High Temperatures. 185 



direct comparison of the two elements, or in other words if T 2 

 be referred to the observations made with the gas thermometer 

 as used with porcelain bulbs. 



In Table XVI in the first column the values of the electro- 

 motive force for T„ calculated from the formula are contained. 

 The column A contains the differences between these values 

 and those obtained from Tj + (T 2 — TJ as suggested above. If 

 now these values of A be compared with the earlier ones (Table 

 XI, former paper) it will be seen that the differences have be- 

 come very small now that the proper expansion coefficients of 

 both porcelain and platin-iridium have been applied. 



From this the conclusion follows that the porcelain bulbs 

 lead to substantially the same results as the platin-iridium as 

 long as the former are not used above 1100°. Under these 

 conditions of course only bulbs glazed inside and out need be 

 used, as the glazing does not melt up to that point. 



Regarding the precautions necessary to be observed in the 

 use of porcelain bulbs and the degree of accuracy obtained in 

 temperature measurements with them, reference is made to the 

 former communication. 



Element Pt—90 Pt / 10 Ru. — The element platinum — platin- 

 ruthenium follows a curve very similar to that of the normal 

 element. Its equation is 



e 3 = — 359 + 9-260;! + 0-00150 ^ 



Element Pt — lr. — For the element platin-iridium a formula 

 has also been calculated which like the others holds from about 

 300° on. Like the following curves it is based on the thermo- 

 electric force of two pure metals and possesses a much greater 

 curvature then those of the alloys. It is — 



e 4 = — 248 + 7-282 i + 0-00554 f 



Element Pt—Rh. — For the combination platinum-rhodium 

 two specimens of pure rhodium were at our disposal. The 

 first was from Heraus, the second was especially prepared in 

 the Reichsanstalt. The two formulae are 



e 5 =— 228 + 7-230 £ + 0-00660 f 



e 6 = — 235 + 7-410i! + 0-00660« 2 



The second equation is derived under the assumption that 

 the differences between the two curves e 6 and e 6 can be repre- 

 sented by a linear relation. The observed results between 

 300° and 1150° — the range of the gas thermometer — are made 

 the basis for the derivation of this as well as of the other 

 formulae. 



Table XYI contains the observations upon these elements as 

 well as the differences between them and the values calculated 

 from the above formulae in microvolts and degrees. 



