N 
372 On the Genealogy of Plants. [June, 
towards the reticulated structure which prevails with the Dicotyle. 
Every one is familiar with many cases of this kind, as in 
Dioscorea, Goodyera, etc., while on the other hand, approaches to 
the parallel venation sometimes occur in the Dicotyle (Plantago, 
etc.). : 
A far greater difficulty is presented by the cotyledons; for 
while there are a few cases in which exogenous plants develop 
but a single cotyledon, I am aware of no case in which an endog- 
enous Angiosperm has been found to develop more than one. It 
is, however, presumable that a more complete investigation of 
this question may reveal transition forms here as elsewhere. 
Such are the principal facts thus far made known which tend 
to encourage the hope of ever tracing the higher class of Angio- 
sperms back to an origin within the lower class. 
Far more satisfactory is the evidence that the Dicotyledons 
have been developed out of the Gretaceæe and perhaps indirectly 
out of the Conifere. The Guetace@, a small but interesting — 
family of only three known genera (Gnetum, Ephedra, and Web 
witschia) possess all the marks of forming a true intermediate link. 
The flowers of both sexes are provided with a sort of half- 
envelope, called the perigonium, which surrounds and protects 
the anther-bearing filament in the male, and the solitary ovule in 
the female flower, and may be regarded either as a rudimentary 
ovary or as a rudimentary perianth. 
It is worth remarking here that the chasm between the Gymno- 
sperms and Angiosperms is at all points greater with respect to 
the floral envelopes (including the ovarian) than with respect to — 
the process and true organs of fertilization. The ovary of the 
Angiosperm is enclosed in an envelope, in the true Gymnosperm — 
it exists but is exposed, in the Gnetacee it is half enclosed and 
half exposed. It matters not whether the perigonium be regarded 
as the homologue of the ovarian envelope or of the outer floral 
envelopes of the Angiosperms, since all the* floral organs, 
including even the essential ones (stamens, pistils, etc.) are simply 
modified leaves. In the passage from the Cycadacee to the Palma- 
cee no such connecting link has yet been discovered, and for the 
truth of such a transition we must rely upon the remarkably 
strong physiognomic resemblance’ coupled with the evidence fur- 
nished by the structure of the tissues and the mode of zstivation 
The Gnetacee, however, while they give us this invincible evi- 
