DISCUSSION AND ANALYSIS OP WINDS. 703 



consequent ascending current cannot be much greater than on the ocean. We should 

 rather expect a great barometrical depression in the treeless llanos of the Orinoco, 

 and in the Pampas of the Argentine State, or in the Campos of southern Brazil, 

 as shown on Plate 14. The last two regions being sub-tropical in greater 

 part, the difference of season is well marked. We do not possess a single annual 

 series of observations in the Pampas and Campos, but already Rio Janeiro, Monte- 

 video and Buenos Ayres, as well as the stations of Chili, have a lower pressure in 

 the warm months of the year. 



In studying the winds of South America, the physical geography of this con- 

 tinent must be borne in mind. It is separated into two very unequal parts by the 

 chain of the Andes, which runs near to the western coast. The mountains are so 

 high, between 9° N. L. and 40° S. L. as not to permit any interchange of air in the 

 lower strata. The eastern part of South America is generally level, having but two 

 mountain systems of any importance, that of Brazil and that of Guiana, which 

 were not inappropriately compared to the AUeghanies and the Canadian plateau in 

 eastern North America. 



These secondary mountain chains of South America have no great influence on the 

 course of ithe winds, the whole extent of the continent to the eastern slope of the 

 Andes being subject to the trade-winds, and the effect of the continental mass is 

 here rather to intensify them. 



This is especially the case on the Amazon, as stated by all travellers who have 

 been there.'' They say the eastern wind is very regular, especially in the dry season, 

 June to November, blowing at times with the strength of a gale. In' the rainy 

 season, especially on the upper Amazon, it is less regular, being frequently inter- 

 rupted by calms and westerly winds. There can be no doubt as to the general 

 accuracy of these facts, notwithstanding the want of long-continued observations. 



We possess, also, an admirable description of the course of the seasons on the 

 llanos (treeless plains) by A. Von Humboldt. The regular blowing of the trades, 

 the clearness of the sky, and want of rain from November to May are particularly 

 noticed there. The appearance of the rainy season is announced by shifting of the 

 wind to S. W. The countries on the lower Orinoco (see Plates 5 and 6) are in the 

 region of the northern trades, while the southern trades are already dominating 

 on the Amazons. 



There is a region between l°-3° N. on the Rio Negro which seems to have 

 prevailing calms and rain in all months, according to Humboldt and Wallace. 



We have observations from Venezuela and Guiana, where the winds are as fol- 

 lows.^ 



' See Hartt's Geol. and Phys. Geogr. of Brazil ; Bates, the Naturalist, on the Amazons ; Wallace, 

 Amazons and Rio Negro ; Martens, Reise nach Brazilien ; Herndon and Gibbon, Explorations 

 of the Amazon. 



» To prevent confusion I give the months of observations for the equatorial regions and the 

 southern hemisphere. In the tables '• Winter" always means December, January, and February, 

 and " Summer" June, July, and August. 



