62 MESSRS. NEWELL ARBER AND J. PARKIN ON 
have already been recognised, on many sides, indications of Angiospermous 
affinity, and thus support is afforded to the theory maintained here. 
Only one discussion of the affinities of the Bennettitean strobilus has 
appeared, so far as we are aware, since Wieland’s full results were disclosed last 
autumn. Professor Oliver *, in a short paper on this subject, has stated some 
of the main points of the argument, to which we had arrived independently. 
He says: “We now come to the question of the morphological interpretation 
of this fructification, whether it is to be regarded as ‘an axis beset with 
sporophylls,’ 2. e. a flower, or whether, on the other hand, it is really a much 
more complex structure, i. e. an inflorescence or branch-system showing 
extreme reduction. . . . .. The view taken by Dr. Wieland, that we have 
here a hermaphrodite flower, wil meet with very general agreement. 
Looked at broadly and having regard to the pteridospermous affinities of the 
Bennettitese this interpretation seems irresistible. To take the other view 
and read a * cyathium? into its structure seems to verge on the gratuitous.” 
The same author adds: * Whatever else one may think of this flower 
it cannot be regarded as that of а quite typical Angiosperm. . . . . Its great 
interest and value seem to be that whilst just missing the Angiosperm 
it shows how close the Cycad line could come to realising it. It is indeed 
the key to the Angiosperms ; when that is recognised the rest is easy .... 
It is possible, no doubt, though it seems almost incredible, that a flower with 
perianth, stamens, and gynzeceum in proper relative position as in Cycadeoidea 
should have been produced except in a line very closely related to that which 
led to the Angiosperms.” 
THE HEMIANGIOSPERME K. 
According to our view, the Tertiary and Recent Angiosperms are directly 
descended from a group of Mesozoic plants to which we apply the new term 
Hemiangiospermec. This group at present is entirely hypothetical. N othing 
is known as to the fructification of any of its members, but we believe that 
its cone approximated so closely to the pro-anthostrobilus of the Bennettitex, 
that the latter, although somewhat removed from the direct line of descent, 
demonstrates emphatically the type of strobilus which gave rise to the eu- 
anthostrobilus, or flower of the Angiospermez. This cone (fig. 4), like that of 
the closely related Bennettiteze +, was ап anthostrobilus of the pro-anthostro- 
biloid type. It was also essentially a Gymnospermic fructification, the pollen 
collection being performed by the ovule itself. — Yet it agreed with the typica] 
flower of the Angiosperm on the one hand, and with the strobilus of the 
Bennettitee on the other, in the juxtaposition of the mega- and micro- 
sporophylls, a feature which is peculiar to the cones of this line of descent, 
* Oliver (1906) pp. 239-40. 
T Hallier (1901 !) p. 105, (1905) p. 154 
