or 



FIG. 1. 



1905.] On the Platinum and Normal Scales of Temperature. 533 



This difference is of the same order as that found by other investi- 

 gators. 



Comparison of the Thermometers at the Temperature of Solid Carbonic 



Acid and Alcohol A mixture of solid carbonic acid and ether 



rectified methylated spirit has frequently 



been employed as a means of maintaining a 



constant temperature. Olszewski states that 



the temperature remains constant so long as 



the mixture has the consistency of butter, 



but that as soon as ether appears on the 



surface, the temperature begins to rise. He 



gives the temperature as - 78-2 C. 



Our experiments lead to the conclusion 



that to obtain the best results the solid car- 

 bonic acid should be free from lumps, and 



should be added to the spirit for there is 



no advantage in using ether till the mix- 

 ture has the consistency of a fairly thin 



paste, and can be stirred with a glass rod. 



The stirring should be repeated at short 

 intervals. 



At the outset of the experiments we were 

 met with the difficulty of exposing the plati- 

 num coil to the action of the mixture, and at 

 the same time protecting it from strain 

 during the process of stirring. When the 

 bare wire was exposed to the mixture, it 

 was found that the values of E before and 

 after an experiment were in every case 

 different, and when the coil was enclosed 

 in a glass tube with holes in the side of it, 

 the circulation of the mixture was obstructed, 

 and the results were discordant. The diffi- 

 culty was at last overcome by surrounding 

 the coil with a tube of wire gauze, as is 

 shown in fig. 1 . The glass and gauze tubes are 

 shown in section, the thermometer in plan. 



a, Glass tube passing inside gauze tube 

 and connected to vulcanite head b of thermo- 

 meter, c, Nickel gauze tube made rigid by 

 binding it with wire on to the glass tube a, 

 and on to thin vulcanite rings d d, which also 

 serve to prevent the coil and leads from 

 touching the gauze, e, Brass ring and wire 

 grating soldered to tube. 

 VOL. LXXIV. 



2 R 



