METEOROLOGY. 315 



above than below. On the other band, it will occasionally happen that 

 the n]>por currents arc slower cr reversed, in which case the summit of 

 the cumulus follows instead of preceding- the main cloud. (D. M. Z., i, 

 p. L'9.) 



203. [Two facts not mentioned by Koppen should be remembered in 

 order to achieve a fuller explanation of cloud phenomena, namely, that 

 the upper part of the cumulus cloud is surrounded at a little distance 

 by the drier air into which it has risen, which, therefore, tends to evap- 

 orate and dissipate the surface of the cloud; and, again, the powerful 

 eflect of the solar rays (which increases slightly with altitude) is to 

 evaporate the moisture at the surface of the cloud and to surround the 

 whole sunny side of the cloud with a thin layer of hot moist air, which 

 must have a special ascensional tendency. This action of the sun is 

 undoubtedly also very efficacious on the upper surface of extensive lay- 

 ers of stratus clouds from which vapor currents rise, which in turn form 

 a second layer at some distance above, while the density of the whole 

 mass of warmed cloud and air is diminished, and an additional ascensive 

 tendency is felt at the earth's suiface over a htrge and extended terri- 

 tory.] 



204. The increasing interest in the observations of clouds has led 

 van Bebber and Koppen to attemjjt a system of corresponding obser- 

 vations, which were planned after consultation with Dr. Klein and 

 Moller and Assmann. The same methods of description are used at all 

 stations, and the complete record of clouds is made a part of the regu- 

 lar entries in the daily journal as well as in the special records. (D. 

 M. Z., I., p. 30.) 



205. Eev. Clement Ley, in an essay on the structure of the cirro-filum, 

 or thread-like clouds, presumably formed of ice spiculte, states that this 

 is one of the most important forms in the prediction of the weather ; he 

 has observed it for twenty-tive years, and oilers a classification of the 

 more reliable observations and an exi)lanation of some of the more prom- 

 inent phenomena. His classification of the clouds relates to their form, 

 their structure, and their relations to each other. The following three 

 generalizations seem to him of importance: 



1. If we turn the face toward the wind as we stand on the earth's sur- 

 face, then in the majority of cases the upper currents of air move from 

 the right to the left. 



2. Winds from the west lie above those from the east in the upper re- 

 gions of the atmosphere much more frequently than is the case at the 

 earth's surface. 



3. The upi)er strata of the air as a rule move much more rapidly than 

 the lower, lie gives a table showing the number of cases in which the 

 cirro-tilum cloud moves from the respective points of the compass. He 

 restricts the name cirrus to the upper clouds formed of bent or twisted 

 bundles of threads, whereas those composed of long lines or filaments 

 have a much greater velocity, and are characterized by him as linear 



