320 



membranes special names, and he therefore employs for the external 

 the name of ectoderm, and for the internal that of endoderm. Each of 

 these membranes retains its primitive cellular structure. In the ecto- 

 derm thread-cells are produced in great abundance ; these are formed 

 in the interior of the ectodermal cells by a process of endogenous cell- 

 formation, and are afterwards set free by the rupture of the mother- 

 cell. The thread- cells in a quiescent state are minute ovoid capsules, 

 but under the influence of irritation, an internal sac is protruded by 

 a process of evagination ; the surface of the evaginated sac is 

 furnished with a circle of curved si3icula, and from its free extremity 

 a delicate and long filament is emitted. The thread-cells of Cordy- 

 lophora thus closely resemble the " hastigerous organs " of Hydra. 

 The polypary is a simple unorganized secretion deposited in layers 

 from the ectoderm. In the endoderm, the author points out a di- 

 stinct and well-developed glandular structure composed of true 

 secreting cells, which are themselves produced in the interior of 

 mother- ceils, and elaborate a brown granular secretion which he 

 assumes as representing the biliary secretion of the higher animals. 

 He describes, as a system of special muscles, certain longitudinal 

 fibres, which may be distinctly seen in close connection with the 

 inner surface of the ectoderm. The tentacula are shown to be 

 continuous tubes communicating with the cavity of the stomach, 

 and thus possess the same essential structure as those of Hydra ; 

 they are formed of a direct continuation of the ectodern of the 

 polype, lined by a similar continuation of the endoderm. The 

 appearance of transverse septa at regular intervals, which is so very 

 striking in these tentacula, must not be attributed to the existence 

 of true septa. It is dae to a peculiar condition of the endodermal 

 layer, but the author has not been able to give a satisfactory ex- 

 planation of it. Through the whole of the canal which pervades 

 the axis of the stems and branches, a constant though a regular 

 rotatory movement is kept up in the contained fluid ; this movement 

 is not due to the propulsive action of vibratile cilia, and is explained 

 by the author as the effect of the active processes going on in the 

 secreting cells of the endoderm, processes which can scarcely be 

 imagined to take place without causing local alterations in the 

 chemical constitution of the surrounding fluid, and a consequent 

 disturbance in its stability. 



The reproductive system of Cordylophora consists of ovoid cap- 

 sules situated on the ultimate branches at some distance behind the 

 polypes ; some of these capsules contain ova, others spermatozoa ; 

 they are plainly homologous with the ovigerous sacs of the marine 

 TubuIariadcB ; they present a very evident, though disguised medusoid 

 structure, having a hollow cylindrical body, whose cavity is conti- 

 nuous with that of the polype-stem, projecting into them below, and 

 representing the proboscidiform stomach of a Medusa, while a sy- 

 stem of branched tubes which communicate at their origin with the 

 cavity of the hollow organ, must be viewed as the homologues of 

 the radiating gastro-vascular canals, and the proper walls of the 

 capsule will then represent the disc. From comparative observations 



