Johnson. — Sphaerococcus coronopifolius , Stackh. 299 
proximal ends from the first. The carpospores escape from 
the ripe fruit through an irregular slit in the pericarp, not 
through a definite pore. On account of the frequent for- 
mation of a fruit-cavity of large size, especially towards the 
apex of the procarpium-branch before there is any indi- 
cation of carpospores, the size of a swelling is not a safe 
guide as to the stage of development of the fruit. Each 
cystocarp is the product of one procarpium and of one only, 
close as the procarpia are to one another and loose as is 
the middle layer of the procarpium-branch. The carpogo- 
nium has in its immediate neighbourhood a number of cells, 
some of which are specialised, and with all of which it fuses 
to produce the central cell of the cystocarp. All these cells 
are auxiliary cells, and being close to the carpogonium do 
not need any ooblastema-thread (connecting-tube) to place 
them in connection with the fertilised ovicell. I am not 
able to throw much light on the fate of the nuclei in these 
auxiliary cells, and cannot say how far their fusion with one 
another, following on that of the hypogynous cell with the 
carpogonium, should be regarded as a second act of fer- 
tilisation (granting this may happen), here many times re- 
peated. Looking at the development of the cystocarp from 
another point of view, Sphaerococcus exhibits the phenomena 
of fecundation as seen in the Florideae at their best. In no 
other genus in which one cystocarp results from one pro- 
carpium is the possibility of the abundant supply of nu- 
triment from a number of different regions in the thallus 
to the central cell of the cystocarp and so to the sporigerous 
filaments insured to such a degree. In Sphaerococcus , not 
only does the carpogonium fuse with the other cells of the 
carpogenous branch, but with the two basal cells of a lateral 
branch and with two joint-cells . of the central axis of the 
whole procarpium-branch. The nearest approach to this 
(after Gracilaria) is seen in Chondria tenuissima , A g., one 
of the Rhodomeleae. In this species 1 , after fertilisation, the 
1 Schmitz, op. cit, p. 28. 
