POLYCJIiETA—BENHAM. 303 
dorsal surface is very convex, and the segmentation is obscured by numerous closely 
set annulations. The ventral surface of the thorax is highly glandular, but no definite 
“ gland shields ” are delimited, as the glandular modification of the integument extends 
across the ventral surface from right to left unicinigerous nueropods; but on the last 
three segments the outlines of the glands are evident.. 
In the abdomen the ventral surface is deeply concave, crossed, however, by 
segmental rounded, transverse ridges. 
The prostomium (figs. 124, 125) consists of two portions, viz— (a) an anterior 
freely projecting flap overhanging the mouth; and (b) a posterior thickened transverse 
fold, which is almost entirely hidden by the basal portion of the second or branchi- 
ferous segment. 
The prostomial flap, or upper lip, is slightly trilobed, the middle lobe being more 
prominent than the later regions, from which it is marked off by a slight notch on each 
side. The middle lobe is inclined forwards and upwards, and has a somewhat thickened 
free edge. On raising the prostomial flap, or on looking into the mouth from in front 
(fig. 127), the base of the prostomium is seen to be continuous, with a transverse lobu- 
lated structure, or “ supra-oral arch,” which is separated from it by a furrow. The 
right and left extremities of this arch touch the upper part of the lower lip on either 
side ; the median portion of it is smooth, and traversed by a number of fine furrows 
radiating from its base forwards towards its edge; the lateral portions are thick and 
swollen. 
The buccal segment or perlstomium is represented dorsally by a rounded trans¬ 
verse ridge, overlapped and concealed by the branchiferous segment. Ventrally, 
however, it is produced forwards to form a great lower lip, which is separated from the 
lateral region of the prostomial flap by a deep, horizontal cleft on each side, and it is 
evidently very mobile (fig. 126). 
Between the upper and lower lips there projects a folded membrane (fig. 127), 
which occupies the entire oral cavity. At first I supposed this to be a part of the gut 
everted, but found on dissection that it has the following relations, from which it is 
clear that it represents the series of tentacles of other Ampharetids. The free edge of 
this tentacular membrane ” is thickened in the median region, but becomes thinner 
towards each side. It is folded much in the way that a partially closed fan is folded, 
but the folds are few and irregular. It arises from the under surface and posterior 
margin of the “ supra-oral arch ” above mentioned (fig. 12S), which is thus seen to 
correspond to the tentacle-bearing ridge of other genera. The line of origin of the 
tentacles is at about the level of the junction of the arch with the prostomial flap. At 
this point—as was seen by slitting open the body wall the buccal cavity (or oesop¬ 
hagus ?) commences ; this is a tube with a thick, muscular wall, whose inner surface is 
thrown into a number of rugae. Its floor is produced forwards to form an internal lip , 
such as that figured by Fauvel (1897) for Ampharete grubei. 
