40 SECTIONAL ADDRESSES. 



being ammonia. The truth of this hypothesis of intermediate co-ordina- 

 tion with the water is strongly supported by the fact that it explains the 

 unusual production of hypochlorous acid from chlorine attached to 

 trivalent nitrogen. 



But carbon tetrachloride cannot react in either of these ways. It 

 has a complete octet, and cannot increase it, and the octet is fully shared, 

 so that it cannot act as a donor. It therefore does not react at all. The 

 remarkable inactivity of carbon tetrachloride has long been regarded as 

 an unexplained anomaly, but we can now see that it is a necessary conse- 

 quence of the theory of co-ordination. If we want to find a similarly 

 inactive halide of an element in a later period, where a valency group of 

 12 is possible, we must obviously choose one in which this group of 12 is 

 fully shared and also is incapable of further expansion. Examples of 

 this are the hexafluorides of sulphur and selenium, whose inactivity is 

 as remarkable as that of their carbon analogue. Tellurium hexafluoride 

 on the other hand is hydrolysed by water, since its valency group of 12 

 can expand to 16, and the tellurium can therefore (like silicon in the 

 tetrahalide) act as an acceptor. 



Now the carbon atoms in an ordinary saturated organic compound all 

 resemble that in the tetrahalide in having fully shared valency groups of 

 the maximum size. They are therefore incapable of the most obvious 

 form of reactivity, which begins by co-ordination with a reagent molecule : 

 if they are to react at all, it must be through some other atom in the 

 molecule. It is a significant fact that one of the most elementary rules 

 of organic chemistry is that a carbon atom united only to other carbon 

 atoms or to hydrogen or the halogens is very slow to react, but that the 

 introduction of a single oxygen atom into the molecule facilitates reaction. 

 The comparison of the paraffins with the ethers or alcohols, of the ethers 

 with the esters and the esters with the acid anhydrides, or of the alkyl 

 halides with the acyl halides, illustrates the effect which an oxygen atom 

 may have on the stability of a molecule. It seems natural to relate this 

 effect to the strong donor properties which oxygen exhibits, and to suppose, 

 for example, that the rapid hydrolysis of an acyl halide is due to the 

 formation through the oxygen of a compound 



R-C^ >0, 



in which the hydrogen of the water is brought into close proximity with 

 the chlorine, while the relative inactivity of an alkyl halide is the result 

 of its inability to form such a compound. 



I make these suggestions (which might easily be extended) because 

 it seems to me that in the intensive modern study of the influence of 

 structure on the reactivity of organic compounds this side of the question 

 has been too much neglected. Great attention has been devoted to the 

 consideration of the effect of other atoms in the molecule on the strength 

 of a particular linkage. A new mechanism and a new terminology — or 

 perhaps more than one — have been invented to account for the results. 

 This mechanism is described in terms of physical concepts, and although 

 it appears to me that the properties which are assigned to these concepts 

 need considerable modification before they can be accepted by the 



