GENERA OF FERNS AND THEIR CLASSIFICATION. 61 



it, though sometimes the prelongation is evidently in 

 advance, and then the mode of growth appears to 

 agree with Eremobrya ; but the non-articulation of the 

 stipes at once distinguishes it. Whatever the character 

 of the stem of Desmobryous Ferns, it is always formed 

 of the united and adherent bases of the fronds, and 

 increases by the successive evolution of fresh fronds, 

 each succeeding one of which is produced on the 

 interior side of the bases of the preceding ones. 



All Ferns are referable to one or other of these 

 two divisions, and in general the difference between 

 them is readily seen, particularly when living plants 

 are examined ; but, as in all attempts to generalize 

 from special organs or structures, there are exceptions. 

 For example, in Elaphoglossum the fronds are neither 

 strictly adherent nor strictly articulate, but have a 

 swelling some distance up the stipes, at which point, 

 though there is no change in structure, the vascular 

 bundles are so weak that the fronds ultimately sepa- 

 rate there; and hence I regard the genus as an 

 aberrant form of Desmobrya (?). In Woodsia, again, 

 the stipes has an elevated articulation ; but the axis is a 

 caudex formed of the adherent bases of the stipes, and 

 this, together with its fasciculate frond, indicates its 

 true affinity to be in Desmobrya. A few also occur, 

 as in the section Ctenopteridece of the tribe Polypodies, 

 in which the articulation is obscure, and a careful 

 examination is required to detect it. 



Notwithstanding these few exceptions, there can be 

 no doubt that the two modes of growth above de- 

 scribed are widely distinct, and the two groups into 

 which Ferns are thereby divided are quite distinct in 

 habit and appearance. The plants too seem to be 



