260 



Sir James Dewar 



[Jan. 16, 



A fairly direct calibration of the instrument can thus be made in 

 terms of actual temperature increase. Since at the boiling point of 

 liquid oxygen the dp/dt per 1° is 81 mm. Hg., an added pressure 

 of 14 to 15 mm. of alcohol (0*8 mm. Hg.) would correspond to an 

 increment of temperature of the liquid air or oxygen of 1 /100th 

 degree. The curves (Fig. 9) give the indications of the thermoscope 

 following increments of pressure of 5, 10, 15, and 20 cm. of the 

 alcohol manometer, the corresponding increments of temperature 

 of the liquid oxygen being 0'023°, 0'067°, 0*10°, and 0*133° Abs. 

 The response of the thermoscope with time after these instantaneous 

 additions of pressure to the oxygen bath are seen to be of the same 

 character in the case of air-saturated charcoal as those produced 



PRESSURE 

 CMS ^ALCOHOL 



I'O 15 



Fig. 9. — Air Saturation. 



by the usual exposure to the Leslie cube, but of much less intensity, 

 as shown by the comparison curve of the ordinary exposure. 

 The limit registered by the thermoscope was in all cases less than 

 that of the bath, as was to be expected from the higher latent heat 

 of the air-in-charcoal as compared with the liquid oxygen. Thus 

 the 5 cm. increment (0*033°) caused a rise of the thermoscope mano- 

 meter to 2 ' 6 cm. in 8 minutes, which on releasing the pressure of 

 the liquid oxygen fell to 1 cm. in 3 \ minutes, and took 9 or 10 

 minutes to re-equilibrate to zero. Similarly for the other increments 

 of 10 ecu., 15 cm. and 20 cm. More extended measurements on 

 these lines would no doubt give much information concerning the 

 latent heat of gases in charcoal under various conditions. 



With hydrogen saturating the charcoal instead of air (Fig. 10) 



