540 



Dr. J. C. Willis. On the Lack of 



in either family which has a proper arrangement for preventing its seeds 

 from washing away. 



In none of the species with flattened highly dorsiventral root does the 

 stem or shoot of the secondary shoots show any sign of flattening out, while 

 in none of those with flattened secondary shoot does the root show sign 

 of flattening. The former are mainly characteristic of India and Ceylon, the 

 latter of South America. 



For the great variety in detail which all these plants exhibit reference 

 must he made to the work of Warming* and to my previous work (loc. cit.) 

 published upon the Indian and Ceylon forms. Tulasne's monographf also 

 contains a fine set of pictures of the families. 



We may now proceed to deal in more detail with the various structural 

 features of these families, which have generally been looked upon as adapta- 

 tions. In the first place, let us take the flattened thallus-like roots or 

 secondary shoots. These have been generally regarded as adaptations to 

 withstand, or to avoid, the rush of the water. But, as already pointed out, 

 the forms that are exposed to it have no marked " Zugfestigkeit." In my 

 monograph of the Indian forms I expressed the opinion that the adaptation 

 was rather to shallow water than to rushing water. The most highly 

 adapted form, on either view, is H. olivaceum, and yet this very form, 

 at the period when the rush of the water is most violent, possesses a tall 

 primary axis with a great bunch of leaves at the end, an axis, moreover, 

 which can hold fast if the flat creeping thallus be removed. Until I came 

 to Brazil I was still in some doubt as to whether these dwarf forms and 

 closely attached thalli might not be regarded as in some degree at any rate 

 adaptations to rushing water, but what I have seen here has completely 

 destroyed that idea, and enabled me to write this paper. Here there are no 

 flat creeping root thalli, but instead there are, as in M. Weddelliana, 

 flattened secondary shoots, forming thalloid outgrowths. Now these thalli, 

 though with far less holdfast than the Indian root thalli, live in more rapid 

 water. Never or very rarely, though for many years I have been familiar 

 with the habitats of the eastern forms, have I seen any in such rapid or violent 

 water as the forms which grow in the State of Bio de Janeiro. I have even 

 found Zophogyne arculifera with the water falling on to it from a measurable 

 height. Even the large Brazilian species of Mourera, Apinagia, and 

 Marathrum live in very swift water without any difficulty, as may be very 

 strikingly seen in Blate 62 of v. Wettstein's ' Vegetationsbilder aus 



* Warming, " Familien Podostemacese, I-VI," ' Kgl. Dansk. Vidensk. Selsk. Skr.,' 

 6 raekke, ii, 1881 ; ii, 1882 ; iv, 1888 ; vii, 1891 ; ix, 1899 ; xi, 1901. 



t Tulasne, ' Monographia Podostemacearum," 'Arch, du Mus. d'Hist. Nat.,' vol. 6(1852). 



