36 DASYCLADEiE. 



Two species, C. barbata and C. rosarium are usually kept up, and Kiitzing has added 

 a third, C. bibarbata, but it seems to me that the differences indicated have reference 

 more to the age and state of individual specimens, than to difference of species. The 

 fringing or non-fringing of the apices with fibrillae surely depends on the state of the 

 specimen. The fibrills are homologues of leaves, and, like leaves, are deciduous when 

 they have performed their functions. I had abundant opportunities of studying the 

 species at Key West, and see no ground for believing that there is more than one as 

 yet known to botanists. 



1. Cymopolia barbata, Lamour. Cor. Flex. p. 293, and C. rosarium, I. c. p. 294. 

 Kiitz., Sp. Alg. p. 511. Corallina barbata, Lin. Syst. Nat. Ed. 12, p. 1305. 

 Ellis and Sol. Zoop. p. 112. Ellis, Cor. p. 54, t. 25, f. C. C. rosarium, Ellis and 

 Sol. Zoop. p. Ill, ^. 21, jig. h. Shane, Nat. Hist. Jamaica, t. 20, Jig. 3. Cymo- 

 polia bibarbata, Kiitz. Phyc. Gen. t. 40,/. 2. Kutz. Sp. Alg. p. 510. (Tab. XLI. A.) 



Hab. Near low-water mark, under the bridge at Key West. W. H. H. (v. v.) 



Fronds tufted, at first simple, till they attain to one or two inches in height, then 

 becoming branched, at first by the development of simple alternate branches. These 

 afterwards fork at their extremities, and throw out lateral branches ; and by continual 

 repetitions of this process of division the frond at length becomes much branched 

 in a di-trichotomous but irregular order. The tendency to become dichotomous is 

 greater in the older specimens ; the branches in all are fastigiate. Every part of 

 the frond, except the young tips of the branches, is invested with a thick calcareous, 

 brittle crust, pierced with innumerable horizontal canals, opening at the surface by 

 pores, arranged in transverse rings, which are so closely placed together that the sur- 

 face appears as if honeycombed. In these canals of the crust the ramelli of the enclos- 

 ed vegetable lie hid, the points only of their divisions protruding through the pores, 

 and this only in the younger parts, which then have a green colour. The calcareous crust 

 is regularly articulated at short intervals ; the internodes in the main stem and 

 branches are about twice as long as broad, those in the young parts of the frond sphe- 

 roidal and bead-like. The iwdes are much contracted throughout, and thus each branch 

 looks like a string of beads. In the older parts the nodes are bare ; but in the 

 younger, toward the ends of the growing branches, they emit whorls of extremely 

 delicate, byssoid, di-tri-chotomous or multifid, membranaceous fibrills ; and whorls of 

 similar fibrills terminate the young branch itself The branches in the developing plant 

 are thus penicillate or barbed at the extremity. When a piece of a frond is macera- 

 ted in acid, so as to remove the calcareous crust, the true frond becomes visible. This 

 we must now describe. It consists of a continuous, tubular axis or filament, seemingly 

 formed of a single, cylindrical, branching cell, which runs through every part of the 

 calcareous covering, and whose growing apices, clothed with byssoid fibres, pro- 

 trude at the ends of the branches. This filament is nodose, annularly constricted at 

 short intervals, corresponding to the articulations of the crust ; but there are no inter- 



