242 



ventrally; none were found in the oesophageal region, and none in 

 the candide (this term seems preferable to » caudal papilla«). Posteri- 

 orly they are more numerous than anteriorly, as Bürger (1890) has 

 shown for other species; but I cannot corroborate this author's state- 

 ment, that those cells situated most posteriorly have two nucleoli, 

 since the 3 or 4 cells I found with two nuclei, were placed in the se- 

 cond third of the body. The cells in each chord are usually at con- 

 siderable, but irregular, distances apart, the dorsal cells not regularly 

 alternating with the ventral ones ; neither are the cells of one chord 

 paired with , nor do they regularly alternate with those of the other 

 chord. The only apparent regularity in the distribution of the neuro- 

 chord cells, is that areas where they are comparatively numerous 

 alternale usually with areas where they are scarcer, or wholly absent. 



In regard to the distribution of the three other species of gang- 

 lion cells, I can corroborate Coe's (1. c.) observations. Structurally, 

 though not necessarily genetically, the first type of ganglion cells is 

 very distinct from the three others ; while the third type is in certain 

 points related to the second type, and in other points, to the fourth 

 type (i. e. neurochord cells). 



In the caudicle, the lateral chords are devoid of ganglion cells, 

 each being surrounded on the outer surface, as well as dorsally and 

 ventrally by a thick layer (4 — 6 cells deep) of mesenchymatic cells; 

 these cells differ in the greater size of their nuclei, and in their mar- 

 ked multipolarity, from the mesenchym cells of the anterior body 

 regions. Since the specimen examined was immature, it is possible 

 that ganglion - cells have not yet been produced in this region, and 

 that they may be present in the caudicular portion of the chord, in 

 the adult animal. 



The anus is situated at the tip of the caudicle, and not, as Coe 

 observes, »just beneath« it. 



The connective tissue elements of the Nemerteans have never 

 been satisfactorily classified. Hubrecht (1887) compares the nemer- 

 tean connective tissue with the medusoid »jellytf, describing it as a 

 gelatinous, homogeneous mass, in which branched cells occur. Ken- 

 ne 1 ( 1 8 7 8) gives a similar discription of the tissue in Malacohdella, giving, 

 further, observations on its histogenesis. Salensky (1884) describes 

 in Monopora the lining of the coelom, as consisting of the somato- 

 pleura (a single layer of flattened cells), and the splanclmopleura (»un 

 tissu parenchymateux«) ; adding: »C'est le tissu conjonctiv qui revêt 

 les cordons nerveux latéraux, qui est le lieu de formation des ovisacs 

 et probablement aussi des testicules«, lî urger (1. c.) divides the 

 nemertean connective tissue into the intramuscular tissue, and the 



