116 Effect of an Aurora Borealis on the Magnetic JVeedle. 



subsequent observations seem to indicate to have been the result of 

 causes foreign to those producing the regular diurnal changes of dip. 



The observed changes for this day, just given, are not entirely- 

 regular, and should be considered, of course, in the light of particu- 

 lar results afifording merely a term of comparison, which is sufficient- 

 ly accurate for the purpose in view. 



I now give a table of some of the observations for May 17th, with 

 a column of remarks, the portion of which relating to the aurora is 

 drawn from the description by Mr. Espy, already given. 



By comparing the fourth column of this table with the correspond- 

 ing column of the first table, we find throughout the day a general 

 accordance in the relative positions of the needle; a minimum of wes- 

 terly variation at about 8 A. M., a maximum between 1 P. M. and 

 4 P. M., a second minimum about 8 P. M., a tendency towards a 

 second maximum, which was interrupted by the aurora. There 

 does not appear to have been any marked change from 8J P. M. to 

 9 ; so that the first part of the phenomenon does not seem to have 

 affected the variation. Observations are wanting to show when the 

 effect began to be felt, and when it was at the greatest, and the near 

 coincidence of the observation at half past ten, with the time of the 

 disappearance of the arch, must, of course, be regarded as acciden- 



