STRUCTURE AND DEVELOPMENT OF SORANTHERA. 483 
found in connection with plurilocular sporangia, the only fruit 
known up to the present for these genera; and Mr. Murray has 
described a similar connection with the unilocular sporangia in 
Adenocystis (Phye. Mem. pt. ii. 1893, p. 59). 
In Soranthera ulvoidea the cells round the eryptostomata 
begin to divide immediately after the formation of the con- 
tinuous epidermal layer and to form, centrifugally, sori of large, 
round-celled paraplyses and unilocular sporangia (P1. 23. figs. 8,9, 
Pl. 24. figs. 10, 12). Round those eryptostomata which have few 
hairs and are not very deep, the sori appear sooner than round 
tbose in which the hairs are more numerous. In a mature plant | 
have found several deep and large cryptostomata (Pl. 24. fig. 11) 
round which the cells were at this lute stage only just beginning 
to subdivide to form a sorus. lt is remarkable that large well- 
formed cryptostomata ave found in connection with groups of 
rhizoids on a mature plant. In one case where the crypto- 
stoma, though among the groups of rhizoids, was not actually 
in close connection with them, the surrounding cells had divided 
and formed the usual paraphyses as in the sori of unilocular 
sporangia. 
In the older plants the sori, by their centrifugal growth, 
sometimes coalesce, and thus large patches ot fruiting tissue are 
lormed, surroundiug two and possibly more eryptostomata. l 
In the foregoing account of the development of S. ulvoidea it 
has been seen that iu the early part of its existence it follows 
the type of plant found in Chordariacez, both im vegetative 
structure and iu its plurilocular sporangia. It bears a strong 
resemblance to Leathesia, and in some respects to Chordari« ; 
indeed, had S. ulvoidea been found only in its young state 1t 
would undoubtedly have been placed in Chordariacee. On the 
other hand, mature plants, with their continuous epidermal 
layer and unilocular sporangia surrounding cryptostomata, show 
so strong a resemblance to genera of Encæliacee, that it is 
impossible to doubt the allinity between Soranthera, Aspero- 
coccus, Colpomenia, ete. Prantl’ 
In Dr. Kjellmau's system of classification (Engler and Prantl’s 
Natürl. Pflanzenfam. Teil i. Abt. 2, 1893, pp. 180, 181) a great 
point is made of the position and mode of develonment of the 
*porangia. 
In the division including Chordariacee the sporangia are said 
