ii8 DRYOPTEROID FERNS [CH. 



species oi Alsophila are often tall Tree-Ferns, but apparently they have shown less ready 

 adaptation of the sorus. 



The result of this whole discussion, as to the origin of the '■^ indusiui/i viferum'' of the 

 Woodsieae and Cyatheaceae, is to show a reasonable chain of argument, favouring its being 

 of the same nature as the dermal appendages borne superficially upon the vegetative 

 organs. The actual source from which it is held to have sprung is from such basal hairs 

 as are a marked feature in the sori of Lophosoria, Metaxya, and Alsophila. The steps 

 suggested by the more rudimentary types of indusium seen in the Woodsieae are: (i) widening 

 of the hairs as flattened scales ; (ii) coalescence of them laterally after the mode of 

 development of a gamopetalous corolla. Thereafter a thickening by periclinal division of 

 the cells to form more than a single layer would produce the state seen in Cyathea. This 

 last is a feature not uncommon in dermal scales. A significant circumstance is that the 

 membranous indusium does not originate in Ferns which bear only dermal hairs (e.g. 

 Lophosoria, Meiaxya), but in those in which dermal scales are present (Cyatheaceae, 

 Woodsieae). But the appearance of an indusium seems to have lagged in the evolutionary 

 history behind that of the dermal scales of the vegetative region: in particular, AlsopJiiln 

 may have dermal scales without any indusium being present. This circumstance seems to 

 support rather than to oppose its suggested origin as the equivalent of those scales. 



From this discussion the final conclusion may be drawn that, however closely the 

 '■'■indusium inferum" of the Cyatheoids and their derivatives may resemble the indusium 

 of the Dicksonioids and their derivatives, the two appear to have been distinct in phyletic 

 origin. They are consequently homoplastic, not homogenetic parts. 



The whole question has been treated here at length, and the argument explicitly stated 

 because it gives a good example of the working of the phyletic method in Comparative 

 Morphology. It is quite clear, from the way in which certain results following from its use 

 have been received, that the method itself has not been fully realised, or the convergent 

 lines of argument followed to their legitimate conclusion. 



The general result of the facts here detailed or alluded to is, in the first 

 place, to demonstrate that the Woodsieae (excl. Cystopteris and Acrophorus) 

 constitute a very natural family, though the genera show some independence 

 of detail. Secondly, they confirm the relation of that family to the Cyathe- 

 aceae, both on the ground of the characters of the gametophyte and of the 

 sporophyte. They point also further to the Gleicheniaceae, and thus consoli- 

 date on grounds of detailed comparison the sequence of Ferns bearing 

 superficial sori as independent, and progressing by gentle steps from a 

 simple to a gradate, and finally to a mixed sorus. Moreover, this sequence 

 is coherent within itself, and takes its course independently of the marginal 

 series, as seen in the Dicksoniaceae and its derivatives ; though both advance 

 along similar but parallel lines. 



If the attempt be made to arrange the Woodsieae in sequence, there will 

 be no doubt that Woodsia itself stands nearest to the Cyatheaceae, as is 

 indicated by its gradate sorus, and its prothallial characters. Diacalpe, with 

 its mixed sorus and basal indusium, together with certain prothallial charac- 

 ters, takes a middle position; while Peranenia,\\\\)ci its lop-sided or zygomophic 

 sorus, mixed sequence of sporangia, curiously specialised basal indusium. 



