THE SPACE RELATIONS OF ATOMS. 139 



we need them as large as can be conveniently handled, and 

 we may then use tetrahedra having corner-caps marked with 

 the formulae of the particular radicals attached to the carbon 

 atom. Models of this kind have been described 1 under the 

 name of "solid formulae," and some of the plates illustrating 

 them are reproduced here to represent the configuration of 

 certain molecules. 



To illustrate the fact that unlike groups are at unequal 

 distances from the central carbon atom, models with unequal 

 sides or limbs would be required, but for ordinary purposes 

 the symmetrical models suffice, and are in fact always used. 



With the aid of the models we may now follow Van't 

 Hoff in his development of the formulae of molecules con- 

 taining!- connected carbon atoms. 



Assuming that the radicals attached to an asymmetric 

 carbon atom have the tetrahedral configuration, we must 

 represent a compound with two connected carbon atoms, 

 CR'R"R'" CR iv R v R vi , by two tetrahedra, each carbon atom 

 occupying at once the centre of one tetrahedron and the 

 corner of the other. In the models the tetrahedra are, for 

 convenience, shown simply corner to corner (see the solid 

 formulae for tartaric acid, figs. 6 and 7, Plates I. and II.). Any 

 other configuration obtained by the rotation of either tetra- 

 hedron about the axis C — C would be equally in accord with 

 the fundamental conception ; but we shall not therefore 

 encounter innumerable isomers, for there will be only one 

 of these configurations realised at any given temperature, 

 viz., the one favoured by the mutual action of the groups 

 R'R'R " and R iv R v R vi at that temperature (" favoured con- 

 figuration "). As change of temperature affects different 

 groups differently, it is evident that if the change be carried 

 far enough another configuration may become the favoured 

 one, and there is evidence that in some cases this happens. 



We may for convenience assumetheconfiguration favoured 

 at ordinary temperatures to correspond to the formula, 



1 Eiloart : Am. Chem. Journ., xiii., 559; also A Guide to Stereochem- 

 istry. London: Clarence Eiloart, 1893. These models are now made by 

 Messrs. Baird & Tatlock, according to samples approved by the writer. 

 Models of the Friedlander type are supplied by the same makers. 



