712 SCIENCE PROGRESS 



thiophene and of their derivatives, taking benzene as a model. 

 The more difficult task of dealing with open chain carbon com- 

 pounds has already been undertaken by Barlow and Fope (Trans. 

 1910, p. 2308). It is not possible here to discuss in any detail 

 this most recent work. Starting from the fundamental principles 

 with which the reader is by now familiar, they have deduced 

 assemblages which, they claim, exhibit geometrical peculiarities 

 characteristic respectively of paraffinoid, ethenoid and acetylenic 

 hydrocarbons. Thus certain peculiarities in ethenoid and acety- 

 lenic assemblages are taken as representative of the double and 

 triple bonds in these hydrocarbons. Of course there are no 

 crystallographic data available for these hydrocarbons, as they 

 are all liquid or gaseous ; however, crystals of their halogen 

 derivatives have been measured and the data afforded by these 

 appear to support the theoretical deductions regarding the hydro- 

 carbons. It will be remembered that hydrogen and the halogen 

 atoms have approximately equal atomic spheres, therefore the 

 displacement of one by the other should produce little morpho- 

 logical change. A cubic marshalling is deduced for methane 

 and in accordance with this, it is known that carbon tetrachloride 

 and tetriodide are cubic whilst the tetrabromide is dimorphous, 

 one form being cubic, the other monosymmetric but pseudo- 

 cubic. The structures for higher paraffins are characterised by 

 long chains of carbon atoms, each attached to two other carbon 

 atoms and to two hydrogen atoms, the points of attachment on 

 each carbon atom being arranged tetrahedrally, in accordance 

 with the Le Bel-van t'Hoff theory. The carbon atoms in these 

 straight chains are arranged zig-zag fashion ; consequently the 

 fourth or fifth atom does not approach the first, as Baeyer 

 supposes in his strain theory, which so admirably explains the 

 formation of 7 and S lactones, etc. 



A few words upon the bearing of the results of crystallo- 

 graphic research upon the unique chemical position of carbon 

 may not be out of place here. It may be stated at once that it 

 appears probable that no element except carbon possesses so 

 high a fundamental valency as 4. Chemically, silicon is the 

 nearest relative of carbon but the differences between the two are 

 great, far greater than those between, say, boron and aluminium 

 or nitrogen and phosphorus. Carbon and silicon appear never 

 to be isomorphously interchangeable. It might be expected that 

 carbonates and metasilicates would be isomorphous but they are 



