E antes — The Morphology of A gat his australis . 17 
resemblances are superficial, not underlying'. These very features bespeak 
not primitive conditions, but strong specialization. Consider the super- 
numerary nuclei supplied for the long, branching pollen-tubes ; a megaspore 
membrane cap to protect young archegonia ; a pollen-tube tip that prepares 
access to several deep-seated archegonia ; the lack of jacket cells at the top 
of the archegonium to give readier access to the male cells since the neck 
has become impenetrable ; a ventral canal nucleus that is ephemeral, and 
about which no wall is formed ; in proembryo formation five or six con- 
secutive free nuclear divisions in a restricted central area —not throughout 
the egg-cell as in Cycadales and Ginkgoales ; a mature proembryo far more 
complex than any other known ; an embryonic cap of remarkably elaborate 
nature for protection and penetration ; and suspensors with a double elon- 
gation, the first to secure a holdfast from which to thrust in the opposite 
direction an embryo suspended freely in the archegonium. 
As possibly primitive features in the history of the male gametophyte 
the large number of microsporangia and the many supernumerary nuclei of 
the pollen-tube have been cited. As concerns the sporangia, the Abietineae 
and the Pcdocarpineae are the only tribes in which the number is strictly 
two. The Taxineae and the Araucarineae universally have more. Among 
the Taxodineae and the Cupressineae several genera, clearly among the 
more specialized within the groups, have three to several microsporangia ; 
for example, Sequoia , Taxodium , Widdringtonia , Juniperus . The two last- 
mentioned tribes are now generally admitted to be recent and specialized. 
Mr. Sinnott (11), in a paper contemporaneous with this, shows the primitive 
character of the Podocarpineae, emphasizing from a study of their gameto- 
phytes and strobilar morphology their abietinean affinities and their close 
relation to the Taxineae, which he considers a specialized branch from the 
group. This presents another instance of multiplication of sporangia, — the 
two of the Podocarps becoming several in their recent branch the Taxads. 
Further, the microsporangia of the primitive Ginkgo are two. Hence the 
occurrence of microsporangia in greater number than two among the 
Conifers seems to be a feature of recent specialization, and the Araucarians 
cannot on that account be considered primitive. 
The excessive amount of prothallial tissue of the araucarian pollen- 
grain has been compared with that found in palaeozoic Gymnosperms. In the 
pollen of the Cycadofilicales and the Cordaitales a number of internal cells 
exist. Is this internal tissue vegetative or spermatogenous ? We cannot 
say at present. Coulter and Chamberlain in their recent text-book (p. 178) 
believe it ‘ a safe inference 5 that these cells are in part, if not entirely, 
spermatogenous. The primitive living Gymnosperms, the Cycads and 
Ginkgo , form but one or two prothallial cells ; in the Araucarians and the 
Podocarps the larger number is the result of proliferation of these. The 
cells within the palaeozoic pollen-grains are not situated at one pole, nor 
C 
