[ 447 ] 



thermometer lo^, or in equivalent Englifh denominations 121 

 cubic inches weigh 37,73 grains, and confequently 1000 cubic inches 

 weigh 311,81. But this determination not being the refult of 

 fingle experiments made in thefe circumftances, but a m.ean 

 colle6ted from experiments made at different barometrical heights 

 and in different, temperatures, is not fufficiently rigorous to fervc 

 as a teft in this cafe. , 



Fourth Experiment. 



Barometer 30,32"^ lo. This is one of ninety-eight exjieri- 

 Thermom. 68^. > ments made by me with an excellent air 

 Hygrom. 75 j pump of Haafs's in the year 1786, on the 

 weight of air at different barometrical heights from 30,36 to 29,03, 

 and at different temperatures from 73° to 41''. The receiver 

 contained 116 cubic inches very nearly, which weighed 34>75 

 grains; confequently 1000 cubic inches fhould weigh 299,56 

 grains. 



2do. 68^ furpafs 32" by 36 degrees, confequently deducing 

 from 1000 what they gained by 36*^ of heat (namely by Schuck- 

 burg, 2,43x36 = 87,48) there remain 912,52, which at tempe- 

 rature 32*^ weigh 299,56 grains. 



3K2 



3tio. 



