57 
1918-19.] The Origin of Anticyclones and Depressions. 
the horizontal pressure gradients in the air are effective not so much in 
increasing its speed as in altering its direction of motion. (2) Mr Shaw 
and Mr Lempfert, in their Life History of Surface Air Currents, gave 
results of following up the movements of particular masses of air. They 
showed that these often move many thousands of miles in very compli- 
cated paths with but little change in speed, being merely deflected from 
their courses in passing regions of low or high pressure. (3) Such results 
are in agreement with the first principles of dynamics. A region of low 
pressure in the air (while it persists and is stationary) may be regarded 
as a centre of attractive force, and by the principle of the conservation of 
angular momentum the moving air cannot pass into it, but must move on 
past. In the process of its deflection the air will be accelerated in that 
part of its course in which it approaches the depression, for here there is 
a component of pressure in the direction of the motion ; but as it recedes, 
if the intensity of the depression is unaltered, all the increase will be lost, 
and it will finally pass away with a speed slightly diminished on the 
whole by reason of the effects of friction and of the conduction of heat. 
(4) We have in the mean pressure of the barometer over a given area a 
very accurate measure of the mass of air overlying it. The measure is not 
perfectly accurate, since a small correction (under J millibar) may be locally 
required at the moments of most violent vertical movements in thunder- 
storms, line-squalls, etc., while the weight of the air itself may be slightly 
varied by a vertical redistribution of mass. But the ordinary rise and fall 
of the barometer by several millibars cannot well be attributed to vertical 
currents, unless we are prepared to consider uniform upward and down- 
ward currents of several hundred miles per hour over areas exceeding that 
of the whole British Isles — a phenomenon not likely to have escaped the 
notice of aviators, who, in ascending to 20,000 ft., have passed above more 
than half the mass of the atmosphere. The second cause of inaccuracy, 
which has been quoted as giving rise to effects “ of the same order as those 
due to differences in latitude,” is likewise capable of producing only trifling 
variations of pressure. It would require a vertical movement equivalent 
to raising the whole column of atmosphere through two miles before the 
combined changes of gravity and centrifugality could diminish the pressure 
by one millibar. 
When, therefore, a depression originates, the diminution of pressure 
must be accompanied by, and may be due to, a general withdrawal of air 
from the area. Thus our postulate is justified by observation: a diminution 
of pressure does not, in natural conditions, result in an immediate “ inrush 
of air.” 
