ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 209 



A. aureuni appear to be acrostichoid derivations of Pieris, from some 

 Litotrorhia type. 10. Thus, in the Pterid series either the outer (adaxial) 

 or the inner (abaxial) indusium may be abortive. Steps of abortion of 

 the inner indusium are seen in the following DicksonioideEe—Z^e/Mis/as^/^/Vz, 

 HypoJepis, PoUjpodium punctatum, and Monachosorum subdigitatum. In 

 Hypolepis and Monachosorum the outer (adaxial) indusium may receive 

 a vascular supply from the receptacle, and appear flattened aa a marginal 

 lobe of the pinnula. Phyletically Hypolepis and Monachos^orwn are 

 derivatives of the Dicksonioid-Dennstjedtioid series. {Acrophorus and 

 Cystopteris are distinguished from these by their scales and advanced 

 dictyostely, and are related to the Nephrodioid ferns.) 11. The large 

 series of the Dicksonioids are characterized by the sori maintaining their 

 identity as discrete developments on the separate endings of the veins. 

 They have dermal hairs, not scales, excepting their Davallioid derivatives. 

 TheDicksonioids probably sprang ultimately from some Schizteoid source, 

 through types of the nature of Lorsomopsis and Thyrsopteris, and cul- 

 minated in the Davallioid sequence. 12. The Pterid series are distinguished 

 from the Dicksonioids by the lateral fusion of their marginal sori, which 

 are linked together by lateral commissures. They are related to the 

 Dicksonioids as a collateral branch, attaching probably in the neighbour- 

 hood of Microhpia ; and they culminated in Acrostichum. 13. The 

 Cheilanthoids, though usually ranked with the Pterids, have yet to be 

 studied as to whether they have any near phyletic relationships. 14. All 

 the ferns studied for the present memoir belong to the Marginales. In 

 some cases the sorus has slid from a marginal to a superficial position. 

 15. In the Superficiales the sorus-slide occurred so early in their descent 

 that the two sequences must be regarded as phyletically distinct, notwith- 

 standing all analogies. 



Evolution of Branching in the Filicales.* — B. Sahni publishes 

 some observations on the evolution of branching in the Filicales. The 

 most important progress of this evolution has been in specialization 

 for vegetative propagation ; but subsidiary efforts have been made in 

 the direction of epiphytism {Nephrolepis volubiUs) and of food or water 

 storage (iV. ttiberosa, etc.). It is found that the branching of ferns 

 may be arranged in a series beginning with those in which the rhizome 

 divides into two more or less equal bra'nches, and ending with forms in 

 which the proximal part of one of the branches attains a great length 

 and bears either reduced leaves (Struthiopteris, etc.), or none at all 

 (Nephrolepis). This portion, a stolon, serves to remove the leafy branch- 

 apex away from the mother-axis and sustain it until it forms an 

 independent root-system. The process is carried to an extreme in 

 Nephrohpis by the production of a large number of lateral branches, 

 each of which is a potential individual plant. The view that this 

 elaborated form of branching is derived from the simple dichotomous 

 type, and is connected with it by an unbroken series of transitions, is 

 corroborated by a study of the branching of ferns from the point of 

 view of their vascular anatomy. In the latter case a series of transi- 

 tions is found parallel to the former. When the growing apex of a 



* N»w Phjtologist, xvi. (1917) pp. 1-23 (figs.). 



