horizontal needle when under the influence of magnets. 391 
terrestrial forces will not be materially affected by the changes 
which may have taken place in the temperature of the mag- 
nets during the times of observation. With regard to the 
observations in the meridian, that is, with the needle pointing 
north or south, they could not be much affected by changes 
in the temperature of the magnets, especially by such small 
changes as I am persuaded alone took place : the extent of 
the deviations may, in some instances, have been slightly in- 
creased or diminished, but their directions could not be 
changed. 
I have before mentioned that both Mr. Barlow and myself 
found some anomalies between the observations in-doors and 
those at the same points in the open air. When the times of 
the maxima shall have been determined independent of the 
temperature of the magnets, I expect it will be found that 
these anomalies have arisen from the difference in the changes 
of temperature in the magnets when in-doors and when in 
the open air : of this I only feel that degree of doubt, which 
should always be entertained until a fact is established. 
If such observations as I have given were continued for a 
length of time, particularly those near the east and west, I 
certainly expect that they would lead to important conclu- 
sions respecting the causes of the diurnal variation, and I 
regret that, as I have not the time to devote to them myself, 
I must leave them to be made by others possessed of more 
leisure. Should they be undertaken, the necessity of ascer- 
taining in the first instance, the effects which changes of 
temperature have on the forces of the magnets employed, 
and of observing the temperature of the magnets themselves 
