64 MESSRS. NEWELL ARBER AND J. PARKIN ON 
The faet that such a cone appears to be wholly unknown at present 
should not militate against the theory, if we bear in mind that the total 
number of Mesozoie fructifications of Gymnospermous affinity at present 
discovered is extremely small, as Wieland has emphatically pointed out in 
the passages quoted above. 
It might be asked why we have not adopted Saporta's * term, Proangio- 
spermer, if some such name is really required. It must, however, be 
remembered that this name was given to fossils, which were regarded as 
primitive Angiosperms, combining characters common to both Dicotyledons 
and Monocotyledons, whereas the hypothetical forms, which we are discussing, 
were the ancestors of these primitive Angiosperms, and were Gymnosperms. 
Further we do not agree that the Bennettitese, in the light of the recent 
researches of Wieland, can be referred to the Ргоаполозрегтег Tf of Saporta, 
as the latter author concluded, for the same reason that their mode of 
fertilization was essentially Gymnospermie $. 
THE ORIGIN OF THE ANGIOSPERME/E. 
We may now proceed to outline the steps by which the typical strobilus of 
the Angiospermez was evolved from that of the hypothetical Hemiangio- 
spermez. We have already (pp. 44—45) indicated what we regard as the 
primitive form of the various organs which compose the flower. 
The amphisporangiate cone of the Bennettiteze was identical, so far as the 
juxtaposition of the mega- and microsporophylls is concerned, not only with 
that of the Hemiangiospermez, but also with that of the Angiosperms them- 
selves. The hypogynous arrangement of the -parts, as in the Bennettiteze, 
was also a primitive feature of the Angiosperms, from which the perigynous 
and epigynous states have been more recently evolved, as indeed has been 
pointed out by several writers §. 
In the cone of the Bennettitez, all the organs are spirally arranged with 
the exception of the microsporophylls. In the cyclic grouping of the latter, 
these plants may be regarded as showing evidences of an early departure 
from the main line of descent of the Angiosperms. In the strobilus of the 
hypothetical Hemiangiospermeze, the organs were all arranged spirally (see 
fig. 4), and this primitive feature is still to be found preserved to some extent 
* Saporta & Marion (1885) vol. 1. pp. 220 & 222, 
T This term is also open to the objection that many fossils have been included under it, 
the nature or affinities of which are wholly doubtful. 
f The term *Angiocycad, provisionally suggested by Oliver (1906) p. 240, does not 
appear to us to be free from objections, for we regard the fructification of this ancestor as, in 
the first place, Gymnospermie, and, in the second, very far removed from that of the living 
Cycad. 
$ Coulter & Chamberlain (1904) p. 15. 
