131 MB. E. M. HOLMES OS CODIOLUM GREOAKIUM. 



amylaceous globules appear. "When the cell (now called a gonio- 

 cytium) takes on the function of reproduction, these globules 

 first aggregate into little groups, and eventually disappear, while 

 the endochrome becomes of a more intense green colour and gra- 

 nulated appearance. The intermediate stages between this granu- 

 lation and the formation of the obovate zoogonidia do not appear 

 to have been observed. Like those of Codium tomentosum, which 

 they resemble in shape, the zoogonidia have their narrow end 

 pointing outwards ; they are, however, smaller than in that plant, 

 being only -^ millim. long. "When mature, the zoogonidia escape 

 from an opening in the apex of the club ; each is furnished with 

 two vibratory cilia at the narrow end. 



Besides this form of reproduction there are found mixed among 

 the tufts perfectly globose cells, larger than the zoogonidia, with 

 the contents (cytioplasm) more evidently granular, and the cell- 

 wall (cytioderm) triple, as described above, with the middle layer 

 gelatinous. Braun believes these to play the role of resting- 

 spores, and to preserve the plant during the winter and spring 

 months, as it does not grow again until the end of summer. To 

 these celb he has given the name of hypnospores. The plant, in 

 this country at least, lasts through the winter and spring, and has 

 been found as late as the month of June, according to Mr. Cress- 

 well's observations, while I have myself gathered it at Christmas. 

 For the above and the following reasons, it appears to me doubtful 

 if the so-called hypnospores belong to Codiolum at all. The tufts 

 of Codiolum gregarium always contained plants in every stage of 

 growth. They frequently have several species growing inter- 

 mixed, especially Hormotrichum Jlaccum and Calothrix scopulorum. 

 I have met with spherical cells answering to Braun's description 

 among Codiolum, and also other cells presenting a similar shape, 

 showing a single division of the endochrome, and evidently re- 

 sembling in diameter and appearance the very young filaments 

 of Hormotrichum Jlaccum with two, four, or eight divisions, which 

 occur in the same tuft. It seems to me highly probable therefore 

 that the so-called hypnospores of Codiolum are only the earliest 

 stage of growth of II. Jlaccum. 



Codiolum occupies a somewhat anomalous position. In vegeta- 

 tion it agrees with the stalked Protocoecacese, while the fructifica- 

 tion corresponds exactly in character with that of Codium (see Ann. 

 des Sc. Nat. ser. 3, torn. xiv. p. 23). Codiolum, however, grows 

 chiefly at the base, and the vegetative cell performs the functions of 



