for the Thermometer, iS\ 



p. S. — Before I became acquainted with the circum* 

 «tance which induced ujC lo adopt the scale which I have 

 now proposed, I endeavoured^ with a view of placing the 

 zero at ihe lowest natural point, to obtain, by inferepce, 

 the utmost degree o^ natural cold. 



In order to effect this, I formed the following table, 

 which is deduced from Mr. Kirwan's table of the ineaih 

 annual temperature o^ every latitude; in which, of course, 

 the latitudes within nine or ten degrees of the pole were 

 deduced, like mine, by inference drawn from the others: 

 and 1 presume, by comparing the results in this table with 

 actual observations as far as these have gone, they will be 

 found to accord sufficiently, to show that the principle \ 

 have adopted, in deducing the results, may be rjiiUed on, 

 (See Tab.* I.) 



Hence it appears, if the resuUs m this table are admitted 

 lo be correct, that the scale would commence at 68 de- 

 grees below the present of Fahrenheit, or 100 degrees 

 below the freezing point of water; and that any greater 

 degree of cold than this, which already has or may here- 

 after be produced by art, would, alone, require the minus 

 sign. 



The greatest decree of artificial cold hitherto upon re- 

 cord is —91 of Fahrenheit: ]iow fur future experiments 

 may carry this point, or whether there be 9. finite point at-r 

 tainable by art, is perhaps impossible to say. 



Knowing that cold is produced by adding snow to alco- 

 hoi, and presuming that alcohol is the last liquid in nature 

 which would freeze, or whose power of producing cold 

 with ice reaches to a lower temperature than any other 

 substance, T formerlv entertained a hope of ascertaining 

 the ultimate point to which this could possibly be carried ; 

 but partly on account of the difficulties of such an under- 

 taking, but more especially in consequence of other pressing 

 avocations which deprive me of the opportunity, I have^ 

 relinquished my intention, at least for the present. 



The method I should have pursued would have been to 

 have formed mixtures of alcohol and nitric acid, increasing 

 the proportion of the alcohol at ihe temperatures at mixing 

 became lower; finally, using alcohol alone with snow, 

 till I should arrive at the points where alcohol itself 

 ceased, when mixed with snow, to produce cold. 



The attempts hitherto made for ascertamnig what is 

 usually called the natural zero^ carry that point so ex- 

 tren^ely low> and the results are so very discordant with 



P M each 



