196 BLECHNOID FERNS [CH. 



pinnae are much broader than in the normal species, but of about the same 

 length: the chief interest lies in the effect of the increasing breadth upon the 

 soral tracts. The effect near to the narrow apex of the pinna may be slight ; 

 and the parallel coenosori may be nearly normal. But passing downwards, 

 as the greater width is reached the coenosorus is seen to be thrown into 

 arches, more or less strongly curved, with or without disintegration. With 

 the more extreme width discontinuity is the rule, interruptions appearing at 

 the summit and base of each arch (Fig. 706, 28). The result of this is a series 

 of isolated and paired soral tracts facing one another: in fact the sorus 

 characteristic of the Hart's Tongue {Phyllitis scolopendrium) (compare 

 Fig. 706, 30). The fertile pinnae of the var. Krebsii have an open venation 

 which also corresponds to that of the Hart's Tongue, though this has a 

 much broader and entire blade. It is reasonable to suggest that so peculiar 

 and at the same time so similar a structure in the two Ferns has been 

 attained along similar lines of increasing breadth: and that it occurred in 

 both genera in a part which, like the pinna of Blechnum, has probably been 

 derived ultimately by condensation from a more elaborately branched an- 

 cestral leaf 



Before this elucidation can be accepted it will be wise to consider other 

 species than the common Hart's Tongue, and in particular those with 

 acuminate leaves. The apex of the leaf in Phyllitis scolopendrium is blunt, 

 but those of Camptosoriis rhizopJiyllus and sibiriciis are elongated into a 

 prolonged tip. If we examine this, its structure may often be found to 

 approximate very nearly to that of a Lomarioid pinna : for there is at the 

 distal end a simple coenosorus on either side of the mid-rib. It has been 

 shown how as the leaf widens downwards an outward arching with interrup- 

 tions of the coenosorus appears, and the details are very like those seen in 

 the var, Krebsii (compare Fig. 707, A, and Fig. 706). But as the leaf widens 

 still farther, the rather complex reticulate venation is seen to bear short soral 

 tracts, which at first sight appear irregular (Fig. 707, E). Examination shows, 

 however, that the vascular loops, which run closely parallel to the mid-rib, 

 bear tracts similar in position to those of Woodwardia: while the rest are 

 frequently paired as in P. scolopendrium. Lastly, in another species, viz. 

 Phyllitis nigripes^ the apical growth is arrested early, and the broadly 

 spathulate lamina is again reticulate: but many of the soral tracts are still 

 paired as before (Fig. 708). The conclusion thus appears justified that 

 Phyllitis, even in its most condensed and derivative leaf-forms, is a natural 

 genus sprung from a Blechnoid source, and that B. punctulattim var. Krebsii 

 gives a true key to its origin. In Phyllitis, if our comparisons be correct, the 

 "indusium" is not by origin a true indusium, but a part of the original 

 leaf-margin reflexed. 



In face of these comparative conclusions we shall ask. What is the relation 



