oct. — mar. 1859-60.] Rooms in Tropical Climates. 315 



enabled, through the kind intervention of Mr. Stirling, C. E., to try 

 the experiment" on as large a scale as could possibly be desired. 



At the Kinniel Iron Works (the nearest works of that descrip- 

 tion to Edinburgh,) air is pumped into a series of blast furnaces by 

 a powerful steam-engine, under a pressure of 3'5-lbs. on the square 

 inch. The air pumps are two in number, double acting, with cy- 

 linders about 5 feet in diameter, and 10 feet stroke ; so that, in so 

 far as a compression of 3'5-lbs. could serve, the volume of air was 

 prodigious, and completely removed all fear of sensible error aris- 

 ing from the frictional heat of the piston, or from radiation at the 

 surface. Mr. Wilson, the owner of the works, very kindly, on the 

 application of Mr. Stirling, gave every facility for trying the expe- 

 riments, and I had a new thermometrical and mercurial gauge ap- 

 paratus constructed for the purpose. 



The observations, which it is needless here to detail in full, as 

 they will appear elsewhere, were made in the presence, and with the 

 assistance of Mr. Stirling, C. E. and Lieut. Driscoll Gosset, It. E. ; 

 and consisted in determining, by a considerable number of trials, — 



1st. The temperature of the air entering the valves of the air 

 pump. 



2nd. The temperature of the air in the large air vessel, into 

 which it had been forced by the pumps under a certain pressure. 

 3rd. The degree of that pressure ; and 4th. The temperature 

 of the air on issuing out into the atmosphere from under that pres- 

 sure. For these last it was necessary to bore a hole into the air 

 chest, and this Mr. Wilson most freely allowed us to do, and the 

 hole being above one inch in diameter, the rush of air out of it was 

 more than sufficient to completely enclose and fully inpress its 

 temperature on the bulb of the thermometer. 



1st. Temperature of entering air 63° Fahr. 



2nd. Temperature of compressed air 92°. 



3rd. Compression 7*2 inches mercury. 



4th. Temperature of escaping air 63° 



The Barometer was about 30 inches at the time. The tempe- 

 ratures may be considered to be determined certainly within a 

 degree less or more, and the pressure within one-tenth of an inch. 

 Hence we have, with a very small probable error, a compression 

 Vol. xx. o. s. Vol. vii. n, s. 



