114 flature's /HMracIes. 



When a season is once established, either as a 

 rainy season or a dry season, it is likely to per- 

 sist in this character until a change comes 

 that is produced by the movement of the sun 

 in its course northerly and southerly, and the 

 change produced from this cause requires sev- 

 eral weeks of time. 



If accurate weather predictions could be 

 made for a long time in advance, or for even a 

 week, they would be of incalculable value. 

 But it is doubtful if ever this will be brought 

 about, as there are too many necessarily hid- 

 den factors which enter into the calculations. 

 If stations could be established all over the 

 oceans with sufficient frequency, and an equal 

 number at a sufficient altitude in the air, I 

 have no doubt that much that is now mys- 

 terious might be made plain. 



