DENNSTAEDTIINAE 599 
whole sorus, and it is on this side that the annulus may be seen to show 
the last traces of continuity at the base. The orientation of the sporangia, 
as described, is not always maintained in the mature state; transverse 
sections of the mature sorus show some latitude in this respect ;~but this 
may be due merely to a twisting of the long thin stalk. This seems 
the probable explanation, since the young sporangia show a fairly accurate 
orientation. 
Sections of sporangia show that the number of spore-mother-cells in 
each sporangium is variable: eight, twelve, and sixteen have been observed. 
Countings of mature spores have shown that in certain cases the full 
number of 64 may be produced. 
Deviations from the basipetal succession have been observed in D. 
Zavallioides (Br.), Moore, and in D. rudiginosa (Kaulf), Moore, Even in D. 
apiifolia isolated cases have been seen of a sporangium seated near the apex 
of the receptacle, apparently arrested in its development, and representing 
a less advanced state than those surrounding it. In D. davallioides, cases 
of this sort are of fairly common occurrence towards the apex of the 
rather elongated conical receptacle. These may be held to be transitional 
forms to what is seen in the allied species D. rudiginosa, the sorus of 
which is represented in longitudinal section in Fig. 332 4s,8. Here, upon 
a relatively short receptacle, and between indusial flaps which are also 
short, the sporangia are disposed with no definite succession; the 
persistent stalks of two old sporangia are near the apex of the receptacle, 
and the younger sporangia below, with spores and tetrads, give a slight 
indication of a basipetal succession, but still younger sporangia are disposed 
irregularly among them. The sorus, though showing some slight signs 
of a basipetal succession at first, is clearly of that type which we shall 
designate “mixed,” that is, with the sporangia produced in no definite 
succession, but the younger interspersed irregularly among those which are 
more advanced; correlated with this we find the receptacle short, but 
wide. The sporangium also shows a difference from D. apzifolia, the 
annulus being definitely interrupted at the insertion of the ‘stalk as is shown 
in Fig. 332 4¢s,D, which represents a dehiscent sporangium seen from 
below, (sz) indicating the insertion of the round stalk. We shall subse- 
quently see that, by these gradations in Denmstaedtia, a transition is indicated 
between two fundamental types of soral arrangement, the Jdasipetal 
succession of sporangia, and the mzxed, where the various ages are irregularly 
interspersed. Dennstaedtia is thus found to correspond to Microlepia in 
the occasional departure from the typical basipetal succession of the 
sporangia in the sorus: this condition leads on to that seen in the 
genus Davadlia. 
The other genera included in the Dennstaedtiinae by Prantl are 
Leptolepia, Saccoloma, and Hypolepis. The first two of these are closely 
related to Microlepia, as their numerous synonyms show: many of 
their species have indeed been included in that genus. yfolepis is 
