[ 284 I 



plains, appears to me much more intelligible and better founded. 

 In this fyflem, firft ftiggefted by Halley, and beautifully illuftrated 

 by Roy, and ftill farther extended by Dodor Hamilton, water is 

 attraded by air, and forms a compound lighter than mere air of 

 the fame temperature, and under the fame preffure ; but the 

 difference of their gravities is not confiderable. Whereas, in Mr. 

 De Luc's hypothefis, water and air, not being chemically united, 

 do not form a compound, but rather what in technical language 

 is called an aggregate, like water and oil mixed together; if fo, 

 vapours, being fpecifically lighter by one-half than air of the fame 

 temperature, fhould rife above the ftratum of air in which they 

 are formed, until they are cooled during their afcent, or meet 

 with air of the fame denfity, betwixt which and air of fuperior 

 denlity they muft remain. If fo, vapours (hould never be found 

 in the lower flrata of the atmofphere, unlefs they be perpetually 

 renewed ; confequently, if the fource of evaporation be cut off, 

 as it is when moiff air is placed under a receiver flanding on 

 metal, or on any other dry fubflance, and the temperature be kept 

 undiminifhed, the -vapour originally difperfed through the whole 

 of its volume fliould gradually colled and float at the top, as 

 happens when water and oil are mixed together. 



But this feparation of air and vapour never happens. For if 

 concentrated fulphuric acid, or dry cauilic tartarin, jor muriated 



lime, 



