3^0 On the hifliiencE of the Butt 



time, f was quite fatisfied, in paffing through this regiftei', 

 that if I had allowed niyftilf to choofe the higher notations 

 about the quarters, and the lower about the new and full, 

 with a latitude of 24 or 36 hours, it would have made the 

 refalis as much more favourable to my Cunclufions as in her 

 former cafe. 



Now, to omit the confideration of other proofs for the 

 prcfent, it appears to me evident, that the atmofphere is fub-» 

 jeft to a periodical change of gravity, whereby the barometer, 

 on a mean often years, is deprclTed at leall one-tenth of an 

 inch while the moon is paffing from the quarters to the full 

 and new; and elevated, in the fame proportion, during the 

 return to the quarter. To whafcaufes fliall we attribute this 

 periodical change,' other than the atlraAion of the fun and 

 moon for the matter compofing the atmofphere ? 



The atmofphere is a gravitating fluid, difiering, in a phy- 

 fical fenfe, from the water, chiefly in pofleffing lefs gravity; 

 and it is demopflrable a priori on the principles of the New- 

 tonian philofophy, that it ought to have its tides as well 

 the ocean, although in a degree as much lefs perceptible as 

 is its gravity. 



I fuppofe, therefore, that the joint attraftions of the fun 

 and moon at the new moon, and the attraftlon of the moon 

 predominating over the fan's weaker attraftion at the full, 

 tend to deprcfo the barometer, by taking off" from the gravity 

 of the atmofphere, as they produce a high tide in the waters, 

 by taking off" from their gravity : and again, that the attrac- 

 tion of the moon being diminiflied by that of the fun at her 

 quarters, this diminution tends to make a high barometer, 

 together with a low tide, by permitting each fluid to prefs 

 with additional gravity upon the earth. 



I am aware that feveral circumfl:ances attending the fa6ls 

 I have brought forward prevent them from making a cafe 

 fairly parallel with the tides ; on which I fliall make fuch re- 

 marks as occur in this. early ftage of the inquiry. 



In the firft place, the waters have daily tides correfpond- 

 ing with the motions of the earth and moon ; and there is 

 additional elevation and depreflion about the phafes of the 

 moon, amountmg, fuppofe, to a fourth part of the whole* 



In 



