262 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



not fully compensated (or offset) by a corresponding increase of the 

 factor 



U - U 



273 + t x 



so that ,9 2 — /? x must attain a maximum which in this special 

 example is to be found at an altitude of about 6500 meters. In 

 this connection I must especially point out the fact that the num- 

 bers given in the table for the quantity of water carried upward do 

 not suggest anything improbable. If we compute the volume that 

 a kilogram of air occupies at different altitudes we find that under 

 the above given assumption the quantity of fluid water in a cubic 

 meter amounts at most to 5 grams, that is to say, 5 mg. per liter, 

 a quantity that can certainly be easily floated in a rapidly rising 

 current. 



The diagram (fig. 23), for the process 

 here considered, differs from that pre- 

 sented in fig. 31, which latter related 

 to supersaturation proper. Whereas in 

 that the adiabat a a t simply intersected 

 the saturation curve S 5 , in the present 

 case it shows a sharp bend at this inter- 

 sec tion, since it changes from the adia- 

 bat of the dry stage to that of the rain 

 stage. On the other hand, in the former 

 case of supersaturation the expansion 

 simply continued along the adiabat of the dry stage even after 

 passing the point of saturation. 



THUNDERSTORMS 



From the above given considerations and developments it results 

 that both supersaturation of air with aqueous vapor and subcooling 

 of the water already condensed must, when these conditions 

 are suddenly dispelled, cause a rapid local rise in atmospheric 

 pressure which in general will last only a short time (except in so far 

 as special circumstances yet to be mentioned do not diminish the 

 restoration of the pressure) and cause a true jump or spring or so- 

 called step up in pressure or a "knick" in the barogram instead of 

 an oscillation at the locality. 



But such oscillations and jumps, as already mentioned, are almost 

 always an accompaniment of thunderstorms and it only remains to 

 investigate whether the processes that come into play in thunder- 



