38 



denly to an opposite point, from whicli quarter it blows with equal vio- 

 lence, till the storm has passed over: it is under this portion of the storm 

 that we notice the greatest fall of the barometer, and the mercury usually 

 begins to rise a short time previous to the change of wind. 



10. On the inner portion of the track the wind commences from a more 

 eastern or northeastern point of the horizon, and afterwards veers by north 

 to a northwest or westerly quarter, where it finally terminates. 



11. Hence Mr. Redfield infers that the portion of the atmosphere which 

 composes the body of the storm blows in a horizontal circuit around a verti- 

 cal axis of rotation, which is carried onward with the storm^ and that the 

 direction of the circuit is from right to left. 



12. The barometer always sinks while under the first portion of the 

 storm, and rises again under the last portion of the gale. 



In 1835 Mr. Redfield published an analysis of several additional storms 

 which visited the American coast, and accompanied his paper with a chart, 

 upon which eleven of these tracks were carefully represented. In subse- 

 quent years Mr. Redfield continued his investigations, and in 184(5 he 

 published an analysis of three additional hurricanes, making sixteen storms 

 whose tracks are delineated upon his chart of the Atlantic coast. The re- 

 sults of all these investigations served to confirm substantially the conclu- 

 sions published in 1833. 



In the Journal of the Franklin Institute for April, 1836;, Mr. Espy 

 commenced the publication of a series of essays, in which he announced a 

 new theory of storms; and he has since continued his researches up to 

 the present time. The following generalizations, given in his own words, 

 are the latest at which he has arrived: 



1st. The rain and snow storms, and even the moderate rains and snows, 

 travel from the west towards the east, in the United States, during the 

 months of January, February, and March, which are the only months yet 

 investigated. 



2d. There is a depression of the barometer near the central line of 

 the storm. 



3d. The central line of minimum pressure is generally of great length 

 from north to south, and moves sideforemost towards the east. 



4th. This line is sometimes nearly straight, but generally curved, and 

 most frequently with the convex side towards the east. 



5th. The velocity of this line is such, that it travels from the Mississippi 

 to the Connecticut river in about twenty-four hours; and from the Con- 

 necticut to St. John, Newfoundland, in nearly the same time, or about 

 thirty-six miles an hour. 

 5 



