NO. 5 HARTMAN : POLYCHAETOUS ANNELIDS 489 



Furcate setae present or absent 



Pygidium a simple collar with terminal aperture and with or without 

 cirriform processes. 

 The most striking disparity concerns the dual nature of the proboscis 

 since it is cylindrical and muscular in Ancistrosyllis, globular and epi- 

 thelial in Pilargis. Furthermore, notopodia in Ancistrosyllis are provided 

 with setae or spines ; notopodia in Pilargis have only embedded acicula. 



Characters common to Loandalia and Talehsapia 



Body long, more or less cylindrical with the anterior region inflated 



Prostomium reduced, lacking eyes or antennae 



Peristomium a simple ring, lacking cirri 



Proboscis cylindrical and muscular 



Parapodia uniquely similar, with small, papillar notopodia that lack dor- 

 sal cirri but are provided with heavy single acicula or also a few 

 fine capillary setae; neuropodia large, truncate with small ventral 

 cirrus near its distal ventral edge 



Neurosetae geniculate, distally pointed, with transverse rows of teeth 



Integument smooth or faintly reticulated 



The most striking difference in these 2 genera concerns the dual 



nature of the prostomium since Loandalia has biarticulated palpi whereas 



Talehsapia totally lacks palpi. 



Characters common to Ancistrosyllis and Loandalia 



Surface epithelium smooth 



Prostomium provided with biarticulated palpi 



Proboscis cylindrical and muscular, without jaws 



Notopodia provided with setae or spines in addition to projecting acicula 



The most striking incongruity concerns the dual nature of parapodia, 

 especially the arrangement of setal fascicles and the microscopic details of 

 setae; in addition Loandalia has an anal plaque that is lacking in Ancis- 

 trosyllis. 



Both Loandalia and Talehsapia (or possibly also Cabira) are char- 

 acterized by extreme reduction or loss of parts, especially concerning the 

 prostomial and peristomial structures. Moreover, in the structure of para- 

 podial parts as also in the inflated anterior region they depart widely from 

 species in other genera. On the other hand, Ancistrosyllis and Pilargis 

 are more nearly related to each other than to the first two. If a super- 

 generic separation were desirable or attempted, it would necessarily be 



