314 POLARITY and the pentagonal dodecahedron. 1 Now when a ray of plane polarized light is passed through one of these crystals the plane of polarization is rotated, the amount of rotation being proportional to the length of the path in FIG. 9. Sodium Chlorate, a, riglit-handed ; b, left-handed. the crystal. The crystals having the form a rotate to the right, those having the form b to the left. They are therefore optically as well as crystallographically enantio- morph. But a solution of sodium chlorate is without action on the plane of polarization, even if the solution be made by dissolving only right-handed or only left-handed crystals, and if a crystal be fused the fused mass is optically inactive, so that it would seem that the optical activity depends on the arrangement of the molecules in the crystal and not on any enantiomorphism in the mole cules. The enantiomorphism of quartz crystals is indi cated by the presence of faces of a tetartohedral form (vol. xvi. p. 389). The two kinds of crystals rotate the plane of polarization equally, but in opposite senses, when a plane polarized ray is passed through a section cut at right angles to the axis of the crystal. Here also the optical activity ceases when crystalline structure is de stroyed by fusion or solution. Eight-handed and left-handed tartaric acids crystallize in enantiomorph forms (fig. 10). Their solutions are optic- FIG. 10. Tartaric Acid, a, riglit-lianded ; 6, left-handed. ally active, the amount of the rotation for the same strength of solution and the same length of path in it being the same in both acids, but the sense of the rotation s right-handed in the one and left-handed in the other. .s clear that here we have to do with enantiomorph molecules. In ordinary physical properties such as den sity, solubility, refracting power in short, in everything not involving right- or left-handednesst^ acids are iden tical. When mixed in equal proportions they unite and form racemic acid which is optically inactive, and from racemic acid we can by various means recover unchanged the right and left-handed tartaric acids. We now know a considerable number of cases where, as in that of the two tartaric acids, both enantiomorphs have been discovered, and many where only one has as yet been found. It is natural that we should ask what peculiarity of titution can give a molecule this helicoidal asymmetry? A very ingenious answer to this question was given simul taneously and independently by the French chemist 1 and the Dutch chemist Van t Hoff. We shall
- tetai *ohedral because the tetra-
" 01 " 1 d decahedr011 MOD * to two different classes Fig. 11. give a short statement of the essential points of this in teresting theory. All the known substances which are optically active in solution are compounds of carbon, and may be regarded as derived from marsh gas, a compound of one atom of carbon and four of hydrogen, by the replacement of hydro gen by other elements or compound radicals. Now we do not know how the atoms of hydrogen are actually arranged relatively to each other and to the atom of carbon in the molecule of marsh gas, but, if we may make a sup position on the subject, the most simple is to imagine the four hydrogen atoms at the apices of a regular tetrahedron in the centre of which is the carbon atom as in the diagrams (fig. 11), where C represents the posi tion of the carbon atom and a, ft, y, 8 that of the four atoms of hydrogen. If these hydrogen atoms are replaced by atoms of other elements or by compound radicals we should expect a change of form of the tetrahedron. If two or more of the atoms or radicals united to the carbon atom are similar there is only one way of arranging them, but if they are all different there are two ways in which they may be arranged, as indicated in the figures. It will be seen that these two arrangements are enantiomorph. In the figures the tetrahedron is represented as regular, but if the dis tance from C depends on the nature of the atom, the tetrahedron, when a, ft, y, and 8 are all different, will not be symmetrical, but its two forms will be enantiomorph. A carbon atom combined with four different atoms or com pound radicals may therefore be called an asymmetric carbon atom. Now all substances of ascertained constitution, the solutions of which are optically active, contain an asym metric carbon atom, and their molecules should therefore, on the above hypothesis, have helicoidal asymmetry. The converse is not generally true. Many substances contain an asymmetric carbon atom but are optically inactive. It is easy to reconcile this with the theory; indeed, a little consideration will show that it is a necessary consequence of it. Let us suppose that we have the symmetrical combina tion of C with a, a, ft, y and that we treat the substance in such a way that one a is replaced by 8. The new arrangement is asymmetrical, and will be right or left as the one or the other a is replaced. But the chances for the two are equal, and therefore, as the number of mole cules in any quantity we can deal with is very great, the ratio of the number of right-handed molecules in the new substance to the number of left-handed ones will be sensibly that of unity. It is therefore evident that by ordinary chemical processes we cannot expect to produce optically active from optically inactive substances ; all that we can get is an inactive mixture of equal quantities of the two oppositely active substances. As these two substances have identical properties in every respect where right- or left-handedness is not in volved, the problem of separating them is a difficult one. We may note three distinct ways in which the separation can be effected. (1) By crystallization. For example, the right and left double tartrates of soda and ammonia crystallize in enan tiomorph forms (fig. 12) and are less soluble in water than the double racemate formed by their union. If therefore racemic acid (the optically inactive compound of equal
quantities of right and left tartaric acids) is half neutral-