DISCHARGK OF KI.IX TKICITY. 241 



cut, tho first a citron <ir(H'ii, the second a more refrangible .i^reeu, and 

 the third a bhie, 1 am inclined to tliink must be due to mercury vapor 

 from the pumj). 



1 am in(lel)ted to Prol'. Liveinj;- for the h>an of a very tine direct- 

 vision s])ectroscope, and to him and Mr. IJobinson, of the Cambridge 

 Chemical Laboratory, for valuable advice in the attempts which 1 made 

 to plioto,i>rai)h the spectra of some phosphorescent glows mentioned 

 ])clow. 



1 should like to call attention to the advantages for s[tectroscopic 

 purposes which attend this method of ])roducing the discharge; it is 

 easily d<uie either by an ordinary electrical machine or an iuducti(m 

 coil. An intensely bright discharge is got, and there is no danger of 

 comi)licatiou arising from the spectrum of the gas getting mixed with 

 tliat of the electrodes. 



[)isc]i(t >•{/(' ill 'hriK/cii. — l>y far the most remarkable ai>])earance is pre- 

 sented when the discharge passi-s through oxygen, tor in this gas the 

 Ijright dischaige is succeeded by a i)hosphorescent glow which lasts 

 tor a considerable time; indeed, with a strong discharge it may remain 

 visible for more than a minnte. When the discharges sncceed one 

 another pretty rapidly, the phosphorescence is so strong that it hides 

 the successive bright discharges, and the tube seems pernmnently full 

 of a bright yellow fog. We can thus, by the use of this gas, convert 

 the intermittent light gi\en by the bright discliarge into a continuous 

 one. 



Perhaps the most striking way of showing this phosphorescence is to 

 use a long tube, about a meter long and (3 or 7 centimeters in diameter, 

 with a bulb blown in the middle, the primary coil l)eing' twisted round 

 this bulb. Then, when the si)arks i)ass between the jars, a bright ring 

 discharge passes through the bulb, from which, as if shot out from the 

 ring, the phosphorescent glow travels in both directions along the tube, 

 moving slowly enough for its motion to be followed by the eye. It can 

 not, therefore, be produced by the direct action of the light from the 

 spark on the gas in the tube, for if it were, the glow would tra^■el with 

 the velocity of light. It is necessary to meution this point, for the 

 light from these discharges has great powers of producing phosphores- 

 cence. 



Tlie glow seems to consist of gas which has been in the path of the 

 discharge, and whose molecules liave been si)lit up by it and i)rojected 

 from the line of discliarge. This gas wliich, when projected, is in a 

 l)e(;uliar state, })y a pro(;ess of chemical combination gradually returns 

 to its original condition, and it is while it is in this state of transition 

 from its new condition, to tlu; old that it phosphoresces. If this is the 

 case we should expect that the period of phosphores(;ence would be 

 shortened by raising the temperature. On trying the experiment I 

 found that this took place to a very marked extent. A disclnnge bulb 

 tilled with oxygen at a low ])!-essure was jtlaced o\t'r a iiunscn burner; 

 li. Mis. 114 i<) 



