692 [June 18, 



obvious that the law of distribution of electricity on the surface of a 

 conductor gives the end of the pipe a great advantage over other 

 portions of it as regards power of collecting electricity from the air. 

 Notwithstanding that the tendency of rain or other downfall is to 

 make the electricity appear weaker than it really is, by virtually 

 shortening the pipe, the observed electricity is generally much 

 stronger during heavy showers than at any other time. 



1 7. The errors resulting from these various causes, though not 

 insignificant in themselves, are very small in comparison with the 

 variations which really occur in the electrical potential of the air, 

 as will be seen by the most cursory glance at the tabulated ob- 

 servations. 



18. My observations have generally been taken regularly at three 

 stated times in the day, viz., between 8 and 9 A.M., between 2 and 

 2| P.M., and between 9 and 9J P.M. ; and in many instances ob- 

 servations have been taken at other hours as well. On Sundays 

 some of the observations have generally been omitted. Either during 

 or immediately before or after each electrical observation, I have 

 also observed barometer, dry- and wet-bulb thermometer, cloud, wind, 

 and state of weather generally. The barometer used is an Aneroid 

 of the usual size, nearly new and in good condition. I have ascer- 

 tained, by experiment, that it is affected by temperature, the disturb- 

 ance being in the same direction as for a mercurial barometer. It 

 has also an index error of about *06, reading too low by this amount ; 

 but I have not applied any correction for either cause ; and it would 

 scarcely be worth while to do so, as very little connexion appears to 

 exist between the fluctuations of the barometer and those of atmo- 

 spheric electricity. 



19. My two thermometers (dry- and wet-bulb) are mercurial with 

 unusually long degrees. They are placed outside the window of a 

 room in the second story, in which there is never any fire ; and are 

 read through the window, from which they are about six inches 

 distant. The window faces the north-west, and the thermometers 

 are well protected from radiation except from the window, while at 

 the same time exposed to a free current of air. 



I have carefully tested the thermometers in melting snow, which 

 showed that at 32 the dry-bulb read '4 too high, and the wet-bulb 

 7 too high. I have also tested them in water at various tempera- 



