58 



CARL SKOTTSBERG, MARINE AI.GM 2. RHODOPHYCEjE. 



illustrated in fig. 29 a — b, it is the same as in G. corallina (comp. Kylin, Entwicklungs- 

 gesch. p. 113 f. 9), but the male branchlets are more compound. Their surface is covered 

 with innumerable spermatanges (fig. 29 a). A small part of a lobe is seen in optical 

 section in fig. 29 b. The involucre develops early from the same joint that carries the 

 male branchlets; it counts about 12 unicellular rays. 



The cystocarps are borne laterally on the branches (fig. 27 b), but the pro- 

 carp is terminal as in G. corallina, pushed aside by a sterile branch. Fig. 29 c shows the 

 first stage of the procarp (F. b.). As in G. corallina it becomes three-celled (fig. 29 d), 



Fig. 27. Orijfithsia antarctica: branches of a male, b female and C tetrasporic specimens, X 12. 



these are the three central cells of Kylin, whose terminology is adopted here. From the 

 second central cell a dorsal pericentral cell (fig. 29 e, pc) is cut off, followed by one late- 

 ral pericentral cell (not two as in G. corallina); the latter is the supporting cell of Kytlin 

 (fig. 29 f, sup c). The supporting cell gives birth to a sterile cell (stc) and to the four- 

 celled carpogenic branch, which is curved inward-upward. After fertilization the auxili- 

 ary cell is cut off (fig. 29 g, aux). The formation of the gonimoblast {gon) is seen in fig. 

 29 h — i. In 29 g the carpogone lies alongside the auxiliary cell, but apparently not in 

 contact with this, and the fusion of nuclei was not observed by the writer, so I can only 

 guess that it takes place as in G. corallina. 



In G. Bornetiana, the supporting cell assumes the character of auxiliary cell and is 

 supposed to fuse with the carpogone, although the actual process of fusion was not ob- 



