Bower.—Studies in ' the Phylogeny of the Filicales . VII. i 3 
the Pterideae in which there is only one indusium, and in which there is no 
reason to believe that there has ever been a second or inner indusium in the 
course of descent. A central typical genus of the latter is Cheilanthes , which 
gives the proposed name to the series. The present memoir will treat only of 
the first, which owe their phyletic origin to some source in which the sorus 
was two-lipped, as in the Dicksonioid Ferns ; and the Cheilanthes series 
will be held over for the present. 
It is impossible to study the Pterideae satisfactorily within the ring- 
fence provided by Pranth The area of comparison upon which their 
grouping is to be founded must include other Ferns as well, such as 
the Dicksonia-Davallia series. On the basis of a curious over-estimate 
of the receptacle as a diagnostic character, Prantl separated the Denn- 
staedtiinae and Davalliinae from one another, and from the Pterideae. 
It has been shown 1 that the form and construction of the receptacle are re¬ 
lated to the arrangement of the sporangia. Basipetal sori have a convex 
vascular receptacle, and that is seen in Dennstaedtia and Microlepia ; mixed 
sori have commonly a flat receptacle, and that is seen in Davallia and many 
Pterideae. But such differences are no bar to a relationship which was 
probably close between the Dennstaedtiinae and Davalliinae and the 
Pterideae, while all of them converge towards the Dicksonioid type. Such 
Ferns as these must therefore come into the circle of comparison, and most 
important of all for our present purposes, Lindsaya and Saccoloina , on which 
a few notes may here be given by way of introduction to the Pterideae 
themselves. 
Anatomically the genus Lindsaya stands on a relatively primitive 
footing. Its typical stelar structure is now known for a large number of 
species, and it may be held as characteristic for the genus, though possibly 
some exceptions may exist. 2 The Lindsaya- type of stele has now been 
observed in the following species, to which the initials of the observers have 
been attached: L . rigida (A. G. T .), gnianensis (A. G. T.), scandens 
(A. G. T.: G.-V.), linearis (G.-V. : F. O. B.), stricta (G.-V.), orhiculata 
(A. G. T.: G.-V.), cuneata (G.-V.: F. O. B.), reniforme (G.-V.), Fraseri 
(G.-V.), decomposita (A. G. T.: G.-V.), davallioides (A. G. T. : G.-V.), 
ensigolia (G.-V.), divergens (G.-V.), heterophylla (G.-V.), microphylla (G.-V.: 
F. O. B.), lancea (A. G. T. : F. O. B.), clavata (F. O. B.), cidtrata (F. O. B.: 
non G.-V.). Its uniformity in these eighteen species suggests that it is a good 
generic character. There are, however, divergences of observation, depend¬ 
ing perhaps on identification of material. Gwynne-Vaughan described for 
L. cidtrata a solenostele with divided leaf-trace. But my own material 
from Singapore shows typical Lindsaya- structure for this species. Probably 
1 Phil. Trans., vol. cxcii, 1899, pp. 91-5. 
2 Tansley and Lulham, Ann. of Bot., vol. xvi, 1902, p. 157; Gwynne-Vaughan, Ann. of Bot., 
vol. xvii, 1903, p. 689; Tansley, Lectures on the Filicinean Vascular System, 1908, p. 46. 
