Daniell on a New Pi/rofneter. 315 



existing in a bath of this fluid, however unequally heated ; but, 

 upon inquiry since, I find that the remark has been often made. 

 The result has afforded another very satisfactory testimony of 

 the delicacy of the pyrometer, and proves that it will very ac- 

 curately indicate the mean of the temperature to which the bar 

 is exposed. It is for the purpose of ensuring the application of 

 heat always to the same point exactly, tliat the shoulder is made 

 ou the tube denoting the depth to which the instrument should 

 invariably be immersed. 



I shall now proceed to enumerate some precautions which 

 are necessary to be taken in the construction and use of the 

 pyrometer, especially when intended for the observation of very 

 high temperatures, such as that of the fusion of iron. I have 

 selected the black-lead earthen ware, after several trials of other 

 materials, on account of the equability of its expansion, its 

 infusibility at high temperatures, and the perfect manner in 

 which it sustains sudden transitions from heat to cold. I have 

 not only repeatedly taken the tubes at once from a white heat 

 into a cold atmosphere, but have plunged them when red hot into 

 cold water without their sustaining any injury whatever. The 

 heat at which this ware is commonly baked is not very high, and 

 I consider it necessary to ensure accuracy, that the tube should 

 have been exposed for some time to a temperature at least equal 

 to the highest which it is intended to measure with the pyro- 

 meter. For this purpose those which I have made use of were 

 placed in an iron-founder's furnace. After this operation they 

 assumed exteriorly a brown appearance, but were as soft and as 

 easily cut as before. When it is required to expose the pyro- 

 meter to a naked fire, it is proper to furnish it with an exterior 

 tube of the same, or some other ware which is easily fitted to 

 the shoulder by grinding ; otherwise the fuel is apt to act as a 

 flux upon the tube, which, becoming vitrified, will crack with 

 sudden transitions of temperature. This precaution is not so 

 necessary when a muffle is used, but is perhaps always ad- 

 visable at very high heats. 



When it is proposed to keep the pyrometer for a long time 

 in a very strong fire, a piece of cloth may be wrapped round the 



