75 
The Seed of Lyginodendron. 
have been borne by such a fern-like plant, but we may perhaps still 
be permitted to doubt whether Lagenostoma was certainly a true “seed” 
or whether it may not have belonged to a more primitive class of 
“ Protosperms” which had evolved some but not all the characters of 
true seeds. This of course is partly a question of definition. The 
two essential points distinguishing a seed such as we find in modern 
Gymnosperms, from the megasporangium of existing heterosporous 
Pteridophytes, are, we take it, on the one hand the existence of a 
pollen-chamber or micropyle, or both, correlated with the germination 
of the microspore on the megasporangium and the habit of partial or 
complete siphonogamy, and, on the other, the formation of the embryo 
completely enclosed in the megaspore (embryo-sac), while still on the 
parent plant, and the subsequent “ germination ” of the seed after a 
resting stage by the growing out of this embryo into the new 
sporophyte generation. 
The first cf these features exists not only in Lagenostoma but 
apparently also in Lepidocarpon 1 and Miadesmia 2 among the 
Lycopodineae, while there is apparently no evidence that Lagenostoma 
exhibited the second. This point is not discussed by our authors in 
the present communication, but no doubt it will be fully dealt with in 
their larger paper. In general structure, of course, Lagenostoma is 
very much closer to the Gymnosperm type than it is to such structures 
as the integumented sporangium of Lepidocarpon or Miadesmia, and 
this may incline Messrs. Oliver and Scott to the view that the embryo 
was formed completely within the megaspore and that germination 
occurred much as in modern Gymnosperms; but the seed certainly fell 
off the plant at an early stage, presumably before fertilisation had 
occurred, and of its subsequent history nothing is known. In this 
connexion it is of interest to note that there is seme evidence pointing 
to relatively late and continuous development of the embryo in 
Cycads, without the intervention of the definite resting period so 
characteristic of the higher Gymnosperms and of the Angiosperms, 
and it is stated that the seed of Ginkgo often falls to the ground 
before fertilization. This type of development of the young sporo¬ 
phyte may perhaps have been characteristic of the primitive 
Siphonogams. 
However this may turn out, it may now be taken as almost 
beyond question, from the general resemblance of Lagenostoma to 
1 See Scott, Phic. Tuans., 1901; and New Phytoeogist, 
Vol. I., No. 2, Feb. 1902. 
2 Miss Benson, New PHYTOEOGIST, Vol. I., No. 3, p.58, March 
1902. 
