* These are the names by which the three substances are best known to English chemists; but the author gives Hesse's names, viz., conchinamine, dihomo-cinchonine, and diconchinine respectively. For the last, apodiquinicine was suggested by Wright, on the ground of its greater resemblance to quinicine, of which it may be regarded as a first anhydride, thus: 2 (C2H24N2O2) — H2O = C40H46N203.-[D. C. R.]

According to the foregoing list, the number of natural active substances known amounts to about 140, of which 65 are leftrotating, 60 right-rotating, and 15 both right and left-rotating.

Of active derivatives, counting all hitherto examined salts of the alkaloids and vegetable acids, there are at least as many known, thus bringing the total number of optically active carbon-compounds up to close on 300, and no doubt many other substances hitherto unexamined possess the power of rotating the plane of polarization.

Substances which display the rotatory power when in a state of solution, and are crystallizable, are not found to exhibit optical activity in the crystalline state, as when a polarized ray is passed through plates cut from them. This is the case with canesugar, tartaric acid, asparagin, camphors, etc. (see Biot,1 Descloizeaux2). Now the phenomenon of circular polarization is only observable in single-refracting or in uniaxial double-refracting crystals, and in the latter only in the direction of the optic axis. But the substances just referred to are all biaxial, and thus in no direction single-refracting; consequently, circular double-refraction 3 could not in any case be observed in them, as it would be overpowered by the more marked phenomenon of ordinary doublerefraction. Whether they are really inactive in the crystalline state is undecided.

1 Biot: Mém. de l'Acad. 13, 39.

2 Descloizeaux: Pogg. Ann. 141, 300.

3 The expression refers to Fresnel's theory of circular polarization, in which the two rays are supposed to vibrate in opposite circular paths. See § 12.—[D. C. R.]

But when these substances are brought into the amorphous solid form their optical activity is retained a fact first observed by Biot with cast plates of sugar and tartaric acid.1

§ 9. As a third and distinct class are regarded those substances which are known to exhibit rotatory power both in the crystalline state and in solution. At present only two such substances are known, viz., strychnine sulphate crystallizing with water in quadrate octahedra,2 and regular amylamine-alum.3

B. Nature of Rotatory Power.

§ 10. The fact that substances in the first of the above classes manifest rotatory power only in the crystalline state and lose it directly they are brought into solution, is proof that the rotation is dependent on crystalline structure-that is, upon a particular arrangement in the groups of molecules (forming the crystal). Dissolution or fusion breaks up this arrangement, and the optical power is consequently lost. In this case then the phenomenon is purely physical.

The second class of substances, on the contrary, exhibit rotatory power in the liquid state. Now there is every reason to believe of matter in this form that the smallest quantities, capable of independent motion as units consist, not of individual molecules, but of groups, and it may therefore be conjectured that the solution of a solid in a liquid does not entail a complete separation of the molecules from each other, but that they still exist in composite groups.4 Whenever, therefore, we find liquids exhibiting rotatory power, we might assume that, as in the case of crystals, the cause lies in the mode in which the molecules group themselves. Thus again the phenomenon would be purely physical.

But for this supposition to be correct the rotatory properties of active substances should vanish when these groupings are really broken up-that is, when the substances are brought into the norma : gaseous state. This important point was first investigated by Biot,5 in 1817. He filled a tin tube, fitted at both

1 Biot: Mém. de l'Acad. 13, 126. Ann. Chim. Phys. [3] 10, 175; 28, 351.

2 Descloizeaux: Pogg. Ann. 102, 474.

3 Le Bel: Ber. d. deutsch. chem. Gesell. 5, 391.

4 See, on this point, Naumann: Ueber Molecülverbindungen nach festen Verhältnissen.

Heidelberg, 1872, pp. 37-49.

5 Biot: Mém de l'Acad. 2, 114.

ends with glass plates, and 30 metres in length, with vapour of oil of turpentine, which he found had still the property of producing a certain amount of deviation in a ray of polarized light. Unluckily, before the observations were completed, the vapour accidentally caught fire, and the apparatus was destroyed. The experiment was next tried by D. Gernez,1 in 1864, who, with the aid of instruments of a superior kind, determined the rotatory powers of various active substances at rising temperatures, and eventually in the gaseous state. The substances thus examined were orange-peel oil (+), bitter orange oil (+), turpentine oil (-), and camphor (+). In each the specific rotation [a], that is, the angle of rotation calculated for equal densities, 1, and equal lengths of layer 1 decim. diminished as the temperature increased; and when the same substances were tested in the gaseous state they gave a specific rotation merely reduced in proportion to the temperature to which they had been exposed. The table appended shows the results obtained with oil of turpentine and camphor :

=

=

It will be seen that the observed densities of the

1 Gernez: Ann. Scient. de l'école norm. sup. 1, 1.

vapours used

in the experiment agreed very nearly with their calculated densities. The polarized ray must therefore have been influenced almost entirely by individual molecules, not by groups. The rotatory

power [a] was, however, manifested to its full extent, and the conclusion is that here optical activity must be a property resident in the molecule itself, and dependent on its atomic structure. The phenomenon is thus seen to be really chemical.

The optical activity of crystals and that of liquids are, therefore, wholly distinct phenomena, and to the latter Biot has given the name of moleculur rotation, indicating that it is a property resident in the individual molecule.

§ 11. Magnetic Rotation.-A rotatory movement of the plane of vibration of a ray of polarized light can be produced in all transparent isotropic bodies, solid or liquid (as glass, water, &c.), by placing them between the poles of a magnet, or within the helix of an induction-coil. This so-called magnetic rotation differs altogether from rotation as seen in naturally active substances. It lasts so long only as the electric influence is continued; it varies in degree with the intensity of the latter; and it takes a right- or left-handed direction, irrespective of the medium, according to the position of the poles of the magnet or the direction of the electric current. There is also this further characteristic difference between the two. Let a polarized ray be transmitted through a naturally active substance, which, for the sake of example, we will say is right-rotating. Then the deviation of the plane of polarization will always be such that, to follow the movement of the ray, the instrument must be turned towards the right of the observer-that is to say, the direction of rotation, with reference to that of propagation of the ray, is invariable. If after passing through the refractive medium the ray is returned into it by reflection, and the analyzer brought round to the same side as the entering ray, it will be found that rotation is annulled. The rotation dependent on magnetism is of a quite different character. The ray transmitted, let us say, from south to north pole, in the direction of the observer, will appear deflected towards the right hand, and, transmitted from the opposite end of the tube, towards the left. If the ray transmitted from south to north pole be reflected back, it will appear farther deflected to the left, so that an analyzing Nicol placed to receive it must be rotated to the left through an angle

equal to double the previous angle of rotation. If again brought back to the north end by a second reflection, this third transmission of the ray through the refractive medium will carry the analyzer placed at the north end through an angle equal to three times the original angle, of rotation, and so on. The same thing happens when circular polarization is induced by an electric current, the rotation always taking the direction of the induction-current from the observer's stand-point.

Magnetic rotation, not being a property of the chemical molecule, need not be further discussed here.

§ 12. The optical theory of circular polarization in quartz is due to Fresnel. According to him there occurs in quartz, in a direction parallel to the main axis of the crystal, a peculiar kind of double refraction, whereby a linear polarized ray on entering is decomposed into two rays, each of which pursues a helical course, the one turning to the right, the other to the left. On emerging, the two circular-polarized rays unite into a single linear-polarized ray again, but if the velocities with which they have traversed the refractive medium have been unequal, the plane of vibration of the emergent ray will have a different direction to what it had originally. It will follow the hands of a watch-that is to say, it will have rotated to the right-if the circular-polarized ray turning in that direction has had the superior velocity, and vice versâ. The existence of these divided rays in rock-crystal was experimentally established by Fresnel, and subsequently by Stefan, and also by Dove, who found that in coloured quartz (amethyst) they were unequally absorbed. The theory of circular polarization has since been treated mathematically by Clebsch, Eisenlohr," Briot," v. Lang and others.

4

Regarding the structure requisite in a crystalline medium to produce rotation of the plane of polarization, a theory has been proposed of an unequal condensation in certain directions of the

1 Fresnel: Ann. Chim. Phys. [1] 28, 147; Wüllner's Lehrbuch der Phys. 3 Aufl., 2, 589.

2 Stefan: Pogg. Ann. 124, 623.

3 Dove: Pogg. Ann. 110, 284.

4 Clebsch: Crelle's Journ. f. Math. 57, 319.

5 Eisenlohr: Pogg. Ann. 109, 241.

6 Briot: Comptes Rend. 50, 141.

7 v. Lang: Pogg. Ann. 119, 74. Erg. Bd. 8, 608.