1814.] Imperial Institute. 22'0 



other known cause of magnetiscment had been carefully removed. 

 The summer of IS 13, indeed, was not favourable to these kinds of 

 experiments, it was so bad. 



Of all the phenomena presented by heat, the dilatation which it 

 produces in bodies is the one the laws of which may be most 

 naturally expressed by mathematical formulas; and the knowledge 

 of these laws, which constitutes an essential part of natural 

 philosophy, is likewise very important in many chemical experi- 

 ments. M. Biot has taken a great deal of pains on this subject, 

 and taking as a term of comparison the dilatation of mercury, he 

 finds that the true dilatation of other liquids may be always expressed 

 by that dilatation, by its square and its cube, multiplying each of 

 these three terms by a particular coefficient, which must be deter- 

 mined for each liquid ; but which, when once determined, remains 

 the same at all degrees of heat. As the substance of the thermo- 

 meter, which contains the liquor observed, undergoes a dilatation 

 itself the apparent dilatation is different from the true one. But 

 M. Biot shows that it follows a similar law. He then calculates 

 from the experiments of M. Deluc the proper coefficients lor eight 

 of the liquids, the dilatations of which it is most necessary to be 

 acquainted with, and shows that these coefficients being once 

 obtained, his formula trives the dilatation for each degree of tempera- 

 ture as accurately as experiment. Finally, lie has made the appli- 

 cation of them to the combined dilatations of the vessel and the 

 liquid, and shows that we may separate the effects which belong to 

 the liquid and to the vessel, and determine the influence of each 

 with sufficient exactness to find by calculation alone all the results 

 observed: so that henceforth calculation may in a great variety of 

 cases dispense with immediate observations, and we may introduce 

 our calculated results without fear into the elements of pheno- 

 mena. This is a greater advantage, because this kind of experiment 

 is excessivelv delicate ; and, unless the utmost possible attention 

 be bestowed, many raises, easily conceived, but almost im- 

 possible to avoid, continually interfere with the accuracy of the 

 result. 



What M. Charles has observed, in a beautiful set of experiments, 



•made with an instrument of his contrivance to render sensible and 



to measure the maximum of the dilatation of water, agrees exactly 



with the formulas of M. Biot, adds to our confidence in them, and 



that thy may lie employed with safety. 



There has existed lor a long time a discussion among chemists 

 respecting the time when alcohol is formed in wine. It was formerly 

 tin general opinion that alcohol was a necessary product of fermen- 

 tation : 'an .'. . , bronj, a corresponding member of the Institute, 

 I.;: - 1 , ported the contrary opinion. According to him it is only 

 tccid . and when there is too much heat, thai fermentation 



product - alcohi I. In 01 vine alcohol is produced only by the 



: of distillation. His principal proof was that alcohol could not 



