[ "5 ] 



and accurate friend whcfe valuable correfpondeijce accompanies 

 it; and whatever its incrinfic value may be, the author has no 

 doubt that, from its having fuggefled the following curious 

 hints and obfervations, it is well worth being offered to the 

 notice of the Academy, He is himfelf convinced, by the ex-' 

 perience of many years, that it will adt effecflually and well in 

 the form he defcribes, which, as being by far the mofl: fimple, 

 he therefore propofes as moft eligible. He muft confefs that in 

 making thefe inflruments a confiderable degree of nicety is 

 required, to adjufl: the cork to the box and tube in fuch a 

 manner, as to allow the air a ready accefs to the furface of the 

 mercury, and at the fame time completely confine it in its 

 box ; and it has often occurred that by compreffing the cork 

 too tightly the rife and fall of the mercury have been more 

 gradual than were to be wiflied. To remedy this, the author 

 thought of, and put in pradlice, fome contrivances that were 

 fully fufEcient ; but as he thinks none of them fo complete as 

 that fuggefted by his correfpondent, he will not increafe the 

 length of this paper by inferting any of them. Having made 

 numbers of the portable barometers in their fimpleft forms, 

 compared their variations with thofe of the very befl open 

 barometers, and found them to correfpond exadlly ; having 

 alfo carried them fome thoufands of miles, moftly in a carriage, 

 but often on horfeback, without injury to any of them, he is 

 inclined to think that, in the improved form, they fliould be 

 adopted for general ufe, and may be readily and univerfally 

 made upon the plan of a corredled fcale, as fet forth in the 



P 2 following 



