136 BOTANICAL GAZETTE [OCTOBER 
The following is a more concise statement of the features of 
resemblance between Heterotheca Grievii and various ovules and 
microsynangia regarded as homologous with it. For the better 
elucidation of the subject, text fig. 8 has been constructed showing 
the possible numerical relations, etc., between the various sporangia 
of the Pteridosperms. We will commence with a comparison of 
Sphaerostoma and Heterotheca. 
Heterotheca Grievii resembles Sphaerostoma ovale in the follow- 
ing particulars. It is approximately of the same form and dimen- 
sions (length 3.5-4 mm., width averaging about 2mm.). It has 
probably six peripherally placed bundles in the buttresses between 
every two loculi as seen in transverse section. Besides these, there 
are two in the foci of the ellipse of the cross-sectional area, making 
eight in all, as in Sphaerostoma, although in the latter all eight are 
peripheral. The bundles are accompanied by enlarged water 
storage elements, and branch as in Sphaerostoma (5), fig. 34. 
There is a central fertile region in each. In Heterotheca there is 
a large amount of vegetative tissue in the region between the 
loculi, forming buttresses vertical to the surface similar to those of 
the canopy of Sphaerostoma. 
The differences can be explained by progressions of well known 
type. Thus the overarching of the central region by the peripheral 
at the apex to form the micropyle and sinus may have been partly 
in relation to the necessity of harboring the pollen grains, and partly 
a direct result of the freer vegetative development of the periphery. 
The later segmentation of synangia is a familiar phenomenon in 
Pteropsids, as may be seen by a comparison of Angiopteris and 
Marattia. The regular circumscissile dehiscence of the pollen 
chamber may be reminiscent of the time when there was a whorl 
of central loculi such as still exist in Heterotheca. 
That there should be so few differences and so many resem- 
blances between the microsporange and the ovule of the same species, 
makes it impossible to homologize merely the nucellus with the 
microsporange; as already stated, we must accept the synangial 
origin of the ovule as a whole, and regard the nucellus as derived 
from the central part of the common ancestor of both Heterotheca and 
Sphaerosioma. ‘That the two structures, borne by the same species, 
