Constants of Igneous Rock. 3 



5. Porcelain Bulbs. — The forms of air-thermometer bulbs 

 of porcelain which I used are shown on a scale of J in. 

 (figs. 1, 2, 3). They were made by Morlent freres, Paris, of 

 the very refractory porcelain of Bayeux. Fig. 1 is the 

 earliest form. Bulb and stem are in one piece, and put 

 together by the maker before firing. In consequence of the 



Fig. 1. — Xon-inglazed, non-reentrant porcelain air-therinonieter bulb. 



Scale 1/4. 



long capillary stem (0' L centim. in bore) , it is all but im- 

 possible to glaze these bulbs internally, and this is not 

 attempted. Their capacity is somewhat over 300 cub. centim., 

 and the mean thickness of the porcelain walls about 0'27 

 centim. The stem, which is, say, 40 centim. long, has a 

 volume of about 0*012 cub. centim. per centimetre. 



To use such a bulb for calibrating a thermocouple, it would 

 be necessary to have a space at an almost unattainably constant 

 high temperature : for the couple is practically a point, and 

 its thermal sensitiveness is instantaneous, whereas the large 

 bulb necessarily lags behind changes of temperature. With 

 the object of eliminating this source of serious error, I had 

 the bulbs made in the shape of fig. 2, the bottom of which 

 is reentrant, projecting inward to form an axial tube, m n, 



Fig. 2. — Non-inglazed reentrant porcelain air-thermometer bulb. 

 Scale 1/4. 



closed at the centre of figure, m. Into this tube the end of 

 the properly insulated thermocouple is introduced, with its 

 junction at m. The insulator, § 10, incidentally subserves 

 the purpose of a plug for the ouen end of the tube ran, 



B2 



