Atmospheric Oi^igin of the Aurora, Sfc. 147 



Proposition 1st, in relation to the relative time of greatest de-' 

 pression of tetiiperature before different meteors. 



The greatest daily depression or decrement of temperature takes 

 place between one and two days previous to the aurora borealis, 

 auroral clouds and halos.* 



Proposition 2d. Relative order of the thermometric and baro- 

 metric changes before different tneteors. 



Previous to the clouds and halos, the temperature changes either 

 earlier than the pressure or nearly at the same time ; previous to 

 the aurora, the pressure changes more than one fifth of a day be- 

 fore the temperature.* 



Proposition 3d. Length of time before the storm^ when its iti- 

 dications appear in case of different m,eteors. 



When the snow or rain is preceded by an aurora borealis or by 

 luminous columns, the thermometer begins to fall and the ba- 

 rometer to rise between three and five days before the storm ; and 

 when the storm is preceded by auroral clouds or halos the same 

 indications are presented between three and three and a half days 

 before it. 



- Proposition 4th. Ificrease of pressure before rain or snow not 

 preceded by these m,eteors. 



Previous to a thunder shower, or a rain or snow not preceded 

 by an aurora borealis, a halo, or auroral clouds or luminous col- 

 umns, the increase of atmospheric pressure for several successive 

 days is less general, but when it does occur, it commences either 

 earlier or later than when the storm is preceded by either of those 

 meteors ; more generally between five and a half and six days 

 before the shower or storm. 



Proposition 5th. Time from different meteors to snoiv or rain. 



The snow or rain descends sooner after a halo than after an 

 auroral cirrus cloud, earlier after this than after a vertical lunar 

 column, and earlier after a lunar column than after an aurora bo- 

 realis. 



Proposition 6th. Theoretical inference in relation to the nature 

 of these meteors. 



* These propositions now stand nearly as they were corrected in the list of 

 errata in many Nos. 



t I use the term storm from the want of a better one equally brief, to signify 

 the descent of rain, snow, or hail. 



