For although in the latter cases no precise conclusions can be drawn regarding the relative arrangements of the atoms in the molecules, yet the study of specific rotatory power may help to throw light on such general questions as the action of solvents, the distinction between chemical and physical change, and so forth.

146. Most of the known compounds which possess the power of rotating the plane of polarisation of a ray of light contain carbon: van't Hoff, following in the steps of Le Bel, has endeavoured to trace a precise connection between the molecular structure of these compounds and their rotatory power. The molecule of every optically active compound, according to this hypothesis, contains one or more asymmetric atoms of carbon. The conception represented by this expression is essentially crystallographic.

An asymmetric atom of carbon directly acts on, and is acted on by, four, chemically different, monovalent atoms, or groups of atoms, within the molecule; thus

To render the conception more definite, the carbon atom is supposed to be situated at the centre of a tetrahedron and each monovalent radicle at one of the summits. Inasmuch as these radicles are chemically different, the action and reaction between each and the carbon atom is supposed to be different; hence each radicle is situated at a different distance from the central atom of carbon; and hence the tetrahedron is irregular, has no planes of symmetry, and is capable of existing in non-superposable (enantiomorphous) forms. Optical activity is regarded as always accompanied by such an arrangement as this of atoms in molecules, which molecules have thus the properties of partially developed non-superposable crystals'.

1 For a more detailed account of van't Hoff's hypothesis, see Dict. 3rd Supplt. 1214-1217; or Armstrong and Groves, loc. cit. 986—-993.

Representing the asymmetric carbon atom (or atoms) in a molecule by an italicised C, we have such formulæ as these;

Lactic acid, H,C-C-CO2H; Malic acid, HO,C — С — С — CO2H;

Mannitol, HOH,C-C-C-C-C-CH2OH.

HHHH

Any compound, the molecule of which contains a single asymmetric carbon atom, may exist in two optically different modifications. When more than one asymmetric atom is present, the number of possible modifications is conditioned by whether the formula of the compound is symmetrical or unsymmetrical: a symmetrical formula, in the nomenclature of this hypothesis, is one in which the radicles directly connected with the different asymmetric carbon atoms are the same; if these radicles, or some of them, are different, the formula is said to be unsymmetrical. Thus the formula

A carbon compound of unsymmetrical formula, containing n asymmetric atoms of carbon, may exist in (2)" modifications; if the compound is represented by a symmetrical formula, it

1 See illustrations of these equations in Dict. 3rd Supplt. 1138-39; or Armstrong and Groves, loc. cit. 987—990.

он он

HO,C−C−C−CO,H,

H H

which is symmetrical, and contains two asymmetric atoms of carbon. The van't Hoff hypothesis asserts the possible existence of three optically different tartaric acids, which may be represented by the symbols

(1) +A+A (=2A), (2) +A-A (=0), (3) -A- A=(2[− A]) ; where each part of the symmetrical molecule,

is represented by the sign A; the symbol + or - being prefixed according as this group is regarded as being dextro- or lævorotatory. One of the hypothetically possible tartaric acids ought therefore to be dextro- and one lævorotatory; the third ought to be inactive, because of the balance of dextroand lævorotatory powers within the molecule. Now three such acids exist1.

When a substance is regarded as being inactive because of the balance of opposite rotatory powers by the structure of the molecule itself, it is said to be inactive by internal compensation. But a substance may also be inactive by external compensation. Thus erythritol is probably represented by the symmetrical formula containing a pair of asymmetric carbon atoms

он он

HOH,C-C-C-CH2OH.

H H

This substance is said to be optically inactive because of

1 Racemic acid is probably a molecular compound of the dextro- and lævorotatory acids.

internal compensation. But by distillation with formic acid erythritol yields the glycol

он

HOH,C-C-C-CH.

H H

This glycol, although containing one asymmetric carbon atom, is inactive. But the hypothesis asserts that two optically different modifications of this glycol may exist: if we suppose that both are actually produced, and produced in equal molecular quantities, we have an explanation of the non-activity of the glycol; it is inactive by external compensation.

This explanation appears to me to require that we regard the glycol in question as owing its optical inactivity to the existence of molecular groups, each composed of two chemically identical but optically dissimilar molecules: it could scarcely be that a mixture of the dextro- and lævorotatory molecules would always give an inactive substance which could not be separated into its optically different parts.

2

Many carbon compounds said to contain asymmetric carbon atoms are optically inactive. This does not seem to me to be a material objection to the van't Hoff hypothesis. Because when we speak of an asymmetric carbon atom, we mean more than is symbolised by the expression CRRRR; we mean not only that the carbon atom is in direct union with four chemically unlike monovalent radicles, but also that the structure of this group, CR,R,R,R,, is of a special kind, and of a kind which, by the very terms of the hypothesis, cannot be distinguished from other possible structures by any chemical methods we at present possess. Whether a structure CR ̧R,R ̧R shall or shall not be associated with optical activity, depends (by hypothesis) on the relative arrangement in space of the five parts which compose the structure; but this again depends on the equality or non-equality of the mutual actions between C and R1, C and R,, C and R,, and C and R.; and this, finally, depends on the chemical nature of

2

R1, R., R, and R., using the expression chemical nature in its widest acceptation'.

We may object to the extremely crystallographic character of the van't Hoff hypothesis, and to the length to which it pushes the vague notion of 'bonds' or 'units of affinity', for it seems to regard these as capable of definite arrangement in space; but I think the hypothesis is worthy of careful consideration, because it draws attention to the inadequacy of the prevalent conceptions regarding isomerism and molecular structure, and because it bases the explanation it has to give of the connection between such structure and the properties of compounds on essentially dynamical conceptions.

But can it be shewn that optical activity is undoubtedly connected with the structure of the molecules of compounds, rather than with that of groups of molecules?

It is said that the specific rotatory power of terpene (CH) from French turpentine oil, and of camphor (CHO), is independent of temperature, and is indeed the same for these substances in the gaseous as in the liquid state: if this is confirmed by further observations we shall have in these compounds instances of undoubted connection between molecular structure and rotatory power. But the great majority of optically active carbon compounds do not certainly exhibit such a connection, or exhibit it only in an indirect manner. Thus it cannot be asserted that the molecule of lactic acid (CH, - CHOH – CO2H) is optically active, because no one has yet obtained molecules of this acid unmixed with molecules of other substances. Again, we do not know the true molecular weights of tartaric or malic acids: even in the cases of amylic alcohol, and valeric acid, which are gasifiable compounds, we cannot assert that the molecule CH ̧. CH2OH, or the molecule CH,. CO2H, is optically active, because the optical activity of the compounds in question belongs to them in

1 For a tabular statement of compounds, shewing in what cases the presence of the structure symbolised by CR2RRR is accompanied by optical activity, see Landolt, loc. cit. 27-9; see also table, loc. cit. p. 36. Just. Annalen 220. 146 has shewn that a number of derivatives of active amylic alcohol are themselves optically active.

2 See Dict. 3rd Supplt. 1209.