G68 SIR c. ELIOT ON NUDiBEANCHS [June 19, 



Tlie two genera are clearly distinguished. Externally the chief 

 difference is that Trevelyana has numerous small branchiae and 

 Netnhrotha a few (3-5) large branchife. In Trevelyana the radula 

 is fairly wide ; the rhachis bare ; the teeth are awl- shaped or 

 slightly hamate, and though the innermost is generally distin- 

 guished from the rest, it is not of an essentially different shape. 

 In Neimhrotha, on the other hand, the radula is narrow ; there is 

 a rhachidian tooth ; the first lateral is large and falciform, the 

 rest are mere plates. Also, whereas in Neimhroiha the hermaphro- 

 dite gland is spread over the liver, as is usual in the Dorididfe, 

 in Trevelyana it is quite separate from the liver and forms two 

 globular masses in front of it. 



This arrangement is very rare in the Dorididse and is charac- 

 teristic of such forms as Scyllcea, Bornella, &c. It might be 

 supposed that it would not occur in a Doridiform animal without 

 being accompanied by other profound structural modifications ; 

 but it is found not only in Bathydoris, but also in Alloiodoris, 

 which, but for this peculiarity, seems to be a perfectly ordinary 

 Dorid. 



It will thus be clear that it is not easy to see how Trevelyana 

 can be derived from JSfemhrotha or vice versa. Nembrotha may 

 be regarded as an animal analogous to Triopa which has lost its 

 appendages, though the dentition is not exactly the same. But 

 Trevelyana cannot be so explained. In its dentition, though not 

 in other respects, it shows greater resemblance to Rotodoris. 



About nine species of Trevelyana seem fairly certain : — 



1. T. ceylo7iica Kelaart. 



= T. ruhromaculata Bergh, 

 = T. picta Pse. 

 ? = Boris impudica Biipp. & Leuck. 



2. T. bicolor A. & H. 



3. T. citrina B. 



4. T. alba B. 



5. T. inornata B. 



6. T. plebeia B. 



7. T. crocea B. 



J 8. T. coccinea Eliot. 

 [9. T. rubropapulosa B. 



Bergh in his ' System,' p. 1144, includes in his list Trev. ? rubra 

 Pease, but in the Opisthobranchs of the ' Siboga ' has inadvertently 

 registered a form under the same name as a new species. The 

 specimen was, however, small, and its state of preservation 

 rendered a full description impossible. It may be the same as 

 T. rubra Pse., which is very incompletely described. T. concinna 

 Abraham, of which nothing is known except the external fea- 

 tures of an alcoholic specimen, does not seem to me sufficiently 

 characterised. 



The remarkable Tr. ? defensa described by Bergh (Siboga, 

 pp. 192-3) must, I think, be regarded as a new genus if not a 



