Prof. T. Carnelley on the Periodic L 



aw. 



19 



were calculated out early in 1877, and have not since been 

 revised in any way. From 1877 to the present time the fol- 

 lowing melting- and boiling-points have been determined by 

 various observers. On comparison it will be seen that, with 

 the exception of the chloride and bromide of tellurium, these 

 agree very well with the values calculated seven years ago. 





Calculated 

 in 1877. 



Found. 



Authority. 



Cu Br 2 , m.p.... 



774-790 



777 



Carnelley and Williams, 1880. 



CsCl, m.p 



below 959 



904 



1880. 



BeCl 2 , m.p. ... 



8-10-870 



858-890 



Carnelley, 1879. 



BeBr 2 , m.p. ... 



802-820 



858-890 



1879. 



AsCl 3 , m.p. ... 



244-245 



Below 244 



Henry, 1879. 



GaCl 3 , b.p 



496-499 



487-492 



Boisbaudran, 1881. 



CC1 4 , m.p 



253-276 



Solidifies at 



Kegnault, Mem. Acad. Sc. 







248°,andwill 



xxvi. p. 658. This deter- 







therefore 



mination was unknown to 







melt a few 



me when the calculation was 







degrees above 



made. 







this tempera- 









ture. 





TeCl2 |jn pP .... 



309-345 



482 



Carnelley and Williams, 1880. 



472-496 



600 



1880. 



™<> fc 



345-363 



553 



1880. 



530-537 



612 



1880. 



It is thought that the difference between the calculated and 

 experimental values for the compounds of tellurium may pos- 



Te=Cl 2 

 sibly be due to the molecules having the constitution || 



Te=Cl 2 

 instead of Te=Cl 2 . With the object of throwing light on 

 this point, some vapour-density determinations of the halogen 

 compounds of tellurium are in progress. 



III. and IV. Application of Melting- and Boiling-points 

 to the Classification and Determination of the Atomic 

 Weights of the Elements. 



The following example will show how a knowledge of the 

 melting-point or boiling-point of the normal chloride, bromide, 

 or iodide of any element may render considerable service in 

 the determination of its atomic weight, or for ascertaining its 

 position in the general classification. For this purpose we 

 will take Ti. Now supposing Ti were a new element, and 

 that its position and atomic weight were unknown; then, in 

 order to determine these latter, we should have to find its 

 equivalent and the melting-points and boiling-points (where 

 possible) of its halogen compounds, and then to collect together 



C2 



