OF THE EARTH’S MAGNETISM. 565 
Trevandrum and Hobarton, than it may seem to merit. This agreement is shown 
in hundreds of other cases; and disagreement is an exception which can be traced 
in most cases to specific causes, as unusual disturbance. 
147. On some days observations were wanting, and the Trevandrum value 
was interpolated : all interpolations are indicated in the tables by brackets [ |. 
At Singapore, also, observations had been omitted in taking daily means, on 
account of disturbances ; all these omissions have been restored by me, and the 
true daily means obtained. 
148. Toronto.*—No attempt has been made to correct the great change of 
scale readings between February and October 1843. It may be noted that the 
temperature of the bifilar magnet was considerably influenced by the heat of the 
stove, as will be seen from a comparison in winter, of the last observation on 
Saturday evening with the first on Monday morning; the latter is generally from 
2° to 10° lower than the former, the stove probably not having been used on 
Sunday, or not so strongly heated. 
149. P.S.—I should have noted, that the conclusions in the preceding paper 
have been confirmed in various ways; and that where other observations discussed 
by me have not shown the same results, it has been from causes that it might have 
seemed invidious to mention. Before any value can be given to results opposed 
to those obtained in the preceding pages, it will be necessary, in the first instance, 
to have the truly corrected daily means; their projection, as in Plate XXVIL., 
will always show where the errors have occurred, and probably from what cause. 
* Lat. 43° 39’ 45” N., Long. 55 17™ 268 W. Observations, vols. i. and ii., published under 
the superintendence of Lieut.-Colonel E, Saninz. 
