c. A. BARBER. lial 
The denser portion of the cortex, laden with masses of starch, 
now occupies a position half way between the surface of the 
haustorium and the host’s root. It has become more sharply 
defined and, in most cases, the outer radiating cells, formerly filled 
with starch, have commenced in their turn to lose their contents. 
The denser portion is moving inwards. A. separation layer will 
shortly be found between the outer, denser part of the cortex and 
the inner, clearer rows of cells. Although this line of demarcation 
is not yet clearly seen, it is frequently possible at this stage to 
form an idea as to where it will be formed. 
The haustorium, as a whole, is becoming crescent-shaped, 
and the lines of growth curve more distinctly towards the median 
line. This is brought about by the expansion of the outer cells 
of the radial rows tangentially, and they are seen to be distinctly 
broader than those within (b and c). The inner, clearer cells are in 
fact smaller in all directions, being confined in a narrower space. 
The oval, central area also shows a certain amount of 
differentiation. Its upper limit is not at present clearly 
defined and the development of this part may be left out of 
consideration for a time. The junction of the axis of the haus- 
torium with the mother root forms, in fact, a transitional region 
at present undifferentiated. In the lower part, however, a dense 
vertical shadow is seen in the middle line, extending almost to 
the surface of contact with the host’s bark. The cells of this 
shadowed area appear to be dividing more rapidly in the vertical 
than in the horizontal direction and thus already appear to be 
elongated transversely to the axis of the haustorium. This elon- 
gation is shown well in figure 2d’ of Plate III, taken from a 
young haustorium of this age. The vertical shadow may be re- 
garded as showing the first stage in the development of the gland. 
The zone of the radiating cells described in the previous 
section (Plate III, fig: 1) has thus undergone considerable 
changes. But the cells of the haustorium, although somewhat 
disturbed, are still arranged in the same rows more or less at 
right angles to the host’s surface and curving inwards to the axial 
line. It is still possible to trace continuous rows of cells from 
