HURRICANES — GENTRY AND SIMPSON 309 



densities and moisture content. Under certain conditions a wave 

 forms along such a boundary and if it becomes unstable it may release 

 enough energy to cause a storm to develop. Such occurrences are 

 frequently observed in middle and northern latitudes in winter. This 

 idea as applied to the Tropics had to be abandoned as it became rec- 

 ognized that true fronts seldom, if ever, exist over the tropical oceans. 

 Another serious limitation of this explanation was presented in a 

 paper by Gordon Dunn in 1940 [5], which showed that many of the 

 hurricanes in the Atlantic formed in easterly waves (deformations 

 in the steady trade- wind flow) entirely within a single air mass 

 where there was not even a suggestion of a front of the Norwegian 

 unstable frontal wave type. 



It is now well known that hurricanes form only over the tropical 

 oceans where the water temperature is high, at least 80° and in most 

 cases 83° to 85° (Palmen [13] and Fisher [8]), and where showers 

 are occurring. The higher ocean temperatures are required for heat* 

 ing the air until it is buoyant and for making it possible for the air 

 to hold a sufficient amount of water vapor. (Warm air will hold more 

 water vapor than cold air ; hence, warm air makes possible the release 

 of a greater amount of latent heat when the moisture is condensed from 

 the rising air.) 



For a hurricane to form it is necessary to have warm moist tropical 

 air and to have a preexisting disturbance in the normally steady 

 trade-wind regime of the Tropics. These are necessary but not suf- 

 ficient conditions. Only a relatively small percentage of the tropical 

 disturbances that frequent the Caribbean area in the summer ever 

 reach hurricane intensity. It is obvious that something else is re- 

 quired. It has been suggested by various investigators, notably 

 Herbert Riehl [16] and J. S. Sawyer [18], that the upper-level flow 

 patterns furnish the key to inception. It seems rather obvious after 

 studying structure of hurricanes that there must be horizontal con- 

 vergence of the air flow in the lower levels and there must be hori- 

 zontal divergence of the air flow at some upper level. Professor 

 Riehl, in his study of formation of hurricanes, has suggested that at 

 about 40,000 feet there are certain flow patterns that help to evacuate 

 air from the incipient storm area and are conducive to the forma- 

 tion of hurricanes. It is hoped that the additional data now being 

 collected will reveal sufficient details concerning the structure and 

 energy processes in hurricanes to make it possible to define both the 

 necessary and sufficient conditions for their formation. 



2. HURRICANE MOVEMENT 



The theories as to what makes a hurricane move may be divided into 

 two classes: (1) Those attributing the movement of a hurricane to 

 internal forces within the hurricane itself, and (2) those attributing 



