294 



SMITHSONIAN MISCELLANEOUS COLLECTIONS 



VOL. 51 



then it is not difficult to study the conditions of equilibrium of a 

 vertical column of air when the curve of condition is known. Let 

 such a curve be represented by Z Z in fig. 35. 



If we now assume that a particle of air (a) for any reason whatever 

 experiences a slight rise in temperature then it will pass over into 

 the condition represented by the location of the particle a 2 . But 

 since the atmospheric pressure must be the same at places of equal 

 altitude therefore it must be specifically lighter than its surround- 

 ings, and must rise higher. If this rise occurs without further addi- 

 tion or subtraction of heat, then it must cool according to the adia- 

 batic law, i.e., the corresponding curve of change of condition will, 

 in so far as no condensation occurs, be a straight line inclined 45 

 to the axis and cutting the curve of condition at a point lying above 

 a. When the ascending air attains this altitude it has attained again 

 the temperature of its surroundings and there is no reason apparent 

 that should cause a further ascent. The equilibrium temporarily 

 disturbed is now again restored. 



-10° +10° +30° 

 FIG. 34 



fig. 35 



When a particle at a experiences a cooling the inverse phenomena 

 occur: instead of an ascending straight line from a t it follows the 

 descending line from a t and the particle warms up until at some 

 lower altitude it attains the temperature of its surroundings and 

 thus again the movement comes to an end. 



We thus perceive that "when the adiabat, whose direction is 

 shown by the straight line A A, ascends less steeply than the curve 

 of condition Z Z the equilibrium is stable." 



When the adiabat ascends more steeply than the curves of con- 

 dition, as is shown in fig. 36, that is to say, when the temperature 

 along the vertical column diminishes with altitude more rapidly 

 than in air that is ascending adiabatically, then the phenomena are 

 quite different from the preceding case. 



In this case no special warming or cooling is needed in order to 

 make a particle rise or sink with increasing speed, but an initiation 



