BRITISH NEMERTEANS, AND SOME NEW BRITISH ANNELIDS. 5361 



with transverse rugse. Towards its first part the duct is surrounded by a 

 minutely granular glandular structure, which usually has a somewhat triangular 

 figure. Several glandular masses lie behind, one to the outer, and another to the 

 inner side in this position. The glandular substance around and behind the posterior 

 part of the ciliated external duct contains numerous granules and finely granular 

 circular cells. From the posterior end of the outer mass in such a view, a structure 

 that appears to be a pale duct passes obliquely towards the superior lobe of the 

 ganglion, crossing this for some distance in a direction inwards and backwards. 

 Traces of a cavity are apparent at its commencement, and, besides, it is distin- 

 guished from the adjoining nerve-trunks under pressure by not being continuous 

 with the ganglion at its edge. In transverse sections of the snout, each sac is seen 

 to occupy a position to the outside of the cephalic blood-vessel, and somewhat above 

 it (Plate IV. fig. 1, m), and to have a special space in the muscular stroma of the 

 head. In large specimens the sacs contain many reddish pigment-granules, and 

 occasionally a large cell filled with coarse granules. Behind the foregoing glan- 

 dular apparatus lie the coiled ciliated ducts (m'), which are sometimes pale and 

 irregularly bulged from included fluid, or else collapsed and minutely granular in 

 aspect. In some specimens of 0. alba the commencement of the duct is tinged 

 of a faint reddish hue. There seems to be no ground for the supposition that 

 the sacs are connected with other organs. In 0. melanocephala they are less 

 dilated than in 0. alba. The coils of the ciliated duct in 0. gracilis are most 

 elaborate, and can be traced for a long distance backwards by the side of the 

 nerve-trunk. In O. purpurea* the external apertures are not so evident as in 

 0. alba and Tetrastemma, because the furrows are less distinct when viewed as 

 transparent objects. They are best seen when the ventral surface is upturned, 

 and occur in the angle of the furrow some distance from the margin of the head 

 in this position. The ciliated pit leading inwards is short. Like other parts of 

 the animal, there is a considerable variation in 0. pulchra from the typical form 

 in the shape and position of these sacs as well as in regard to the furrows. The 

 latter species has numerous short longitudinal or accessory furrows on the 

 front of the ventral grooves, and in this respect is allied to the Borlasia 

 sphndida of Prof. Keferstein. Instead of lying in front of the ganglia (in the 

 ordinary position under examination), the sacs are situated laterally and pos- 

 teriorly, forming somewhat elongated pyriform organs, which adapt themselves 

 to the curves of the ganglia. Each sac is filled with rounded granular cells, 

 reddish pigment and other granules, has a ciliated duct anteriorly, which opens 

 at the constriction or lateral dimple of the head just in front of the ganglia, and 

 posteriorly ends in a ciliated tube which by-and-by bifurcates and extends for a 



* I have a strong suspicion that this is the same species as the Borlasia camillea of M. de 

 Quatrefages, which he places next B. anglice, an association, if I am correct, founded on erroneous 

 principles. 



VOL. XXV. PART II. 4 Z 



