Kopp on the Boiling-points of Bromine andChlorineCompounds. 191 



place of Cl„ boil also by .a? x 32 degrees higher than the latter. 

 From this, he reasoned, a criterion was obtained as to how many 

 equivalents of chlorine or bromine were contained in correspond- 

 ing bromine and chlorine compounds. As bromide and chloride 

 of silicon differ by 3 x 32 degrees in their boiling-points, Kopp 

 inferred that their formulae were respectively Si Br^ and Si CF. 



The observations on boiling-points made at that time in cor- 

 responding chlorine and bromine compounds justified the assump- 

 tion of that relation, but Kopp thinks now that that remark in 

 its original generality is no longer true. In the last few years 

 many new compounds have been discovered whose boiling-points 

 do not agree with this rule, and the boiling-points of many 

 bodies known at that time have since been determined in a more 

 accurate manner. Kopp gives a list of such boiling-points, from 

 which it appears, that, although in many cases of corresponding 

 compounds containing bromine or chlorine the difference of boil- 

 ing-points is a constant one, yet this is not universally the case, 

 and it cannot be said that the difference of boiling-points is 

 always the same for the same difference of formulae. Hofmann 

 has lately announced that the boiling-point of bromide of tita- 

 nium differs from that of chloride of titanium by 95°, =3 x 31°'7, 

 and reasons from this that the formula of bromide of titanium is 

 Ti Br3, and of the chloride, Ti CF. 



Kopp says, that although the difference in the boiling-points 

 of chloride and bromide of silicon affords a support to the idea 

 that their formulae are Si CP and Si Br^, yet it by no means 

 proves it. From a purely chemical point of view we cannot 

 decide which of the formulae. Si CP, Si CP, Si CI, for chloride of 

 silicon is correct. Relations are found to exist between the 

 chemical composition and certain physical properties, some of 

 which may be urged for the formula Si CP, and some for the 

 formula Si CP or Si^CP. 



In the case of titanium, the analogy of several of its com- 

 pounds with tin, the atomic weight of which must be considered 

 as fixed, and especially the isomorphism of rutile, TiO^, and of 

 native tinstone, SnO'^, as well as of Brookite, and of the crystal- 

 lized oxide of tin obtained by M. Daubree, appear to establish its 

 equivalent, and the formula of many of its compounds. That 

 the bromides and chlorides of titanium and silicon exhibit the 

 differences in boiling-points is undoubtedly interesting ; but it 

 does not appear allowable to change the formulae Ti CP, Ti Br^, 

 &c., which are estabhshed on sure foundations, to Ti CP, Ti Br^, 

 only on account of a difference in the boiling-points of two com- 

 pounds which has no connexion with any general rule. Relations 

 between the physical qualities and chemical composition may be 

 consulted for the determination of atomic weights and of for- 



