THE 



AMERICAN JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



Art. XLIII. — The Nitrogen Thermometer Scale from 300° 

 to 630°, with a Direct Determination of the Boiling Point 

 of Sulphur • by Arthur L. Day and Robert B. Sosman. 



1. Purpose and Plan of the Investigation. 



In our recent investigation of the fundamental high tempera- 

 ture scale with the gas thermometer,* attention was chiefly 

 directed to the temperature region above 1000°. The lower 

 temperatures had been determined with considerable accuracy 

 in an investigation undertaken in the Reichsanstaltf some 

 years earlier (1900) and there was no reason to suspect, any 

 uncertainty in it of greater magnitude than the errors of 

 observation determined at that time (2° to 3°). This does not 

 mean that it was considered unnecessary to reduce the mag- 

 nitude of the errors, which were known or suspected to exist 

 in the Reichsanstalt instrument, — quite the contrary. All our 

 efforts throughout an investigation which has now continued 

 since 1904 have been put forth in the direction of reducing 

 these errors to a minimum, but this was done more especially 

 because of their much greater influence upon temperature 

 determinations above 1000° than because of any specific 

 demand which then existed for a new or better temperature 

 scale in the region below that temperature. 



The situation from our viewpoint at the time of publication 



* Preliminary Publications : 



Day and Clement, Phys. Eev., xxiv, 531, 1907 (Abstract). 

 Day and Clement, this Journal (4), xxvi, 405-463, 1908. 

 Day and Sosman, " (4), xxix, 93-161, 1910. 



E. B. Sosman, " (4), xxx, 1-15, 1910. 



Final Publication : 



Day and Sosman, Carnegie Institution Pub. No. 157, 1911. 

 f Holborn and Day, Ann. Physik (4), ii, 505-545, 1900 ; this Journal (4), 

 x, 171-206, 1900. 



Am. Jour. Sci. — Fourth Series, Vol. XXXIII, No. 198. — June, 1912. 

 34 



