1913-14.] Principia Atmospherica. 87 
circulation along parallels of latitude from west to east in which the air 
nearer the poles is the colder is a circulation which may remain practically 
identical at all heights, and is suggestive of durability and persistence. 
The distribution of pressure at the 4-k. level given by M. Teisserenc de 
Bort suggests that the actual circulation in the upper air is not a circu- 
lation along parallels of latitude, but yet is an approximation thereto, being 
something intermediate between a circle and a figure-of-eight. 
That the circulation at the 4-k. level remains of the same general character 
up to the 8-k. level is suggested by the fact that in those regions distribution 
of temperature is such as to cause very little change in pressure-differences, 
in accordance with the formula of Proposition 2. 
It may be remarked that the distribution was calculated by M. L. 
Teisserenc de Bort from the distribution of pressure and temperature at 
the surface, and is subject to two uncertainties : first, the reduction of the 
original pressure readings to sea-level ; and secondly, their further reduc- 
tion to the 4-k. level. The uncertainties arise from the uncertainty in the 
values of the temperature of the air “ below the ground ” in the reduction 
to sea-level, and above the ground in the reduction to the 4-k. level. To a 
certain extent these two uncertainties compensate each other in the 
important features of the result, and the conclusion as to the circulation 
at which M. Teisserenc de Bort had arrived, is supported by the results of 
Hildebrandsson’s discussion of the international cloud observations (see 
Hildebrandsson and Teisserenc de Bort, Les Bases de la Meteorologie dyna- 
mique, vol. ii., Gauthier- Villars, Paris), and by other considerations of a 
more general character. 
The statements of these two lemmas are based upon observation and 
are, therefore, liable to modification or correction in detail as the results of 
observation become more conclusive. They are, however, sufficiently well 
established to justify their use in the further consideration of meteoro- 
logical problems. 
Section III. — Propositions. 
We now proceed to the consideration of the propositions which are set 
out in the Synopsis. I shall deal in detail with only three of the pro- 
positions, numbered 1, 5, and 6 respectively, because the remaining three, 
numbered 2, 3, and 4, have already been dealt with in a paper communicated 
to the Scottish Meteorological Society, with the title of “ The Calculus of 
the Upper Air, and the Results of the British Soundings in the International 
Week of May 5-11, 1913.” The paper is published in the Journal of the 
Society for 1913. 
