16 THÉEL, HERMAPHRODISM IN HOLOTHURIDS. 



the neighbourhood of tlie madreporic canal wliere it turns up- 

 wards in order to accompany the latter and, finally, to open exte- 

 riorly on the dorsal surface of the animal. Of course, during 

 its whole course it is attached to the left side of the dorsal 

 mesentery. Posteriorly and below, it is supported by the two 

 raesenteric membranes which limit the genital sinns of 

 Semper. 



The outer investment of the efferent diict is, of course, a 

 continuity of the coelomic epithelium, its cells being not very 

 high and not distinctly limited but being in possession of an 

 oval nucleus deeply stained with iron-höematoxyline. Separa- 

 ted from this layer by a very thin stratum, obviously of 

 muscular nature, a layer of areolar connective tissue forms 

 the main portion of the wall, the filaments crossing each other 

 in larger or smaller bundles in all directions. In the meshes 

 b)etween the bundles can be seen numerous migratory cells 

 of several diiferent appearances, which may be treated of 

 further on in a succeeding chapter. 



Inside the connective tissue and separated from it by a 

 very thin basement-membrane occurs a ciliated epithelium, 

 presenting a high columnar shape and having the oval nuclei 

 placed close to the broad and superficial end of the cells. 

 Where the duct is in a contracted state, the nuclei are very 

 orowded, so that they present themselves in two layers or 

 more. As a rule, the cells are much higher on the left and 

 upper sides. The peripheral nucleus is surrounded by a 

 layer of finely granulate protoplasm which appears to be 

 unenclosed in a definite cell-wall. Towards the basement- 

 membrane the cells taper into long fine threads which, treated 

 with iron-hsematoxyline of Haidenhain, become of an almost 

 black colour. The hair-like cilia are long and powerful, espe- 

 cially at the left and upper sides of the duct (Pl. II, fig. 13). 



In transverse sections of the efFerent duct there appear 

 between the thread-like basal parts of the ciliated cells, large 

 juxtaposed bodies of a fine granular structure (fig. 13 a). Here and 

 there may be perceived a nucleus surrounded by a clear space, 

 but it always has its place close to the basement-membrane 

 and has evidently nothing to do with the bodies in question. 

 Treated with different stainiug fluids. they still remain påle 

 -and almost colourless. At first one is given the impression 

 that the bodies in question represent mucous-secreting cells, 



