especially as defined by Helmholtz. 29 



may have been attached to the existence of alkylic compounds 

 of particular types : thus lead, judging from its general 

 chemical behaviour, would appear to be a dyad ; yet the 

 existence of the tetrethide Pb(C 2 H 5 ) 4 is commonly held to he 

 a proof that it can function as a tetrad. But the properties 

 of lead are such that I am tempted to suggest that it is one 

 of the metals in which the " charges " have but a small 

 degree of freedom ; and it is conceivable that the tetrethide 

 is actually a compound of dyad lead, each charge serving to 

 bind two ethyl groups, thus : — 



H 5 2 ~ B L C 2 H 5 



The same may be true of tin, although in this case the fact 

 that we are dealing with an element of the carbon-silicon 

 family tends to favour the conclusion that it may be a tetrad. 

 Also too much importance must not be attached to the 

 existence of stable volatile chlorine compounds : thus tellu- 

 rium tetrachloride may well be a compound of dyad tellurium, 

 thus : — 



Iron, and the other members of the family which boron 

 heads, in like manner, I feel convinced, are triads even in 

 their ic compounds : recent vapour-density determinations all 

 support this conclusion. 



It is even conceivable that chlorine may form closed-chain 

 compounds, and that a tetrachloride may exist, such as is 

 represented by the formula : — 



Te 



^ CI C I 



I think it is especially noteworthy that so many well 

 characterized and comparatively stable double chlorides exist 

 formed by the union of chlorides, of which one at least is 

 per se very unstable ; the tin-sulphur chloride SnCl 4 . 2S01 4 , 

 and the remarkable series of aurous compounds recently de- 

 scribed by Lepetit (Ann. Chim. Phys. 1887, p. 11) may be 

 cited as examples. 



If my contention in this and previous papers be correct, 

 that residual affinity thus plays a far more important part 

 than has hitherto been supposed, and that it must be taken 

 into account in all discussions on valency, it follows of 

 necessity that our views regarding the constitution of the 



