9 
Eames . — The Morphology of Agathis australis . 
bundles continue to function. Fig. 37 is a photograph of a transverse 
section of an axial bundle from a part-grown cone. Tubes are seen 
throughout the bundle, the older ones full of a mucilaginous substance and 
the younger ones appearing white and empty. The tubes reach the central 
cylinder either through the cortex or along the scale traces. Within the 
axial bundles they seem to pass chiefly downward and depart only with 
the traces. They may accompany the branches of the latter well out 
into the scale. 
These tubes of so strong haustorial nature appear in many cases to be 
the most rapidly growing branches. The time of approach of the male tubes 
to the embryo-sac is probably very variable. In A. australis , Thomson ( 14 ) 
has found tubes beside the very young embryo-sac ; the writer’s material 
of this stage did not show them. Many ovules ready for fertilization show 
tubes just entering, and in others the erosion of the embryo-sac — by a 
single unbranching tube — has clearly occurred only since the maturity 
of the latter. Thus apparently the fertile branches of the tubes reach the 
ovule at widely different times, a fact due either to varying times of 
departure from the pollen-grain or to greater or less amounts of meandering 
in sterile tissue. Perhaps at times the first branches enter the nucellus, at 
others the cone axis. 
Tube branches pass from the cone axis and from the base of the 
adjacent scales to the nucellus, crossing the space that exists between them 
in those cases where nucellar enlargement is not great. Whether the 
space at this time is filled with fluid or with air is not known. Thus 
another Angiosperm-like feature is added to so-called * protosiphonogamic ’ 
fertilization. 
The great exposure of nucellus by the large micropyle, and the sinus 
in the wing of the cone scale directly over the ovule, are probably means of 
providing readier access to the ovule for the pollen-tubes. The openings in 
the scales suggest that the fertile tube branches may come from pollen 
deposited on other scales than that bearing the ovule. 
Entering the nucellus a tube may continue to branch and cause much 
erosion, or its course may be straight down to the gametophyte. After 
that point is reached, the course is normally downward between the nucellus 
and the megaspore membrane. This is followed, often without erosion, 
until the thinner portions of the megaspore membrane are reached. Then 
these advance tube-tips — the fertile nuclei are not yet within the ovule — 
erode the tissue about the necks of the archegonia, forming depressions 
or cavities above them (PI. I, Fig. 9 — the archegonium on the left). PI. IV, 
Fig. 38 shows a transverse section of this cavity. After one archegonium 
has been thus prepared for fertilization, the tube passes on to others. 
The male elements follow down the pathway so well prepared before- 
hand. The latter were not found within the ovule in any case until just 
