10 BOTANICAL GAZETTE [JULY 
whether conifers have sprung from ferns or lycopods, one sort of 
sporangium must have migrated to the other side of the sporophyll. 
This is true, of course, only if one accept the author’s view that the 
cone scales of conifers are really simple. 
The presence of seeds in Lepidocarpon and Miadesmia is held 
to prove that “there is thus abundant evidence that the potentiality 
of seed production existed in this phylum as well as in the fern 
phylum.” Any differences that exist in the vascular supply of the 
sporophylls between podocarps and lycopods “‘is to be accounted 
for by the greater relative importance of the ovule as compared 
with the sporophyll.”” Small, narrow, uninerved leaves are char- 
acteristic of both conifers and lycopods, but are unknown among 
the Cordaitales. Those podocarps and araucarians with broad 
parallel veined leaves are not primitive, but have derived their 
leaves from narrow-leaved ancestral forms. 
One of the most interesting points in the argument is the attempt 
to show that, while a siphonostele with leaf gaps is certainly char- 
acteristic of the fern alliances, it is not necessarily limited to them. 
It merely represents a goal toward which vascular plants of all 
sorts have tended. The ferns reached it early, while the paleozoic 
lycopods did not quite reach it. They did actually attain the seed 
habit, another one of the milestones of plant evolution, but attained 
only to a medullated siphonostele in which nearly all the metaxylem 
had been obliterated and which had become broken up in some 
forms into separate strands. These bundles were still exarch, 
however, and the leaf traces did not produce leaf gaps in the stele. 
These would have been the next logical steps in the evolution of 
the lycopod stele. The inference probably is that they were 
actually attained by the yet unknown lycopodialean ancestors of 
the conifers. While admitting that the presence of bordered pits 
in the secondary wood of conifers is a point against the lycopod 
theory, he thinks that the presence of a modified sort of pit in 
Sigillariopsis Decaisnei (48, 53) shows the possibility of their de- 
velopment in this phylum. The double leaf trace of the Abietineae 
(and Araucarineae), which has been used as an argument for the 
cordaitean origin, he thinks is offset by the single trace of the 
primitive podocarps. 
