364 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



then the existence of an ascending current in this axis itself is not 

 conceivable. 



On the contrary, I would rather hold it to be probable that in 

 the axial tube no important vertical movement takes place, but 

 that this is essentially a moving rarified space, whereby new par- 

 ticles of air are being drawn into the motion and thus subjected to 

 the rarefaction. 



The assumption of an ascending current in the axial tube is also 

 by no means necessary, since Ferrel has proved that the rarefaction 

 of the air produced by the centrifugal force, when no addition of 

 heat comes in, is sufficient to explain the condensation and the 

 origin of the filmy cloud. This is quite natural when the trunk of 

 the cloud axis is first recognized as an appendix hanging below a 

 cloud and appears to gradually descend lower, since in the earlier 

 stages of the development of this phenomenon where the friction at 

 the earth's surface does not come into consideration, considerable 

 velocities will occur, which might cause rarefaction of the air and 

 hence the condensation. Moreover, the air must be most nearly 

 saturated directly beneath the cloud and hence it requires only very 

 slight rarefaction of the air to bring about condensation. 



It is only when the velocities in the lower strata of the atmosphere 

 have acquired a corresponding increase, that the rarefaction pro- 

 ceeds to such a degree that the cloudy film extends down to the 

 earth. 



But we are not to conclude from this that the cause of the whole 

 phenomenon is to be sought in the upper regions; rather is it to be 

 expected that in even the cases where the process is initiated by over- 

 heating of the lowest strata of air and the unstable equilibrium 

 produced thereby, still the larger velocities will be attained at great 

 altitudes sooner than below. 



For since after breaking up the unstable equilibrium the accelerat- 

 ing forces increase with altitude, therefore not only does the ascend- 

 ing current (which is to be thought of not as exactly in the axis of 

 the subsequent tornado but as extending over a large area) itself 

 acquire steadily increasing velocities, but this is also true especially of 

 the currents of air streaming inward from all sides, since with increas- 

 ing height slighter resistances oppose its motion. 



In general, the fact that the axial cloud appears to descend from 

 above justifies no conclusion whatever as to whether the true cause 

 of its origin is to be sought for above or below. 



Neither are we to draw any conclusion from the apparent descent 

 of the cloud-axis as to any descending movements in its interior. 



