66 MESSRS. NEWELL ARBER AND J. PARKIN ON 
Wieland * gives no clear information. But the fact that the integument, 
enclosing the micropyle, is produced f beyond the united outer surfaces of the 
interseminal scales for some 2 mm., or, as Wieland describes it, “ projects 
stigma-like a little beyond the pericarp,” seems to show that the latter, what- 
ever their homologies may be, played no part in the pollen-collecting 
mechanism, a duty no doubt performed by the ovule itself. This also lends 
probability to the view that pollination was effected by means of anemophily. 
These points appear to us to be of great importance, since we regard the 
likelihood that, in the Bennettiteæ and the Hemiangiosperme, the pollen- 
collecting office was performed by the ovule itself (as in the Coniferales and 
the Pteridospermex), and the additional probability that the microspores 
were brought into position, as it were, by the wind, as being two features 
eminently characteristic of these groups, as opposed to the Angiosperms. 
With regard to the precise homologies of the seed-pedicels and interseminal 
scales of the Bennettitean fructification, there are already several theories in 
the field. Lignier's conclusions have been mentioned (p. 56) and Wieland's { 
views will be found discussed at length in the IXth Chapter of his book. We 
do not intend to pursue the matter at length here. On our view, the homologues 
of the interseminal scales of the Bennettiteze in the cone of the Hemiangio- 
spermeæ were simple earpellary leaves, bearing several ovules on their margins, 
much like the megasporophylls of the living genus Cycas. We conceive 
that the ancestors of the Bennettiteze themselves also possessed this type of 
sporophyll, though in Bennettites this structure has become highly modified, 
perhaps even divided ; for there is a possibility that the seed-pedicels may, 
in part, represent a lobe of the carpellary leaf. Also the megasporophylls, 
or portions of them, have become united to form the pericarp of the fruit. 
In these features we recognise clearly that the strobilus of the Bennettiteze 
departs considerably from the lines along which the Angiospermes have 
descended. The evolution of the pericarp of the Bennettitez represents one 
path of advance, and one wholly gymnospermic. On the other hand, the 
Angiospermee, with their closed carpels, form another line of descent, called 
into being by the adoption of the entomophilous habit, in conjunction with a 
shifting of the mechanism for pollen-collection from the megaspore itself to 
the closed megasporophylls. 
Since the adoption of entomophily, by means of closed carpels, as the 
mode of fertilisation, evolution has taken place in many different directions, 
and thus the great cohorts, families, and orders of Angiosperms have been 
called into being. Among the more important changes have been reduction 
and suppression in the number of the floral members, leading in extreme 
cases to the monecious and diccious conditions, often, as we have pointed 
L* Wieland (1906) p. 122, fig. 63, t Wieland (1906) pp. 121, 234. 
1 Wieland (1906) p. 230, «е, 
