at the Linnean Society. 143 
purely cryptogamic. We had to remember that the anatomy of the 
seed-bearing Lepidodendreae was exactly like that of the cryptogamic 
forms, whereas it was the peculiar anatomy of the seed-bearing 
Pteridosperms that first suggested that they were not typical 
Filicineae. Further the seed-like organs of carboniferous Lycopods 
were clearly derived immediately from cryptogamic sporangia, while 
the seeds of Pteridospermeae were far removed from anything of the 
kind. As regarded the other great question under discussion, the 
origin of the Coni ferae, the data were still very inadequate. The fossil 
Coniferae were very abundant and extremely unsatisfactory. Palaeo- 
botanists were, in fact, without adequate knowledge of the structure 
of the earlier Coniferae, upon which the question must depend. 
The view that the Coniferae or, at least, the Araucarieae had des¬ 
cended from Lycopods was brilliantly stated by Mr. Seward at the 
previous Meeting. It was difficult to follow Mr. Seward if one did not 
quite agree with him. A fundamental point was the existence of the 
Cordaiteae, a synthetic carboniferous group. The structure of the 
stem and root was Coniferous and resembled that of the Araucarieae, 
while the leaves were Cycadean, and the male and female fructi- 
b-45 fications to some extent resembled those of G ink go, as he would 
shew by reference to diagrams. As Mi - . Seward admitted, there 
was a series of forms which, anatomically speaking, connected the 
Cordaiteae with the Pteridosperms. Thus the Cordaiteae possessed 
characters pointing in no less than four directions. The structure of 
the wood, which he would illustrate, was admittedly indistinguishable 
from that of the Araucarian Conifers. This was of some importance 
because it was of the type characterising the Conifers, Cordaiteae, 
Pteridosperms and Botryopterideae. It was the great characteristic 
type of the Fern series throughout the Palaeozoic. Lycopod wood 
on the other hand was scalariform. This fact tended in the strongest 
possible degree to ally the Cordaiteae with the Ferns and to separate 
them from the Lycopods. Mr. Seward had not made his position 
quite clear. Did he really mean to sweep all the Coniferae into his 
Lycopod net? If so Mr. Seward was, he thought, in a position of 
considerable difficulty, because he would have to include the Taxaceae 
and Ginkgo. Ginkgo was universally admitted to have a Cycadean 
and therefore a Filicinean affinity, and it also had a real affinity 
with the Taxaceae, for instance with Cepluilotaxus, established before 
the multiciliate spermatozoids were discovered. Was this to be dis¬ 
regarded ? On the other hand, if the argument were intended to apply 
to the Araucarieae alone, where were we going to draw the line 
between Lycopodinean and Filicinean Conifers ? Most botanists 
were agreed that the Conifers were a natural group, and it was a 
<>reat deal to demand that they should be split up. Much had been 
made of the peculiarities of the Araucarieae, but it would be very 
difficult to maintain that they had no affinity with the Abietineae, 
for instance, or with the Podocarpeae. He must not keep them too 
long, but there was one point Mr. Seward had mentioned to which 
he must refer. He did not dispute the great geological age of the 
Araucarieae, nor their relatively primitive position, in relation to the 
Abietinea: for instance. But the gradual transition between sporo- 
phylls and foliage leaves on which Mr. Seward had laid stress was 
paralleled among the Pteridospermeae in which the sporophylls 
indeed were scarcely different from the foliage leaves. The male 
sporophylls of Araucarieae were admitted to be unlike those of the 
Lycopods, but they were said to resemble the sporophylls of Chciro - 
