical charge. It still retains this 
 signification in electro-chemistry and is used in the same 
 sense by a great many physicists in the subject now under 
 discussion. But some authors define the ion as an elementary 
 quantity of electricity which is not bound to a similar quantity 
 of opposite sign. Thus they speak about molions, atomions 
 
60 CHAPTER iv. THEORY OF THE DISCHARGE.' 
 
 and electronions, meaning an electrically charged molecule, 
 atom or electron. 
 
 This word electron is defined by them as the elementary 
 quantity of matter though it is still used by the majority of phy- 
 sicists in its original and more appropriate meaning, i. e. the 
 elementary quantity of electricity. 
 
 To this must be added that, in order to avoid the confusion 
 resulting from the use of these terms, other authors entirely 
 dispense with them and employ only such terms as carry with 
 them their own meaning: thus for instance Lenard speaks of 
 electrical carriers and elementary quantities of electricity. This 
 seems to be a better method than using the other words 
 without going to the trouble of defining them but it has a draw- 
 back in this that it again introduces another set of terms into 
 the subject. It would seem preferable to retain the words ion 
 and electron, as has been done by representative physicists in 
 the few last years, in their time-honored original meaning of 
 electrically charged atom or molecule and elementary quantity 
 of electricity respectively. It is this sense that will be given 
 them throughout the following pages. 
 
 Since the structural character of the chemical atom may now 
 be looked upon as definitely established, it would also be 
 convenient to have a name for those extremely minute parts 
 whose grouping constitutes the atom. Several terms have been 
 suggested:- energon by Reuterdahl and dynamide by Lenard 
 The latter word is founded on the strongly sustained hypothe- 
 sis that these minute parts constitute electrical fields of force 
 through the juxtaposition of a negative and a positive electron. 
 The word corpuscle extensively used by J. J. Thomson would 
 be an appropriate appellation were it not for the fact that this 
 author uses it in a slightly different sense. 
 
 The term kathode ray is not interpreted in the same manner 
 by all physicists. Some introduced a distinction between 
 kathode and Lenard rays:- A kathode ray for them is the path. 
 
CHAPTER iv. EARLIER SYSTEMS. 61 
 
 of any particle that carries negative electricity whereas the 
 Lenard ray is the one outside the kathode tube after all the 
 particles larger than the electrons have been sifted out by the 
 aluminium window, But there is no necessity for making this 
 distinction, and we may, with the majority of physicists, define 
 a kathode ray as the path of the negative electrons. 
 
 II. EARLIER SYSTEMS. 
 
 This subject is relatively very recent and it is largely on this 
 account that there is so much embarrassing confusion. Another 
 potent factor in bringing it about is the variety of systems set 
 up for the explanation of the facts that were gradually dis- 
 covered. These systems are scarcely less numerous than the 
 investigators, although quite frequently they differ only in 
 some minor details. 
 
 One of the earliest theories was that advanced by G. Wiede- 
 mann, according to whom the molecules are electrified near 
 the charged electrodes and then repelled according to well- 
 known electrical laws. These molecules do not travel from one 
 electrode to the other but by their impact on others they give 
 them their charge which is thus transmitted until it is neutral- 
 ized on meeting an opposite charge on molecules coming from 
 the other electrode. Moreover if the gas may be dissociated 
 electrolytically, the transfer of electrical energy may be accom- 
 plished by the separation, the convection and the reunion of 
 ions as in the electrolytic process. 
 
 A somewhat similar theory was adopted by Sir W. Crookes 
 and supplemented by a revival of Faraday's ideas concerning a 
 fourth state of matter. Puluj proposed another theory in 
 which the particles that effected the transfer of electricity were 
 not the molecules of the gas but the electrode-dust thrown off 
 from the kathode at a great velocity and carrying with it a 
 strong negative charge. 
 
 Another class of hypotheses was based upon the supposition 
 
62 CHAPTER iv. ELECTRON THEORY. 
 
 that the electric discharge was a phenomenon in ether. This 
 view was favored by a great many careful experimenters. But 
 gradually the electrolytic dissociation theory already proposed 
 by G. Wiedemaim gained ground, thanks to the efforts of such 
 able defenders as Schuster, GKese, Elster and Geitel, J. J. 
 Thomson and many others. To day all physicists seem to admit 
 that gases may become conductors in an electrolytic sense and 
 that consequently electricity may be conveyed through them 
 by the migration of ions. 
 
 But this dissociation is insufficient to account for the pheno- 
 mena that occur in rarefied gases. Moreover it could not possi- 
 bly be applied to conductors of the first order, although it seems 
 natural to suppose that whenever there is a current of electricity, 
 whether in metals, electrolytes or gases, the essential nature of 
 this current is always the same. The following theory which 
 may be called the electron theory seems to offer a fundamental 
 idea applicable to the several kinds of electrical conduction. 
 
 III. -ELECTRON THEORY. 
 A. FOUNDATIONS OF THE ELECTRON THEORY. 
 
 The electron theory which may be defined as the theory of 
 electric dissociation is founded on several branches of physical 
 science, the most important of which is electricity itself. The 
 ideas of scientists with regard to the nature of this important 
 agent have gradually undergone a change. The single- and two- 
 fluid theories were the earliest conceptions of electricity, but 
 for many great physicists they were only a convenient mathe- 
 matical expression which proved very efficient in the analytical 
 treatment of the various problems connected with electrical 
 forces. 
 
 By laying down the fundamental laws governing electro- 
 lysis, Faraday furnished the fundamental idea on which a new 
 theory was soon to build up. The first of these laws is the one 
 stating that "the chemical decomposing action of a current is 
 
CHAPTER iv. ELECTRON THEORY. 63 
 
 constant for a constant quantity of electricity," or that "the 
 chemical power of a current of electricity is in direct propor- 
 tion to the absolute quantity of electricity which passes". From 
 this it follows that the quantity of electricity which passes is 
 the equivalent of and therefore equal to that of the particles 
 separated. The second law states that "electro-chemical equi- 
 valents coincide with and are the same as ordinary chemical 
 equivalents". 
 
 These laws suggest the idea that the electrical charge pertain- 
 ing to any valency of an ion may be a fixed quantity having a 
 separate existence, so that there may be atoms of electricity as 
 well as of matter. Weber and Helmholtz were the first champions 
 of this theory which was soon to supplant the older fluid-theo- 
 ries. But their views were deficient in as far as they failed to 
 sufficiently account for the action of electricity beyond the space 
 occupied by the small particles. By conceiving these electrons 
 imbedded in the cosmic ether and surrounded by an electro- 
 magnetic field of force, this objection is completely removed, 
 while at the same time the exigencies of atomistic structure are 
 satisfied. 
 
 This view of electricity still leaves the question as to its real 
 nature an open one. It may be a separate substance different 
 from what is ordinarily called matter or it may simply be a 
 localized condition of the ether. The latter hypothesis is cer- 
 tainly the one by which the transmission of electrical force 
 through the ether would be most easily accounted for. 
 
 ZEEMAN EFFECT. The electron hypothesis received a fur- 
 ther development from two other important sources:- the kath- 
 ode rays and the Zeeman phenomenon. As the former will 
 be discussed later on it will be enough to state the main facts 
 concerning the latter. 
 
 It is well known that incandescent gases in general give a 
 line spectrum and that each line represents a definite period of 
 vibration. But if the gas is subjected to a strong magnetic 
 
64 CHAPTER iv. ELECTRON THEORY. 
 
 field, an important change takes place in the spectrum. When 
 the propagation of the light which causes a certain spectrum 
 line is in the direction of the magnetic force, this line disappears 
 and in its stead two new ones appear at an equal distance from 
 the position of the original line. When the field is perpendi- 
 cular to the light-wave, there are in general three lines of which 
 the middle one occupies the original position. In some cases 
 these general phenomena are even more complicated, the D i 
 line for instance yielding four new lines, while D 2 is resolved 
 into six by a magnetic field. 
 
 A careful study of these phenomena revealed the fact that 
 the new vibration periods are due to a negatively electrified 
 particle, for which e/m is about 1000 times greater than for a 
 hydrogen atom. Since we may assume that the electrical 
 charge is equal in both cases, it will follow that the mass of 
 the vibrating particle is very small and that the Zeeman 
 effect is due to a magnetic influence on the negative electrons 
 which possess a greater freedom of motion than the positive 
 electrons. In the case of kathode rays this greater freedom 
 results in a total liberation from the influences of the atom 
 and of the positive electrons. 
 
 B. APPLICATION OF THE ELECTRON THEORY TO THE DIS- 
 CHARGE OF ELECTRICITY THROUGH GASES. 
 
 The foregoing views which to a great extent are only a 
 statement of facts, furnish a new explanation for the con- 
 duction of electricity. In this new theory the negative 
 electrons are looked upon as separable from the atom. They 
 possess a certain freedom of vibration within the atom, which 
 vibration may be increased by the absorption of radiant energy, 
 by electrical tension, etc., to such an extent as to overcome the 
 attraction of the positive electron and the rest of the atom. 
 In such a case the negative electron breaks away from its 
 former vibrating position and travels into space with a velocity 
 
CHAPTER iv. ELECTRON THEORY. 65 
 
 that is determined by its own energy and the amount of 
 exterior forces acting on it. At the same time the rest of -the 
 atom, which has now a positive charge, moves in the opposite 
 direction. These motions or migrations constitute the electric 
 current. The only difference between rarefied gases on the one 
 hand and electrolytes and gases of a higher pressure on the 
 other lies in the fact that in the former the negative electron 
 may travel on for a great distance while in the latter, some 
 atoms or molecules immediately condense on it, thus forming 
 the electrolytic negative ions. In conductors of the first order 
 conduction is similarly accounted for by a motion of electrons. 
 We assume that in these solids there is always a number of 
 free electrons which move in the interatomic space as soon as 
 new electrons enter from any of the several sources of electri- 
 city. These free electrons might be both positive and nega- 
 tive, but since we have no sufficient evidence for admitting 
 the separate existence of positive electrons, we may assume 
 that the conduction is effected only by negative electrons. 
 This hypothesis sufficiently accounts for all the known facts by 
 assuming that when a negative ion is being deposited on the 
 anode, it gives up its negative electron while a positive ion 
 arriving at the kathode takes a negative electron from the kath- 
 ode itself. For entirely metallic circuits, a current of elec- 
 tricity would be nothing but a motion of negative electrons in 
 one direction. 
 
 Let us now apply the electron theory to the passage of elec- 
 tricity through rarefied gases with which we are mainly con- 
 cerned. 
 
 Generally a gas contains some ions although their number 
 may be very small. As soon as a difference of potential is estab- 
 lished between the two electrodes, these ions move in opposite 
 directions, thereby causing the weak current which is 
 noticed long before the difference of potential becomes high 
 enough to produce the well-known phenomena of the discharge. 
 
66 CHAPTER iv. ELECTRON THEORY. 
 
 This current is purely electrolytic in character and in conse- 
 quence of the relatively greater velocity of the negative ions it 
 produces a greater rarefaction near the negative electrode. As 
 the difference of potential rises the vibration of the negative 
 electrons on the kathode becomes more intense until it tinally 
 reaches a value which allows these electrons to break their con- 
 nection with the atom. They are then thrown off from the kath- 
 ode at a high velocity. By impinging on the atoms of the 
 gas they produce other negative electrons and a corresponding 
 number of positive ions. 
 
 FIRST KATHODE LAYER ^The initial path of these new car- 
 riers has every possible direction but all the positive ions are 
 gradually bent back by the electrical field towards the kathode. 
 
 The great fall of potential near the latter gives them a velo- 
 city sufficient to produce a vivid fluorescence when their path 
 is stopped by the solid electrode. This gives rise to the so- 
 called first kathode layer. 
 
 KATHODE DARK SPACE AND KATHODE LIGHT. The relative 
 darkness of the next layer is due to the few impacts of the nega- 
 tive electrons on account of their high velocity and the relative 
 scarcity of the positive ions. These impacts gradually increase 
 in number up to a place well within the negative light. The 
 negative electrons, like the positive ions, at first travel in every 
 direction, but those which return towards the kathode, being 
 gradually retarded by the rapidly increasing force acting 
 against them, are finally stopped and repelled towards the ano- 
 de. The third kathode layer is chiefly due to the secondary ne- 
 gative electrons. Its greater luminosity is accounted for by 
 the larger number of impacts for slow electrons in a given cross- 
 section; its sharp definition is a natural consequence of the 
 rapidly increasing strength of the electrical field near the 
 electrode, the point where this field will be able to 
 destroy the oppositely directed velocities will be practically the 
 same for all negative electrons. It may not be evident at first 
 
CHAPTER iv. ELECTRON THEORY. 67 
 
 why the absorption of slower electrons should be greater than 
 that of swifter ones in the same length of the gas-column, .but this 
 is a direct consequence of the concept of matter which under- 
 lies this theory. The atom has structure and its component 
 parts are at least mainly, if not essentially, electric fields of 
 force resulting from the juxtaposition of two electrons of opposite 
 sign. In such a field of force the true radius of the component 
 electrons and the efficient radius of the field are two different 
 things. By true radius is meant the radius of the space 
 occupied by the electrons while the term "efficient radius" 
 denominates the radius of the electric field which is sufficient 
 to completely counterbalance the velocity of a negative electron 
 crossing it. This effective radius is naturally different for 
 different velocities, and beyond it the negative electrons are 
 merely retarded. 
 
 DARK SPACE. The great production of negative electrons 
 and positive ions within the negative glow results in a rapid 
 rise of the potential gradient. The ionization decreases and 
 finally ceases almost entirely, thus bringing about the dark space. 
 
 POSITIVE LIGHT. The relative scarcity of electrons and ions 
 due to this absence of ionization causes a decrease of conduc- 
 tivity and consequently another drop in the potential gradient. 
 This difference of potential again accelerates the negative 
 electrons and gives them energy enough to produce new ions 
 and electrons by impinging on the atoms. The place where 
 this new ionization sets in is the beginning of the so-called 
 positive light. This light appears under two distinct forms: 
 the striated and the unstriated. 
 
 UNSTRIATED POSITIVE LIGHT. If the ionization and the 
 electrical forces in this part concur in producing negative 
 electrons and positive ions of all velocities, these will continue 
 producing an ionization whose amount will be practically the 
 same for all cross-sections in the column of light. In this case 
 there will be no appreciable difference in the potential gradient 
 
68 CHAPTER iv. ELECTRON THEORY. 
 
 between any two points; as a result the luminosity of the gas 
 will be nearly the same throughout the whole length of the 
 positive light and the discharge will be unstratified. 
 
 STRIATION. But if a majority of the electrons have, or 
 acquire, nearly the same velocity, their mean free path will be 
 the same. Within this free path very little ionization will take 
 place and consequently a dark layer will be produced 
 and the potential gradient will fall again. When they 
 have reached the end of their free path, new electrons are pro- 
 duced in great numbers; this gives rise to a greater degree of 
 luminosity and to an increase in the potential gradient. The 
 luminous layer will be succeeded by a dark one, the latter again 
 by a brighter one and so on until the anode is reached. This is 
 the striated discharge. The fundamental idea underlying this 
 explanation was first advanced by Goldstein who considered 
 each brilliant layer of the positive light as the starting point 
 of a new current. It appears readily how a new period of 
 ionization really supplies the elements for a new current. 
 
 ANODE LAYER. The conditions near the anode are mainly 
 dependent on the state of ionization of the gas in its 
 immediate neighborhood. If few electrons and ions are 
 present there is a rapid change in the potential gradient, the 
 velocity of the existing electrons is accelerated and these will 
 produce ionization and a brilliant layer on the surface of the 
 anode. If on the other hand this electrode is in the negative 
 glow or in any part of the tube where there is a high con- 
 ductivity, the anode fall is small, very little ionization is pro- 
 duced and the anode layer becomes faint or vanishes com- 
 pletely. 
 
 At the beginning of this discussion it was supposed that 
 the increase of vibration which resulted in the separation of 
 the electron from the atom, was due to the high potential at 
 the kathode and that thus an electric current was established 
 in the gas. This however is not the only cause that may 
 
CHAPTER iv. ELECTRON THEORY. 69 
 
 produce the same result. If the kathode absorbs energy from 
 ultra-violet light or an exterior source of heat, this energy 
 passes to the electrons and increases their vibration Ultra- 
 violet light is an electro-magnetic vibration propagated 
 through the ether in very short wave-lengths. These vi- 
 brations excite a stronger activity in the negative electrons 
 and give them sufficient energy to allow them to overcome 
 the attraction of the positive electrons with which they are 
 associated. These negative electrons then pass into the sur- 
 rounding gas and thus bring about the discharge of the plate 
 on which ultra-violet light is impinging. If this plate is not 
 charged initially, the effect of ultra-violet light is still the same 
 but in such a case the throwing off of negative electrons will 
 result in leaving it with a positive charge. Heating the kathode 
 has a similar effect because heat is a vibratory motion commu- 
 nicated to the electrons. 
 
 When a gas surrounding a charged body is directly ionized 
 by an external agent such as ultra-violet light, kathode rays, 
 Roentgen Rays, Becquerel rays, etc., the ions thus produced 
 receive an acceleration from the electric field near the charged 
 surface. The current thus established lasts as long as there 
 are free ions in the gas. If the field is strong enough to in- 
 crease the velocity of the ions to such a degree as to allow them 
 to ionize more gas in their turn, a new current directly produced 
 by the electro-motive force of the electrode is established. 
 Hence as a general rule, currents may be divided into two main 
 classes:- the dependent and independent currents. The depen- 
 dent currents are those that consist entirely in the motion of 
 ions produced by an external agent. The independent currents 
 again are of two kinds:- completely so or only partially so. The 
 former are established by the electro-motive force of the elec- 
 trodes while the latter need the help of some exterior agent for 
 beginning but are then maintained by the difference of potential 
 between the electrodes. 
 
70 CHAPTER iv. ELECTRON THEORY. 
 
 C. THE ELECTRON THEORY AND THE DIFFERENT KINDS 
 OF RADIATION. 
 
 1. KATHODE RAYS. 
 
 Kathode rays are the path of negative electrons. This view 
 is universally held to day with one unimportant exception. 
 Some authors apply the term kathode ray to the path of any 
 carrier of negative electricity, whether this carrier be an electron 
 or an ion. It is only a question of words and it seems prefer- 
 able to adopt the first definition because it is almost universally 
 recognized. 
 
 For a long time, the most notable experimenters v. g. Hittorf , 
 Hertz, Goldstein, Lenard (in the beginning of his work) and 
 many others held an entirely different view. For them, a 
 kathode ray was a process in the ether, pobably very refrangible 
 ultra-violet light. The deflection of the ray by a magnet was 
 explained by saying that it was an effect similar to the rotation 
 of the plane of polarization. This at best was not a very clear 
 explanation. These views naturally led to the conclusion that 
 the kathode ray was merely a phenomenon accompanying the 
 electric discharge without taking any part in it. 
 
 But gradually the electrical character of these rays was estab- 
 lished, and they came to be looked upon as one of the most 
 important factors of the discharge through their power to pro- 
 duce ionization. 
 
 The real nature of the electrons, those quantities which 
 constitute the kathode rays is still under discussion. To 
 explain them, one of two hypotheses must be admitted :-either 
 the kathode quantity is part of the material atom with an 
 elementary charge of negative electricity or it is this elementary 
 quantity existing independently of matter. This second view 
 has been held throughout this discussion. No convincing rea- 
 son can be given against it. The strongest argument adduced 
 by the adversaries of this theory is that it is an altogether new 
 
CHAPTER iv. ELECTRON THEORY. 71 
 
 assumption . But this is likewise true of their own hypothesis 
 for although radium emanation shows after some time the 
 spectrum of helium, the faintest trace of which could not be 
 previously detected, thus showing that some change had taken 
 place in the atomic structure of the expelled gases, still these 
 changes involve immense groupings of the elementary consti- 
 tuents of matter and do not justify the conclusion that one of 
 those small parts with its negative charge, can be separated 
 from the atom. An analogous case is found in the more com- 
 plicated chemical molecules: when such a molecule is broken up 
 by heat or otherwise, the resulting bodies are never one solitary 
 atom and the former molecules with only a slight change, but 
 new molecules are formed which represent a grouping that 
 already existed in the first molecules. Thus oxalic acid for 
 instance, may in several ways be broken up by heat, yielding 
 carbon monoxide, carbon dioxide, water and formic acid* 
 occordiiig to the following formulae: 
 
 (OOOH)o = CO* -|- H. COOH 
 
 2(COOH) 2 = 2CO 2 -|- CO -|- H. COOH -|- H 3 O 
 
 But no single atom ever breaks off , leaving the rest of the 
 compound in an incomplete grouping. 
 
 By assuming that the electron is nothing but an elementary 
 quantity of electricity, this difficulty is avoided. This is not 
 the only advantage of the electron theory as exposed above. 
 Another difficulty against assuming that there is a material 
 mass connected with the elementary quantity of the kathode 
 ray comes from the fact that we have r.o evidence of its mass un- 
 less it be in motion. Thus the highest charge on a conductor 
 does not bring about the slightest difference in mass, nor is any 
 transfer of mass ever noticed in a conductor of the first order 
 through which a current is passing. This, in connection with the 
 fact that the mass of an electron is a function of its velocity 
 warrants the conclusion that this mass is entirely electro- 
 
72 CHAPTER iv. ELECTRON THEORY. 
 
 magnetic in character. It is easy to see how self-induction and 
 the resistance of the surrounding field against deformation by 
 any external cause may sufficiently account for this apparent 
 mass ; for a rapidly moving electron is to all intents and pur- 
 poses an electric current which, on account of the extremely 
 small radius of its carrier, creates an intense field of force in its 
 vicinity. Any force tending to deform the field will expend 
 an energy that will be resisted by the electron. This inertia is 
 sufficient to explain all the facts that have been observed up to 
 the present. 
 
 A very interesting but complex phenomenon occurs when a 
 kathode ray strikes a solid obstacle. In such a case the ray 
 may be:- 1. transmitted, 2. reflected, 3. deflected, and 4. 
 absorbed and changed into a new form of energy. These pheno- 
 mena may all happen at the same time and several of them are 
 always produced simultaneously. This behavior is easily 
 explained by the electron theory. 
 
 Thin sheets of metal allow the kathode rays to pass through 
 them because of the almost infinitesimal radius of the electron 
 and the relatively large space between the small quantities 
 whose grouping constitutes the atom as well as between the 
 atoms and molecules themselves. But the electrons are at the 
 same time subjected to the influence of the parts of the atom, 
 which influence tends to destroy their velocity and to change 
 their original direction. In fact kathode rays are retarded in 
 going through an aluminium window and they issue from it in 
 all directions. 
 
 But some electrons will impinge on the space really occupied 
 by the obstructing body and because of their elasticity be 
 reflected. This reflection will necessarily be diffuse because 
 the surface on which they strike will always be rough for such 
 small quantities as the electrons. This real reflection may be 
 accompanied by something of a quite similar nature. Instead 
 of striking against the solid parts of the substance, 
 
CHAPTER iv ELECTRON THEORY. 73 
 
 the kathode quantities may be gradually brought to rest and 
 then accelerated in tlie opposite direction by the opposing 
 field of force, since a body on which they strike becomes 
 thereby negatively charged. 
 
 Deflection may also take place on account of this same 
 negative charge, especially if the angle of incidence is great. 
 This follows from the very nature of the electron viewed as an 
 elementary quantity of negative electricity. 
 
 Ultimately, whenever the kathode rays impinge on a solid 
 some of their energy is transformed. They may thus cause an 
 increase of vibration and consequently heat and fluorescence. 
 And again when they impinge on any particle, they will be 
 suddenly stopped, at the same time throwing the particle out 
 of its ordinary vibratory path. This will result in a stress in 
 the ether and in a wave that may be characterized as an elec- 
 tro-magnetic pulse, since it is of very short duration. This 
 phenomenon is known as the Roentgen ray. It is evident why 
 the intensity of this ray should increase as the velocity of the 
 producing kathode ray increases since the disturbance created 
 in the ether will be more intense for greater velocities. 
 
 The Zeeman effect shows how light is probably due to the 
 vibration of the negative electron around its point of rest. 
 This vibration, i. e, the constant varying between the velocities 
 O and a certain maximum, sends out into the surrounding ether 
 a continuous train of electro-magnetic or light-waves. When 
 the negative electron is suddenly stopped in its path by striking 
 on a solid surface, the same sudden, or rather, a more sudden 
 variation of velocity occurs, and consequently an electro- 
 magnetic disturbance is propagated through the ether. After 
 having been thus stopped, the electron may not return or may 
 return with an acceleration quite different from what it had 
 before arriving. If, therefore, we look upon a wave of light 
 as produced by a regular full or half vibration, we will not 
 be justified in calling the Roentgen ray a wave, since it lacks 
 
74 CHAPTER iv ELECTRON THEORY. 
 
 the regularity which this name implies: hence the name elec- 
 tro-magnetic pulse. 
 
 2. CANAL RAYS. 
 
 In discussing the first layer of the kathode light we assumed 
 a continuous electrode. If this is not the case, some of the 
 positively charged particles instead of impinging on the 
 kathode, continue through the openings and by their high 
 velocity are constituted canal rays. These rays may also be 
 produced by accelerating the positive carriers in a strong 
 electro-static field. 
 
 The nature of the canal rays is not known so well as that 
 of the kathode rays owing partly to their more complex 
 character. In kathode rays the so-called magnetic spectrum 
 which is due to differences of velocity. But in canal 
 rays there is also different deflectibility which can not be 
 accounted for by different velocities. There are some rays 
 that can not be deflected even by the strongest electro-magnets 
 although their velocity is of the same order as that of those 
 which are deflected. 
 
 Again, these rays show widely different values for their 
 specific charge, e/m. This might be accounted for in two 
 ways:- neutral atoms or molecules combine with the charged 
 particle, the charge meanwhile remaining constant; 
 or the mass remains constant while the charge varies. 
 The latter hypothesis is the more probable owing to the general 
 behavior of the rays. The electron theory offers a ready expla- 
 nation of this. There is no reason to suppose that only one 
 negative electron may be separated from the atom for all the 
 electrons enjoy an equal freedom of motion and, consequently, 
 it may happen that several of them are separated from the 
 atomic group at the same time. But each negative elementary 
 quantity of electricity that is thrown off makes it harder for 
 the others to break their connection with the atom. As a con- 
 
CHAPTER iv ELECTRON THEORY. 75 
 
 sequence, a limited number can only be separated and, more- 
 over, the atom positively electrified in this way will tend towards 
 neutralization by combining anew with the negative elementary 
 quantities which it may meet in its path. This accounts for 
 the change of the specific charge of the rays during their 
 propagation and for all their known properties: 
 
 The rays that are not deflected are the path of those particles 
 which have become completely neutralized before entering the 
 electric or magnetic field applied to the tube. The several 
 groups produced by a deflecting force are formed by such 
 particles as retain an equal charge. 
 
 Finally, this recombination explains the fact that in very 
 highly rarefied gases the current can pass much longer than 
 could be expected from the computation of the number of 
 molecules in the tube, for the gas is being constantly regen- 
 erated, thus rendering possible the production and transfer of 
 new carriers, or in other words allowing the current to continue. 
 
 3. BECQUEREL RAYS. 
 
 There are two differences between the Becquerel rays and 
 those hitherto discussed but neither of them is essential. The 
 one is quantitative while the other concerns their source. 
 Qiiantitatively, Becquerel rays and in particular radium rays 
 are generally stronger than the corresponding rays produced 
 in an ordinary vacuum tube. Thus rays correspond to canal 
 rays. rays are strong kathode rays or in other words, they are 
 negative electrons of high velocity. Y rays show all the charac- 
 teristics of "hard" Roentgen rays. But while canal, kathode 
 and Roentgen rays are obtained from ordinary bodies under the 
 influence of some well defined exterior agent such as strong 
 heat, ultra-violet light, a high electro-motive force, etc., they 
 seem to be emitted by the radioactive substances without the 
 absorption of any external energy. 
 
 The production of the Y rays is explained on the same prin- 
 ciple as that of the Roentgen rays which they resemble in 
 
76 CHAPTEB iv ELECTRON THEORY. 
 
 every property. But here the negative electrons strike against 
 the particles of the radium itself and being thus suddenly 
 stopped, produce the samte electro-magnetic pulse which is 
 generally very intense on account of the high velocity of the /? 
 rays. 
 
THE CONSTITUTION OF MATTER. 
 
 The study of the electric discharge through gases and more 
 particularly that of the kathode rays has led to a new concept 
 of the nature of matter. The existence of electrons both posi- 
 tive and negative, grouped so as to neutralize one another, 
 while allowing at the same time the negative electrons more 
 freedom of motion, has led physicists to ask the question 
 whether there is anything else but these electrons. If all the 
 properties of matter can be explained on such a supposition, it 
 will be logical to admit that the so-called material atom is noth- 
 ing but a group of electrons. Inertia can be explained very 
 well as has been shown for the negative electrons- 
 Moreover it is possible to conceive that, the small 
 parts of the atom composed of one positive and 
 one negative electron rotating around a common centre 
 possess an apparent mass quite different from that which would 
 have to be attributed to each separately because the radius of 
 the rotating double elementary quantity would be different from 
 the sum of their two radii, and the rotation itself would consti- 
 tute a very important factor of this inertia. 
 
 The systems of electrons, or whatsoever the component parts 
 of the atom may be, are the same in all substances. This seems 
 to be an inevitable conclusion from the law that the absorption 
 of rapid kathode rays is directly proportional to the density of 
 the absorbing medium without being influenced by any of the 
 physical or chemical conditions which this medium may assume. 
 A quantitative study of this absorption shows that the real 
 radius of the impenetrable system of two electrons must be less 
 than 0. 3 X 10' 10 mm, and that the relation between the volume 
 
78 CONSTITUTION or MATTER. 
 
 of the atom and that of a system of electrons is less than 10" 9 . 
 But the efficient radius is always greater, and consequently the 
 above law which practically states that the efficient radius is 
 the same for equal mass-lengths of all substances, supposes 
 that the strength of the electric field is the same for the compo- 
 nent parts in all atoms, In other words, the study of kathode 
 rays has taught us the probable unity of matter, 
 
 Even more than this, it leaves us to suppose that the one and 
 only thing which exists as a substance in the physical world is 
 the universal ether. The electrons may be nothing but a local- 
 ized condition of this same ether and consequently matter itself 
 would be another modification of the universal substance which 
 fills all space. Thus a grand unity which may eventually even 
 lead us to a better understanding of gravitation possibly the 
 most mysterious force of nature, would be at last established in 
 our concept of the physical world. 
 
BIOGKAPHTCAL. 
 
 Nicholas M. Wilhelmy was born in Bech, Luxemburg, 
 February 23, 1880. His first education was received in the 
 public school of the same place. His secondary instruction 
 was begun at Differt, Belgium. In September 1896, he entered 
 the noviciate of the Society of Mary at La Bousselaye, France, 
 and the following year began to study philosophy in the 
 scholasticate of the same Society at Paignton, South Devon, 
 ED gland. He came to the United States of America in 
 September 1900, and continued his studies successively in the 
 Marist College, Washington, D. C., and in Jefferson College, 
 La., from which he was graduated in 1903 with the degree of 
 Bachelor of Arts. He was ordained to the priesthood in June, 
 1904. After matriculating at the Catholic University of 
 America in 1903, he followed the courses in physics and 
 chemistry under the Faculty of Philosophy and that of applied 
 mathematics in the School of Technology. 
 
INDEX. 
 
 CHAPTER I. FIRST PERIOD. 
 
 Earliest Observations and General Phenomena 1 
 
 Valve-Tube and Funnel-Tube 5 
 
 Striation 7 
 
 Influence of a Magnet on the Electric Discharge 8 
 
 Spectroscopic Study of the Light in Vacuum Tubes 9 
 
 Factors of the Discharge 9 
 
 Effects of the Discharge 11 
 
 CHAPTER II'. SECOND PERIOD. 
 
 I. W. Hittorff ,..14 
 
 II. E. Goldstein 18 
 
 III. Other Observers 21 
 
 1. General Phenomena and Properties 22 
 
 2. Factors of the Discharge ...... r 23 
 
 3. External Influences , 27 
 
 4. Effects of the Discharge 28 
 
 Heating 28 
 
 Fluorescence 30 
 
 Mechanical Effects , 32 
 
 Chemical Effects 32 
 
 CHAPTER III. THIRD PERIOD. 
 I. Conductivity of Gases. lonizatiori 34 
 
 1. Influence of Ultra- Violet Light on Gases 34 
 
 2. Influence of Heat on Conductivity 39 
 
 3. Influence of Kathode Rays on Conductivity 40 
 
4. Influence of Roentgen Rays on Conductivity 41 
 
 5. Influence of the Becquerel Rays on Conductivity 42 
 
 6. Chemical Sources on lonization 43 
 
 7. lonization of Liquids by Zerstaeubung 44 
 
 II. Phenomena Connected with the Discharge 45 
 
 1. Kathode Rays 45 
 
 Path of the Kathode Rays 46 
 
 Effects of the Kathode Rays 48 
 
 Kathode Rays from Ultra- Violet Light 49 
 
 Reflection of the Kathode Rays 51 
 
 2. Roentgen Rays 53 
 
 3. Canal Rays 54 
 
 III. Potential Phenomena 56 
 
 1. Fall of Potential . 56 
 
 2. Kathode Fall 57 
 
 CHAPTEE IV. THEORY OF THE DISCHARGE. 
 
 I. Vocabulary 59 
 
 II. Earlier Systems 61 
 
 III. Electron Theory 62 
 
 1. Foundation of the Electron Theory 62 
 
 2. Application of the Electron Theory to the Discharge 
 
 of Electricity through Gases 64 
 
 Kathode Dark Space and Kathode Light 66 
 
 Dark Space 67 
 
 Positive Light, Striated and Unstriated 67 
 
 Anode Layer 68 
 
 3. The Electron Theory and the Different Kinds of 
 Radiation 70 
 
 Kathode Rays 70 
 
 Canal Rays , 74 
 
 Becquerel Rays 75 
 
 The Constitution of Matter . . . . 77 
 
 THf 
 
 UNIVERSITY I 
 
YC 1075; 
 
i