[ i68 ] 



experiment, to the number of i o 5 

 fub-divifions. 



Befides the fources of errors al- 

 ready mentioned, there may happen 

 fome others from accidental circum- 

 itances, which may be principally 

 three. 



I. The degree of heat of the 

 common air may change during the 

 time you make your experiment, 

 and may occation a greater or lefs 

 exteniion of the column of air in 

 the great tube. 



II. The weight of the air, or its 

 preffure, may alfo change in the in- 

 terval of inflituting the experiment. 



III. The difference of heat com- 

 municated to the tube by the body 

 of the obfervator himfelf in the 

 time he is near it to make the 00- 

 fervation. 



Thefe 



