57 
Trigonocarpus Shorensis , 
as shown by the great difference in the area of sclerotesta sectioned at 
the two successive levels. These facts, taken together with the position 
of the sarcotestal bundles, seem to justify the assumption that the seed was 
T rigonocarpus P arkinsoni. 
In both of the sections a few secretory sacs can be seen. Dr. Scott has 
permitted me to examine his best preparations of T. Parkinsoni with 
the sarcotesta preserved, and particularly Wild’s section S. 1952, figured by 
Scott and Maslen, 1 in which there are present some dark bodies surrounded 
by a clear space 0-045 mm. -0-052 mm. in diameter. Neither for these nor for 
similar bodies in the other preparations can one definitely assign a secretory 
nature, but a comparison with ill-preserved secretory sacs of T. Shorensis 
discloses a striking similarity between them. Probably, then, secretory sacs 
were present in T. Parkinsoni , both at the base and apex of the sarcotesta, but 
in very much smaller numbers than in T. Shorensis. 
The central region of the lowest of the two sections cuts the single main 
chalazal bundle obliquely, at the level where the sclerotesta begins to be 
differentiated. 
The general outline of the vascular bundle was very slightly triangular, 
with a diameter of about 0-3 mm., the angles corresponding in position 
to the secondary ribs. From each of the two upper of these angles (PL V, 
Fig. 14, v.b.) a strand is seen in process of being given off into the sarco- 
testa. 
The main bundle is surrounded by a thin-walled parenchymatous tissue 
{par.) consisting of vertically elongated elements. In the centre is a solid 
mass of xylem, throughout which are scattered short tracheides horizon- 
tally directed, exhibiting scalariform or pitted thickenings. The section is 
sufficiently oblique to show that the longitudinally directed tracheides 
were scalariform. 
In the centre of the xylem a single secretory cell can be detected 
(PL V, Fig. 14, m.s). 
The periphery of the bundle is formed of a band of short tracheides 
1-2 elements in width, which for the most part are separated from the 
central xylem mass by what appears to have been parenchymatous tissue 
similar to that surrounding the bundle as a whole. But this tissue is 
interrupted at several points by junctions between the outer and inner 
tracheides. The smallest xylem elements lie mostly at the periphery of 
the central core, but others are situated more interiorly. 
From an examination of the angle which is cut at the highest level, 
it can be seen that from this peripheral zone pass out the xylem ele- 
ments of the sarcotestal bundles. In this particular case, the portion has 
partially separated off from the central mass, and the corresponding 
bundle, owing to the double obliquity of the section, is again cut further 
1 Loc. cit., PI. XII, Fig. 11. 
