THE SMITHSONIAN INSTITUTION. 205 



about 14 inches below its ordinary height. Now it is easily seen that 

 if the warm and light wind from the southward, which blows with so 

 much regularit}^ in summer, were as light as hydrogen gas, with the 

 same elasticity, it would cause an immense fall of the barometer along 

 the whole of the Mississippi valley when it displaced the cold air which 

 has come over the Rocky Mountains. The southerly winds from the 

 Gulf of Mexico are much warmer and lighter than those which have 

 poured down the eastern slopes of the Rock}'- Mountains, and been 

 cooled by radiation on the plains to the east. Consequently when 

 they displace these colder winds in the valley of the Mississippi, a fall 

 of the barometer ought, to a certain extent, to take place. I am, there- 

 fore, of the opinion that the fact of the displacement of the warm and 

 moist current from the Gulf by the colder air from the west and north- 

 west is sufficient to account for the fluctuations of the barometer, in a 

 great man}- instances, in places to the west of the Alleghany chain of 

 mountains. 



Professor Espv informs me that he has traced a central line of mini- 

 mum pressure, which precedes the eastern storms on the Atlantic coast 

 first in the western States. This line is of great length from north to 

 south ; and from an examination of his charts of the weather, in his 

 Report on Meteorology, I was at once struck with the fact that its direc- 

 tion corresponds with the trough of the Mississippi valley, and with the 

 course of the moist winds from the Gulf of Mexico. In Europe, I have 

 often been enabled to trace the connexion between the fluctuations of the 

 barometer and changes in the temperature and moisture of the air ; but 

 I have never found the connexions so regular and intimate as in the 

 Mississippi valley, which is more removed from influences whicli tend 

 to disturb this action. 



In my last lecture I showed that the temperature of the air at sun- 

 rise is, as a general rule, a close approximation to the dew point. A 

 high dew point, or, in other words, a large amount of vapor, has the 

 effect, as we have before said, of maintaining the warmth of the air 

 during the night. 



The observations collected by the Smithsonian Institution give the 

 temperature at 7, a. m., and 9, p. m. As a rise or fall in the tempera- 

 ture at these hours may be considered as indicating an increase or de- 

 crease in tlie amount of moisture in the air, and as the fluctuations of the 

 barometer are also given for the same hours, I have, therefore, in the 

 diagrams, [figures 1, 2, 3, 4,] shown the connexion which exists 

 between the heat and moisture of the air and the changes in the 

 pressure. 



I have compared the temperature taken at 7, a. m., and 9, p. m., 

 leaving out the day temperature altogether. The temperature at these 

 hours, as I have stated, is a close approximation to the dew point. 

 And this fact, which, as a general rule, holds true, especially in 

 autumn, is the only means I possess to ascertain the hygrometrical 

 state of the air during the storms of this country. The amount of 

 vapor is an essential element in the investigation of atmospheric dis- 

 turbances, and hence even an approximate estimate of the quantity is 

 important. 



