184 SECTIONAL ADDRESSES. 
Arber derived the Sphenopsida from Algz-bearing whorled branches 
of limited growth, converted into leaves, which were originally and 
always microphyllous. The Pteropsida, with which he associated his 
Paleophyllales (Psygmophyllum, with foliage like the Maiden-hair tree), 
were descended from Alge in which the branches were large, numerous, 
scattered, and not whorled, eventually metamorphosed to megaphyllous 
leaves. The Lycopsida, on the other hand, were derived from Algz 
in which the usually dichotomous axis bore emergences, metamorphosed 
to microphyllous leaves. 
Thus, as regards the origin of the leaf, Arber was in general agree- 
ment with Lignier, while he differed from the French author in the 
important point that he did not derive the Sphenopsida from the 
Fern-stock, but kept them as an independent line. 
A remarkable feature in Arber’s hypothesis is his treatment of the 
Psilotales. | He made this problematic family ‘a quite independent 
race, also of Algal origin, which appeared on the scene long after the 
other races . . . possibly in Mesozoic times or even later’ (p. 87). 
Thus he rejected both the connection with Psilophytales, suggested by 
Kidston and Lang, and the affinity with Sphenopsida, once maintained 
by the present writer. 
We thus see that, on Arber’s view, there were altogether four 
ow lines of descent, running back independently to ‘ Thallophytic 
Algee.’ 
Dr. Church, from quite a different point of view, arrives at some- 
what similar conclusions, but he goes further. He says: ‘ Speaking 
generally, it appears safer to regard a ‘‘race’’ or ‘‘ phylum ’’ as the 
expression of a group of organisms which derived their special 
attributes from the equipment of a preceding epoch, if not in one still 
further back. Thus all the main lines of what is now Land Flora must 
have been differentiated in the Benthic Epoch of the sea (i.e. as algal 
lines), as all algal lines were differentiated in the Plankton phase. The 
possibility is not invalidated that existing groups of Land Flora may 
trace back their special line of progression to the flagellated life of the 
sea, wholly independently of one another (Pteridophyta).’ (‘ Thalassio- 
phyta,’ p. 41.) 
Taking the Lycopods and Ferns as an example, and arguing from 
their different types of flagellated spermatozoids, Dr. Church states: 
‘It appears impossible to avoid the conclusion that the Lycopod phyla 
only merge with those of the Filicines in a distant Plankton phase, 
even beyond an independent origin as benthic sea-weeds’ (l.c., p. 82). 
Thus the idea of independent parallel lines of descent is carried to its 
extreme limit. ‘Hach phylum goes back the whole way, without any 
connection with anything else.’ Of course, this thorough-going poly- 
phyletic conception is involved in the doctrine already mentioned—that 
morphological differentiation was attained in the sea before the 
transmigration. 
I have cited Dr. Arber and Dr. Church as independent representa- 
tives, approaching the question from quite different sides, of the poly- 
phyletic or parallel-phyla hypothesis. The opposite view, of convergent 
monophyletic races, is also well supported. Some reference has already 
eT ee 
