536 
and may remain simple or branch by dichotomy. Sometimes the ramification is lateral 
as in fig. 519 A, C, which may perhaps be due to the checking of one of the pro- 
ducts of the dichotomy. Besides this apical ramification a branching by prolifera- 
tions regularly takes place. The latter, one or more, arise from the plane surface of the 
frond, usually near the end, and a series of generations of proliferations is con- 
sequently produced, representing probably an equal number of years (fig. 520). At least 
5 or 6 generations are normally produced in the Danish 
individuals. The growth of the shoots normally ceases 
when a proliferation arises near its end. The ramification 
by proliferations is usually more pronounced than the 
apical forking in the specimens from the Danish waters, 
where the proliferations are very often simple, in par- 
ticular from the waters within Skagen. The membrana- 
ceous frond reaches a breadth of 10 mm; in the southern 
Kattegat and the Sound it only becomes 5 mm broad 
at most. The thickness diminishes gradually from the 
middle towards the margin. As to the anatomical struc- 
ture of the frond reference may be made to DARBISHIRE’S 
paper (1895, p. 12). As shown by this author, small 
interstitial cells are wanting between the medullary cells. 
The shoots are often furnished with a midrib in their 
lower part. It is formed by local thickening of the cortex, 
produced by activity of the outer cortical layer on both 
sides (comp. DARBISHIRE ].c. p. 12). 
The antheridia are produced in small andro- 
phores on particular male specimens. The androphores 
are spherical or subspherical bodies, about 0.5 mm in 
Fig. 520. diameter or a little more, borne on a short pedicel on 
Phyllophora epiphylla. Fromthe9re- the plane surface of the frond near the border (fig. 521). 
sund; north of Kronborg. Sept. 1848. 5 ~ ‘rc 
Caroline Rosenberg. 2:1. They were first pictured by DERBÈS et SoLIER' in the 
nearly related Ph. nervosa which is probably not specific- 
ally distinct from Ph. epiphylla, later on they were briefly mentioned by THURET (Et. phye. 
1878, p. 82) who found that “les corpuscules males sont contenus dans des cryptes 
tapissées d’anthéridies fort semblables à celles du Gracilaria confervoides’. In 1883 
they were described and figured by BurrHam (1883, p. 292, pl. XIII, figs. 5—7). This 
author observed the cavities, from the walls of which numerous tufts of thin fila- 
ments spring which produce at their extremities the spermatia, and he presumed 
that these bodies escape through one common issue at the top of the androphore 
(comp. l.c. fig. 7). My observations on the structure of the androphores however, are, 
not in accordance with those of BurrHam. A vertical section of an androphore 
shows a layer of cavities or crypts covering the whole surface of the globular body 
" Ann. sc. nat. III: S. tome 14 p. 278, pl. 37 figs. 10—11, 1850. 
