408 F. Dellmanu on Atmospheric Electricity. 



former ring of brown shell-lac was replaced by one of white. 

 A remark agamst Coulomb, made by Riess in his memoir " On 

 the determination of the Density of Electricity in the Torsion 

 Balance *," was here corroborated in a surprising manner. The 

 tirst series of measurements on employing the new ring gave the 

 following numbers . — 



0-59 (28), 0-56 (28), 0-54 (29), 0-50 (32), 0-58 (32), 0-56 (51), 

 the angle of deflection at the same time diminishing from 71^° 

 to 9'. To give an example of the capriciousness of even this 

 shell-lac, I may mention that after the above measurements were 

 made the sphere was again charged, without being otherwise at 

 all touched, when in the first succeeding 34 minutes a loss of 

 0-78 per cent, per 25 seconds took place, and in the following 

 30 minutes a loss of equal amount. Why, without the least 

 apparent cause, should the loss in 1 hour 4 minutes be now so 

 much greater than that which before took place in 3 hours 

 20 minutes ? 



In all probability some electricity still adhered to the new 

 shell-lac ring, since it was made use of immediately after being 

 constructed On this account the sphere was placed in metallic 

 connexion with the earth for eight hours before being used for 

 measuring the atmospheric electricity ; when so used, a few drops 

 of rain fell, and the air was strongly negatively electric. The last 

 positive charge which the sphere brought down was allowed to 

 remain in order to measure the loss. Three times successively, 

 at intervals of 34, 34, and 69 minutes respectively, this loss 

 amounted to 0'30 per cent, per 25 seconds f ; in the next fol- 

 lowing 61 minutes it was 0"46, and in the succeeding 62 minutes 

 as high as 1*27. It was here, for a second time, observed that 

 slight moisture by rain-drops renders shell-lac more and more 

 conducting in presence of electricity ; for in a former series of 

 measurements the sphere, with drops on it, had stood uncharged 

 in the room from 10 p.m. to 6 a.m., and still the first measure- 

 ments in the morning only indicated the ordinary loss. It is 

 true that after the first 14 minutes a slight increase in the loss 



* PoggendorfF's Annalen, vol. Ixxi. p. 369. 



t This small loss appeared extraordinary, but it was explained as in the 

 text, since in all probability the quality of the atmospheric electricity had 

 meanwhile changed. As this fact was not, however, established by ex- 

 periment, the positive electricity which the sphere brought down at a sub- 

 sequent period was also reserved for the measurement of the loss. The 

 numbers obtained were 



0-44(11). 0-48 (15), 0-50(23), 0-50(22), 0-50(23), 0-49(79). 

 During the 2 hrs. 53 mins. occupied by these measureroeats, the loss, there- 

 fore, was as usual ; the quality of the atmospheric electricity, however, had 

 not altered during this time, — a fact which, though easily anticipated from 

 the state of the weather, was nevertheless verified. 



