22 BOTANICAL GAZETTE [JANUARY 
CYCADALES.—It appears that the cycads, ancient and modern, 
are closely related to Cycadofilicales. This relationship is not 
contradicted by the pollination devices, for they are so similar as 
to afford little difficulty in bridging over the gap. The method of 
deposition, as well as the presence of a pollen chamber in the 
nucellus, are as near alike as one could well expect. The greatest 
difference is the presence of a pollen tube in modern cycads. It 
has already been pointed out that this tube grows away from the 
female gametophyte and is exclusively haustorial in function. 
The pollen chamber itself provides access to the archegonia in just 
the way that we have conjectured for the preceding group. The 
real difficulty arises in supplying a convincing reason for the origin 
of a tube at all. If we are to homologize the pollen tube of cycads 
with the rhizoid of the germinating spore of their pteridophyte 
ancestors, it means that an organ that had been completely lost 
has been revived. Admitting the possibility, about which I am 
very dubious, of the revival of this ancient structure after the lapse 
of geologic ages, it is evident that it would be useful, subject to the 
laws of selection, and likely to be preserved. If pollination pre- 
ceded fertilization only a short time in the earlier seeds, and the 
remaining processes took place on the ground, it is evident that 
what these ancient plants had attained was not the ‘‘seed habit,” 
in the sense that we employ the term with reference to modern 
plants. They had merely attained the ovule and pollination 
habit. A real seed could be developed only if the seed structure 
were retained on the plant until its maturity (except the growth 
of the embryo itself). An advantage would certainly lie in early 
pollination of the ovule that would permit the further growth of the 
ovule even while the gametophytes were maturing. That this 
habit of pollination long before fertilization is an advantage is 
indicated by the fact that all modern seeds practice it, although 
it is difficult to imagine that the first seeds or ovules that were 
pollinated furnished food and protection on the exposed nucellus 
sufficient to maintain the male gametophyte for a year or more, 
as is commonly the case in modern conifers. 
CorpalTaLes.—The Cordaitales differ from the Cycadofilicales, 
among other things, in that their seeds occur in cones and not on 
