[ I20 ] 



Steel cannot be too highly tempered for the needle of a fea- 

 conr-pafs, as the more it is hardened the more permanent is the 

 wagnetijm it receives ; but, to preferve the magnctifm and confe- 

 qnently the polarity of the needle, I recommend to have the needle 

 caftd with thin, well-polilhed, foft iron, or elfe to have it armed 

 at the poles with a bit of foft iron. 



I HAVE found, from many experiments, that the cajed needle 

 preferved its magnetifm in a much more perfeft degree than the 

 needle not caied ; and I have fomctimes thought that the mag- 

 netic power of the cafed needle had encreafed, and the magnetic 

 power of the uncafed and unarmed needle always lofes of its 

 polarity. 



Some time ago I placed a cafed needle, an armed needle, and one 

 without either cafe or armour, in a room, for three months, having 

 at that time precifely the fame direction, and nearly the fame de- 

 gree of force. At the expiration of the three months I found that 

 the cafed needle and the armed needle had not in the leaft changed 

 their direction ; but the other had changed two degrees, and had loft 

 very confiderably of its magnetic power : If there were any changes 

 in the other needles, it was too inconfiderable to be perceived. 



These obfervations appear to me to be new, and may tend to 

 great improvement in our fea-compafs : They are here -fubmitted 

 to the confideration of philofophers. 



Cove of Cork, li^th April, 1788. 



