3 i8 A YEAR IN BRAZIL. 



country, where the " tropeiros," or mule-drivers, are barely covered 

 with cotton clothing, and often lie under open sheds, or even 

 outside. 



" At first sight, the explanation of this abnormal phenomenon 

 is difficult, for the lower lying currents of air (vents inferieures), 

 coming from the far-distant southern regions, cannot reach these 

 latitudes at a low temperature, as they are warmed along the whole 

 of their course by contact with the soil under the influence of 

 solar radiation. A direct descent of cold air from the higher 

 regions of the atmosphere cannot take place without a consider- 

 able increase in the heat of that air, in virtue of the compression 

 it sustains (compression eprouvee}, and consequently one cannot 

 have recourse to the pure and simple hypothesis of an atmospheric 

 current descending, especially as the phenomenon in question 

 would then be frequent. The only possible explanation is, there- 

 fore, to admit that in a much more southern latitude, where, con- 

 sequently, the winter might be very severe (pouvait shir avec 

 rigueur) for the month of June is a winter month in the southern 

 hemisphere a great mass of cold air, at a temperature far below 

 zero (centigrade), and due to a very strong radiation from the 

 earth's surface and to southerly winds, was carried, by a cause 

 whose origin we will presently examine, to a great height above 

 the surface. By expansion, owing to such an elevation, its tem- 

 perature was again lowered to a great extent; but this would once 

 more attain its original condition if the mass of air descended 

 again to its former level. Then, driven northwards at its high 

 elevation, the current approached the equator without becoming 

 much warmer, contrary to what would have happened had it passed 

 over the soil, for the solar rays raise the temperature of the air in 

 passing along, and we know that, above all, it is warmed by going 

 over the soil, and by the ascending currents which its passage 

 occasions. But, on approaching the latitude of the table-lands of 

 Minas Geraes, this cold mass of air descended to the level of the 

 plateau, and the warming resulting from the descent could only 

 bring back its original temperature, and even that only by supposing 

 that it did not primarily come from a lower level than the plateau. 

 Therefore, it could then have carried there a still lower tempera- 

 ture than its original condition, except the small increase gained 



