RF.MARKS L'PON METEOROLOGY. 7 



which would not reach us, or that to the N W there was & 

 decom position of some part of the atmosphere, commencing 

 in the higher regions, to winch the upper current, in which 

 these clouds moved, was rushing. The appearances that 

 afterwards took place proved, that this latter supposition 

 was correct; for, although the wiud went to the W, the di- 

 minution of the weight of the atmosphere still continued, 

 while the appearance which I have described, of white clouds white clouds 

 on a gray ground, is one of the most certain indications of on a gray 

 the mixture of two currents possessed of different eonstitu- 6 

 tions. It is to be observed, that, before the storm attained 

 its greatest violence, the barometer began to rise; and as it 

 depended merely upon the tendency of the air to produce 

 an equilibrium of pressure, it was probable, that it would 

 neither be of long continuance, nor very extensive. 



I am here led to notice the difference between this kind of stormsfrom 

 storm, which is produced by an abstraction of part of the diaiinullun 



, . . . . „ aiul inoeaseof 



atmosphere, and that which appears to originate from atheatmos-. 



contrary cause, from a sudden increase of the volume of the P here « 

 atmosphere. Of this latter species a well marked instance 

 occurs while I am writing this paper. Yesterday, March 

 25th, we had a strong gale from the E; the barometer had 

 been before about the medium height, but it rose during 

 the continuance of the stoim; to day the wind is more mo- 

 derate, and the atmosphere is becoming lighter. I may ' 

 here observe how much insight would be gained with respect 

 to the theory of the weather, had we a number of accurate Comparative 

 comparative observations made in different places on the observation- 

 progress of such storms as the one that took place on the 

 18th of September. Were we to ascertain exactly at what 

 hour the barometer got to its minimum, when it began to 

 rise, how long after this storm acquired its greatest violence, 

 when it began to abate, and when the barometer arrived at 

 its maximum, we should probably have gone a great way 

 towards obtaining a correct theory on the subject. One 

 important point might be ascertained, whether a storm be 

 retrograde or progressive in its formation. 1 apprehend, 

 that, where the current is rushing forwards to supply a va- 

 cuum, the storm will be retrograde; i. e. supposing the par- 

 tial vacuum tc have taken piace over the extern part of the 



county 



