Decembee 18, 1896.] 



SCIENCE. 



895 



up their study of the alkali metals in 1894.* 

 Cells containing metallic Na, K and E.b 

 were prepared by methods described in 

 earlier papers.f An atmosphere of hydro- 

 gen was used in each case, the pressure be- 

 ing reduced to such a value (about ^ to ^ 

 mm.) as would make the apparatus most 

 sensitive. All measurements were made 

 with a galvanometer, current passing be- 

 tween the illuminated surface and a ring of 

 wire placed just in front of it. 



90. Rubidium was found much more sen- 

 sitive than either Na or K. The maximum 

 effect was also further toward the red than 

 in the case of ISTa and K, as shown by ex- 

 periments with various absorbing solutions. 

 For E,b the maximum was in the yellow, 

 with a very noticeable effect in the orange 

 and red. With both Na and K the max- 

 imum action was in the blue. 



91. Elsterand Geitel draw the conclusion 

 that '^not only does the actino- electric sen- 

 sitiveness increase with the electro-positive 

 character of the metal, but the electro-posi- 

 tive metals also show an increased tendency 

 to be affected by light of greater wave- 

 length." 



92. In this same article ;{; observations are 

 also described showing the effect of illumi- 

 nation in helping the development of Hertz 

 vibrations in vacuum tubes. The effect is 

 especially marked when one of the alkali 

 metals is present. These observations are 

 intimately connected with work by Zehn- 

 der, Warburg and others on the behavior of 

 Geissler tubes containing sodium. To fol- 

 low out the matter in detail would require 

 the considerable literature of this subject to 

 be consulted. I shall therefore merely state 

 the conclusion reached, viz : that in the 

 presence of one of the alkali metals the 

 effect of illumination is to render the gas 



*Wied. Ann. 52, p. 433. 



fA special mode of procedure was necessary in the 

 case of El). 



X 1. c, p. 445. 



capable of conducting the rapidly oscilla- 

 ting currents from a Hertzian oscillator.* 

 Taken in connection with Arrhenius' dis" 

 covery that rarified gases conduct electroly- 

 tically when illuminated, f this result has 

 an important bearing upon J. J. Thomson's 

 suggested explanation of actino- electric phe- 

 nomena.]; 



BEHAVIOR OF FLUORESCENT MINERALS. 



93. The theory advanced by Arrhenius, 

 by which many phenomena of atmospheric 

 electricity are explained as a result of the 

 action of sunlight in discharging negative 

 electricity from the earth's surface, seems 

 to require that other substances besides the 

 metals should be acted upon. A number 

 of minerals, woods, etc., were therefore 

 tested by Lampa,§ with negative results. 

 Preliminary experiments by Elster & Geitel || 

 also failed to show any action of sunlight 

 upon such substances. The sensitiveness 

 of one non-metallic substance (Balmain's 

 luminous paint) had however already been 

 proved.^ This substance being phosphor- 

 escent, it seemed not improbable that other 

 phosphorescent materials would show the 

 effect. Later experiments by Elster and 

 Geitel** confirm this view. Various arti- 

 ficial phosporescent powders (e. g., the sul- 

 phides of Ca, Sr, etc.) were found to show 

 greater or less actino-electric sensitiveness, 

 according as they were more or less strongly 

 phosphorescent. 



94. To observe the action of sunlight upon 

 phosphorescing minerals, more sensitive 

 apparatus was needed than that previously 

 used. The arrangement of apparatus is 

 shown in the figure. A dish, D, of oxy- 

 dized iron was enclosed in a grounded 



* See Wied. Ann. 52, p. 454. 



t Wied. Ann. 32, p. 545 ; 33, p. 638. 



J See 'i I 112, 113, below. 



I Wiener Anz. 1890. Beibl. 15, 384. 



II Wiener. Berichte 99, p. 1008. 

 i Wied Ann. 38, p. 507. 

 **Wied Ann. 44, p. 722. 



