METEOROLOGY. 325 



which he indicates a few rules additional to the ideas previously pub- 

 lished by others. Thus he states that when a thin cirrus arc appears 

 upon the western horizon witli its summit iu the west, this surrounds a 

 depression still further west moving eastward, and will be followed in 

 four to eight hours by a heavy rain. Again, if for twelve hours the 

 cirrus clouds moving from the west occupy only the southern horizon, 

 then for the next day we expect an east wind ; but if the cirrus move 

 from the southeast many dry days of east winds are to be expected, 

 and if the cirri move from the east-northeast, or the east, a continuous 

 dry weather may be expected. 



229. [The importance of studying the movements of both upper and 

 lower clouds which had already been abundantly indicated by Espy 

 and Redfield and otlieis, led the present writer to include all of these 

 iu his daily telegraphic dispatches for the Daily Weather Bulletin of 

 the Cincinnati Observatory, the publication of which began in Septem- 

 ber, 1869. The daily manuscript weather map of the Army Signal Office 

 began in 1871 to show the direction and kind of clouds, and the util- 

 ization of cirrus and other cloud movements for the anticipation of 

 storms, aud even as a first indication of the formation of a new storm- 

 center, growing up immediately under one's eyes, became a matter of 

 daily practice from that time.] {Z. 0. Q. M., xviii, p. 220.) 



230. J. Damian, of Innspruck, by careful observations of the clouds 

 during June, 1882, has endeavored to determine the relation between 

 the rainfall and the upper aud lower currents of air. He finds the 

 quantity of rain greatest when the lower current is from the southwest 

 and the upper current from the north, but least when the lower current 

 is from the north with the upper from the south. When the angular 

 deviation between the direction of the two currents is 23° he finds one 

 case with rain and six without rain ; when it is 45° the numbers are 1 

 and 9, respectively; when it is 90° the numbers are 2 aud 8; lor 135° 

 the numbers are 8 and 3 ; for 180°, 4 and 1, which results are equiva- 

 lent to saying that the more nearly the upper are opposed to each other 

 the more likely is rain to follow\ [Z. 0. G. J/., xviii, p. 345.) 



231. G. Mantel gives in the Swiss Annuaire for 1882 a study into the 

 connection between the number of clear and cloudy days in any period 

 and the mean cloudiness of that period. His investigation is partly 

 mathematical and partly based on actual observations. He shows that 

 assuming a mean daily cloudiness of 1 on a scale of 10 for the clear 

 days, 9 for the cloudy days, and 5 for the remaining days, we can com- 

 pute the mean cloudiness even for a period of months in close agree- 

 ment with direct daily observation. [Z. 0. G. M., xix, p. 340.) 



232. Grossman has studied the connection between the mean cloudi- 

 ness of any period and the number of clear and cloudy days, in a man- 

 ner similar to the study of G. Mantel for Switzerland. Grossman has 

 tested Mantel's Swiss formula by application to the German stations; 

 he finds that the monthly mean cloudiness (O) can be expressed as a 



