THE ROBERTS* RANGE-FINDER. 
367 
sights on the exterior of the tripod-telescope are made nse of. The 
telescope on the rod is removed, and the base made as long as possible. 
In other respects the range is taken as already described. 
Neither of the prisms are placed over the index points ; but the 
distance (about lOins.) from the prism on the rod to the 1 on the rod, 
is equal to the distance from the index on the tripod-telescope to the 
prism in its sunshade. This arrangement enables the observer to 
hold the rod steadily on his shoulder, and does not effect the accuracy 
of the instrument. 
If the parallel lines in the field of the telescope are cut on glass, the 
instrument is absolutely independent of any adjustment. Should, 
however, spider webs be employed, it will be necessary to test their 
distance apart occasionally. This can be easily done by measuring off 
on the ground a distance of 40 yds. (twenty times the length between 
| and 8 on rod) and ascertaining if, at that distance, they subtend the 
length from 16 to I on scale. Should they not do so, they can be 
corrected in the usual manner. 
Another method of correcting the error arising from this cause will 
be described further on. 
The clear definition of distant objects afforded by the telescopes 
admits of the use of a comparatively short base, which should be 
selected so as to lie between ^Vfih and ^jth in. of the estimated range. 
The following are recommended :— 
Approximate range. Base. 
1000 yds. 30 yds. (about), 
1500 „ 40 „ 
2000 „ 50 „ u 
8000 „ 65 „ 
Above 3000 a . 75 « « 
I took six observations to the same object the second time I saw the 
instrument. The results varied from 1600 to 1640 ; the distance, as I 
measured it on the ordnance map, was 1615 yds. The base used was 
about 35 yds. 
With a 6 ft. rod (graduated from 800 to 4000), the base limits 
are from 80 yds. to 16 yds. The range limits are 16,000 yds, and 
128 yds ; so there can be no difficulty in obtaining any ordinary range. 
In connection with a plane-table, which is interchangeable with the 
telescope, the instrument appears to be well adapted for making rapid 
survey sketches for military purposes—more especially over ground, 
such as an enemy's position, which cannot be approached, or in a close 
country where a plane-table would be of but little use. 
From a limited number of ranges taken with the instrument, the 
probable error at 2000 yds. distance was found to be 1*3 per cent . It 
is, however, hardly fair to compare this result with the probable error 
of other range-finders, as the conditions under which the observations 
were made may not have been the same in each case. 
I hope, shortly, to furnish a statement showing the results obtained 
by the range-finder in the hands of non-commissioned officers. 
