226 Royal Society : — 



the part due to the expansion of mercury, as well as the two last 

 terms, which contain the part due to the expansion of glass. 



Let r = the ratio in which the coefficient of glass has to be re- 

 duced, and X the ratio of the change that has to be made simulta- 

 neously in the first term. We have at 100°, tm =ta ; and by division, 



Bx /lOOOO , 1 \ , „ 



hence 



«=^{'-(T"-r)} 



The coefficient of glass being taken at M. Pierre's minimum, fixes 

 >'=|, and log.r=9'9S41437, and the formula becomes 



BU ft.^ L\ 



If we now compute the difference between this thermometer and the 

 former, we shall find that it is j J-^th of a degree behind at 50°, and 

 l°-52 in advance at 300°. 



This seems to prove that the influence of a difference in the quality 

 of the glass of which thermometers are composed can have no sen- 

 sible effect on the graduated difference brought out by the projection 

 (fig. 2). 



In conformity with what M. Regnault remarks as to glasses with 

 the lesser coefficient experiencing a less rapid augmentation of dila- 

 tation with the temperature, it is probable that the term involving 

 the cube should be changed to a lower power of the temperature 



when the coefficient — is diminished. 



22 London Street, Edinburgh, J. J. Waterston. 



January 7, 1858. 



XXIX. Proceedings of Learned Societies. 



ROYAL SOCIETY. 



[Continued from p. 1 52.] 



June 18, 1857. — The Lord Wrottesley, President, in the Chair. 



The following communications were read : — 

 " Remarks on a New Class of Alcohols." (Second Note.) By 

 A.W. Hofmann, LL.D., F.R.S., and Augustc Cahours, F.C.S. 



In a former communication (Phil. Mag. Oct. 1856, p. 309), we 

 endeavoured to delineate the characters of a new alcohol — the AUylic 

 alcohol, which is the prototype of a new class of alcohols. We have 

 since continued these researches in order to complete the history of 

 this remarkable compound. 



In submitting to the Royal Society the full account of our expe- 

 riments upon the subject, we beg leave to mention in this abstract 

 briefly some additional compounds which we have examined since 

 o\ir last communication. 



