184 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 3 5 



part, in which the temperature falls with increasing height, is 

 known as the troposphere and the upper part as the stratosphere. 

 The conditions in these two regions do not remain constant from 

 place to place nor from day to day. Over the Equator the lower 

 air is naturally warmer than that over the polar regions, but over 

 the Equator the troposphere extends to an average height of some 

 17 to 18 kilometers. As the temperature is falling at a rate of some 

 6° C. for every kilometer throughout all this height, it is very low 

 by the time the stratosphere is reached, being about 80° C. below 

 zero. In polar regions the stratosphere comes relatively low down 

 and begins at a height of 6 to 8 kilometers, with the result that 

 though the temperature of the air at ground level is very low there 

 is no great difference between the air at the surface and that in the 

 stratosphere, and the latter has a temperature of about 40° C. below 

 zero. The temperature of the stratosphere over the polar regions is 

 thus much warmer than that over the Equator, and as there is little 

 change of temperature with height in the stratosphere we find 

 the rather surprising result that at a height of 16 to 20 kilometers 

 the air over the polar regions is much warmer than that at the same 

 height over the Equator. Indeed, the coldest air in the atmosphere 

 is probably that at a height of 18 kilometers over the Equator. 



The conditions of the air in both the troposphere and stratosphere 

 also vary from day to day, and these variations are closely asso- 

 ciated with the meteorological situation. When a well-defined de- 

 pression has just passed across the country, it will be found that 

 the troposphere is colder than usual, that the stratosphere is ab- 

 normally low and that its temperature is above the average. On 

 the other hand, when a well-marked anticyclone covers the country 

 the troposphere will be warm, the stratosphere cold, and the tropo- 

 pause high. It will be seen that these changes are similar to the 

 changes with latitude, so that the conditions in middle latitudes in 

 the rear of a depression are similar to those in polar regions and 

 the conditions in an anticyclone are similar to those in equatorial 

 regions., 



Meteorology of the upper atmosphere. — ^What has been described 

 above has now been known for several years, but until recently little 

 was known about the conditions above 20 kilometers, and it was 

 usually supposed that the temperature would remain roughly con- 

 stant up to very great heights. This view was questioned when 

 the results of observations of meteors passing through the upper 

 atmosphere were used to give a rough idea of the temperature at 

 these heights. Contrary to expectation, it appeared that the tem- 

 perature rose again above the stratosphere and became as warm 

 or warmer than the air at ground level. These results were, how- 



