T Ic 7 ] 



■wet to the touch on the furface, and an the frefh furfaces of the 

 fractured parts. I then placed a part of it in a glafs funnel before 

 the fire, to melt, and found the water ftrongly faline to the 

 tafte, but not near fo faline as equal parts of fea and river-water 

 mixed. 



" Another portion of. this ice, which was wrapped up in filter- 

 ing paper, and left to drain on a heap of dry mow during four 

 days, when melted, was faline to the tafte, and not fenfibly 

 different from that which had drained only fix or feven hours. 

 Whence it appeared, that ice formed in the fea-water, in circum- 

 itances fimilar to thofe which attend natural congelation, is, 

 neverthelefs, faline to the tafte. 



" The feveral portions of water obtained in the foregoing ex- 

 periments, from the wafhed ice of the fea-water in A and B, 

 being preferved in glafs- ftopper- bottles, were not examined. 

 Although they were frefh to the tafte, it appeared by the quan- 

 tity of luna cornea^ which they all formed with faturated nitrous 

 rfolution of filver, that they were ftrongly impregnated with 

 marine fait, comparatively with Thames and New River water, 

 ^examined in the like manner. 



44 Mr. Barrington obferving, that fait hx water is an impedi- 

 ment to the congelation of that water, prefumed, that fait 

 an water would accelerate the thawing of ice immerfed in it ; ; 

 •and that in equal temperatures ice would be thawed in fea-water 

 dfooner than in frefh water. I therefore made the following ex- 

 periment. 



" Januaiy the 20th, when the Thermometer pointed to 

 twenty-three, about nine o'clock at night, I placed five ounces 

 •and half a drachm, averdupoife, of Thames water in a half pint 

 glafs tumbler ; and the like quantity of the fame water diftilled 

 in another half pint glafs tumbler, of equal figure and capa- 



P 2 city 



