NOTES ON NEUROSORIA PTEROIDES. 55 



the evidence points to the Cheilanthince as the section of 

 the Pteridece to which Neurosoria belongs, and to Cheil- 

 anthes and Hypolepis as its nearest relations. That is to 

 say, that Neurosoria must find a place among the Cheil- 

 anthince with thickened nerve-ends. For, to return once 

 more to a point already touched upon, there can be no 

 longer any doubt as to the nerve-ends of Neurosoria being 

 thickened (see Plate IV b); indeed Kuhn himself, while 

 describing the nerve-ends as "vix incrassati," yet, when 

 he comes to the comparison of Neurosoria with Allosorus, 

 says, of the latter, nerve-ends "non incrassati," and, of the 

 former, nerve-ends "paullulum incrassati." 



3. The testimony derivable from the structure of the 

 vascular bundles of the rhizome, while used by Kuhn to 

 distinguish Neurosoria from Allosorus, with its one 

 horseshoe-shaped fascicle, may perhaps be disregarded for 

 the purposes of this paper. The writer has no specimen 

 of the rhizome of Neurosoria, but what he takes to be a 

 structure corresponding to the "fasciculi 3 sphaerici" of 

 Kuhn's description of the new genus is found in the rhizome 

 of Cheilanthes tenuifolia Sw. 



4. A fern that has a striking superficial resemblance to 

 Neurosoria pteroides, and might easily be mistaken for it, 

 is Brown's Cheilanthes caudata, which Domin (op. cit.) 

 makes a subspecies of his amplified and emended Ch. tenui- 

 folia (Sw.); but close examination shows this plant to be a 

 true Cheilanthes with submarginal confluent sori, and a 

 distinct indusium, though this latter, unlike most spp. of 

 Cheilanthes, is practically continuous, as in Pteris. It 

 comes from Tropical Queensland, and like Neurosoria, is 

 very rare. 



As the result of these considerations, the writer ventures 

 to suggest that the following scheme, based partly on that 

 of Diels in the "Pflanzenfamilien," indicates the place of 

 Neurosoria in fern classification. 



