388 
fact that the Sigillarieze have not even a remote 
connection with those plants. 
Stigmaria is held to represent a rhizophore, of 
which the stigmarian appendages would there- 
fore be the roots. From this point of view 
these remains must be regarded as representing 
organs comparable with the rhizophores and 
roots of existing Selaginellas. 
Dating from Silurian time, the ferns gained 
special prominence in the Carboniferous, where 
it is even yet difficult to separate the true ferns 
from fern-like plants. The author nevertheless 
adopts the conclusions of Bower and Campbell 
respecting the relative antiquity of the Huspo- 
rangiate and Leptosporangiate as amply justi- 
fied by paleontological evidence, which also gives 
support to the classification of ferns by Bower 
on the basis of the development of the sorus and 
the output of the spores. Im this, the division 
into Eusporangiate and Leptosporangiate is sub- 
ordinated to the development of the sporangia 
with reference to time and place. 
One of the most remarkable and significant 
results of paleontological research in recent 
years has been the recognition of the Cyeado- 
filices as established by Potonié, thus forming a 
connecting link between the ferns and the 
Gymnosperms through the Cycads. Dr. Scott 
rightly places special stress upon an elucidation 
of the characters of this important group and 
shows: 
1. That hitherto supposed forms of Alethop- 
teris, Sphenopteris and Neuropteris types really 
represent the foliage of Cycadofilices. 
2. The anatomical characters of the stem are 
in close agreement with those of the Cycadacez. 
3. While the exact character of the fructifica- 
tion is not known with certainty, that which 
in all probability belongs to those plants is 
widely different from that of the ferns and: ap- 
proximates to that of the Cycads. 
Poroxylon is shown to be a transitional form 
between the Cycadofilices and Cordaites through 
Lyginodendron. With respect to the Cordaiteze 
to which he deservedly devotes a large measure 
of space, he fittingly summarizes prevalent 
views respecting this most important group 
when he says that ‘‘ Further investigation will 
doubtless modify greatly our conception of the 
Cordaitez, and display a much greater variety 
) 
SCIENCE. 
[N.S. Von. XIII. No. 323. 
among the members of this family then we are 
at present prepared for. But whatever the 
future may have in store for us in this respect, 
there can be no doubt that the revelation of 
the existence of this fourth family of Gymno- 
sperms was a discovery of the first magnitude, 
which reflects the greatest credit upon the 
investigators to whom it was due, and pro- 
foundly modifies our whole conception of an 
important sub-kingdom of plants.’’ 
Among the Mesozoic Cycadales, the genus 
Benneittites gains special prominence and impor- 
tance, not only because of the great perfection 
with which important structures are preserved, 
but because of the character of the fructification 
and the relatively high development attained 
by these plants. The stem structure agrees 
closely with that of the Cycadaceze, though 
representing a more primitive type. On the 
other hand, the fructifications of the two groups 
are totally different, and the greater complexity 
attained in the Bennettitese points to a consider- 
ably higher degree of development. Thus in B, 
Gibsonianus, the seeds, many of which are to 
be found in a remarkable state of preservation, 
are dicotyledonous and exalbuminous, while 
the whole character of the fruit approximates 
to that of an Angiosperm. Here again we 
observe a repetition of that parallelism in de- 
velopment between various branches of the 
phylogenetic tree which was so well exhibited 
in Cardiocarpus and Lepidostrobus, and which 
goes far to sustain the idea so frequently sug- 
gested throughout the plant world, that in the 
general progress of development, the various 
branches are all extending forward in the same 
direction, whence arises a parallelism which 
indicates approximation to, without actual con- — 
nection with, other yet higher phases of develop- 
ment; or, in other words, that deviation of a 
branch from the main line of descent involves 
certain inherent deficiences which, while per- 
mitting development in the same general direc- 
tion, impose a definite limitation of such a 
nature that the phylum is incapable of further 
variation, and hence can not extend beyond a 
limit which is always much inferior to that 
attained by the main line. 
The work throughout affords one of the best 
evidences among recent contributions of the 
