ON THE REFKACTIVE INDEX OF GASEOUS FLUORINE. 



325 



experiments by which the index has, we believe, been measured with some approach 

 to accuracy. 



Fig. 2 shows, in a diagrammatic form, the arrangement of the apparatus. A is 

 the copper electrolytic tube, from the right side of which issues the fluorine. In 

 order to prevent the escape of the vapour of hydrofluoric acid the exit tubes were 



Fig. 2. 



carried upwards, and surrounded by vessels, B 1( B 2 , containing alcohol, cooled to 

 78 C. As a further precaution, the gas next passed through a length of 73 centims. 

 of platinum tube immersed in a solution at the same temperature (C), and in many of 

 the experiments a guard tube, filled with NaF, as recommended by M. MOISSAX, was 

 added at the point F. The fact that the presence or absence of this salt did not 

 appear to affect the refractivity of the mixture of gases is evidence that the other 

 precautions were effectual. 



It was probable that some part of previously observed discrepancies arose from the 

 presence of ozone produced when the oxygen was liberated ; and, in order to destroy 

 this, the gases next passed through a spiral of platinum tube, 49 centims. in length, 

 heated to from 250 to 300 C. (E). The bends, D and F, were immersed in iced 

 water to prevent conduction of heat from the spirals to the condenser on one side and 

 the refractometer tube on the other. 



The refractometer tube, H, was of platinum-iridium. 4578 centims. long and 

 0'65 centim. in diameter. Its volume, with the leads, was 15 '01 cub. centims. Each 

 end was furnished with a collar of platinum 0'25 centim. broad. The plates with 

 which the ends were closed were of fluor spar, and were secured to the tube by a 

 shoulder of shellac, melted round the outside of the circumference of the collar, so as 

 to be as far as possible from the fluorine within. Thus, after leaving the electrolytic 

 tube, the fluorine was never in contact with anything but platinum and fluor spar. 



The plan adopted for measuring the volume of fluorine present was to allow it to 



