064 Scientific Proceedings, Royal Dublin Society. 
Tas_E showing the Effects of Wave-elevation of the signal on the 
observed altitude. 
Height of Waves, Distance in Miles. 
erest to hollow : 
in feet. 1 1k 9 3 
2 46” 34” B83” lq? iil” 
4 1’ 32” WV 46" 34" 23" 
6 2uellou 1 42" WP 51” 34” 
8 3 OF BY GY Ww Bil? etait 45" 
10 | ag" 2! 50" 1 Gey iY Og 56” 
12 4! 32" 3) 23” Zellow V’ 42” fey? 
In rough water it is more important to increase the distance 
from the signal than in smooth water. While half a mile, or 
even less, might give an accurate result in still water; in a sea-way, 
1 mile or 13 miles would be desirable. It is to be noticed that 
some correction reducing the amount of the error will generally be 
possible. Thus, if a six-foot sea is estimated erroneously to be an 
eight-foot sea, the error introduced is that arising from the sub- 
traction of 4 feet instead of 3 feet from the height of eye. This, 
at 1 mile distance of observation, gives an error of about 34” in 
the altitude, the altitude being over-estimated by this amount. 
If the distance from the signal be erroneously determined, of 
course the dip will be incorrectly taken out. But this source of 
error need not be serious; for, with a good log, there is no reason 
to expect even as much as 5 per cent. error. If there was 5 per 
cent. error in a one-mile run the error will be about 40” in the 
case of observations made from a height of eye of 24 feet. The 
error might be completely eliminated in smooth water by attach- 
ing the signal to a line, and paying-out the line, so that the signal 
was finally at an accurately known distance from the ship. ‘This 
procedure appears to be needless in order to arrive at results 
possessing the degree of accuracy required. 
An observation of the altitude of a known star will, in general, 
be an “ex-meridian,” and may be reduced to give latitude in the 
