THE 



AMERICAN JOURNAL OF SCIENCE 



[FOURTH SERIES.] 



Art. YIII. — The Nitrogen Thermometer from Zinc to Palla- 

 dium ; by Arthur L. Day and Robert B. Sosman ; with 

 an Investigation of the Metals, by Eugene T. Allen. 



Contents : 



1. Introduction and Plan. 



2. Apparatus. 



3. Details, Errors, and Corrections. 



A. Temperature of Gas in Bulb. 



B. Definition of Temperature by Measurement of Pressure. 



C. Transference of Temperature by the Thermoelement. 



D. The Fixed Points. 



4. Experimental Data and Calculated Eesults. 



A. Expansion Coefficient. 



B. Gas Thermometer Data and Fixed Points. 



5. Interpolation between the Fixed Points. 



6. Analysis of Metals. (By E. T. Allen.) 



7. Conciusion. 



1. Introduction and Plan. 



The measurements of absolute temperature here offered 

 were undertaken in direct continuation of those published 

 from the Geophysical Laboratory two years ago,* with the 

 purpose of extending the gas scale to 1600°, or as near it as 

 might prove practicable. Except in explanations of new or 

 particularly important features, descriptive details have accord- 

 ingly been omitted here and must be sought in the first paper. 

 Substantially the same methods and apparatus have been 

 employed throughout. 



One conclusion in particular which was brought out at that 

 time is entitled to even greater emphasis, namely, that the 

 existing uncertainties in the absolute temperature scale at 

 1000° and above are the result of experimental limitations and 

 not of any failure of the principles involved. The experi- 

 mental conditions were scrutinized with great care throughout 

 the first investigation, and not only were the known correction 

 factors all redetermined, but their total magnitude was reduced 

 nearly 75 per cent. This success, after so long and painstak- 



* Arthur L. Day and J. K. Clement, this Journal (4), xxvi, 405-463, 1908. 



Am. Jour. Sci. — Fourth Series, Vol. XXIX, No. 170. —February, 1910. 



7 



