102 Proceedings of the Royal Society of Victoria. 



plug (c. pi.) forms a relatively narrow stalk, which seems to 

 have a sliglitlj^ tibrous structui-e. On i-eachiiig the cavity of 

 the proboscis sheath it swells out into a rounded mass of 

 cells, covered, I believe, with a flattened epithelium continu- 

 ous with the epithelium which lines the proboscis sheath, 

 and projecting into the rhynchocoelom. 



The median vessel seems, according to my observations, to 

 be connected with both of these cellular plugs. It appeal's to 

 be directly continuous with the stalk of the one (Fig. 18) 

 and to send off a branch to the other. Whether there is any 

 constant distinction between the right and the left in this 

 respect I am not able to say. 



These cellular plugs appear to me to be probably vestigial 

 structures, and to indicate two things — (I) a formei- more 

 intimate connection of the median vessel with the proboscis 

 sheath and the rhynchocoelom, such as at the present day 

 exists in many marine Nemer tines, "■ and (2) the former 

 existence of a pair of vessels one on each side of the middle 

 line beneath the proboscis sheath, instead of a single one as 

 at present. Such a pair of vessels exists at the present day 

 in what Oudemans-|- calls the " Palseo-type." I have been 

 unable to determine whether the right or the left vessel 

 constantly persists in Geonemertes or whether it is some- 

 times one and sometimes the other. Not expecting to meet 

 with an}^ distinction between right and left sides I did not 

 take sufHcient care in orientating my section series to justify 

 me in forming a definite conclusion on the point in question. 

 In other words, although the proper sequence of the sections 

 has been rigidly maintained I am not absolutely certain 

 that all the series have been mounted with the same side 

 u])j)ermost. It is some time since many of the sections 

 were cut and I do not like to trust to my memory on such 

 a point. 



At its posterior extremity the median 'vessel is continued 

 into a vessel of smaller diameter, which in histological 

 structure presents a veiy interesting transition between the 

 main median vessel and the network of excretory tubules 

 to be presently described. A portion of this part of the 

 median vessel is shown in Figure 20. It will be seen that 

 the outer layer of large vesicular cells is absent and that 



* vide Oudemans : — " The Circulatory and Nephridial Ap])aratus of the 

 Nemcrtea," Quarterly Journal of Microscopical Science, Vol. XXV., Suppl. 

 t Log. cit. 



