19 
1918-19.] Researches in Optical Activity. 
ditrichloracetyltartrate exhibits a minimum, exactly the opposite of what 
had been found for the simple ester (J.C.S., 1912, 101, 378 ; 1913, 103, 152). 
But the curves for these two substances must .surely have something in 
common, and, taking what has been said above into account, it appears 
probable that, just as the change of constitution in passing from ethyl 
tartrate to 'isobutyl tartrate causes a slight change in appearance of the 
T-R graph, and the change from the homogeneous condition to solution 
in nitrobenzene causes one much greater, so the considerable change of 
constitution in passing from ethyl tartrate to ethyl ditrichloracetyltartrate 
causes a great shift in the position of the maximum, in fact, removes it 
from the region of ordinary temperatures altogether, bringing into view 
a minimum which would occur in the ethyl tartrate curve at temperatures 
at which the ester has not yet been investigated. This induction is 
strongly supported by the fact that a number of T-R curves are known 
in which a point of inflection occurs in addition to a maximum (or 
minimum), for example in those for certain carbinols (Pickard and Kenyon, 
1912, 101, 623), or else in curves obviously tending towards a 
maximum (or minimum), as is the case for ethyl tartrate in water 
(Patterson, J.G.S., 1904, 85, 1129), and for homogeneous ethyl di-o-nitro- 
benzoyltartrate (Frankland and Harger, J.G.S., 1904, 85, 1571). In 
these, therefore, we have direct evidence of the connection between the 
maximum and the minimum. 
4. Thus by piecing together evidence of various kinds — the behaviour 
of the homogeneous ester up to the highest temperatures possible ; the 
behaviour of homologous esters or derived esters ; the behaviour of 
these substances in different solvents — it seems reasonable to conclude 
that a fundamental form of T-R curve is common to all the tartrates, 
and that different tartrates at ordinary temperatures exhibit different 
regions of this fundamental curve, thereby appearing at first sight to 
have no connection with each other. This may be stated in the two 
following propositions : (a) the T-R curve of an active substance over 
a wide range of temperature would probably show several maximum 
and minimum values ; (6) the influence of change of constitution or the 
effect of a solvent is, apparently, to displace not merely the maximum, 
but the whole T-R curve in one direction or the other, as the case 
may be (J.C.S., 1913, 103, 158 et seqq.\ 1916, 109, 1140-1142). 
5. These conclusions, it may be noted, were arrived at by the study of 
data obtained by using only one colour of light. They are strengthened 
and justified when data for other colours are taken into consideration. 
The extended curves for light of different refrangibilities are similar in 
