1819.] Prof. Leslie- on Heat and Climate. 17 



whence y - (2 x - x &) i£L = (*■ - j x «) 14-27°. This quan- 

 tity must be augmented in the ratio of A F to O F, in order to 

 find the proportional excess of temperature, which is, therefore, 



= H-27 (V^*j. 



To ascertain the actual quantities, which is a most important 

 point, I took another glass receiver, whose capacity was one- 

 third of that of the former, and its surface -f^ths, allowing for 

 that of the circular pieces of leather on which it stood. The 

 utmost rise of the included thermometer was observed to approach 

 to 7-5°, and the other ascents were proportional to the numbers 

 already discovered. But the quantity of heat lost on the glass 

 will depend on the extent of surface compared with the capacity 

 of the receiver. In the present instance, the effects would have 

 been the same in receivers of equal capacity, but whose surfaces 

 were as 27 to 40. The heat communicated to the surfaces is, 

 therefore, as 27 x 10-7 to 40 x 7-5, or as 288-9 to 300. Hence 

 as 300 — 288-9 is to 288-9, so is the difference between the heat 

 lost on the glass in the two cases, or 10*7° — 7-5°, to 83*3°, the 

 heat lost on the smaller surface, or on the large receiver. 

 Whence the whole heat extricated was 94°. Augmenting, there- 

 fore, the preceding formula in the ratio of 94° to 10-7°, we 



obtain 125 a? \-^~) r ° r tne decrease of temperature arising 



from a diminution x of the air's density. If .r be made negative, 



125 x ( i + ** ) will denote the augmentation of temperature due 



to an increase x of density.* 



This last formula affords a satisfactory explanation of several 

 curious phenomena. If I blow with a common bellows against 

 the bulb of a thermometer, the mercury will rise three or four 

 degrees ; because the air striking forcibly is suddenly condensed 

 and its temperature elevated. But the effect must be precisely 

 the same whether a body is exposed to a stream of air, or is carried 

 with equal celerity through the still atmosphere. To verify this 

 position, I took an hollow brass ball about an inch and half in 

 diameter, and filled the cavity with mercury, having previously 

 rubbed the inside with oil to prevent the action of this metal. 

 I then fastened a string of a convenient length to the ball, and 



• I have since carefully repeated this experiment on a much larger scale, and 

 with an air-pump of the best construction. The results are somewhat different and 

 much simpler. If d ex press the density of air, while that of the common standard 



is denoted by unit, the corresponding difference of temperature is 45° ( d) 



on Fahrenheit's scale ; or putting x = 1 — d, the formula becomes 45° x ( ) 



in which, chiefly, the coefficient is altered. See the article Climate in the Supple- 

 ment to the Kncyclopaedia Britaiinica. — A. 



Vol. XIV. N° I. B 



