THE ANEROID BAROMETER 
109 
at 5 p. M. it was 29.100. That difference in pressure 
corresponds to nearly 150 feet in elevation, and height 
observations made during the day would be uncertain to 
very wide limits if the change could not be allowed for. 
THURSDA Y FRIDA Y 
The possibility of correction rests in two suppositions: 
(1) that at any moment of time the air pressure is constant 
over a considerable horizontal area, and (2) that the field 
barometer and the station barometer work together, and 
that they both follow exactly and quickly the change of air 
pressure. The latter point may be expressed in this way — 
that the field barometer, if left at the base station, would 
have followed the same course as did the instrument which 
in fact was left there. 
The field barometer may not read the same as the 
barograph when they are brought together, but that 
“ index error,” as it is called, does not matter if the differ¬ 
ence between the two remains constant. In this case the 
field barometer at camp in the morning read 29.350 and at 
night 29.200, .1 inch higher than the barograph. One 
may, therefore, when he gets to computing, draw on the 
