454 METEOEOLOGY. 



eighteen, and cirrus, seventeen. Two rotations, from 29tli of June to 

 lOtli of October, were not accompanied with those of the wind. 



Sometimes we find that all the strata of clouds, up to the cirrus, com- 

 plete their rotation at the north on the same day and at the same hour. 

 At other times, and these are the most numerous, the wind gains upon 

 the cumuli, these on the cirro-cumuli, and these latter on cirri ; that is 

 to say, from below upward instead of from above downward, as before 

 their rotation. This fact seems to contradict the hypothesis that the su- 

 perior currents determine gradually the inferior currents. This, how- 

 ever, is owing to the fact that the currents are inclined, forming very 

 nearly an angle of 45^ with the surface of the earth, so that they are 

 first felt at a point more to the north, falling by degrees until they at- 

 tain all the points of their journey toward the south, through which 

 they have passed above, till their extinction naturally, or by the shock 

 of opposite currents. This appearance of the inferior current before the 

 superior is especially frequent in low regions. It is presented five 

 times against four between the wind and cumulus, and four other times 

 simultaneously. In the higher regions, six times to five, the cirro- 

 cumulus appeared before the cirrus, and in three other cases at the same 

 time. The cumuli, in their turn, gained eleven times against two upon 

 the cirro-cumuli, and twice again they hajopeued at the same time. 



By reason of the opposition or inversion of temperature between land 

 and sea, the wind tends toward the south in the evening, and in the 

 morning toward the north. The influence of these local movements of 

 the breeze is such, in the general circulation, that it may retard the 

 rotation of the wind, which terminates at the north, not only many 

 hours, but besides from 9(P to 180° in azimuth. The action of the sea- 

 breeze seems to be more considerable than that of the land-breeze ; but 

 the breezes are much less sensible upon the cumulus and cirro-cumulus, 

 especially when these latter are elevated, and they do not seem to reach 

 the region of the cirrus. The continuance of each rotation varied con- 

 siderably in 1863, as follows : 



Days. Hrs. Days. Hrs. 



For tlio ci/Tus, from 5 5 to 49 11 



For the cirro-cumulus, from 3 8 to 62 5 



For the cwmw^HS, from 3 3 to 36 22 



For the wind, from 4 to 71 9 



The month of July did not present a single rotation of any of these 

 four elements. It should be remarked that the greater number of the 

 wind's rotations are accompanied by another rotation in the cumulus ; 

 that those of cirro-cumulus are more rare and less corresponding with 

 the first ; and that, in fine, those of cirrus are still further removed. It 

 also appears that the rotations are less frequent in the higher regions 

 than at the surface, and that the first of the cirrus, due to the equato- 

 rial current, are borne rather to the west, and especially to the south- 

 west ; while the second of the cumulus, arising from the i^olar current, 

 are confined more to the region of the east, from north to southeast. 



