270 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 51 



declining of the sun and the accompanying cooling of the lower air 

 gradually enfeeble the conditions that favor the process and thus 

 during the night time bring about a gradual extinction of the 

 thunderstorm. 



As a special indication of the existence of well-developed whirls 

 and horizontal axes, must be mentioned the fact that the wind then 

 blows nearly perpendicularly to the isobars, constituting an 

 apparent exception to the basic law of the winds and for which 

 Moller 21 has given an explanation. 



The preceding explanation does not exclude the existence of 

 individual small vortex whirls around a vertical axis, since it may 

 be that the thunderstorm roll is not continuous or that irregular- 

 ities in special places develop such vortices. But it seems to me 

 that such details are not sufficient to justify designating such storms 

 as vortex or cyclonic thunderstorms. The description here given 

 corresponds substantially with the presentation of the subject of 

 the propagation of thunderstorms as given by Koeppen 22 ten years 

 ago in his "Untersuchungen, etc.," "Investigations Relative to 

 the Thunderstorm of August 9, 1881." 



I must return, however, to this subject again because I recognize 

 the characteristics of "heat thunderstorms" in this method of 

 origination and propagation, which, indeed, has been subsequently 

 confirmed in the case of many great thunderstorms both by the 

 investigations of Ciro Ferrari 23 as also by the investigations carried 

 on in Bavaria and the neighboring States and later also in North 

 Germany. 



I have also intentionally attached small importance to the pres- 

 ence of areas of depression or barometric lows, but more rather to 

 the fact that there exists a region in which neither the cyclonic nor 

 the anticyclonic character is especially well marked. 



I have therefore entirely omitted to consider the circulation of 

 the upper atmosphere as dependent on the general distribution of 

 pressure over large areas and have made this substantially pro- 

 visional study under the assumption that the general circulation of 

 the atmosphere does not come into consideration. 



It seems important to me to make it clear that under this assump- 

 tion we must expect a propagation of the thunderstorms from west 



21 Zeitschrift d. Oest. Gesell. fur Met. 1884, XIX, pp. 80-84. 



22 Ann. d. Hydrographie, 1882, X, pp. 595 and 714. Compare also Sprung, 

 Lehrbuch, etc.: "Treatise on Meteorology," pp. 294 et seq. 



23 Annali dell Ufficio Centrale di Meteorologia, 1883, Vol. V, part i.and 

 1884, Vol. VII. 



