0)1 an Austndkui Land Nemertine. 103 



the whole vessel closel}'' resembles one of the excretoiy 

 tubules with the additioii of a thin layer — a])i)areiit]y 

 discontiuuo\is — of circular fibi-es around the outside. 



The position of the two lateral vessels is shown in Fig. <S 

 (J. v) They are not so distinct and easily recognisable as the 

 median vessel and never seem to attain to such complexity of 

 histological structure. I have not been able to recognise 

 either the circular muscles or the outer coat of vesicular cells 

 in their walls and they seem even more like a specialised 

 portion of the network of tubules to be described ])resently. 

 Nuclei can be distinguished in their walls and occasionally 

 the vessel dilates into irregular lacunae (Fig. 23, I. I. c.) 

 What happens to the lateral vessels at the anterior and pos- 

 terior extremities I cannot say, but in the head region there 

 are a number of wide, irregular lacunas into which they 

 probably open. 



In sections which have been stained with Kieinenberg's 

 hoematoxylin a system of tine, apparently intra-celluJar 

 tubules (Figs. 7, 8, ex. t ) is very distinctly visible, ramify- 

 ing through all parts of the body between the muscle layers 

 and the in-oboscis sheath. I'he.se tubules run in all directions 

 and blanch freely, but they are especially developed in the 

 region of the bod}^ above the proboscis sheath and they 

 generally, though hy no means alwa3^s, run in a direction at 

 right angles to the long axis of the body, forming a series of 

 irregular loops curving over the proboscis sheath from side 

 to side. They are also to be found running transversely 

 beneath the alimentary canal. As already stated the}^ 

 branch freely and some of the branches run in the direction 

 of the long axis of the body. These tubules open into 

 the lateral vessels (Fig. 23) and probably also into the median 

 one. I have not been able to demonstrate any opening of 

 the tubules into the latter but they can be traced very 

 close to it and the transitional condition of the median 

 vessel at its posterior end in regard to histological structure is 

 indicative of a close connection between it and the network 

 of tubules. The histological structure of t.he tubules (Fig. 

 19) points to an intra-cellular nature. Tlie3^are very narrow 

 and at iairly regular intervals present very distinct swellings. 

 Each of the swellings is caused by the presence of a nucleus 

 which curves [)a-rtially round the tubule and which stains 

 very darkly with ha3matoxylin, thus rendering the tubules 

 very conspicuous in sections. Sometimes the tubules appear 

 to be empty and sometimes they appear to be filled with a 



I 2 



