XERJLLA AX ARCHIANNELID. 
419 
differing genera of that group. The affinities of Nerilla are 
not with one genus in particular, but with several. It thus 
binds the Archiannelida together, while at the same time to 
some extent bridging* over the gap between them and the 
Polychaeta. 
Comparison of Nerilla with other Archiannelids. 
In common with Dinophilus and Protodrilus, it possesses a 
ventral ciliated tract, which forms a distinct longitudinal 
groove as in Protodrilus. These three genera also are pro- 
vided with very similar ciliated rings. Simple parapodia are 
found in Saccocirrus similar to those of Nerilla in structure, 
but differing in the presence of only one bundle of cliaetae 
instead of two ; a point in which the latter approaches the 
. true Polvchaetes. Possibly the tentacles of Polygordius, 
Protodrilus and Saccocirrus are homologous with the lateral 
tentacles of Nerilla, in spite of structural differences; but it 
is undoubtedly the Histriobdellids which come nearest to it 
in the possession of shorter but similar processes on the head 
and trunk segments. The nervous system, closely connected 
with the epidermis, bears a marked resemblance to that of 
Protodrilus. The alimentary canal of Nerilla is very like 
that of Dinophilus ; as in all Archiannelids, with the single 
exception of Polygordius, 1 there is a muscular veutral 
pharynx. The pharynx is unarmed, as in Dinophilus and 
Saccocirrus. The nephridia are more like those of. Poly- 
gordius, Protodrilus, or Saccocirrus in general structure; in 
distribution and specialisation in the two sexes they resemble 
those of Dinophilus and Histriobdella. Certain specialisations 
in the glands and ducts of the genital organs and the. presence 
of a median male pore seem also to point to affinity with these 
two genera. 
In a work on Saccocirrus published some years ago (8), I 
pointed out that the absence of chaetm in some Archiannelids 
cannot be considered as a primitive character, since Sacco- 
1 See footnote on p. 404. 
