16 MESSRS, NEWELL ARBER AND J. PARKIN ON 
especially valuable in the consideration of the origin of the Angiospermous 
type of leaf, which we suggest was initiated by a change in the branch habit. 
We regard the Angiosperms as essentially a monophyletic group, the 
Monocotyledons having branched off from the Dicotyledonous stock at an 
early period, probably from the Ranalian plexus. In both these groups 
entomophily was a primitive feature. We consider that the change, from the 
assumed generally anemophilous habit of the Mesozoic Hemiangiospermeze, 
and the Bennettitese, to entomophily, by means of a shifting of the pollen- 
collecting mechanism from the megasporangium to the megasporophyll, and 
the consequent formation of an ovary, has supplied the “motive force," 
which not only called the Angiosperms into existence, but laid the foundation 
of their future prosperity. 
If these conclusions have weight, then it is now possible to trace back the 
line of descent of the Angiosperms to a very early geological period. This 
may be shown in tabular form as follows :— 
E . Mesozoic and Tertiary (Recent), — 
5. Angiospermea:........................ . 
i Eu-anthostrobilateæ. 
4. Hemiangiospermeæ .................. 
‚о. Mesozoie,— Pro-anthostrobilateze. 
(unknown fossils). 
3. Pteridospermes ..................... | . 
9 л . Palmozoic, —N on-strobilate 
2. Heterosporous Fern-like Ancestor. Ancest 
. Ancestors. 
1. Homosporous Fern-like Ancestor.. 
Numbers 1, 2, and 4 are unknown fossils, but the key to numbers 1 and 2 
is given by the Pteridospermem (No. 3), and to number 4 by the Bennettiteze. 
Numbers 3, 4, and 5 were Spermophytes. 
Such a theory of descent will permit us to venture rather further afield. 
In both the homo- and heterosporous, primitive, fern-like ancestors, there 
is every reason to believe that the sporophylls were arranged in a lax manner, 
and not aggregated into definite strobili. This condition still remained a 
feature of the fern-like seed-plants, or Pteridospermes. From this Palæ- 
ozoic plexus, however, strobilate lines of descent were probably evolved 
in Mesozoic times by two distinct methods. In the one, like sporophylls 
were aggregated into monosporangiate cones. In the other, both male and 
female sporophylls were massed in one amphisporangiate strobilus, the 
sporophylls however, for a time at least, retaining their primitive, fern-like 
form, as is clearly seen in the male organs of the Bennettitez. The mono- 
sporangiate Strobilateze led to the modern Cycads. This conclusion receives 
support from the fact that, in the genus Сусаз itself, only the male sporophylls 
are aggregated into cones. The female may be regarded as having remained 
more or less in their ancestral condition, especially with regard to their 
distribution on the axis. Such a сазе would be difficult to explain on the 
