104 Proceedings. 



the different thermometers used as intermediate standards in 

 an experiment carried on exactly as in an actual test : — 



Kew, B reading, 101*4. 



Clinical thermometer, found at Kew to be 



correct ... ... ... ... ... 10 r 5 



(1) Clinical thermometer, found correct at 



Berlin ... ... ... ... ... 101*4 



(2) Clinical thermometer, taking account of 



corrections supplied by the Technische 

 Reichs Anstalt ... ... ... ... ioi - 5 



" The thermometers agree, therefore, to the limits of 

 accuracy which can be attained. 



" The College has tested about 300 thermometers in the 

 last two years, and, as a general rule, it may be said that 

 the corrections have been small ; but it has occasionally 

 happened that thermometers were found to be wrong by 

 o°'4 and o°'5, which shows that no thermometer can be trusted 

 to be sufficiently accurate which has not been compared 

 with some standard. 



" The result of testing also has shown that the more 

 expensive kinds of thermometers have errors as great as the 

 cheaper ones. The advantage which the more expensive 

 thermometers claim, of being more rapid in their indications 

 is often illusory. When a clinical thermometer is plunged 

 into water of ioo°F. it takes up the temperature almost 

 immediately, and as to the time required when the ther- 

 mometer is placed into the mouth of the patient it is the bad 

 conductivity of the tissues of the mucous membrane which 

 causes the lag in the rise. The skin or tongue is, under 

 ordinary conditions, below the blood temperature, and is 

 further chilled by the introduction of the cold thermometer- 

 By reducing the mass of the thermometer the first chilling 

 effect may be diminished and the instrument would indicate 

 more quickly the correct blood temperature. But the gain 

 is not as great as is generally supposed." 



