34 
HINTS TO TRAVELLERS. 
This operation must be repeated several times in order to get a good 
mean value. The bubble of a level is very susceptible to changes of 
temperature (heat makes it lengthen and cold contracts it), so care 
must be taken that it is not exposed to such changes while this operation 
is being performed. Should there be any chance of the bubble altering 
its length while you are determining the value of the divisions of the 
scale, it will be necessary to read and record both ends of the bubble. In 
observing, as described previously, for each vertical angle taken, the 
readings of both ends of the bubble must be recorded. To apply the 
correction the rule is as follows 
Divide the difference between the sums of the readings of the object 
end and eye end by the total number of readings, and the result will be 
the dislevelment in terms of divisions of the scale. Multiply this result 
by the angular value of one division of the scale, and you obtain the 
angular correction for dislevelment to be applied to the mean vertical 
angle. Supposing two observations ara taken to a point one face left and 
one face right, and the readings are as follows 
o. E. 
F. L. 
F. R. 
. 5 8 
. 7 6 
12 14 
In this case the sum of the readings of the eye end exceeds that of 
the object end by two. The number of readings of ends of the bubble is 
four. So to get the dislevelment in terms of division of the scale we 
must divide 2 by 4 = Suppose the value of one division of the level 
scale is 16 seconds, then to get the correction .we must multiply 
\ by 16" = 8 seconds. The eight seconds of arc must be applied to the 
mean of the two observed angles. With regard to its sign, the eye end 
being in excess, the correction must be subtracted from an eleva¬ 
tion and added to a depression. If the object end were in excess, the 
process would, of course, be reversed, or correction to altitude = 
Q_ 
+ - 1 -«--tt— x value of 1 division. 
number of readings 
The magnetic needle is used in the following manner:—Attach it 
