537 
and surrounding a globular parenchymatous medullary tissue (fig. 521). This parenchy- 
matous columella seems not to have been observed by Burruam, but it has been 
distinctly represented by DERBÈS et Souter (i. c. fig. 
10—11). The crypts are deep, 120—150 w, nearly pris- 
matic but somewhat irregular in a transverse section, 
and with rounded ends, 45—63 w in diameter. The 
separating walls between these cavities are built up of 
very thin filaments from which the spermatia producing 
filaments spring. When the cavities are more irregular 
in transverse section, the separating walls may form 
folds in the cavity like incomplete septa, or the cavities 
may be partly confluent. The ostioles of the crypts 
were not observed by me but there is no doubt that 
the latter have each their particular opening, not a 
common one as supposed by BUFFHAM, and this species 
is therefore in better accordance with the other species 
of the genus than must be supposed from Burr- Fig. 521. 
HAM’s representation. The spermatia are very small, a en 
about 4 w long, 2 w broad; one nucleus was easily section of androphore. 30 : 1. 
recognisable in the dried material. The antheridia 
were met with in September (Kn and Su). According to Kyrın (1907, p. 125) 
antheridia-bearing specimens collected in May are present in the Riksmuseum at 
Stockholm. 
The procarps are developed in particular globular carpophores which have a 
similar position to that of the androphores. To begin with 
they are like these globular, sessile or furnished with a 
very short stalk but early become beset with anastomosing 
crests (comp. Harvey, l.c. and DARBISHIRE 1895, p. 33). The 
procarps arise in the inner cortex in considerable number in 
each carpophore. The carpogonial branch is, as in the other 
species of the genus, three-celled and borne on a large cell 
that probably becomes the auxiliary cell (fig. 522). The 
trichogyne was observed to protrude through the surface in 
some cases; the narrowing between the ventral part of the 
carpogone and the trichogyne was very distinct at the level 
Fig. 522. of the surface of the carpophore. The development of the 
Phyllophora epiphylla. Pro- cystocarp was not examined. A great number of the pro- 
IR ee carps observed in a carpophore collected in October were 
pogonium. 390 : 1. not fully developed and many of them would probably never 
have reached maturity. Carpophores with procarps were 
observed in September and October, cystocarps in winter (November to March); as 
late as May cystocarps may be met with but more or less empty. 
D.K.D. Vidensk. Selsk. Skr., 7. Række, naturvidensk., og mathem. Afd., VII, 4. 69 
