in Fluids. 367 



ference of temperature), can remain in contact, I leave 

 others to determine ; in the mean time, it is most cer- 

 tain that the existence of this impossibility of any im- 

 mediate communication of Heat among the particles of 

 a Fluid renders the distribution of Heat very -unequal ; 

 and it seems highly probable that many appearances 

 which have been attributed to very different causes are 

 in fact owing to intense Heat existing and producing the 

 effects proper to it in situations where its existence has 

 not even been suspected. 



If Fluids are non-conductors of Heat, no situation 

 can possibly be more favourable to its preservation than 

 when it exists in them ; and it is not only evident, a 

 priori^ that the most intense Heat may exist in a few soli- 

 tary particles of some Fluids without its being possible 

 for us to detect it, or to discover the fact, either by our 

 feeling or by the thermometer ; but there are many ap- 

 pearances that strongly indicate, and others that prove, 

 that intense Heat actually does exist in that concealed or 

 imperceptible state very often. 



There is no reason to suppose that it is possible for 

 ice to be reduced to steam without being previously 

 melted ; and it is well known that ice cannot be melted 

 with a lower degree of Heat than that of 32 of Fahren- 

 heit's scale : but in the midst of winter, in the coldest 

 climates, and when the temperature of air of the atmos- 

 phere, as shown by the thermometer, has been much below 

 32, ice, exposed to the air, has been found to evaporate. 



How can we account for this event, except it be by 

 supposing that some of the particles of air which acci- 

 dentally (as we express it) come into contact with the ice 

 are so hot, as not 'only to melt the small particles of ice 

 which they happen to touch, but also to reduce a part of 



