11 
avoid any uncertainty due to their thickness, the scale n is pivoted on the pro¬ 
longation of its graduated edge, and the sliding index of the scale m has its point 
running parallel to the scale, and passing through its pivot as shewn by the 
dotted line. A third scale, p, may be employed for measuring the range, the 
point r being the position of the battery. 
One of these position-finders is installed at Fort Hamilton, in the roadstead of 
New York. For two months it has remained exposed to the weather without any 
covering, and subsequent thereto has given a series of ranges from 2000 to 5500 
yards, with a mean error of '33 per cent., the base being 270 yards. 
The theoretical figure 11 applies to the case where, for whatever reason, the 
chart D is placed with one side at an angle to the base, instead of parallel to it. 
The base AB is then represented by the line &B. We readily see that, in order 
to apply the rules we have already examined, it will suffice for the telescope L' 
and its scale m to be placed at an angle in horizontal projection equal to that 
between the lines AB and &B. The triangle of the chart will be similar to 
the triangle ATB in space when the galvanometer marks zero, this similitude 
depending in reality, not on the parallelism of the scale n and the telescope L, 
but on the equality of the angles which they intercept on their corresponding 
arcs, the arc h' being a fixture on the chart. 
Fig. ll. 
/ 
/ 
\ 
\ 
\ 
\ 
IY. Telemetric Position-Finder. 
Captain Fiske’s most recent improvement, not hitherto published, consists in 
adapting his automatic range-finder to the finding of a position. He unites the 
two telescopes by a second Wheatstone’s Bridge which gives the direction, whilst 
the first one continues to measure the range. This arrangement is obviously of 
great utility for indirect fire. Let L and L' be the two telescopes, installed at the 
observatories A and B (Fig. 12), and moving round the conducting arcs li and Ji'. 
