432 Umversity of CaUfornin I'lihlicatioiis in Botany [Vol.7 



I have exaniiiK'd a great iiuinber oi' species belonging to the present 

 tribe Delesserieae as to the mode of growth, anatomical structure, 

 position of cystocarps and sori, and I believe that I can refer every 

 species to one of the two principal types. There are of course many 

 I have not seen or of which the material was less satisfactory. In all 

 the species of simtosa-growth. the cells of the main axis, as told above, 

 get divided by cross-walls, so that the central lamella of the costa is 

 formed by comparatively short cells. A cortex is formed by tangential 

 and radial divisions, the cells being arranged in a regular manner 

 (fig. 1) recalling the cortex in Nitophylluni, though tlie difference 

 between a "medulla" and a "cortex" is more evident than in the 

 latter genus. \n Hie hypolossum-tyipe, the costal cell-rows are formed 

 by very long cells: these cut off cortical cells, that divide by radial 

 walls, the cells thus formed get stretched in longitudinal direction, 

 are again divided by a tangential wall, and so forth, so that we observe, 

 on a length section, very long tubes in the center and gradually shorter 

 cells toward the surface. Even in a cross-section of a young costa 

 the radial arrangement is much less distinct than in the other type, 

 and later the difference becomes more marked. The long tubes make 

 the costa a rather weak structure, which is strengthened by the devel- 

 opment of numerous narrow hyphae (fig. 5). I have not found such 

 hyphae in any species belonging to the sinuosa-type. Those different 

 kinds of anatomical structure have been described and figured before 

 (see Wille, "Beitrage zur Entwickelungsgesehichte der physiologischen 

 GeW'Cbesysteme bei einigen Florideen" — Nova Acta der K. Leop.- 

 Carol. Akad. LII [1887]), but they were not, as far as I am aware, 

 combined with the different modes of apical growth. It should be 

 mentioned that, among species belonging to the h ypoglossum-type, I 

 have found two with non-typical anatomy, D. Montagneana J. Ag. 

 and B. ricscifolia Ag., both referred to the genus Apoglossum by J. G. 

 Agardh. In these small species hardly any pyphae at all get developed, 

 the differences between "medulla" and "cortex" are marked, but the 

 radial arrangement of cells less distinct than in sinuosa, etc. I may 

 add that the larger species, A. decipiens J. Ag., has typical hypoglos- 

 sum-strncture in the costa. In spite of these apparent exceptions, the 

 anatomical distinction between the two types is obvious. 



If Delesseria sanguinea be recognized as the type of the genus, D. 

 sinuosa must of course be excluded and Phycodrys of Kiitzing be 

 restored. We have seen that Scliizoneura quercifolia differs from other 

 members of this dubious genus while it agrees with Phycodrys sinuosa 



