l8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 1 10 



To explain that relationship, Archenhold points to the possibility that 

 solar neutrons, because of their special qualities, penetrate much deeper 

 into the earth atmosphere than do the solar alpha particles, protons, 

 and negatrons. Only in those layers which contain a sufficient amount 

 of water vapor, e.g., in the cirrus level, would they undergo a con- 

 siderable retardation, and even absorption. A necessary provision 

 for the occurrence of condensation phenomena would in all such cases 

 be the presence of an atmospheric layer saturated with water vapor 

 and relatively free from other condensation nuclei. As has been 

 shown on different occasions, these conditions occur not infrequently 

 (A. Schmauss and A. Wigand (15)). Even then, if the neutron 

 hypothesis could not stand the test, there would be a possibility of 

 explaining the presence of solar-produced condensation nuclei in the 

 upper troposphere. According to the investigations of H. Peter- 

 sen (23), E. Palmen (24), and A. Refsdal (25), a drop of the 

 tropopause produces a cyclonal circulation. This flow may continue 

 up to the high stratosphere and may suck down air from there in the 

 center of the cyclone. This is possible because the kinematic viscosity 

 of air in the tropopause level is very small, according to Chapman 

 and Milne, and only in heights of about 60 km. again reaches the 

 sea-level value. In that scheme there is considered the important 

 fact discovered by Palmen in 1932 that the upward movement of 

 the air in the cyclone and the downward movement in the anticyclone 

 are confined to the lower and middle troposphere, and that the vertical 

 movements in the upper troposphere and in the stratosphere have the 

 opposite direction. 



The assumption of a separate existence of the troposphere, inde- 

 pendent of the stratosphere, had been definitely destroyed by these 

 findings, A down-transportation of condensation nuclei might be 

 possible in such a way, and the question now arises, to what extent 

 could an additional supply of condensation nuclei act upon the tropo- 

 spheric processes? As is known, the liberated condensation heat 

 inheres into the water droplets themselves, and, as the expansion of 

 fluids compared with that of gases is extremely small, the temper- 

 ature increase becomes evident only when the energy has been trans- 

 mitted to the surrounding air. This energy transfer is performed 

 much faster if a certain amount of water vapor condenses into many 

 small rather than into a few large droplets. In such a way, according 

 to C. Braak (26), a greater number of nuclei can accelerate the 

 transformation of condensation heat into intensified convection. 



A local turbidity of the stratosphere, produced by nuclei, can 

 become important even without any condensation phenomena, because 



