52 Salisbury . — On the Structure and Relationships of 
3. The Inner Flesh. 
All those sections which pass through the body of the seed, even 
where the preservation of soft tissues is excellent, show no indications of 
an inner flesh. It is certainly true that in most of the preparations the 
internal surface of the sclerotesta exhibits an irregular outline, but in the 
transverse series S.33, a- S. 33,^, in which the internal preservation is almost 
perfect, the inner margin of the sclerotesta in some places forms a perfectly 
even surface. This latter affords positive evidence of far more value than 
the negative evidence of a ragged surface, and still more so since we find in 
the badly preserved material that the sclerotesta does show signs of superficial 
disintegration. 
In dealing with the nucellus we shall find that it was possessed of 
a highly differentiated and probably strongly circularized epidermis, and 
there is evidence that it was in close contact with the sclerotic tissue. 
Having due regard to all these facts, one is forced to the conclusion 
that in this species of Trigonocarpus an inner flesh was not present in the 
middle part of the seed cavity. 
When we come, however, to about the level of the insertion of the 
micropyle, the zone of sclerization begins as it were to shift slightly 
outwards, and we thus find within the sclerotesta several layers of cells 
which, though possessing thick walls, have not the dark brown contents of 
the sclerotic fibres themselves, and, moreover, become thinner- walled as we 
pass inwards. This tissue reached its maximum thickness, of some four 
layers, midway between the angles, and on its inner margin exhibits a broken 
surface that here might well indicate a greater extent of probably more 
parenchymatous elements (PI. IV, Fig. 8, if.). 
Even then, if we regard this micropylar lining as homologous with the 
inner flesh of other Trigonocarpeae, our seed is as extreme a member on 
the one hand, as Pachytesta on the other. 
IV. The Vascular Organization. 
(a) The Chalazal Bundle and Nucellar System. 
The lowest of our transverse sections yields us no information as to 
the vascular structure, but from S. 33, b y Man. R. 1 161, b, and the longitudinal 
section S. 32, e, we find that a single vascular bundle entered the chalazal 
end of the seed. This passed up to the base of the nucellus, at which level 
it is cut by the section S. 33, £ (PI. V, Fig. 16). The bundle is there seen 
to have been a solid mass of xylem very slightly triangular in outline, the 
angles corresponding in position with the three main ribs of the seed. It 
was formed of from sixty to seventy elements, and attained a diameter of 
about 0-18 mm. The larger tracheae are situated mostly at the periphery 
