The Hymenophyllacecc. 
119 
the lower side of the stele, and we have the vascular system of the 
rhizome consisting of a truly “collateral” strand. These types 
are found in species with very slender rhizomes. Boodle gives 
very weighty reasons for disagreeing with Prantl, who held that 
these simple types were primitive, and for believing that they form 
a reduction-series correlated with extreme adaptation to epiphytic 
life in damp air. The final stage in reduction is reached in T. 
labiutuin and T. Motleyi, in which one tracheid only and no phloem, 
or even no proper vascular elements at all, are present. There 
seems, in fact, every reason to accept Boodle’s theory that the 
more primitive forms of Hymenophyllaceous stele are those which 
belong to what I have called the central type, found in certain species 
of both genera, that from this central type there have been derived, 
on the one hand, those species of Tricliomanes with an increased 
amount of xylem in the stele, and on the other, in parallel series, 
the simpler types of both genera. To this I should add the hypo¬ 
thesis that the petioles with a basal structure identical with that 
of the stem represent a really primitive form of leaf-trace, which, 
in the most primitive type of all, T. renifonne, is continued right up 
to the lamina and there undergoes simple dichotomy. From this 
the more or less arched form of trace, with more or less localised 
protoxylems, is derived by specialisation of lateral channels for the 
supply of rows of pinnae. 
Thus we should conclude that the central Hymenophyllaceous 
type of vascular system is probably nearly primitive for the Ferns. 
The case for this hypothesis rests largely on the fundamental 
identity of their stem structure with that of the central Botryo- 
pteridese, in spite of the quite different frond-structure and 
presumable conditions of life. If we suppose that the ancestors of 
the Hymenophyllaceae became adapted to life in a very damp 
atmosphere at an early period of their history, it is quite intelligible 
that they should have retained a very early and simple type of 
structure, since the demands of water-conduction would be much 
less than in the progressive ferns which became adapted to more 
exacting conditions involving improved facilities for water-transport. 
This view may be met by the contention that if we admit the 
probability that the filmy types of ferns are reduced from forms 
with normal leaf-structure it is much more reasonable to suppose 
that the vascular system has been likewise reduced. It is not 
denied that the vascular system of the central Hymenophyllacese is 
reduced, but it is suggested that the reduction has affected the size 
