STRUCTURE.] HARVEY's ACCOUNT OF THEM. 125 



in the plants in which it occurs, seems by the regularity with 

 which it takes place to be as essential to the propagation of 

 the species as that of the sporules or primary seeds. How- 

 ever distorted the form of the receptacle may be in which the 

 granules are immersed, there is no reason to suppose that 

 they originate in disease, for they are produced with so much 

 constancy that they furnish us with some of the best dis- 

 tinguishing characters for genera. But there is a really 

 anomalous structure connected with an imperfect attempt at 

 fructification not uncommonly found on several FlorideaB, 

 especially those of the structure that we have above compared 

 to that of exogens, as Chondrus, Gigartina, &c. This, to which 

 Agardh gives the name of nemathecium, is a wart-like pro- 

 tuberance, of a very irregular figure and generally large size, 

 consisting entirely of concentric filaments with coloured 

 joints, in all respects resembling those that form the peri- 

 phery, but much longer. To the naked eye these warts often 

 strongly resemble capsules, and as such have been frequently 

 described, but they never contain any seeds. The so-called 

 capsules of Chondrus dilatatus are of this nature. It is rare 

 that dependence can be safely placed on bodies of so anoma- 

 lous a nature as furnishing specific, much less generic 

 characters, but in Gigartina plicata and Grimthsise, plants in 

 which no other effort at fructification has yet been noticed, 

 they afford good specific marks. In Griffithsia setacea, bodies 

 (noticed in the description of that plant, p. 103), sometimes 

 occur in the position of capsules, which have apparently the 

 structure of nemathecia; but, judging from their position 

 and size, I am more disposed to consider them viviparous 

 capsules, in which the sporular mass has been converted into 

 minute filaments whilst attached to the parent. Another 

 anomalous body simulating fruit, if it be not a male flower, 

 frequently occurs in some of the filamentous tribes, especially 

 in the genus Polysiphonia (P. fastigiata, fibrata, fibrillosa, 

 &c.), to which Agardh gives the name of antheridium. It is 

 a minute pod-like or lanceolate body, of a yellow colour, con- 

 taining a granular fluid, borne on the colourless, long-jointed 

 fibres, that at particular seasons are found issuing from the 

 tips of the branches in several, if not all Polysiphoniae, 



