Dischidia rafflesiana ( Wall ,). 253 
ence in thickness. The green layer showed the ordinary 
chlorophyll absorption-spectrum ; the purple layer gave 
a spectrum which differed most obviously from that of 
chlorophyll in the absence of the characteristic absorption- 
band in the red, from B to beyond C. 
More accurate observations were made on an alcoholic 
extract of the purple pigment. The absorption-spectrum, in 
bright sunlight, was visible from -71 to *435 on the scale used, 
which was divided according to wave-lengths, that is, from 
midway between a and B, nearly to G. 
From *71 to *589 (the D line) there was little or no absorption, 
the lines B, C and D being very clear. From *589 to *516 
(about b) there was considerable absorption, the lines E and b 
being faint but distinct. From -516 to -435 (that is, from b 
nearly to G) the absorption was much greater, and no Fraun- 
hofer’s lines were visible. The whole violet end of the spectrum 
was completely absorbed. 
We see then that the green and red layers between them 
cut off almost all light from reaching the interior of the pitcher; 
in particular, the blue and violet rays are most completely 
absorbed. 
It is evident that the pitcher thus forms a dark chamber, 
into which the negatively heliotropic root is likely to be 
attracted, the darkness being most complete as regards those 
rays by which the direction of growth is influenced. The 
darkness of the pitcher may serve, not only to attract the 
adventitious root in the first instance, but also to ensure that 
its branches remain within the recesses of the organ. We do 
not overlook the fact that the presence of moisture and food 
must also be of influence in determining the direction of growth 
of the pitcher-root, and its branches. 
Development of the Pitcher. 
The development of the pitcher was studied in some detail. 
The youngest undoubted pitcher available was about | in. 
(3 mm.) in length. The lamina was curved like a shell as in 
Treub’s Fig. t a, PL IV, the apex pointing downwards, and 
