220 
THE SCIENCE OF FRONTIER DELIMITATION. 
must be carefully fixed in latitude, so that the work may be verified. 
Time observations should also be taken, in high latitudes the question 
of curvature must be taken into consideration if a parallel is being run 
with a theodolite. 
(3) A meridian of longitude .— 
A meridian is run in the same way as a parallel, the difference being 
that the starting and closing points must be carefully fixed in longitude. 
This means that about a fortnight must be spent at each place observed. 
The results must be computed on the spot. A good triangulation, if 
accepted by both commissions, would save the necessity of the closing 
point being observed, provided good observations were made for 
azimuth of the base. Or, if time is important and the country is of 
no great importance, both sides might be content to accept a line run 
from the starting point with magnetic bearings. 
A meridian also may be run by latitude and azimuth, i.e., by taking 
the true bearing of some distant point nearly due north or south, 
moving to this point and observing the latitude from which it is easy 
to.calculate the distance of the object east or west of the line. This 
method, however, is only applicable to country where long bearings 
can be taken, and what with the possible error in latitude, and the 
setting off of an east and west line from the object to the meridian and 
measuring it, it is probably not very accurate. 
(4) A straight line joining two points at some considerable distance 
apart .— 
If the two points are not accurately fixed, this is a difficult line to 
follow, unless the country should be very open with one or two elevated 
positions, in which case flags might be aligned between the terminal 
points. But if the country is broken or covered with bush, this is out 
of the question. The best course is then to make a careful triangulated 
survey of the intermediate country, from which the relative position of 
the terminal points can be fixed and the azimuth from one to the 
other calculated. If this cannot be done, I know of no other method 
but to follow the true bearing of one point from the other, as nearly 
as can be judged, or as taken from the best map available, to note 
where this line cuts an east and west or a north and south line, as the 
case may be, passing through the second point to measure the distance 
of the point so obtained from the second point and also to carefully 
record the distance traversed from the first point. From these data a 
corrected bearing of one point from the other can be worked out. 
The line must then be delimited. This involves going over the 
ground twice. 
In all these forms of delimitation the perambulator should be used 
to give the distances along the line, and these should be checked fre¬ 
quently by times and latitudes. The commissioners must follow the 
actual line as closely as possible, and it will generally be necessary to 
cut a road. If a triangulation is not made, the azimuth will, at any 
rate, give one co-ordinate to which to refer the topography. The 
perambulator distances will enable the topographer to obtain a second 
