in the Velocity of Propagation of Light in Water. 279 



Further, the temperatures were compared with an excellent 

 normal thermometer by Greiner, divided into half degrees, be- 

 longing to Professor Hankel, and which was properly corrected. 

 The comparison gave : — 



I. Reading 



on the normal thermometer. II. Simultaneous 

 reading on the thermometer employed in the experiments. 



I. 



II. 



A. 



I. 



II. 



A. 



I. 



II. 



A. 



80-0 



79-5 



5 



40°5 



40°0 



5 



194 



190 



4 



69-8 



69-4 



4 



36-7 



36-2 



5 



16-8 



16-5 



3 



68-5 



68-1 



4 



30-5 



30-2 



3 



100 



100 







66-7 



66-2 



5 



30-4 



29-9 



5 



8-7 



8-5 



2 



65-6 



65 



6 



272 



26-4 



8 



7-2 



70 



2 



64-8 



64-2 



6 



270 



262 



8 



5-9 



5-8 



1 



601 



59-4 



7 



26-8 



26-0 



8 



40 



4-0 







57'3 



56-4 



9 



24-4 



23-9 



5 



36 



3-4 



2 



54-8 



54-2 



6 



23-7 



23-1 



6 



31 



31 







52-6 



521 



5 



23-2 



22-7 



5 



2-8 



2-8 







50 5 



500 



5 



21-7 



21-2 



5 



1-6 



1-6 







48-7 



48-3 



4 



18-5 



18-0 



5 



00 



0-0 







43-8 



43-4 



4 















Hence the following corrections are deduced : — 



80° 65° 60° 55° 50° 35° 30° 25° 20° 15° 10° 5° 0° 



0°-5 0°-6 0°-7 0°-6 0°-5 0°'6 0°'7 0°'6 0°-5 G°-4 0°-2 0° 



(6) The slit. — In order to obtain a straight line of light, an 

 ordinary moveable interference-slit was employed, the two edges 

 of which were about \ a millimetre apart. This was so fastened 

 to a holder on a screen as to be moveable in two directions at 

 right angles to one another. 



(7) The source of light. — The choice of the source of light 

 also demanded much consideration and many experiments. It 

 would have been best, in order to come to a clear conclusion 

 about the alteration of the refraction with the temperature, to 

 have used sunlight, and to have taken Fraunhofer's lines as the 

 fixed data. As, however, I made my experiments in winter, and 

 as considerable time was necessary for each one of them, I was 

 obliged to dispense with this light, because in our climate the 

 sun is very often invisible for weeks together, and I did not 

 choose to be so dependent upon the weather. Fraunhofer's lines 

 in diffused daylight, even when a condensing-apparatus is used, 

 are so feeble that they could not be employed. I contemplated 

 next making use of the absorption spectra of gases, such, for in- 

 stance, as nitrous acid, which Hankel, Grailich, and Weiss had 

 used in their experiments. The coincidence of some of their 



U2 



