232 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1939 



Figure 7. — East-west vertical cross section through the atmosphere, showing conditions 

 aloft shortly before the time of the map in plate 3, figure 1 (cf. explanations in legend to 

 figure 4). 



frequently are obscure and may not exist in all of the physical prop- 

 erties; broad and diffuse transition zones sometimes take the place 

 of true fronts. Success in the analysis requires that appropriate and 

 adequate data be available to a trained and experienced personnel. 

 In particular, a network of aerological soundings to fairly great 

 heights is indispensable, and until upper-air data could be made 

 quickly available by the introduction of airplane soundings, air-mass 

 analysis could not receive extensive application in daily work. The 

 current characteristics of an air mass depend on both its source and 

 the modifications it has undergone during its progress away from the 

 source region ; these modifications, which usually tend to weaken the 

 distinctions, are in general much more rapid and extensive near the 

 surface of the earth than aloft. Furthermore, surface observations 

 alone frequently do not reveal conditions and physical processes aloft 

 which on many occasions are the principal cause of important phe- 

 nomena at the surface, and without a knowledge of which the latter 

 cannot be explained or foreseen, nor the surface map interpreted. 

 Methods are now being rapidly perfected for obtaining aerological 

 observations under all conditions of weather, and from a denser net- 

 work, by means of the radiometeorograph or "radiosonde" — a mech- 

 anism which is carried aloft by a small balloon, and by which 

 pressure, temperature, and humidity are caused to actuate a short- 

 wave radio transmitting apparatus that conveys a record to radio 

 receiving apparatus on the ground. Vertical cross sections through 

 the atmosphere (figs. 4, 7), so far as they can be constructed from 

 data now available, are being used to supplement the surface map 

 and the mere pilots of upper-air data against height (Willett, 1935, 

 1935a). 



Examples of analyzed maps are shown in plate 3, figures 1 and 2. 

 When the movements of the currents are such that colder air is ad- 



