CONTRACTION OF WATER BY HEAT, 343 



In 1788, Sir Charles Blagden added the curious obferva- |^"j. B ^- 

 tions, that water, which by flow and undifturbed refrigera- wat „ ma y be 

 lion permits its temperature to fall many degrees b eloVv its cooled^many de- 

 freezing point, perfeveres in expanding gradually as the tem- 3i? w i t h©„t 

 petature declines ; and that water having fome muriate of freeaing, and 

 foda or fea-falt diflblved in it, begins to expand about the ^ *™ t% M w * 

 fame number of degrees above its own term of congelation 

 that the expanfion of pure water precedes its freezing, that 

 is, between eight and nine degrees. More lately, (Philofo- 

 phical Tranfaaiom, 1801), he, or rather Mr. Gilpin by his 

 direction endeavoured to afcertain, by the balance and weigh- 

 ing bottle, the amount of this change of denfity caufed by a 

 few degrees of temperature. 



Every one mult be familiar with the ufe which Count Rum- Count Rum- 

 ford has made of this peculiarity in the conftitution of water, tions o{ ^- ir 

 in explaining many curious appearances that prefented them- doctrine to the 

 felves in his experiments upon the conducting power of fluids,^" y 

 and in accountiug for certain remarkable natural occurrences. 

 The Count, with his ufual ingenuity, has endeavoured to point 

 out the important purpofes which this peculiarity ferves in the 

 economy of nature, and to afiign the final caufe of fo remark- 

 able an exception from a general law. 



In recording the obfervations and opinions that have been Mr * p*lton*s 

 ,,. n , • • ' .. • \ t • L f j *• experiments, 



pubhihed concerning this point, I might now, in order, notice 



thofe of Mr. Dalton of Manchefter, related in the fifth volume 



of the Moncheftcr Memoirs, which tended to confirm and enlarge 



our knowledge of it. But as Mr. Dalton himfelf has called in 



<{ueftion the accuracy of the conclufion which had been drawn 



from his experiments, and from thofe of preceding obfervers, I 



fliall only remark, that they are of the fame nature, and nearly 



to the fame purport, as thofe of M. de Luc. 



It was in contequence of a communication with which Mr, w ^° 1 ue ft' oa * 



Dalton favoured me, three months ago, that my attention was conc lufions, 



directed to this fubjeft, He informed me, that after a long 



train of experiments he was led to believe that he, and his 



predeccflbrs in the fame field of invefiigation, had fallen into 



a mifiake with regard to the contraction of water by heat, 



and its expanfion by cold, in confequence of overlooking or 



underrating the effect which the change in the capacity of the 



ihermometeNfliaped apparatus employed, rnuft occafion on the 



apparent volume of the fluid. He ftated, in general terras, 



that 



