414 Prof. E. Wiedemann on the Behaviour of Gases 



mirror. It is, further, of importance that the quantity of 

 electricity produced can be varied simply by a change in the 

 velocity of revolution of the machine and without alteration of 

 the connexions. The introduction of resistances (which would 

 cause complications in the processes of discharge) is therefore 

 rendered unnecessary. 



By the introduction of resistances into the circuit the mode 

 of discharge is in general not altered, but the quantity of elec- 

 tricity which passes suddenly in each discharge is increased. 

 Conditions are much more complicated when the induction- 

 coil is employed*. 



At each breaking and closing of the primary current we 

 have, besides the principal discharge, a whole series of partial 

 discharges of decreasing intensity. That we have not only 

 a single discharge traversing the tube is seen by the use of the 

 rotating mirror. But each of the partial discharges lasts a 

 certain length of time, during which the electricity flows gra- 

 dually from the coil to the electrodes ; it will therefore again 

 either be itself composed of a series of partial discharges, or 

 correspond to a more or less continuous discharge of electri- 

 city. The images of the discharges do not consist of separate 

 sharp lines, as with the electrophorus-machines, but of more 

 or less broad bands in which maxima and minima occur. 



The total heating in the tube is, as shown above, always 

 dependent upon the total quantity of electricity transmitted, 

 without reference to the quantity in each discharge; so that 

 the result obtained by Naccari and Bellatif is intelligible 

 without further explanation. 



Hence it results that at different pressures of the gas in the 

 discharge-tube the duration of the induced currents, and the 

 quantity of electricity furnished by the inductorium in each 

 discharge, is essentially different, since the potentials at the 

 ends of the coil must rise to different magnitudes before the 

 flow of electricity takes place. If the pressure of the gas be 

 very high, then no doubt only the principal discharge passes 

 through. 



From all these considerations, it follows that the induction- 

 coil is not suitable for the quantitative study of the discharge 

 or of spectral phenomena ; for each of the partial discharges, 

 being of a different strength, will produce a different spectral 

 phenomenon, so that one may produce a line spectrum and an- 

 other a band spectrum. 



Exactly similar remarks hold good also for the discharges of 

 large batteries of Ley den jars, only that here another circum- 



* Compare G. Wiedemann, Pogg. Ann. clviii. p. 286 (1876). 

 t Naccari and Bellati, Beibl. ii. p. 720 (1878). 



