HYMENOPHYLLACEAE 585 



( T. scandens) : there is reason to think that these, being more specialised 

 in their general habit, have undergone secondary modification also of 

 their stelar structure. 



The petiole receives in all cases a single strand, which passes off from 

 the stele without a leaf-gap, and widens out upwards into a collateral 

 structure with more or less clearly curved xylem. 



The comparison of this structure with that of other living Ferns leads 

 to the recognition of no near relation. Perhaps the nearest is with 

 the ancient family of the Botryopterideae, and especially with the genus 

 Zygopteris. It will be seen that the correspondence is closest between 

 the structure seen in Zygopteris Grayi (Fig. 270, p. 500), and that seen in 

 T. reniforme and H. scabrum, that is, with the species which show the 

 nearest resemblance to one another of the living forms of Hymenophyllaceae. 

 The chief features of resemblance are in the structure of the stele, and 

 the mode of origin of the leaf-trace, and of the supply to the axillary 

 bud. As to the former, if the small central tracheides of Zygopteris are 

 protoxylem, as seems highly probable, the agreement with T. reniforme, 

 or the larger Hymenophyllums is very close, the differences being such as 

 would be due to a fths arrangement of the leaves on the one hand, and 

 distichous on the other. The mode of origin of the leaf-trace without 

 any foliar gap, and the insertion upon it of the supply to the axillary bud 

 are also points of similarity. This resemblance to a very ancient form 

 appears to confirm the recognition of the living species named as being 

 probably primitive, while from that central point the remaining species 

 of Hymenophyllum diverged slightly in one direction, but those of Tricho- 

 manes diverged much more strongly along their own lines, either of reduction 

 or of other specialisation. Such a conclusion appears to emerge clearly 

 from the anatomical comparison. 



The Hymenophyllaceae have undergone vicissitudes of classification : 

 Brongniart first separated them as a special family, though the name of 

 Endlicher is usually connected with their recognition as ranking on an 

 equal footing with the Cyatheaceae or Polypodiaceae. Presl regarded them 

 as a connecting link between Mosses and Liverworts, and somewhat far 

 removed from the Filicineae ; but the actual separation of them from the 

 Filicineae was opposed by Mettenius. The comparison of their vegetative 

 structure with that of certain Bryophytes was, however, maintained, and 

 strengthened by fresh observations : it was pursued subsequently by Prantl, 

 with regard to the sporophyte, and especially to the sorus, and by Goebel 

 as regards the gametophyte. Consequently they were held to illustrate 

 the phylogenetic connection between Mosses and Vascular Cryptogams, 

 and to bridge over the gap between these circles of affinity in the sequence 

 of development of the Archegoniateae. Further, the Hymenophyllaceae have 

 been held to be at least as near to the Polypodiaceae as to any other family 

 of the homosporous Leptosporangiateae : in accordance with such views we 



