THE LIVING CYCADS AND THE PHYLOGENY OF SEED PLANTS I5I 
our two questions: "What was the origin of the cycads?" and "Have the 
cycads left any progeny?" 
But it would emphasize the answer to the second question if we could 
show that the visible progeny could be referred to some other ancestry. 
In case of murder, the victim constitutes a concrete fact to be accounted for. 
The defendant may claim he didn't do it; but it adds weight to his claim if 
he can cast suspicion on some one else. 
So let us ask another question: "Could some other group have given 
rise to the Coniferophytes and the Angiosperms " ? We shall consider the 
two groups separately. 
If the Coniferophytes have not come from Cycadophytes, they must 
have come from the ferns or from the lycopods. This is a problem, in the 
discussion of which leaf gaps are balanced against leaves, pinnate leaves 
against simple leaves, and abaxial sporangia against adaxial. I believe the 
evidence is sufficient to establish a Pteridophyte origin; but the comparative 
claims of ferns and lycopods do not appear the same to me as they did 
several years ago. 
As far as the seed is concerned, some of the Paleozoic lycopods, like some 
of their living descendants, had progressed so far that their megasporangia 
are separated from seeds by arbitrary definitions rather than by facts. 
We separate the Gymnosperms from the Angiosperms by the ovules 
on open carpels and ovules enclosed in an ovary; and the distinction is 
good and very useful in a taxonomic key; but rigid definitions may harden 
our ideas and may prevent us from getting an unbiased view of the facts. 
In most Angiosperms, except epigynous forms, the ovules appear on 
open carpels, the closed ovary developing later. In cases like the Ranun- 
culaceae, the integuments of the ovule appear and the embryo sac is well 
started while the carpel is just as open as in any Gymnosperm. In the 
Amentiferae, the ovules are well started before the carpels close; and in 
Podophyllum, sometimes the carpels do not close at all, the ovules being 
borne on perfectly open carpels, as in the Gymnosperms. 
In considering this whole subject, we must remember that the extinct 
forms which have been preserved are mostly woody, especially in the 
Mesozoic. Has there been an extensive herbaceous flora which has dis- 
appeared? Have we lost herbaceous Gymnosperms which may have given 
rise to herbaceous Angiosperms? And could such herbaceous Angiosperms 
have given rise to the woody Angiosperms which became prominent in the 
Cretaceous? 
Unless such an herbaceous flora has arisen and disappeared, it is neces- 
sary to derive the Cretaceous Angiosperms from woody forms; and this 
means from more or less well known Cycadophytes or Coniferophytes. 
Such attempts have been made. We have already paid some attention 
to^the claims of the Cycadophyte line. 
In looking for the origin of the Angiosperms, the claims of the Coniferales 
