258 Slopes.— Roots in Bennettites . 
traverse this break, the clear mineral matter filling it is felted with root-hairs. 
PL XIV, Fig. 1, shows only one focal level in a thick slide, so that their 
quantity may be imagined. Individual hairs may be traced for nearly 8 or 
9 mm. across the gaps, and the expanded bases of the hairs, where they are 
attached to the tissues of the rootlets, are well seen in many places ; cf. 
PI. XIV, Figs. 1, 2, and 3, b. 
Description of rootlets . Most of the rootlets are roughly 1 mm. in 
diameter, but a few branches are present about half this size. Branching is 
infrequent ; though several lengths of rootlets from 1 to over 3 cm. are to 
be seen, there is only one clear instance of dichotomy, and a couple of 
apparently lateral branches being given off. 
The individual rootlet is composed of a central vascular strand, of which 
in longitudinal and oblique sections three or four good-sized scalariform 
tracheides are well preserved. Transverse sections showing the vascular 
elements clearly are not to be seen. Surrounding the tracheides is a several- 
layered zone of small, very dark, apparently thickened cells. Outside of the 
central strand are 3-4 layers of large cortical cells, bounded by similar but 
smaller cells, vertically elongated, from which the numerous root-hairs 
spring. All these layers are to be seen in PI. XIV, Fig. 3. 
The rootlets which pass through the unbroken ground-tissue of the 
Bennettites differ in this region only in having no root-hairs . Such a hairless 
rootlet, passing through the mass of parenchyma, can be seen in PI. XIV, 
Fig. 4. It is impossible to offer precise proofs, but the impression created 
in my mind is that the rootlets originate in some way connected with the 
gum canals of the ground tissue. One must simply await the discovery of 
further specimens of root-bearing Bennettites before the mode of origin of 
these rootlets can be determined. There remains also the question whether 
they belong to the plant itself, or are an invasion from the outside. I incline 
to the view that they are adventitious roots of the Bennettites itself. In a few 
places it appears as though the leaf-base tissues had contributed to the 
rootlets ; and the tracheides of the rootlets are entirely similar to these 
scalariform elements which form such a characteristic feature of Ben- 
nettitalean wood. 
Far from the ‘ histological details of both wood and bast ’ ‘ agreeing 
precisely with the corresponding structures in a recent Cycad ’, the secondary 
wood of Bennettites spp. has a type of scalariform tracheides which is, so far 
as I am aware, not found in the secondary wood of any other of the families 
of higher plants. The perfect agreement of the tracheides of the rootlets 
with those of the Bennettites type seems to me, while not conclusive, con- 
tributory evidence favouring the view that they belong to the plant in whose 
tissues they occur. 
No root structures — neither primary nor adventitious — have hitherto 
been described for Bennettites so far as I am aware. Such roots as these, 
