370 Pteridophyten. 



of which have reverted to an upright habit. In this connection stress 

 is laid upon the fact that the relatively primitive species of Alsophila 

 A. {Metaxya) blechnoides and A. {Lophosoria) pruinata are creeping. 

 It is held that further support for the origin of the dendroid Cya- 

 theaceae from an ancestry with Gleicheniaceous habit is seen in the 

 so-called aphlebiae of Hemitelia; these are in reality the lowest pair 

 of pinnae left behind owing to the localization of the intercalar} 7 

 growth of the petiole above instead of below the basal pinnae. This 

 localization of growth occurs sporadically in all three genera of the 

 Cyatheae. As already pointed out the anaton^ and soral characters 

 are quite in harmony with the view that Lophosoria presents a 

 further development of the Gleicheniaceous type — this develop- 

 ment proceeding along lines involving elaboration of the stele and 

 leaf trace, loss of median dehiscence of the sporangia owing to 

 crowding and reduction of the spore Output per sporangium. 



The author then discusses the affinities of four pairs of genera; 

 the first, Struthiopteris and Onoclea, he regards as closely related to 

 the Cyatheae, Onoclea having apparently reversed to the creeping habit. 

 Cystopteris and Acrophorus form the second pair; the affinity of the 

 former seems to be with Struthiopteris, while the position of the 

 latter remains doubtful untill it has been studied more in detail. The 

 third pair, Peranema and Diacalpe, are treated in detail by Davie 

 in the same number of the Annais of Botany and the author agrees 

 whith the latter botanist in placing them at the base of the Aspidiae. 

 The last pair to be considered is Woodsia and Hypoderris, the for- 

 mer seems to show affinities with the Cyatheaceae, retaining their 

 gradate sorus, while Woodsia, though related to it, shows a Pro- 

 gression, especially an anatomical progression in having not only 

 leaf gaps causing dictyostely but perforations of the axial vascular 

 tissue and in the complexity of its four-stranded leaf trace. 



Finally the interest of Lophosoria as a synthetic type represen- 

 ting a phyletically early condition of Cyatheoid development is 

 emphasized; of the Cyatheaceae the genus nearest to it is Alsophila, 

 but on the whole the latter is more advanced. Hemitelia has pro- 

 gressed still further while Cyatheae seems to be the most specialized 

 t}rpe of the series. The strengthening of the comparison of Cystopteris 

 to the Aspidiae is also important for the sorus of Cystopteris is simi- 

 lar in some points to that of Nephrodium filix rnas and the "Nephro- 

 dioid" type of sorus "may be held to represent a fiattened modifi- 

 cation of the Cyatheoid type". Thus it becomes clear that the 

 Cyatheaceae were not a blind or final evolutionary line, though the 

 existing genera may be the ultimate exponents of such a line. 



Isabel Browne (University College London). 



Davie, R. C.« The Structure and Affinities of Peranema 

 and Diacalpe. (Ann. Bot. p. 245-268 with PI. XXVIII-XXIX. 

 April 1912.) 



In Peranema cyatheoides D. Don. the stock is upright; in Diacalpe 

 aspidioides Bl. it may be upright from the first or creeping at first 

 and upright later. In both plants the anatomy is dictyostelic. but 

 there are no perforations, or gaps in the cylinder independent of 

 the departure of leaf traces; in this the author disagrees with 

 Schlumberger who records numerous perforations in Diacalpe. The 

 xylem is centrifugal in development though there is no definite 

 protox} T lem. In Peranema each meristele is lined internal!}' and 



