1 6 E antes. — -The Morphology of A gat his australis . 
down through the endosperm. In Figs. 33, 34, and 35 are seen progressive 
stages in the growth of the embryo and the disappearance of the cap. 
The latter is spread out and flattened by the enlarging embryo. In Fig. 35 
all traces are lost. This figure also represents the latest stage in embryo 
development obtained. The embryo must be left as a small, somewhat 
spherical mass in the lower central section of the endosperm. 
Soon after embryos begin to pass down into the endosperm the entire 
upper portion of the ovule inside the integument begins to shrivel, all of the 
archegonium- bearing section being concerned. This destruction of gameto- 
phytic tissue is very extensive, including as it does nearly or quite all of 
the larger upper section of the mature gametophyte. Fig. 3 shows the 
beginning of this : all that section above the centre of the suspensor coil 
gives evidence of disintegration. The lower portion has already increased 
in diameter, thus reversing the club- like shape of earlier stages, the basal 
portion being now the larger. The archegonial region in all gymnospermous 
ovules is destroyed during the development of the embryo, but the destruc- 
tion is naturally not great owing to the limited distribution of the arche- 
gonia. The ripe endosperm of Agathis thus consists in large part of the 
smaller portion of the mature gametophyte. Fig. 4 shows a longitudinal 
section of a ripe seed. Embedded in the endosperm is a distinctly large 
embryo with two cotyledons. The embryo extends nearly throughout its 
length. A dry beak of collapsed tissue represents the fertile section of the 
gametophyte. The shrivelled nucellus is still conspicuous, free from the 
integument. The endosperm shows no signs of the c ruminating ’ outline 
reported to occur in the Araucarineae. Neither was such a condition found 
in the large seeds of Agathis vitiensis , which were also examined. Erosion 
by the voracious pollen-tubes in regions which become permanent endo- 
sperm would doubtless cause a condition simulating that in the ruminated 
seeds of the Taxaceae, and it is possible that such is the condition in 
Araucaria. 
Affinities. 
There have been presented above, in the history of the gametophyte 
and of the embryo of Agathis , a number of features sufficiently striking in 
their contrast to those known to obtain in other Conifers to justify from this 
standpoint — as well as from that of the anatomy and morphology of the 
mature sporophyte— the isolated position of the Araucarineae. The 
important question then arises — what is the phylogenetic relation of this 
group to the other Conifers ? Are these features indicative of a primitive 
condition, as those already known have led some to believe, or is some 
other explanation more consistent with the facts ? The writer believes 
that they are certainly not primitive— that though at first glance they may 
suggest the conditions prevailing in the more primitive Gymnosperms, the 
