100 PHENOMENA ACCOMPANYING 



At 4""™ the rotation is too rapid to allow its degree to be observed ; at 3™™ 

 it appears completely to cease. The rose-colored aureole is very vivid when 

 the summit of the rod of soft iron is positive. When there is no longer any 

 rotation, there is observed, as with atmospheric air, a movement of depression 

 and of ascension under the influence of magnetization. 



The presence of vapor modilies in some important particulars the results 

 obtained with dry gases. The following is an experiment made with ordinary 

 air subjected to a pressure of 2™™, into which vapor of water has been intro- 

 duced in successive quantities, so as to increase that pressure solely by the 

 eflect of the presence of the vapor: 



Pressure. Number of rotations in a minute. 



Rinff, positive. Ring, negative. 



2 mm 

 ..........••••••--•- ,, ...................... ,, 



4'"'° „ M 



6°>m „ 92 



8™™ 140 70 



10™™ 120 52 



12™™ 90 50 



14™™ 80 48 



We sec that at an equal pressure the rapidity of rotation is greater with 

 vajjor of water than Avitli dry air, which is attributable probably to the greater 

 facility with which the electric discharge is transmitted. With the external 

 air of a mean humidity, we have, with a pressure of 14™™, 72 revolutions 

 instead of 80 Avhen the ring is positive, and 44 instead of 48 when it is 

 negative. 



But the most characteristic fact produced by the presence of watery vapor 

 is the division, under the influence of magnetism, of the single current into 

 several small distinct and equidistant currents, which turn like the radii of a 

 wheel. This division is only observed when the ring serves as a positive 

 electrode. At a pressure of 6™™ the single current begins with turning, then 

 (ixpands, whereupon the rotation is no longer perceptible ; but at the pressure 

 of 8, of 10, and of 12™™ this current, from the commencement of its rotation 

 under the action of magnetism, divides into five or six small sti-eams which 

 turn, as was just said, like the radii of a wheel ; while, when the air is dry, the 

 current never divides ; but, under a weak pressure, it merely expands into a 

 sector or a circle of which all the parts arc continiious. 



When the riug is negative, and there is vapor present, it will be seen that 

 the current which issues from the summit of the iron rod presents, where it is 

 in contact with the iron and at the moment when this is magnetized, instead of 

 a continuous smface, a series of small brilliant points, which seem points of 

 emanation for as many small currents, too little distant from one another to 

 hecome distinct. Here, then, this current, which does not divide into separate 

 filaments, simply undergoes dilatation or expansion at the point where it is in 

 contact with the iron. 



The vapor of alcohol produces similar effects with the vapor of water. The 

 single current is, in this case, much more brilliant than with dry air or with the 

 vapor of water ; it presents fine stratifications, Avhich give it an appearance not 

 unlike that of a caterpillar. Magnetization expands and divides it into several 

 currents, sensibly larger than those observed Avith the vapor of water. If, 

 however, the diameter of the ring is too large, greater, for instance, than fifteen 

 centimetres, the subdivision of the current is not effected without difficulty, 

 unless the intensity of the discharge and that of the magnetization be very 

 considerable. 



The following is an experiment in which, the rarefied gas being hydrogen, 

 different portions of alcoholic vapor were successively introduced. The pressure 



