184 
Palaeontologie. 
sporophylls and megasporangia, are to be found in the first two 01 
the above mentioned Orders. It is shown how syncarpy has arisen 
from the primitive apocarpy. The replacement of the spiral grou- 
ping of the organs of the cone by the verticillata, is due partly to- 
a tendency to cohesion and adhesion, which has always been 
marked among the Angiosperms, and partly to a proneness to a 
dissimilarit}^ in the size and shape of the different organs of the 
strobilus. In Magnoliaceae the androecium is primitive. The stamen 
is regarded as a sporophyll bearing two synangia. The perianth is 
an ancient organ which was completely differentiated from the spo¬ 
rophylls on the one hand, and from the foliage leaves on the other, 
before the existing Angiosperms came into being. The perianths of 
the Magnoliaceae and Ranuncnlaceae are described in this connection, 
and three possible origins are discussed by which the perianth in 
the higher cohorts of the Dicotyledons may have been derived. 
The second portion of the paper relates to the fossil evidence. 
It is shown that the fructification of the Bennettiteae combines many 
of the peculiarities which the authors regard as primitive among 
living Angiosperms. A short account of the amphisporangiate stro¬ 
bilus of Bennettites, recently described by Wieland, is given, 
and the previous interpretations of its structure are reviewed. The 
authors conclude that the cone of Bennettites is a simple strobilus,, 
and not an inflorescence. Its parts are homologous with the carpels , 
stamens and perianth of a typical amphisporangiate Angiospermous 
flower. It is a Pro-anthostrobilus, differing from the Eu-anthostrobilus 
or flower of the Angiosperms in the absence of a seminal eollection- 
mechanism, and in the form of the microsporophylls. This Interpre¬ 
tation has all the merit of simplicity, but it also involves certain 
difficulies, especially as to the homologues of the interseminal scales 
of the Bennettiteae, which are discussed. 
The Authors conclude that the Tertiary and Recent Angiosperms 
are directly descended from a group of Mesozoic plants, as yet 
entirely hypothetical, to which they apply the new term Hemiati- 
giospermeae. The cone of these ancient plants so closety approxima- 
ted to the Pro-anthostrobilus of the Bennettiteae, that the latter, 
although somewhat removed from the direct line of descent, demon- 
strate emphatically the type of strobilus that gave rise to the An¬ 
giosperms. It was essentially a Gymnospermic fructification, the 
pollen-collection being performed by the ovule itself. Yet it agreea 
with the typical flower of the Angiosperm on the one hand, and 
with the strobilus of the Bennettiteae on the other, in the juxtapo- 
sition of the mega- and microsporophylla (a feature which is peculiar 
to the cones of this line of descent) as well as in the possession of 
a primitive perianth. It differed from the Bennettitean strobilus in 
that the megasporangia were seated on the margins of the carpels r 
(the homologues of the interseminal scales) which were free from 
one another, and not united at the apex. Also the microsporophylls 
were spirally arranged, and perhaps more reduced than those ol 
that group. Such a strobilus would be all but Angiospermic, were 
it not that the task of pollen-collection was still performed by the 
ovule, and that it lacked the precise form of microsporophyll which 
is called a stamen. The general form of megasporophyll would cor- 
respond more closely with that of the living genus Cycas, than 
with the corresponding structure presented b}^ the known Bennet¬ 
titeae. A figure is given showing the structure of the hypothetical 
Pro-anthostrobilus of the Hemiangiospermeae. 
