in highly Rarefied Elastic Fluids. 251 



greater or less resistance which the gaseous medium opposes to 

 the passage of the discharge, since the oscillations are larger 

 with nitrogen than they are with hydrogen, and they diminish 

 with the pressure after a certain point, namely that at which the 

 discharge is able to take place completely and the intensity of 

 the oscillations attains its maximum. 



The stratification of the electric light appears, therefore, to he 

 a phenomenon analogous to the production of sound- waves — that 

 is to say, a mechanical phenomenon resulting from a succession 

 of isochronous impulses acting upon a column of rarefied gas, 

 and produced by the series of electric discharges following each 

 other in very rapid succession. A further confirmation of this 

 view is afforded by the disturbance of the stratifications caused 

 by a displacement of the gaseous matter and the consequent 

 alteration of the equilibrium of the elastic fluid in which the 

 stratifications appear. In order to produce such a disturbance, 

 it is sufficient to introduce into the tube containing the rarefied 

 elastic fluid an additional quantity of the same gas, so as to in- 

 crease the pressure by |, or at most J a millimetre. The fol- 

 lowing is what takes place with hydrogen, the effects being the 

 same with tubes 15, 50, or 100 centims. in length. 



First of all the gas is rarefied as far as 2 millims., so as to 

 produce the phenomenon of stratifications in as marked a manner 

 as possible. A small quantity of hydrogen is then allowed to 

 enter. If the gas enters near the negative electrode, striated 

 slices of a beautiful rose-colour are immediately seen to form in 

 the dark space ; their diameter is the same as that of the strati- 

 fied column — that is to say, of the tube ; but they are very narrow 

 and very sharp. They advance gradually into the tube, be- 

 coming mixed up with the original much wider and less well- 

 defined striae; then, on stopping the entrance of the gas, the 

 luminous column is seen to recede slowly from the negative 

 electrode, and gradually to resume its first appearance. When 

 the gas enters near the positive electrode, in place of strise occu- 

 pying the whole width of the tube, there appears a brilliant jet 

 of very small diameter (2 or 3 millims.) sharply striated, and 

 exactly like a small spiral spring, which advances along the axis 

 of the tube in the comparatively dark interior of the luminous 

 column, which last, when the gas begins to enter at the positive 

 electrode as well as at the negative, immediately advances so as 

 to occupy the whole of the dark space as far as the negative 

 electrode, from which it is separated only by a stratum of 2 mil- 

 lims. in thickness which it cannot pass. Afterwards, when the 

 entrance of the gas is arrested, everything soon returns to its 

 normal condition. At whichever extremity of the tube the gas 

 is allowed to enter, its entrance is seen to be accompanied by 



