5°4 
Arber and Parkin . — Studies on the 
however, on the view that the stamen of the Angiosperms is a £ sporangio- 
phore bearing two synangia \ 1 
As regards the evolution of these organs we are inclined to derive them 
by reduction and fusion from more complicated sporophylls bearing many 
synangia. Such sporophylls are known in Bennettites , and were also 
characteristic, as we have shown, of the hypothetical Hemiangiosperms, 
the direct ancestors of the Angiosperms, though in a more reduced form . 2 
Wieland 3 has pointed out that there is no ‘ unbridgeable gap between 
the staminate disk of Cycadeoidea and that of Welwitschia ’. This explana- 
tion will also be found to account for the variation in the number of the 
synangia, and of their component sporangia, on the basis of a progressive 
reduction, which in the case of Gnetum has reached its furthest limits. 
The Megasporangium. 
The morphological interpretation of the female fructification is as 
puzzling as that of the male. There is only a single megasporangium 
(ovule) in all three genera, which is situated terminally, and is surrounded 
by two, or in Gnetum by three, envelopes. We have already concluded 
that the outer envelope is in reality a perianth. In Ephedra and Welivit- 
sckia } the inner is evidently integumentary, as it has been generally regarded. 
In Gnetum , however, there is a further difficulty as to the homology of the 
middle envelope. The innermost corresponds so exactly to the inner 
covering of Ephedra and W elzvitschia in its adaptation as a spout-like 
micropyle, that there is every reason to consider it as homologous. 
We thus adopt the usually accepted conclusion, that none of the 
envelopes are equivalent to a megasporophyll (carpel). That some sort of 
foliar structure bearing the megasporangium was present in the more primi- 
tive Gnetaceae appears to us to be highly probable. The apparently 
terminal, ‘ cauline ’ ovule possessed by the three existing genera is not, 
however, a point which causes us hesitation. We agree with Goebel , 4 
as opposed to the older school of morphologists, in regarding leaf-borne and 
axis-borne organs as not necessarily of a different morphological value, 
when in their other peculiarities they appear to be similar. It therefore 
follows that it is quite legitimate to trace phylogenetically such differently 
situated organs to the same place of origin, and to follow ‘ the path by 
which this axial position has been acquired ’. 
It must be remembered that the megasporophylls of the Gnetaceae 
were never so fundamentally essential as the Angiospermous carpels, for in 
the former group the pollen-collection was in all probability always per- 
Arber and Parkin (’07), p. 68. 
Wieland (’06), p. 245. 
2 Ibid., p. 63, Text-fig. 4. 
4 Goebel (’05), part ii, p. 556. 
