66 SCIENCE PROGRESS 



purpose by the parts of the curves in Fig. 6, lying between 

 E and H. In most, if not in all, cases that have been in- 

 vestigated, the T-R curves for three or more colours of light 

 do not intersect at one point, but over a range, and it is this 

 relative displacement of the T-R curves which gives rise to 

 what is known as anomalous rotation dispersion, a phenomenon 

 which will be referred to in the sequel. In the meantime 

 it is to be observed that the point at which two T-R 

 curves cut one another, as, for example, at G in Fig. 6, may also, 

 like a maximum, be regarded as a singular point. It is therefore 

 of great interest that, in much the same way as the maximum 

 is displaced in accordance with the changes of constitution of 

 an active substance or with a change of solvent or of concentra- 

 tion, the point at which two given temperature rotation curves 

 cut one another shifts at least in the case of tartaric acid {J.C.S., 

 1913, 103, 167) and I'sobutyl tartrate {J.C.S., 1916, 109, 1147), 

 in a somewhat similar manner ; but with this difference, that 

 although the intersection of the T-R curves may occur at some 

 different temperature with change of solvent, etc., its rotation- 

 value remains either constant or nearly so (see J.C.S., 191 6, 109, 

 1 158). Much the same thing obtains with regard to the whole 

 region of so-called anomalous dispersion as far as the matter has 

 been investigated, and since the region of anomalous dispersion 

 shifts about with change of constitution or solvent in much 

 the same way as the maximum, it is a further important piece 

 of evidence that the whole T-R curve is being subjected to this 

 kind of change. Exactly the same sort of thing as had been 

 found for ethyl tartrate occurs in the case of i'sobutyl tartrate. 

 In z5obutyl tartrate a maximum rotation occurs at a tempera- 

 ture of about 1 50° for all the different colours of light, and since 

 the ester will not remain superfused for a sufficiently long time, 

 it is not possible in the homogeneous ester to observe the region 

 of anomalous rotation dispersion. But by using a solvent such 

 as acetylene tetrachloride, which has a depressing effect upon the 

 rotation of the tartrate, it is possible to bring the region of 

 anomalous dispersion within the ordinary range of temperature. 

 Thus, the maximum rotation and the regions of anomalous 

 dispersion can both be moved about in accordance with the solv- 

 ent used, and, as far as can be judged at present, on esters of 

 similar constitution a given set of solvents would have much 

 the same relative influence (J.C.S.y 1916, 109, 1147). 



The T-R curves for ethyl tartrate for eight different colours 

 of light have been examined over a wide range of temperature 

 {J.C.S., 1916, 109, 114s). They intersect one another as has 

 been described, and the question then naturally presents itself, 

 that if the curve for one of these colours, say sodium yellow, has 

 the sinuous form shown in Fig. i , which appears to be indicated 



