95 
1913-14.] Principia Atmospherica. 
case which is referred to in Mr Cave’s book on the Structure of the Atmo- 
sphere in Clear Weather as a frequent precursor of weather of the thunder- 
storm type, accompanied by the setting in of the south-westerly wind. 
The distribution of temperature is such as to change the direction of the 
pressure-gradient near the surface. Consequently the outflow from high 
to low goes from under the upper “ low ” to under the upper “ high.” The 
necessity for the thickening of the southerly current is therefore not re- 
lieved by the outflow, but accentuated thereby. At the same time the 
north-easterly current has to get thinner, so it is gradually replaced by the 
south-westerly current settling down to the surface. The appropriate re- 
distribution of pressure at the surface accompanies the redistribution of 
air-currents in the vertical section. 
These examples are adduced because it seems not improbable that they 
give us the opportunity of watching the operation of the causes of change 
which are inherent in any actual state of atmospheric motion. 
Let me summarise the attitude which seems to me to be appropriate 
for the meteorologist to take up in face of the complexities of the atmo- 
spheric circulation, by again referring to the position of the astronomer 
before the final enunciation of the laws of motion. Imagine the perplexity 
of the astronomer who, finding the heavenly bodies moving in all sorts of 
directions with all sorts of velocities, set himself to explain the motion 
which each possessed. To him the laws of motion bring the assurance that 
it is not necessary for him to explain why a body moves ; it is the changes 
of motion which should occupy his attention. So the meteorologist, looking 
at the circulation of the atmosphere in obedience to the distribution of 
pressure, has not to ask himself why the pressure is high here or low 
there, but rather, “ Is the distribution persistent, and if not, are the causes 
of change inherent in the existing circulation sufficient to account for the 
changes ? ” If it be said that, after all, the problem remains the same and 
the point of view is immaterial, it is right to remember that in astronomy 
the change in the point of view has simply reduced chaos to law. 
From what has been already said, it appears that a steady state of 
persistent motion of the earth’s atmosphere is in the highest degree im- 
probable, because it can only occur in a combination of circumstances which 
are independently fortuitous ; but it is desirable to call attention to a possible 
case of motion which is quasi-persistent in consequence of two concurrent 
and persistent infractions of the conditions of steadiness. 
If we suppose the south-to-north and north-to-south currents of fig. 2 
placed back to back so as to form an anticyclonic section instead of the 
cyclonic section represented in fig. 2, we find in juxtaposition a south- 
