soon reaches the summit of the fruit. In the meantime the 

 base of the style has split into two parts, and through the 

 aperture the plumule emerges into the water. The soft tissue 

 of the pericarp decays, probably by a sort of maceration in the 

 sea water, and there remains only a 4-lobed comb, each lobe 

 cut into 10-20 subulate teeth, and the hardened tube gripping 

 the base of the embryonic plant. The comb breaks away from 

 the end of the branch and sinks to the bottom, where the teeth 

 of the lobes catch in the fibres of Posidonia australis, or in 

 other material lying on the sea-floor, and so anchor the new 

 plant while it roots and grows. 



There are still several points which require clearing up, 

 such as the history of the ovule in the very young carpel, 

 which has not yet been seen. At the fall of the stigmas, the 

 embryo already occupies nearly the whole of the ovary, and 

 the integuments of the ovule seem to have been absorbed, per- 

 haps in the store of albumen. Or it may be that the pollen 

 finds great difficulty in reaching the stigmas, and that in order 

 to ensure reproduction of the species, the ovary develops a bud 

 in place of an ovule, as is said to occur sometimes in Crinuni 

 and Amaryllis. Against this hypothesis are the facts that the 

 growth in the ovary resembles an embryo, that the anthers 

 are fertile, and that no fruits of a different and more normal 

 character have been found. Nor do we yet know what period 

 elapses between the emergence of the embryo from the fruit 

 and the fall of the comb, but probably flowering takes place 

 from September to January, and the young plants are firmly 

 rooted before the beginning of winter. 



It is also probable that there are two forms of P. 

 antarctica. One, very numerous at Henley Beach, and which 

 may be considered the type, has shorter leaves (sheath, 8-10 

 mm. long ; blade, 12-35 mm. long), and a much rarer variety, 

 found at the same place, has longer leaves (sheath, 15-20 mm. ; 

 blade, 40-70 mm.). No male flowers or young females of the 

 type were found, January being apparently too late for them, 

 but both were gathered on the long-leaved variety, which 

 flowers later, and the development of the carpel, apart from 

 the fact that the membranous bracts are fewer, although rarely 

 quite obsolete, is exactly the same as in the short-leaved form. 



There are certain analogies between the fruiting of 

 Pectinella and that of two genera of a distant family, the 

 Rhizophoracea. In BJiizo'phora (the tropical mangrove) there 

 is likewise no seed in the ordinary sense of the term, as the 

 embryo germinates in the fruit while the latter is still growing 

 on the tree, and without a period of rest ; in Bruguiera the 

 similarity is increased by the fruit itself falling to the ground 

 along with the growing embryo. 



