along in the direction of the arrow will trace out a plane (see figure 21) between 6 and b'. If this polarised beam be made to travel now through a solution of sugar, the net result is that the plane so traced out is twisted or rotated; the two spokes, as in bb', do not trace out a plane, but we must consider that they rotate as they travel along, as though guided by a spiral or screw thread cut on the axis, so that after a certain distance the vibrations take place as in b"; later in b'', and so on. This effect on polarised light is due to the molecules in solution, and the amount of rotation will depend on the strength of the solution, and on the length of the column of the solution through which the light passes; or in the case of a quartz plate on its thickness.

If a plate of quartz be interposed between two nicols, the light will not be extinguished in any position of the prisms, but will pass through various colours as rotation is continued. The rotation produced for different kinds of light being different, white light is split into its various constituent colours; and the angle of rotation that causes each colour to disappear is constant for a given thickness of quartz plate, being least for the red and greatest for the violet. These facts are made use of in the construction of polarimeters. Polarimeters are instruments for determining the strength of solutions of sugar, albumin, &c., by the direction and amount of rotation they produce on the plane of polarised light. They are often called saccharimeters, as they are specially useful in the estimation of sugar.

Soleil's Saccharimeter. This instrument (see fig. 22) consists of a nicol's prism d, called the polariser; the polarised ray passes next

B

D

FIG. 22.-Soleil's Saccharimeter.

through a quartz plate (b) 3·75 m.m. thick, one half (d in fig. 23) of which is made of dextrorotatory, the other half (g in fig. 23) of lævorotatory quartz.

The light then passes through the tube containing the solution placed in the position of the dotted line in fig. 22, then through a quartz plate cut perpendicularly to its axis (g in fig. 23), then through an arrangement called a compensator (r in fig. 23), then through a second nico! (a) called the analyser, and lastly through a telescope (L in fig. 23).

The compensator consists of two quartz prisms (RR, fig. 23) cut perpendicularly to the axis, but of contrary rotation to the plate just in front of them. These are wedge-shaped and slide over one another, the sharp end of one being over the blunt end of the other. By a screw the wedges may be moved from one another, and this diminishes the thickness of quartz interposed; if moved towards one another the amount of quartz interposed is inc reased.

The effect of the quartz plate (d, g) next to the polariser (c in fig. 23) is to give the polarised light a violet tint when the two nicols are

FIG. 23.-Diagram of optical arrangements in Soleil's Saccharimeter.

parallel to one another. But if the nicols are not parallel, or if the plane of the polarised light has been rotated by a solution in the tube, one half the field will change in colour to the red end, the other to the violet end of the spectrum, because the two halves of the quartz act in the opposite way.

The instrument is first adjusted with the compensator at zero, and the nicols parallel, so that the whole field is of one colour. The tube containing the solution is then interposed; and if the solution is optically inactive the field is still uniformly violet. But if the solution is dextrorotatory the two halves will have different tints, a certain thickness of the compensating quartz plate which is lævorotatory must be interposed to make the tint of two halves of the field equal again; the thickness so interposed can be read off in amounts corresponding to degrees of a circle by means of a vernier and scale (E in fig. 23) worked by the screw which moves the compensator. If the solution is lævorotatory, the screw must be turned in the opposite direction.

Zeiss's Polarimeter is in principle much the same as Soleil's; the chief difference is that the rotation produced by the solution is corrected not by a quartz compensator, but by actually rotating the analyser in the same direction, the amount of rotation being directly read off in degrees of a circle.

Laurent's Polarimeter is a more valuable instrument. Instead of using daylight, or the light of a lamp, monochromatic light (generally the sodium flame produced by volatilising common salt in a colourless

gas flame) is employed; the amount of rotation varies for different colours; and now all observations are recorded as having been taken with light corresponding to the D or sodium line of the spectrum. The essentials of the instrument are as before, a polariser, a tube for the solution, and an analyser. The polarised light before passing into the solution traverses a quartz plate, which however covers only half

the field, and retards the part of the ray passing through it by half a wave-length. In the 0° position the two halves of the field appear equally illuminated; in any other position, or if rotation has been produced by the solution when the nicols have been set at zero, the two halves appear unequally illuminated. This is corrected by means of a rotation of the analyser, that can be measured in degrees by a scale attached to it.

Specific Rotatory power of any substance is the amount of rotation in degrees of a circle of the plane of polarised light produced by

1 gramme of the substance dissolved in 1 c c. of liquid examined in a column 1 decimetre long.

w=weight in grammes of the substance per cubic centimetre. l-length of the tube in decimetres.

(a) specific rotation for light with wave-length corresponding to the D line (sodium flame)

In this formula + indicates that the substance is dextrorotatory ; - that it is lævorotatory.

If on the other hand («), is known, and we wish to find the value of w; then

The specific rotatory powers of a few of the more important optically active substances found in the body are as follows :—

The first work in this direction was performed by Pasteur, and it was his publications on this subject that brought him into prominence. He found that racemic acid, which is optically inactive, can be decomposed into two isomerides, one of which is common tartaric acid which is dextrorotatory, and the other tartaric acid differing from the common variety in being lævorotatory. The salts of tartaric acid usually exhibit hemihedral faces, while those of racemic acid are holohedral. Pasteur found that, although all the tartrate crystals were hemihedral, the hemihedral faces were situated on some crystals to the right, and on others to the left hand of the observer, so that one formed, as it were, the reflected image of the other. These crystals were separated, purified by recrystallisation, and 1 As solution may cause condensation of volume, the density (d) or specific gravity of the solution should be also taken; then (a) = ± and w= ± 2 Ann. Chim. Phys. (2) xxiv. 442; xxviii. 56. Poggendorff's Annalen, lxxx. 127; xc. 498, 504.

a

uld

a

(a) D× ld.

Comptes rend. xxxvi. 26; xxxvii. 162.

those which exhibited dextro-hemihedry possessed dextrorotatory power, whilst the others were lævorotatory. Pasteur' further showed that if the mould penicillium glaucum be grown in a solution of racemic acid, dextro-tartaric acid first disappears, and the lævo-acid alone remains. The subject remained in this condition for many years; it was, however, conjectured that probably there is some molecular condition corresponding to the naked eye crystalline appearances which produces the opposite optical effects of various substances. What this molecular structure was, was pointed out independently by two observers-Le Bel2 in Paris, and Van't Hoff3 in Holland—who published their researches within a few days of each other. They pointed out that all optically active bodies contain one or more assymmetric carbon atoms, i.e. one or more atoms of carbon connected with four dissimilar groups of atoms, as in the following examples :

C2HS
H-C-CH,

CH.OH

Amyl alcohol

CO.OH

H-C-OH

CH-CO.OH
Malic acid

The question, however, remained-do all substances containing such atoms rotate the plane of polarised light? It was found that they do not; this is explained by Le Bel by supposing that these, like racemic acid, are compounds of two molecules- -one dextro-, the other lævo-rotatory; that this was the case

was demonstrated by growing moulds, the fermenting action of which is to separate the two molecules in question. Then the other question-how is it that two isomerides, which in chemical characteristics, in graphic as well as empirical formula, are precisely alike, differ in optical properties?--is explained ingeniously by Van 't Hoff. He compares the carbon atom to a tetrahedron with its four dissimilar groups, A, B, C, D, at the four corners. The two tetrahedra represented in fig. 25 appear at first sight precisely alike; but if one be super-imposed on the other, C in one and D in the other could never be made to coincide. This difference cannot be represented in any other graphic manner, and probably represents the difference in the way the atoms are grouped in the molecule of right- and left-handed substances respectively.

THE SPECTROSCOPE

When a ray of light passes from one medium such as air into another such as water or glass, it is bent out of its original course, or as it is termed refracted. The ratio of the sine of the angle of incidence to that of the angle of refraction is called the refractive index. When white light is passed through a glass prism or a triangular bottle containing a substance like carbon bisulphide with a high refractive 1 Compt. rend. li. 153. 2 Bull. Soc. Chim. (2) xxii. 337.

3 La chimie dans l'espace.