THE SCIENCE OF FRONTIER DELIMITATION. 
213 
some method of determining longitude absolutely must be adopted. 
This means that about a fortnight must be spent in carrying out 
observations, unless Capt. HilFs camera is used, when three or four 
days would suffice ; the moon also should be in the first quarter. With 
Capt. HilFs camera, it should be within a few days of full moon. It is 
best to arrange the march up with reference to these conditions, but in 
case the boundary is reached at a time when the moon is not in a 
favourable quarter, a longitude of some other place on the boundary 
may be determined, provided that it is carefully connected by triangu¬ 
lation or by the best means available with the initial point of the 
boundary. 
The latitude of this point should be determined by one or two 
circummeridian observations of north and south stars with about the 
same elevation. In taking circummeridians, about eight pairs of 
observations should be made of each star, and each pair should be 
worked out for latitude independently. In this way any very dis¬ 
cordant results can be eliminated, and it is easy to see whether the 
watches time of transit has been correctly found. 
Longitude observations need not be worked out on the spot, unless 
a meridian is being run. 
For carrying out the survey of the boundary, a triangulation is by 
far the best method. It involves a considerable amount of labour, and 
if the country is covered with bush, it becomes a matter of great 
difficulty; nevertheless, if it can be managed at almost any cost, it pays 
well. It gives a permanent value to the work, and puts every plan and 
every physical feature into its place with great accuracy. With the 
fixed points obtained from a triangulation, the filling in of the topo¬ 
graphy becomes a very simple matter. Such a triangulation cannot, 
of course, compare with the geodetic survey of a civilized country, a 
work which is executed at great cost and which takes a long time to 
complete; but it will be found quite accurate enough for the scale on 
which boundary maps are drawn, and if made with reasonable care, it 
will fit on to fixed points with great exactness. 
In carrying out such a triangnlation, and indeed in all other depart¬ 
ments of the delimitation work, one has to make a compromise between 
what is desirable and what is possible. Just as in a soldier’s training, 
many regulations strictly enforced in peace time have to be thrown 
overboard on service, so in boundary survey work, and delimitation 
many matters which are considered indispensable at home must be 
sacrificed to the difficulties of country and the exigencies of time. For 
an account of the modification of the rules in force on high class 
geodetic surveys, which may be adopted in boundary work, I know of 
no better advice than that contained in a small pamphlet on “ Eapid 
Triangnlation/’ by Major Hon. M. G. Talbot, E.E., compiled at the 
Intelligence Division. It is very desirable that the country to be 
triangulated should be reconnoitred first, but I doubt if this is often 
possible. If it is unexplored, and is covered with bush or otherwise 
difficult, a reconnaissance involves much labour and the expenditure of 
much time. It means the deciphering of the boundary more or less 
