318 M, P. Riess on the Passage of Electric Currents 



meter uo longer suffices for its measurement. Again, when the 

 discharge takes place in free air between a knob and a very fine 

 point, whose distance asunder is so regulated that no spark 

 ensues, the heating effect in the wire is very small ; whereas it is 

 very great if the fine point is replaced by a blunt one, so as to 

 cause a spark between the two electrodes. In both cases where 

 the spark appeared, the time during which the same quantity of 

 electricity was discharged was greatly diminished ; hence, accord- 

 ing to a well-known law, the heating effect was greatly increased. 

 Exactly in the same manner as we are able to change the mode 

 of discharging a battery through water and through free air by 

 bringing the electrodes nearer to one another, and by changing 

 the form of one of them, we can, as the new experiments show, 

 attain the same end in very I'are air by diminishing the negative 

 electrode. The battery charge can pass through rarefied air, 

 between arbitrarily-shaped electrodes, in two essentially different 

 ways. In the first, which might be called the glimmei'ing man- 

 nei', a reddish luminous cone of air takes part in the discharge, 

 its summit touches the positive electrode, and its base lies always 

 at some distance from the negative electrode. The stratum of 

 air in contact with a great part of the negative electrode takes 

 part in the discharge in a less degree, and glows with a peculiar 

 blue light. The second manner of discharge is that which I 

 have called discontinuous. A narrow cylinder of air touching 

 both electrodes takes part in the discharge, is rendered white 

 hot, and burst asunder in the same manner as a metallic wire is 

 which connects both electrodes, and whose diameter does not 

 exceed a certain limit dependent upon the charge of the battery. 

 The glimmering discharge takes place so slowly, that the sur- 

 faces of both electrodes remain uninjured, and only a small 

 amount of heat is generated in the whole discharge circuit, 

 whereas the discontinuous discharge tears away parts of the elec- 

 trodes, scatters them red-hot on all sides, and heats the discharge 

 cii'cuit. Whatever may be the distance between the electrodes, 

 the glimmering can be converted into the discontinuous discharge 

 by increasing the density of the discharged quantity of elec- 

 tricit)^, by the admission of air, or, as our late experiments justify 

 us in concluding, by diminishing the magnitude of the negative 

 electrode. That is to say, if the surface of the negative elec- 

 trode be very small in comparison to the surface which, with the 

 employed density of the discharged quantity of electricity, would 

 have been covered with a glimmering light, the discharge instead 

 of being glimmering will be discontinuous. The recorded ex- 

 periments are in complete harmony Avith this remarkable con- 

 clusion. When the direction of the discbarge current was from 

 the small platinum surface to the naked knob, the brass rod, and 



