615 
of Edinburgh, Session 1874 - 75 . 
uppers are rapid, while the surface winds are retarded, the 
numerous intermediate layers betwixt these, must all move at 
different rates of speed attended by much friction and consequent 
retardation. 
The direction assumed by the curve, is therefore one which is 
nearly opposite to that of the motive inflow , which produces it. In 
a somewhat similar manner within the tropics, oceanic currents are 
in certain cases produced, moving in nearly an opposite direction 
to that of the North East trades, on which they depend. 
Winds representing the different Modes of Inflow . — On the west 
segment of a barometric depression, Polar winds prevail, which are 
dry, cold, and dense, and are fed by descending currents, with a 
vertical source of supply. They may be regarded as surface winds. 
On the east segment are found equatorial winds, which are warm, 
moist, and less dense; they are weakened by their ascending ten- 
dency, — they have not so much the character of surface winds, 
but assume more the character of rapid uppers, and instead of a 
vertical, have- a horizontal source of supply. 
Progress. — When a barometic depression is formed, a spiral 
inflow towards the centre takes place ; if this were equally uniform 
in every direction, the great central fall of the barometer would 
extend itself all round, gradually diminishing as it proceeds out- 
wards towards the circumference, and lowering the surrounding 
gradients as it proceeds; but if, as is usually the case, the inflow is 
not uniform, the depression will then extend itself in one particu- 
lar direction, in the manner above described. This extension, 
which is due to the mode of the central inflow, takes place mostly 
in front, and in an easterly direction : it will there create a 
scarcity of supply, towards which the low central barometer will 
advance. What thus takes place may be illustrated in this 
manner : — Suppose that the ascent of a balloon, situated near the 
surface of the ground, is retarded, though not arrested, by a chain 
passing over it. This chain, where it reaches the ground on each 
side, is not fixed to it, but is laid outwards along its surface, one 
end extends for a short distance, the other for a considerable 
distance. Under these circumstances, the ascent of the balloon 
will not be vertical, but in a direction inclined towards that in 
which the chain extends for the greatest length over the ground, 
