324 SCIENTIFIC RECORD FOR 1884. 



been formed, the eqnilibriiim is soon disturbed, the diminution of tem- 

 perature under the cloud is slower and the average temperature there- 

 fore higher. This suffices to inaugurate an inflow of air from all sides 

 and an ascending motion and a falling barometer. Warm, moist air 

 blows up from the south and we have the conditions for the winter rains. 

 If this is correct we must consider the quiet calm air as the initial con- 

 dition, but this calm is due to the existence of the high chain of mount- 

 ains that surrounds Northern India on the east, north, and west. Were 

 these absent the strong, dry northeast winds would carry away the va- 

 por and the weather would be similar to that of Southern China. 



226. [This is also a very fair description of the growth of a storm- 

 center in the region west of the Mississippi, such as occasionally takes 

 place directly under the observation of the Signal Office.] {Z. 0. G. if., 

 XIX, p. 452.) 



227. A. Richter, of Ebersdorf, has discussed the observations made 

 at that place for six years on the direction of movement of the upper 

 clouds, with the following results: The mean direction of the move- 

 ment is from the point south 84° W., but for observations made one or 

 more days befor<^, rain the direction is from a more northerly point, and 

 for observations made a few hours before rain the direction is from a 

 more southerly point. The probability of rain within twenty-four hours 

 is greatest (65 per cent.) when the clouds move from the southwest; the 

 general probability that rain will follow within twenty four hours after 

 an observed cirrus is 56 per cent. ; but rain fell on 58 jjer cent, of all 

 the days in these six years 5 therefore, in general, the cirrus precedes 

 non -rainy weather. This is explained by the fact that cirrus observations 

 cannot be made when low clouds obscure the heavens, on which days 

 the probability of rain is greatest. The direction of cirrus n)ovement 

 is for winter from the N. 82° W., and for summer S. 74° W. Using the 

 wind direction as observed at a more favorable station five miles dis- 

 tant, it is found that the cirrus deviates from the wind most frequently 

 by angles of plus or minus 45o, but the probability of rain is less when 

 the wind is opposed to the cirrus than when the directions agree, and 

 similar results are found by conij)aring the cirrus and lower clouds. On 

 6i)ecial days of low barometer the direction of the cirrus has no appar- • 

 ent influence upon the rainfall, but when the barometer is simply fall- 

 ing with a southwest wind the occurrence of cirrus movements nearly 

 coinciding with that of the wind is in 73 per cent, of such cases followed. J 

 by rain, while with northeast winds only 43 per cent, are followed by « 

 rain. (D, M. Z., i, p. 319.) J 



228. Prof. H. Klein proposes to make more use than hitherto of ob- f 

 servations of the cirrus clouds for purposes of weather prediction. The ;J 

 difficulty lies mostly in the fact that all observations must be made at | 

 the earth's surface, and we therefore know but little of their altitude ' 

 and thickness. Systematic observations, however, have lately' been 

 made both by himself at Cologne and by Neumeyer at Hamburg, from 



