85 



after Further transverse (Hvisions il becomes a ;5- lo .Vcelled lilamenl giving rise lo 

 a number of branches. The trichogyne or a small remnant of it may often be 

 seen some time afterwards on the convex side of the second cell in the main fila- 

 ment of the gonimoblasl (figs. 18 C; 20 E, H \ {V2E,F,H). Unfortunately, I have not 

 been able to follow the development of the cystocarp in Ch. hallandica , where it 

 seems lo be somewhat different (figs. 21, 22). In four of the five sexual species men- 

 tioned the carpospores arise only in the terminal cells of the branched gonimoblast. 

 In Ch. gyiiandra, rhipidandra and Thuretii the branches are numerous, the cystocarp 

 capituliform ; in Ch. hallandica the number of the branches and the carpospores is 

 very low. Ch. efflorescens is also in this respect different from the other species, 

 the carpospores arising not only in the terminal cell but also in one or two of the 

 cells lying behind in the filaments of the cystocar]), thus seriately (fig. 63). 



Sporangia occur in all known species of Chantransia. For some time it was 

 generally accepted that monosporangia only occur in this genus, the older state- 

 ments of tetrasporangia by Harvey being supposed to be due to some error. In 

 later years, however, tetrasporangia have been pointed out with certainly in some 

 species by Schmitz and Hauptfleisch (1896), Borgesen (1903) and Kylin (1906 and 

 1907), and I have been able not only to confirm these statements but also to find 

 tetrasporangia in five other species, so that the occurrence of tetrasporangia is now 

 established in eight of the species mentioned below (Ch. Thuretii, Daviesii, virgalula, 

 polgblasta, cytophaga, Dumontiw, efflorescens, pectinata). In Ch. Dumontice and poly- 

 blasta tetrasporangia only have been met with, in the others also monosporangia. 

 The division of the tetrasporangia is always cruciate, the tirst division being hori- 

 zontal. Amoeboid movements of the monospores immediately after the liberation, 

 similar to those described formerly for Helminthora dioaricata, were observed in 

 Ch. Thuretii (fig. 30). 



In most of the species provided with sex-organs sporangia occur in the sexual 

 plants, in the monoecious species as well as in the dioecious Ch. rhipidandra. On 

 the other hand, as the sex-organs are not present in all the plants, individuals 

 bearing only sporangia will always be met with. In Ch. efflorescens only there is 

 a sharp distinction between sexual plants and sporangia-bearing plants. This is 

 perhaps connected with the fact that tetrasporangia occur in this species. As shown 

 by Yamanouchi ' the tetrasporic plants of Polgsiphonia violacea show double the 

 number of chromosomes to that of the sexual plants, and a reduction in the number 

 of chromosomes takes place by the formation of the tetraspores. If that is general 

 for the Floridese, a similar alternation of tetrasporic plants with sexual plants must 

 be supposed to exist in Ch. efflorescens , and that is supported by Ihe fact that the 

 sporangia-bearing plants occur in the Danish waters chietly in spring, the cystocarp- 

 bearing plants in summer. In the sexual species with monosporangia such alter- 

 nation of generations does not occur, and the reduction of chromosomes must be 

 supposed to take place not in the sporangia but probably in the cystocarps, as in 



' S. Yamanouchi, The life-history of Polysiphonia violacea. Botanical Gazette Vol. XLIl. 19()(). 



