SABBLLIDES AND SERPULIDES 257 



ing the uncial plates of Spirorbis borealis Daudin as having the lowest 

 tooth larger than the others, as this does not appear to be the case 

 in any of the species studied. The embedded portion does not extend 

 to the end of the thicker surface plate, on which the teeth seem evenly 

 spaced. He bases his subordinate divisions on the differences in form, 

 not only of the collar setae, but of all the others, with the result that 

 species of Spirorbis are widely separated and referred to other genera, 

 as instanced in the following : 



Spirorbis, with S. borealis Daudin as type, is placed under his 

 second grand division with Filograna Oken, Salmacina Claparede, 

 and Filogranula Langerhans. Spirorbis granulatus Linn6 is re- 

 ferred to Pileolaria Claparede, type P. militaris Clap. ; S. pagen- 

 stecheri Quatrefages to Janua gen. nov. as type; S. corrugatus 

 Montagu and S. lucidus Montagu to Circeis gen. nov., type C. ar- 

 moricana Saint -Joseph ; S. Icevis Quatrefages to Leodora gen. nov. 

 as type. 



It can be readily seen by this and a further study of the relative 

 positions of these genera that such an arrangement can be adopted 

 only by ignoring the generic importance, not only of the presence of 

 a calcareous plate in the operculum, but also the number of thoracic 

 segments. 



The few odd setae long blades with deeply serrate (comb-like) 

 ends, similar to those of Apomatus* Philippi 1844 (type, A. ampulli- 

 ferus), and short setae like those of Salmacina Claparede 1870 (type, 

 S. incrustans) said to be found on the second and third thoracic seg- 

 ments are also given generic value. In the large number of animals 

 examined, however, only the long form has been noticed in the third 

 fascicle, although in a few instances these varied in length in the same 

 bundle, but they were not always found. Whether they simply failed 

 to show in the mounting, or really do not always occur, or occurring 

 only during the breeding season are easily lost, has yet to be deter- 

 mined ; therefore it seems undesirable to make them of much im- 

 portance. Caullery and Mesnil (1897) arrived at the same conclu- 

 sion, and, finding it impossible to follow Saint- Joseph, discarded five 

 of his generic names, but proposed other subgencric ones based on 

 different characters (see p. 256). 



As the form of the superior collar setae is of first importance in 

 identifying species, and with care can be invariably determined even 

 in very poorly preserved specimens, an attempt has been made to dc- 



1 These are also very like some of the dorsal sct found in Amphitrite, as 

 figured bj Saint-Joseph (1894). 



