238 DISCOVERY REPORTS 



it was quite possible for a thermometer only to have been brought into use for a few stations at the 

 end of a commission (normally twenty months), or not to have been used at all. 



Nevertheless, it is perhaps significant that the constant Vq for three of the German thermometers 

 least used is >200° (max. 265^''). This large value for V^ inevitably entails a large and often badly 

 shaped expansion chamber at the top of the bore, and so to a tendency for the mercury to stick in this 



Fig. 3 



Fig- 4 



Fig. 5 



chamber, instead of running back to the main bulb, when the thermometer is reset. It has already 

 been mentioned that at the N.P.L. check in 1953-4 three other German-made thermometers of the 

 earlier age groups (24, 22 and 20 years) failed to meet the requirements for a certificate. All were 

 condemned for the same failure, i.e. the mercury column did not break at the constriction unless the 

 thermometer was tapped. For these instruments the value for V^ was less than half the figure men- 

 tioned above, but was still considerably larger than that of present day thermometers. 



