604 FILICALES 
of the spores produced from each sporangium. If in a sorus of the type 
of Gleichenia dichotoma the receptacle were elongated to receive a basipetal 
succession of sporangia, which retained their form, but showed a diminished 
spore-output, and lateral dehiscence in accordance with their basipetal 
sequence, the sorus of Ad/sophila would be 
the result. 
Such changes are inherently probable, 
and it has been seen in the sorus of 
the Hymenophyllaceae how the greater 
number of sporangia goes along with a 
fall in their individual productiveness. 
This is carried further in Cyathea than in 
Alsophila, for there the sporangia ‘are 
smaller, and the output in C. dealbata 
may fall as low as 16, or even 8 spores 
per sporangium, though in C, medullaris 
the number may remain at 64. The 
development of the sorus in this genus 
Cyathea dealbata, Sw. The upper figure pas ae been solewad so CNHs in 
shows a very young sorus, with receptacle essential point from that of Alsophitla, 
and indusium already indicated. The lower i ‘ 
shows the indusium (é) more advanced, and excepting in the presence of the basal 
the sporangia s, s, arising in basipetal ‘ , i 
succession. X 200, indusium, which appears before any of the 
sporangia (Fig. 335). The inconstancy of 
occurrence of the indusium in a group of closely related plants indicates 
clearly that, however large, or early in appearance, or biologically important 
it may be, it is not to be held as an essential part of the sorus, nor trust- 
worthy as a phyletic character. 
Fic. 335- 
ANATOMY. 
Py 
Anatomically the Cyatheae show very great complexity of structure, 
though it can be referred, even in the most complex examples, by comparison 
to a simpler source: the conclusions are, however, rendered less certain 
by the lack of graded intermediate conditions. A relatively simple state 
was found by H. Karsten! in the western species A/sophila pruinata, a 
Fern which grows with an upright stem some three feet or more in height. 
In transverse sections of the axis a solenostelic structure is seen, which 
opens here and there with a foliar gap, from the margin of which the 
leaf-trace is given off, apparently as a simple strand, with the usual horse- 
shoe-like transverse section : after leaving the axis the leaf-trace soon breaks 
up into a number of strands. As the internodes are of perceptible length 
the leaf-gaps do not overlap, and the stele often appears as a complete ring 
(Fig. 336). A peculiar feature is seen in this Fern in the leafless 
runners, which originate below the leaf-bases, and grow like roots downwards 
into the soil: it is interesting to note that they have at first a solid stele, 
1 Vegetationsorgane ad. Palmen., p. 123. 
