2 Cephalozia. 



Lined. Thus lie unites to Cephalozia the Jungermania capitata of Hooker, 

 on account of its rather large involucre, whereas it really belongs, by all 

 its characters, to his own § Lophozia of Jungermania; but he excludes a 

 very characteristic Cephalozia, C. co7inivens (Dicks), solely because of the 

 cihated mouth of the perianth, and relegates it to his genus Blepharoa- 

 toma, whose character is " Perichsetium polyphyllum undique imbri- 

 catum, phyUis articulato cihatis. Plantte exstipulatae foliis transversis 

 vel verticahbus." It hardly needs pointing out that C. connivens, having 

 neither transverse leaves nor ciliated bracts, could not, on Dumortier's 

 own shewing, be a Blepharostoma. He enumerates but 3 species of this 

 genus; Bl. trichojjhylluyn, Bl. connivens, and Bl. setacea. If the first be 

 considered the type of Blepharostoma, the second is a Cephalozia, and the 

 thirdi Si Lepidozia ; that is, they belong to three distinct genera — all of 

 Dumortier's own proposing ! 



Dumortier's generic character does not exactly correspond to any 

 single species of Cephalozia, yet he indicates by its name a peculiarity 

 which, although found also in some other HepaticcB, is normally never 

 absent from any species of Cephalozia, namely the capitate or comate 

 female flowers. The bracts are usually much larger than the stem- 

 leaves, and are closely set in 3 or 4 transverse rows, and along 3 sides 

 of the stem or branch — i.e. they are tristichous ; for, although in a great 

 majority of the species underleaves are wanting to the stem, they are 

 never absent from the flower heads, and (at least in the innermost row) 

 they mostly equal the side leaves in size and form, being often connate 

 with them into an outer cup, or perianth. Moreover, although in certain 

 species the stem leaves may be round and entii-e — in a few others reduced 

 to mere papillsform projections, or scales, and in one species entirely 

 absent — in place of the bilobed leaves that prevail in the great majority 

 of Cephalozia — ^the bracts of both male and female flowers are, as a rule, 

 in every species, deeply cloven, mostly bifid — ^in a few species occasion- 

 aUy 3—5 fid. 



Within the involucre, quite free from it, and usually protruding a 

 long way beyond it, is a na,rrow trigono-prismatic perianth ; only in 

 certain species of the subgenus Cephaloziella does it acquire supple- 

 mentary angles and become 4 — 5 — or 6-angled. This three- sided pe- 

 rianth differs essentially from that of Lophocolea in the origin of its 

 angles, or keels. In Lophocolea the angles are at the marginal sutures 



