136 Bihlio(/raj>hical Xotices. 



primary divisions are: — 1, those ChtJopJionr which arc attaclied 

 ■when young, and afterwards float freely in the form of tufts ; 2, 

 those which are always attached ; and, li, those which are at first 

 attached, and afterwards form jjlobular masses which are often 

 free. This last division includes ('Judophora mf(i;/ropiJa. It 

 is doubtful whether this third division c.n be m; intained. The 

 eight forms of Chtdophora cPiidijropUa here described all grow in 

 still water, and may possibly be oidy forms of C. t/Iomerata, modi- 

 fied by their place of growth". Professor Henfrey* considered Cla- 

 dophora mjfKjropda to be identical with C. ijlomerata ; and Mr. 

 HassaU, in* his ' Freshwater Alga\' took the same view with regard 

 to C. Broimii, which is classified by Dr. Uabenhorst as one of the 

 forms of C. crgcK/ropiJa. On the other hand, Dr. Harvey states 

 that llobei-t Brown (who first d(>scribed it as Conferva p)ulvinnta), 

 Mr. Kalfs, and himself agreed in considering it a perfectly distinct 

 species, at least as well characterized as any other specific form in 

 the genus Cladophom, and better characterized than sever;d reputed 

 speciesf. 



The fourth family, the (Edorioniaceiv, comprises three genera, 

 (Edoqonium, Ci/inatonema, and Bidboclmte. Much interest has at- 

 tached to the former genus, on account of Pringsheim's beautiful 

 observations upon the mode of impregnation + in some of the 

 species. ^^lany species remain, amounting in this work to upwards 

 of thirty, in which the antheridia and oogonia are either unknown 

 or require further investigation, aff'ording a fine field for the atten- 

 tion of algologists. With regard to C>/mafonema, the genus seems 

 quite unnecessary : and Dr. Rabenhorst is apparently of this opinion; 

 for although it is figured in the 'Conspectus (Jenerum,' the descrip- 

 tion in the text (p. 351) makes Cijmatonetna a synonym of the 

 original name of the plant, (Edogonhnn uudidatum, Breb. 0{ Bid- 

 hochirte there are nine weU-established species, besides seven others 

 unknown to the author or of doubtful validity. 



The fifth family, the UlotrichefB, seems to require further con- 

 sideration. It contains the genera Jformiscla, Uhtliviv, Hormidium 

 (which is only the terrestrial form of Ulothriv), and Schizogonium. 

 Except for the occurrence of two kinds of zoospores (mcgazoo.spores 

 and microzoospores, as Areschoug§ has called them), Honniscia 

 might well have been placed in the Confemtcece. Ulothrix has 

 hitherto been considered an ally of Draparnaldia, and Stigeoclo- 

 nium and Schizogonium might with great pro])riety be placed in 

 the Ulvacece. It does not appear that the occurrence of two kinds 

 of zoospores (one of the main features of the family) has been noticed 

 in Uhihrix or in Schizogonium. 



The Chroolepidiece (fam. 6) comprise only two genera, Chroolepus 



* Micr. Diet. p. loO. 



t ' Phycologia Dritannica,' rpniarks under plate xxx. 

 I Jahrbiiclier fiir wisscnschaftliche Botanik, vol. i. p. 1. 

 § Aresch. Obs. phvcol. in Act. Keg. Soc. Scient. tips. ser. 3. vol. vi. 

 fasc. 1 (1866), 



