952 



MR A. CRICHTON MITCHELL ON THE 



The following table contains nearly all the substance of the further calculations, 

 and sufficiently explains itself : — 



Thermometer. 



Distance in Feet 

 along Bur. 



Temp. Excess 



■c, 



dv 



dx ' 



Area of Curve of 



Cooling to next 



Value of x. 



Area of Curve of 



Cooling to end 



of Bar. 



Area corrected for 



Change of Specific 



Heat. 



A 

 B 

 C 

 D 

 E 



00 



025 



05 



075 



1-25 



189-2 



1082 



617 



341 



115 



35-32 



20-29 



11-86 



711 



6-625 

 3321 

 1-636 

 102 



12-603 

 5-977 

 2-656 

 102 



14-653 

 6-619 

 2-841 

 1065 



From these data, it follows that the thermometric conductivity of manganese-steel is 

 represented by the following numbers : — 



(1) 



From uncorrected areas, 









0° 



•00221 



100° 

 00254 



200° 



•00287 



(2) From corrected areas, 



•00233 



00285 



00337 



It now remains to correct the values of the tangents and the areas of the curve of cooling 

 for the error involved due to variable heating of the thermometer stems. The method by 

 which this error is estimated and applied is detailed by Professor Tait in his introduction 

 to my former paper. Applying this correction, the above results become — 



(1) From areas not corrected for change in specific heat, 



0° 100° 200° 



00211 -00246 -00281 



(2) From areas corrected for change in specific heat, 



•00219 00272 -00325 



For the purposes of comparison with the corresponding results for iron* (Forbes' 

 wrought iron bar, cooled midway), the conductivity of manganese-steel may be taken 

 as above, the results being corrected for change in specific heat, and also for error due to 

 unequal heating of thermometer stems. The figures are as follows : — 





0° 



100° 



200° 



Manganese-steel, . 



•00219 



•00272 



00325 



Iron, .... 



0119 



•01274 



01358 



Hence it appears that the presence of 10 per cent, of manganese in iron or steel 

 lowers its conductivity at 100° to one-fifth, and that the rate of increase of conductivity 

 with temperature is, in manganese-steel, little more than half the corresponding 

 coefficient in iron. That such a proportion, comparatively small, of manganese should 

 have such a distinct effect upon the conductivity of steel, is remarkable, and, while it 

 is on a parallel with the other rather anomalous properties of this substance, it points 



* Trans. Roy. Soc. Edin., vol. xxxiii. p. 555. 



