( 113 ) 



have been lighted, tlie thennostal .-kIJiisIs itself to a tern peral ure that 

 is practically determined by the quantity of mercury |)resent in ihe 

 apparatus. The utility of the apparatus i;^ therefore greatly increased. 

 To give an idea of the capabilities of the apparatus temperatures 

 taken on some of the measuring days are given below. The temperature 

 was taken immediately before the water was allowed to flow into 

 the manometer jacket. 



J 2 h. 15 m. 12 h. 45 m. 2 h. 15 m. 3 h. 5 m. 4 li. 55 m. 

 March 22"^ '11 16.04 18.07 16.07 16.07 16.07 



3h.40m. 8h.55m 4h.20m. 4h.45m. 5h.25m. 6h. 

 March 30^1' '11 19.00 19.00 19.00 19.00 19.00 19.05 



So that the temperature as can be seen remains for hours at a 

 time constant to less than 0.01°. 



How far this constancy of the temperature can be utilised to keep the 

 temperature of the volumenometer constant depends upon the constancy 

 of the room temperature; its heat insulation however can still be 

 improved. If sufficient care is taken to keep the room temperature 

 constant, one is usually successful in keeping the gradual change of 

 the temperature of the volumenometer to within O'^.OS per hour ^), 

 and any single temperature measurement remains certain to 0^.02. 



Physics. — "Further Experimeyds inith Liquid Helium. D. On the 

 Change of the Electrical Resistance of Pure Metals at very 

 loio Temperatures, etc. V. The Disappearance of the resis- 

 tance of mercury" By Prof. H. Kamerlingh Onnks. (Commu- 

 nication N". 122^* from the Physical Laboratory at Leiden). 



As was mentioned in a former Communication (April 1911) I have 

 made a more accurate examination of the resistance of pure mercury 

 at helium temperatures, in which I have once more had the assis- 

 tance of iMessrs. DorsmaxN and Holst. The resistance was now mea- 

 sured with the differential galvanometer by the method of the over- 

 lapping shunts (KoHLRAUscH; and also by the method of the measurement 

 of current strength and of potential difference. By this it was con- 

 firmed that at 3° K. the value of the resistance sinks to below 

 0.0001 times the value of the resistance of solid mercury at 0° C. 

 extrapolated from the melting point. But from the present measure- 

 ments it has also been ascertained that the actual value of the 

 resistance is very much smaller than this upper limit which I was 

 able to ascribe, to it from my former measurements. 



^) For the comparison of mercury columns as in Table II constancy to within 

 10 times this value will be sufficient. 



8 



Proceedings Royal Acad. Amsterdam. Vol. XIV. 



