314 Brooks and Stiles .— The Structure of 
are mostly either solitary or associated in pairs, and which includes the 
genera Dacrydimn and Podocarpits. A remarkably different condition 
exists in two species of Podocarpus, P. andinus and P. spicatus , in which 
the megasporophylls bearing ovules occur at intervals along the fertile 
branch. In all a single ovule is present on the upper surface of the 
megasporophyll, and is borne in a more or less reversed position. 
Assuming that the Podocarpeae form a natural group, a conclusion 
which from general considerations seems inevitable, there appears to be 
every reason to suppose that the state found in Dacrydium and Podo¬ 
carpus is derived from that in Saxegothaea and Microcachrys by reduc¬ 
tion in the number of scales in the cone, from an indefinite number to 
two or three pairs decussately arranged in Podocarpus. Here very 
occasionally both scales of two pairs may contain ovular bundles (27) j 
in some species normally both bracts of the lower pair are fertile, while 
in many species, e. g. P. alpimis and P. elatus , only one scale of the whole 
fructification is fertile. The female fructification is then to be regarded 
as a very reduced cone. In P. andinus and P. spicatus it would appear 
that the axis, instead of becoming very much shortened, has lengthened 
out a good deal, and the sporophylls are now separated by comparatively 
long internodes. The insertion of the ovule differs markedly from that 
found in Saxegothaea and Microcachrys. In these genera the single ovule 
is sessile on the upper surface of the megasporophyll, whereas in Podo¬ 
carpus it is borne on a stalk which apparently arises from the upper 
surface of the bract which bears it. Dacrydium exhibits transitions between; 
the two cases, for in some species the ovule is borne on the upper surface 
of the scale, in others it is more or less as in Podocarpus. The internal 
structure of the megasporophylls of the Podocarpeae also shows a similar 
series of stages. In Saxegothaea one bundle leaves the axis of the. 
cone and gives off the ovular supply by branches arising from it; in 
Microcachrys the two bundles are presumably separate from the cone axis, 
but very close together (30); 1 in Podocarpus one bundle of the peduncle 
passes up into the megasporophyll, while two others, one on each side of 
the sporophyll, join up and supply the ovule. 
It seems fairly clear that Saxegothaea is primitive for the Podocarpeae, 
and the state of things found in Podocarptts is derived from that found in 
Saxegothaea. The much smaller amount of centripetal xylem and the loss 
of function of the resin canals in some species support this view. 
The independence of the vascular supply of the ovule from the vascular 
bundle of the subtending bract can be easily, and it seems to us reasonably, 
explained as due to basipetal evolution of the vascular system, as the ovule 
has become more and more important in relation to the subtending bract., 
1 But Thomson (24) states that the sporangium bundle arises in the same way as that of 
Saxegothaea, by branching from the sporophyll bundle. 
