1922] BENSON—HETEROTHECA GRIEVII 137 
should be homologous is undoubtedly the simplest explanation of 
their origin. ‘They both separate from the frond, and in this respect 
differ from later forms, where, on the attainment of dry dehiscence 
in the microsporange, the pollen escaped before the fall of the 
sporange. The indusium or cupule which surrounds the ovule as 
long as it remains on the parent plant is one of the distinctive 
features of microsporange and ovule, for there is no evidence of an 
indusium surrounding Heterotheca. ‘This difference, however, is 
really confirmatory of the synangial origin of the inner integument, 
as such an origin does not involve a double indusial formation in a 
relatively primitive ovule. On grounds such as these, it may be 
assumed that in the particular case of Heterangium Grievii its 
microsporangia and ovules are homologous, both being transforma- 
tion products of a common ancestor. 
Those who accept this thesis will undoubtedly be prepared to 
accept it for all the other Lagenostomales, such as Lagenostoma, 
Conostoma, and Physostoma. A necessary corollary as respects 
Physostoma is that there has been a meristic variation in the 
periphery. The bundles are approximately, or we might say, 
potentially twice as numerous as in the periphery of such a spo- 
range as Heterotheca, and each dominates one radially symmetrical 
“tentacle” instead of a dual compartment as in Spharostoma (text 
fig.8,a,d). Thusinstead of regarding Physostoma as a Lagenostoma 
with unfused chambers (OLIVER 13), we should, in the light of 
Heterotheca, regard it as a Lagenostoma with a lobed canopy. 
Turning from the Lagenostomales to the Trigonocarpeae, 
which are regarded as the ovules of some at least of the Neurop- 
terideae, we may ask if they bear any internal evidence of origin 
from such a type of synangium as Heterotheca. They show obvious 
series in the structure of the integument, series which in some 
respects run parallel to those in the Lagenostomales. Excellent 
details of the integument will be found in SALIsBuRY’s (14) work 
on Trigonocarpus Shorensis. In the Trigonocarpeae the nucellus is 
free from the inner integument except in the plane of its basal 
attachment, through which runs a vascular bundle which eventually 
forms a tracheal envelope round the nucellus. Six other bundles 
enter the integument and travel in its periphery. Polylophosper- 
