1009 
Multiseriate Ray of the Dicotyledons . 
ray is well developed, one finds the evidence equally conclusive. Among 
the Proteaceae, Grevillea and Hakea have been examined. The wood of 
Grevillea robusta for some distance from the pith possesses typical large 
rays of the oak type, which extend vertically for long distances. They are 
not confined to the vicinity of the leaf-traces as in the Ericaceae, but are 
uniformly distributed throughout the wood. A short distance from the 
pith they become divided into two or three smaller rays by the intrusion 
of fibres. Fig. n shows a broad ray at the bottom being split in this 
manner into two smaller rays. In any given section two or three such rays 
can usually be seen undergoing division. 
In Hakea sp. the conditions are almost identical with those described 
for Grevillea robusta. Fig. 12 shows the division taking place in this 
species, three smaller rays being formed at the top. 
These two examples are sufficient to show that the dissection process 
occurs in the Proteaceae and that it is probably of widespread occurrence 
in the family. Importance is added to this by the fact that the broad rays 
are much better developed here than in the Ericaceae. 
In regard to the Casuarinaceae the story is much the same. The 
early formed wood of Casuarina glauca possesses very large solid com¬ 
pound rays, as may be seen in PI. LXXVIII, Fig. 13, which is a photograph 
of a tangential section of this species taken very close to the pith. The 
homogeneous rays immediately become traversed by strands of fibres in the 
form of a network (Fig. 14). The strands increase in size and number, with 
the result that the remnants of the original broad ray gradually become 
smaller and scattered uniformly through the wood. The ultimate condition 
is shown in Fig. 15, in which there is no indication of a broad ray, but in 
its place a typical multiseriate-rayed wood. This condition is reached at 
a distance of two or three centimetres from the pith. 
In Casuarina equisetifolia the broad rays are never so well developed 
as in C. glauca, and become broken up much sooner. Fig. 17 shows the 
typical broad ray for this species, and Fig. 18 the aggregation of smaller 
multiseriate rays resulting from its dissection. A section still further from 
the pith would show a uniform distribution of small multiseriate rays. 
An undetermined species of Casuarma presented the same pheno¬ 
menon. It appears, therefore, that the conditions described are normal 
for the family. The obvious conclusion is that the Casuarinaceae formerly 
possessed broad rays and are replacing them by smaller multiseriate ones. 
In the broad-rayed genera of the Fagaceae the breaking-up process 
is sometimes apparent and resembles slightly that found in Casuarina. 
At the pith of Fagus ferruginea broad rays extend for extremely long 
distances vertically, as may be seen in Fig. 19. The two rays visible here 
extend far beyond the limits of the figure. A short distance from the pith 
strands of fibres begin to cross the rays obliquely and divide each into 
