INDIAN SUMMER. 137 



ter, or land-breeze, and also the summer, or sea-breeze. This point we 

 must state at somewhat greater length, in order to show its application 

 to the explanation of the partial repose prevailing during the Indian 

 Summer. 



At about the autumnal equinox and a little afterwards, the retiring 

 sun rapidly loses his influence and his power to keep up the tempera- 

 ture, which his rays had previously produced in the higher northern 

 latitudes. That portion of the atmosphere belonging to those latitudes 

 must, by the process of cooling, contract in bulk and increase in density, 

 so that the whole mass will not now extend as high above the earth as 

 it did before. The first consequence will be, that, at some elevation, 

 the air will overflow from the tropics to fill up this depression in the 

 North ; and the second, that, by the increased pressure thus produced 

 in the latter regions, the lower portions of the air, near the surface of 

 the earth, will tend toward the place of the sun, forcing before them 

 that which before occupied our latitude, and supplying its place with 

 that which is colder. This is the cause of the cold and unpleasant 

 winds of October. They are the first tribute, from the North, to the 

 sun in his new southern home. After this first rush, by which much 

 of the heated air of the northern temperate zone has been transported 

 southward, an equilibrium gradually takes place. The surface of the 

 land being yet warmer than that of the adjacent ocean, keeps up the 

 temperature of the air over it at a point a little higher than that over the 

 ocean, so that there would yet remain a tendency of pressure towards 

 the continent, were it not counteracted by the constantly increasing 

 pressure from the rapidly cooling regions of the northern land, and the 

 modifying influence of the great circulating system, of which the trades^ 

 and our south-westerly currents are the counter-parts. Since, however, 

 the surface of the land is now cooling more rapidly than that of the 

 adjacent Atlantic ocean, an equality of temperature is gradually estab- 

 lished, and the atmosphere consequently becomes nearly quiescent; for 

 the diurnal variations of temperature and local causes must produce 

 gentle currents, so that an absolute calm cannot exist. This state of 

 quiescence, during which the oscillations of the barometer are small and 

 slow, may continue two or three weeks, but sometimes it lasts only a 

 few days. It occurs not earlier than the middle of October, nor later 

 than the second week of November. It is the period of the Indian 

 Summer, and the cause of the smoky state of the atmosphere at that 

 time. 



But in the space of from about ton to twenty days, the earth, which 

 is the better radiator, reduces the temperature of its surface and that of 

 18 



