VOL. LXXV.] PHILOSOPHICAL TRANSACTIONS. 70] 



that there is a limit even in the rarefaction of air, which sets bounds to its con- 

 ducting power ; or, in other words, that the particles of air may be so far sepa- 

 rated from each other as no longer to be able to transmit the electric fluid ; that 

 if they are brought within a certain distance of each other, their conducting 

 power begins, and continually increases till their approach also arrives at its 

 limit, when the particles again become so near as to resist the passage of the 

 fluid entirely, without employing violence, which is the case in common and 

 condensed air, but more particularly in the latter. These experiments however 

 belong to another subject, and may possibly be communicated at some future 



time 



It is surprizing to observe, how readily an exhausted tube is charged with elec- 

 tricity. By placing it at 10 or 12 inches from the conductor, the light may be 

 seen pervading its inside, and as strong a charge may sometimes be procured as 

 if it were in contact with the conductor : nor does it signify how narrow the bore 

 of the glass may be ; for even a thermometer tube, having the minutest perfora- 

 tion possible, will charge with the utmost facility ; and in this experiment the 

 phenomena are peculiarly beautiful. Let one end of a thermometer tube be 

 sealed hermetically. Let the other end be cemented into a brass cap with a valve, 

 or into a brass cock, so that it may be fitted to the plate of an air-pump. When 

 it is exhausted, let the sealed end be applied to the conductor of an electrical 

 machine, while the other end is either held in the hand or connected to the 

 floor. On the slightest excitation the electrical fluid will accumulate at the sealed 

 end, and be discharged through the inside in the form of a spark, and this ac- 

 cumulation and discharge may be incessantly repeated till the tube is broken. 

 By this means I have had a spark 42 inches long; and, had I been provided 

 with a proper tube, I do not doubt but that I might have had a spark of 4 times 

 that length. If, instead of the sealed end, a bulb be blown at that extremity of 

 the tube, the electric light will fill the whole of that bulb, and then pass through 

 the tube in the form of a brilliant spark, as in the foregoing experiment ; but in 

 this case I have seldom been able to repeat the trials above 3 or 4 times before 

 the charge has made a small perforation in the bulb. If again a thermometer 

 filled with mercury be inverted into a cistern, and the air exhausted in the 

 manner I have described for making the experiment with the gage, a Torricel- 

 lian vacuum will be produced ; and now the electric light in the bulb, as well as 

 the spark in the tube, will be of a vivid green ; but the bulb will not bear a fre- 

 quent repetition of charges before it is perforated in like manner as when it has 

 been exhausted by an air-pump. It can hardly be necessary to observe, that in 

 these cases the electric fluid assumes the appear nee of a spark,* from the nar- 



* By cementing the string of a guitar into one end of a thermometer tube, a spark may be 

 obtained as well as if the tube had been sealed hermetically. — Orig. 



