48 Bower.—Shcdies in the Phytogeny of the Filicales. VII. 
required. A specimen of a young pinna of a narrow type from Martinique, 
collected by Hahn, 1868, showed at the distal end a purely P teris-struc- 
ture (Fig. 35, a), with well-marked indusial flap, a narrow receptacle 
traversed by a vascular commissure, and the relatively few sporangia 
strictly limited in their insertion to the receptacle. Lower down the 
receptacle itself widens if), and the sporangia are at first still restricted to 
it; but lower down they encroach in their insertion upon the free surface 
of the pinna (< c). This may extend in broader-leaved examples fully half¬ 
way to the midrib (Fig. 36) ; or indeed the whole distance, as in Hooker’s 
Plate 58. It will be noted from the surface view that the indusium is 
continuous, and destitute of vascular tissue, and that the vascular com¬ 
missure, which is continuous, is considerably widened, but that the soral 
area extends far beyond it. Thus 
it is not merely a widening of the 
receptacle, as in P. podophylla , but 
an actual overflow on to the leaf- 
surface. 
Such comparisons are obviously 
insufficient without anatomical and 
other evidence. Unfortunately it 
is improbable that material of A. 
praestantissimumwiW be available at 
present. A. aureum has been ex¬ 
amined by Miss Thomas 1 and by 
Frau Schumann. 2 Neither of these 
accounts is wholly satisfactory, and the matter is being reinvestigated by 
Dr. Thompson, with a view to comparisons with such types as Saccoloma 
and P. podophylla. 
Fig. 36. Margin of a fertile pinna of Acro- 
stichum praestantissimum, showing in surface view 
the indusium, receptacle, and soral area. ( x 10.) 
Conclusions respecting the Genus P ter is. 
The genus Pteris , in the old comprehensive sense of the ‘ Synopsis 
Filicum included a number of sub-genera. Some of these are now 
distinguished as substantive genera, as indeed they had been previously 
by certain writers. They, together with the genera Lonchitis and Actinio- 
pteris , are associated in Christensen’s Index (pp. xliii-xlv) under ‘VI. 4. 
Pteridinae’. With them should now be associated phyletically the genus 
Acrostichum in the original sense of Linnaeus, now adopted by Christensen 
(l.c., p. 4). Possibly some others will be added when the details necessary 
for phyletic conclusions have been sufficiently worked out. These Ferns 
appear to constitute a natural group, which probably have had a common 
ancestry, though this was not recognized by the older systematists who 
classified them together. They have probably all originated from an 
1 NewPhyt., vol. iv, 1905. 2 Flora, 1915, p. 208, &c. 
