630 PROCEEDINGS OP THE AMERICAN ACADEMY. 



luminous masses are shot out from the anode and move toward the 

 cathode with increasing strength of current, and retreat to the anode 

 with decreasing strength of current. A striking fact, however, is this : 

 The movement of the luminous masses is toward the negative pole when 

 the pressure in the tube is suddenly increased by a small amount. This 

 movement, however, is dependent upon the strength of the current flow- 

 ing through the tube, and cannot be perceived with feeble currents. I 

 have employed currents from 10 milliamperes to 40 milliamperes. The 

 number of luminous masses increases with the strength of current. 

 The method of experimenting was as follows. 



The current was led to a reservoir of water. At the base of the 

 reservoir the water flowed out through a long glass tube of 1 cm. inter- 

 nal diameter. A wire introduced in this tube could be pushed in or out, 

 thus modifying the resistance of the circuit. This high resistance was 

 found sufficiently exact for the character of the experimentation. A 

 milliamperemeter (Siemens and Halske) was introduced in the circuit of 

 this water resistance and the exhausted tube while the water was kept 

 running. The indications of the milliamperemeter were perfectly steady. 

 Other forms of high resistance which I have tried heat under the power- 

 ful currents necessary for the production of the movements of the 

 luminous masses. Running water gave, as far as the indications of the 

 milliamperemeter was concerned, a perfectly steady resistance ; when 

 a telephone, however, was introduced in the circuit, rumbling crepita- 

 tions were heard, due probably to electrolysis in the liquid. The 

 exhausted tube also produces a note which is superposed on the rum- 

 bling of the liquid resistance. This latter note is especially noticeable 

 in comparatively large tubes. I experimented first with a tube 60 cm. 

 long, 8 cm. wide, provided with circular flat electrodes 3 cm. in diameter. 

 The striae were in constant movement, producing a loud hum in the 

 telephone introduced in the circuit. It was noticeable that the striae 

 were straight ; not curved toward the positive electrode, except near 

 the sides of the tube, where a curvature like that of a liquid meniscus 

 was formed. 



The tubes were maintained in connection with a mercury pump, by 

 means of which the pressure in the tubes could be changed to any desir- 

 able extent. 



The slow movement of the luminous masses was best shown in tubes 

 of 4 cm. internal diameter. Figure 1 is a photograph of a typical form of 

 such slow moving luminous effects. It will be noticed that the ends 

 of the mass are curved as if the centre of the mass constituted a positive 



