SPORE-PRODUCING MEMBERS 635 



This elongation and structural adaptation of the receptacle becomes much 

 more pronounced where there is a continued succession of sporangia, as 

 in the Gradatae. The basipetal sequence seen in the Cyatheae, Dicksonieae, 

 Dennstaedtiinae, Hymenophyllaceae, would hardly be possible without a recep-- 

 tacle having intercalary growth : the continual moving upwards of the more 

 mature sporangia, so that they can freely shed their spores, leaves space 

 for the initiation of younger ones below, while the latter are in their early 

 stages close to the source of supply, and protected by the basal indusium. 

 This seems to be the raison d'etre of the elongated receptacle in such 

 cases : given a basipetal succession, its origin has been a response to the 

 need of space, by upgrowth from the base, not the result of " metamorphosis " 

 of any pre-existing vegetative part such as a lamina of a leaf (Prantl). 



An indirect support for this opinion is found in the fact that the 

 receptacle disappears in those Ferns where the mixed sorus is acquired ; 

 for here the receptacle is commonly flat, though some exceptions do 

 occur. A peculiar interest attaches to those closely allied Ferns of the 

 Dicksonia-Davallia affinity, in which a transition from the basipetal sequence 

 to the mixed sorus is believed to have taken place. In Dennstaedtia and 

 Microlepia the receptacle is clearly conical, though it does not show an 

 elongation at all comparable to that of the Hymenophyllaceae (Fig. 332). 

 Occasionally in Denn. davallioides and in Microlepia hirta, but more com- 

 monly in Denn. rubiginosa, the strict basipetal succession which is seen in the 

 first stages of the sorus is departed from, though the receptacle still maintains 

 in some degree its conical form. But in Eudircallia, which is without doubt 

 closely allied to the above, the sorus is a mixed one, and the receptacle 

 is almost flat, though still provided below with a considerable mass of 

 tracheides (Fig. 339). Here, since the basipetal succession is not maintained, 

 an elongated receptacle is not required, and since this difference occurs 

 between Ferns which are certainly of a common stock, it is probable that 

 a reduction has taken place. Accordingly, an elongated receptacle is not 

 in itself an important morphological feature ; it is to be correlated with a 

 continued basipetal succession of sporangia, and it is this last which is the 

 essential morphological feature in such cases. 



There are, however, instances where there is an enlarged receptacle on 

 which the sporangia are borne not in basipetal succession, but in mixed 

 order ; as an example, the familiar sorus of Nephrodium may be quoted, 

 with its large receptacle and internal mass of tracheides (Fig. 4). It may 

 be urged that these facts are inconsistent with the view expressed in the 

 last paragraph, but the large receptacle may here be a survival, which, 

 though the sorus has become a mixed one, may still be a convenience ; 

 as it certainly is in Nephrodinm, where its size allows space for the 

 sporangia between the indusium and the leaf-surface. The general view 

 may then be formulated thus : the Simplices have as a rule a receptacle 

 of small size ; the Gradatae have a more or less elongated receptacle, and 

 it may attain very considerable dimensions where the succession is long 



