THK PLUMULARID.T:. 41 



leaves have become palmate or lobate. their ends no longer being open, and the distal end of the 

 corbula twig is also closed. Two or three budding uematophores are now seen on the inner prox- 

 imal edge of some of the leaves, bnt they can uot at this stage be distinguished from the budding 

 gonangia except by their position. 1 In the ten-leaved corbula. (tig. 114) the gonangia have so 

 increased in size that they can readily be distinguished from the uematophores. Up to the time 

 when about twelve corbula leaves have been acquired (tig. 113) the individual leaves are thick, 

 fleshy lobes filled with coniosarc, except in their central cavities, which are apparently homologous 

 with the central cavity of the stem, lined with endodermal ciliated cells which cause rapid currents 

 of water to pass to and fro. Shortly after this stage the leaf becomes much Hatter and thinner, 

 the central part is solidified by the increased deposition of chitine, which is partially opaque and 

 finely punctate in appearance, and the cu-uosarc is pushed away from the center to the edges, 

 where it forms a flattened tube of ectoderm with an open central cavity running around the 

 periphery of the leaf and connecting with the sarcostyles. In the latter large nematocysts have 

 appeared, the neinatophores are completed, and the defensive zooids are functional. 



In the plate of chitine which now occupies the oblong oval space in the middle of each leaf 

 there appears a longitudinal streak near the posterior edge of the plate (fig. 115, s). This streak, 

 called the " septum'' by Alltnan, is perfectly clear and transparent, being composed of a structure- 

 less chitine, and may serve the purpose of strengthening the leaf by the hard flexible rod. 2 



From this time on the leaves are vascular along their edges. As they grow the edges meet 

 and their distal ends curve over toward the center of the rachis until they meet above. 'Ihe 

 coalescence of the corbula leaves along their edges is, as has already been described, aided by 

 the sacrostyles, which send forth processes from one leaf to another, thus holding the edges together 

 until a permanent connection of chitine and sarcode is established, and the cit'lomic cavities of 

 the tubes of adjacent leaves are united, and currents are established running from one to the other. 

 During all this time the gonaugia are growing and the sexual elements reach maturity shortly 

 after the corbula is completed. 



At an early stage in the history of the gonangia the sex cells migrate from the corbula twig 

 or rachis into the blastostyle, where the gonophores are formed, as already described. 



Embryology. The embryology of the Plumularidie has not been very extensively studied. 

 The following facts, however, seem to be well established: In speaking of the development of the 

 ovum in hydroids which produce planuliv, including, of course, the Plumularidn-, Allman says: 3 



In such cases the ovum, which is mostly destitute of vitellary membrane, after passing through a regular or 

 nearly regular segmentation in accordance with the usual binary law of embryonal development, becomes trans- 

 formed into a solid spherical mass of cells (bastosphere), from which a peripheral layer soon becomes separated by a 

 process of delamination. The embryo now as a rule becomes more or less elongated, and the central cavity makes 

 its appearance in it. 



At this stage the embryo is in the form of a hollow oviform body whose walls are composed of two layers, an 

 external or ectoderm, and an internal or endoderm. It is by delamiuation, never by invagination, that the two ger- 

 minal layers, ectoderm and endoderm, are formed. The embryo lias now usually escaped from the confinement of 

 the gouophnre, and its ectoderm becomes clothed with vibratile cilia, by the aid of which it moves about as a free 

 larva in the surrounding water. It would seem to be about this time that the mesosarc shows itself as a very line 

 structureless membrane between the eudoderm aud the ectoderm. To the larva thus formed Dalyell, by whose 

 observations it was first made known, has given the name of I'lamila [lig. 116]. 



The planula is still a completely closed sac. After enjoying for a time its free locomotive life it loses its cilia 

 and fixes itself by one end the aboral pole. A delicate chitinous pellicle, the foundation f theperisarc, isexnvted 

 over a greater or less extent of its surface; the free or oral pole becomes perforated by a mouth round which a 

 circle of tentacles has become developed. The larva may now be recognized as the primordial hydranth of the 

 colony, aud it only remains for this to become complicated by the budding of other hydrauths and of the sexual 

 zooids in order that it may attain the condition of the fully developed dendritic colony. 



Hainann 4 agrees with Altaian regarding the important point involved in his declaration that 

 the separation of the embryonal layers is by delamination, and states that he has examined vari- 

 ous species of Aglaophenia and J'luiniihtria and finds the same to be true in all cases. Ill Aglao- 



'This refers to their external appearance; satisfactory sections were uot secured. 



2 The analogy between this structure and the rod found in graptolites is quite, suggestive. 



'-' Report on Hydroida dredged by H. M. S. rlmlli-nyer, during the years 1S73-76, Pt. .'. IMS*, p. \ 



4 Der Organismus der Hydroidpolypeu, .Jeuaische Zeitschrii't fiir Xaturwisscnschaft, XV, IXX'2, p. 31. 



