24 DAVALLIOID FERNS [CH. xxxvii 



(iii) Linkage of sori to form Coenosori 



A special interest attaches to the progression leading from primitively 

 separate sori to the fusion-sorus, partly because it is a change of this nature 

 that accounts for the soral state of the whole series of the Pteroids; but also 

 because there is ample evidence that the coenosorus has originated poly- 

 phyletically among the derivatives of the marginal Dicksonioids. Already 

 a partial linkage of sori has been described for Saccoloma (Vol. II, p. 272, 

 Fig. 541, A): but numerous and more complete examples are presented by 

 certain Davallioid derivations, such as Ncphrolepis and Diellia, and these 

 culminate in the condition seen in Lindsaya and Dictyoxiphium. The lateral 

 fusion may involve not only the indusia, but the vascular supply is also 

 connected up by commissures parallel to the margin of the blade, while 

 sporangia may be seated upon these, so that the receptacle itself becomes 

 continuous along the whole coenosorus. In the examples which follow, the 

 linkage appears to be in close relation to condensation of the blade, marking 

 a transition from a highly cut towards an entire form. 



Tapeinidium (Presl, 1849) C. Chr. 1906 



This genus comprises four species from the Malayan region. It was founded 

 to receive the Fern described in 1802 by Cavanilles as Davallia pinnata, 

 but the species has passed under many synonyms, such as Wibelia, Saccoloma, 

 Microlepia, Lindsaya, and even Dicksonia. This at once draws attention to 

 it as a probable synthetic type among Ferns, with sori seated at or near to 

 the margin. T. pinnatum (Cav.) C. Chr. is the species upon which detailed 

 observation has been chiefly based. It is a Fern with a creeping rhizome 

 which frequently dichotomises. The leaves are alternate, and simply pinnate, 

 though the pinnae are sometimes forked, and may also be cut into separate 

 pinnules at the base, thus indicating a relation with Ferns of more complex 

 pinnation, from which it is probably a derivative. A type named T. pinnatum 

 (Cav.), var. tripinnatnin Rosenstock, from New Guinea, supports the idea of 

 simplification from a Dennstaedtioid type of foliage. The rhizome is described 

 by Diels as hairy (E. and P. I, 4, p. 216): but Prantl {Syst. d. Fame, p. 16) 

 ranks it with his Davalliinae, with the words "Haare stets Zellflachen." The 

 specimens at Kew show dark brown ramenta. 



The anatomy as described by Gwynne-Vaughan is very similar to that 

 of Odontosoria aculeata. Below each leaf-insertion there is a pocket that 

 fades out downwards, giving a Lindsaya-iXxwcXxxxQ till the next leaf-base is 

 reached (Fig. 594). This imperfect state of solenostely compares closely with 

 that of Lindsaya and Odontosoria, linking these forms with those that are 

 fully solenostelic. As in these also the origin of the leaf-traces is marginal. 



