'*.■ ••• V <1^ low Tejnperatures, "^^^ ^^ 187 



marked for low temperatures, and having provided a quan- 

 tity of murij^te of lime and snow, I reduced the temperature 

 of the alcohol and metals, as in my former experiments the 

 subject of M. Tardy de la Brossy's observations, and ob- 

 tained the results expressed in the following table. 



In an examination of the 1st and 2d columns of this 

 table relating to silver, it being previously known that 

 10*362 is the specific gravity of silver, and that -8141 is ihe 

 specific gravity of alcohol at 50^ of Fahrenheit, where watej^ 

 \s !-000, it appears, following the general mode of calcu- 

 lation in which the loss of weight is made the divisor 

 pf the quantify weighed, and the quotient expresses the 

 specific gravity, thai as at 100 degrees of heat the divisor 

 for the quantity of metal is less, apd at zero greater, that 

 at zero the specific gravity is less, and at 100 degrees above 

 it the specific graviiy is greater : — A conclusion opposing 

 a general law, which therefore we cannot admit. 



'I'o account for this apparent inaccuracy of result, we 

 must direct, our attention to the alcohol employed, and 

 see if, in the comparison of the /?//./"(y alcohol and the solid 

 gilver, this appearance of contradiction to an established law 

 does not arise from the contraction of the fiuid. Of this it 

 will be ditficidt to bring visible proof to the extent required; 

 tor if I put into a glass tube alcphol, of wljich the weight is 



known. 



