STEPHANOMIA. 73 



The polypites are of two kinds, large and small ; the former are attached at regular 

 intervals along one side of the coenosarc by a short peduncle, measure six tenths of an inch 

 to one inch in length, and resemble those of other Phi/sophorida. 



Their ventricose enlarged portion is marked by numerous red striae, which are simply 

 elevations of the endoderm, containing thread-cells and colouring granules, and are 

 homologous with the villi of the Calijcoj^horida (PL VI, fig. 11). 



The small polypites (hydrocysts ?) are attached to the coenosarc between the large ones. 

 They possess no coloured villi, and I am not sure that their extremities are open ; they are 

 frequently as long as the large stomachs, but never nearly so wide ; and their apices are 

 commonly provided with large thread-cells. A rudimentary tentacle is developed from 

 one side of their base (fig. 12). 



The tentacles are attached to the pedicle of the large polypites, or to their base close to 

 the pedicle, but which I could not determine. They consist of a stem, which gives rise on 

 one side to a series of branches, which carry at their extremity deep-red sacculi. Tn structure 

 most of these parts resemble those of the Cali/copJwridce, but the sac requires especial notice. 

 It is very long, and is usually coiled up and partly inclosed within a cup-shaped, hood-like 

 process of the ectoderm of the peduncle. On each side of the ordinary imbricated series of 

 thread-cells there is a single row of larger thread-cells ; the latter are oval, and ^th of an 

 inch in length, while the others are elongated, somewhat curved, and not more than 

 5^th of an inch in length. The thin wall of the sac contains a quadruple series of peculiar 

 bodies, represented in PI. V, fig. 9. They are elongated, slightly curved, and pointed at 

 one end, while the other end is obtuse, and presents three or four elevations. They are 

 disposed transversely to the axis of the sacculus. 



The filament, which terminates the sacculus, is long and cylindrical, and contains a vast 

 number of thread-cells of j^^i\\ of an inch in length. 



The mode of development of these organs exactly resembles that of their horaologues in 

 the Cali/copJioridce. They commence as csecal processes, which elongate and become constricted 

 into three portions — pedicle, sac, and filament. The thread-cells are at this time all alike, 

 and not more than jja^th of an inch in diameter. The involucrum commences as a circular 

 projection of the ectoderm of the peduncle. 



The gonophores (fig. 12) are attached by branched footstalks to the coenosarc, between 

 the larger polypites. There was no very great regularity in their occurrence, but I think 

 that, as a general rule, there were both ovarian and seminal organs between every two 

 polypites. The same stem that supported them frequently gave rise to a small polypite, 

 or hydrocyst. 



The androphores are distinguishable from the gynophores by their greater absolute 

 length, their more elongated form, and by their greater clearness and transparency. The 

 largest androphore I measured was one tenth of an inch long ; the largest gynophore three 

 fortieths of an inch. 



The calyx of the androphore is terminated by four obtuse elevations, containing large 

 thread-cells. In essential structure it exactly resembles the corresponding part in Biphycs. 

 The thickened outer wall of the spermsac contains round vesicles, about -^U^^ of an inch in 

 diameter, none of which were developed into perfect spermatozoa in the specimen I observed. 

 The androphores were developed from simple papillae, in the same way as those of Diphjjes. 



10 



