2 PERCY SLADEN TEUST EXPEDITION. 



In a collection of Nemertea from any region the genera Lineus and Cerebratulus are 

 sure to bulk largely, and the determination of the various species on preserved material 

 is the most harassing part of the nemertinologist's task. For this reason we have 

 collected together in the form of an Appendix the scattered observations dealing with 

 anatomical points which we feel to be likely to prove of service. We trust that the 

 recording of these points wiU commend itself to future observers, and that the numerous 

 gaps in the table will become gradually filled. 



But though the validity of a Lineid species may be established on its anatomy the 

 systematist is yet confronted with the task of referring it to one of the three genera 

 Lineus, Micrura, or Cerebratulus. In the days when comparatively few forms were 

 known this was easy, but with the rapid increase in the family of the Lineidse it has 

 become more and more difficult to frame definitions for these three genera. The 

 absence of a caudicle has been relied on by most authors for the separation of the genus 

 Lineus from the other two ; but a delicate structui-e like this is a most unsatisfactory 

 feature to work with — it is frequently missing in the living and almost always in the 

 dead. Nor do we consider it to be a good generic criterion, apart from these practical 

 difficulties. Its use in classification may unite forms which are widely difi"erent in other 

 respects, and at the same time may relegate to different genera forms which differ in 

 little else but the presence of a caudicle *. 



In most zoological families it would appear that the constituent species fall into 

 several groups, such that all the members of a given group are differentiated from the 

 members of another group by several constant peculiarities, and these generic characters 

 are held to differ from specific ones in their greater antiquity. It is not possible to point 

 to any such grouping of characters among the Lineidte. Though there is considerable 

 diversity of structure among the members of the family, and though the systematist has 

 to deal with a number of characters, there is yet no association or gi-ouping of these 

 characters to admit of our establishing genera with obvious claims to be considered 

 natural. Nevertheless, the family already contains more than 150 species and for 

 cataloguing purposes some form of subdivision seems to be called for. In the present 

 paper we have referred to the genus Lineus such Lineidse as possess a gelatinous 

 connective-tissue layer and to the genus Cerebratulus such as are without it. But in 

 assigning a species to one genus or the other we have been guided purely by convenience, 

 and we do not wish to imply that in our opinion there is any peculiar phylogenetic 

 significance in the criterion which we have chosen. 



To one further point we wish to call attention in this connection. Since future 

 sy sterna tists will doubtless bring fresh ideas to the arrangement of this mass of species, 

 we venture to suggest that in the naming of new forms care should be taken not to 

 apply to a new species a name which has already been used for a member of any of the 

 three genera in question. 



* Thus Mkrura varicolor and Lineus ruber (olim guserensis) differ in little else than the possesaion of a caudicle 

 by the former (Punnett, Bergons Mus. Aarbog, p. 21, 1903). 



