294 



MR. CYRIL CROSSLAXD OX THE 



[Feb. 16, 



filaments, viz. four to six. In the larger specimens from East 

 Africa the number may rise as high as ten, though in no case are 

 the gills large enough to meet over the back. Posteriorly, over the 

 greater jDart of the body the gills, though remaining of nearly 

 the same length, are simpler, being usually composed of two or 

 three filaments only. 



In the very small specimens from the Maldives the gills are 

 proportionately small, their degree of development depending 

 roughly upon the size of the worm. Thus the smallest, which is 

 incomplete and but 1 mm. broad, has only two pairs of gills 

 of three filaments, the others anteriorly being of two and pos- 

 teriorly of only one. Those of 2 mm. broad have gills of three or 

 two filaments anteriorly and of one posteriorly, while two of the 

 fovir which are 3 mm. broad attain to gills of four filaments in the 

 anterior part of the body. 



The remaining Maldivan specimens are too iiniform to afford 

 further evidence of this dependence. The following table describes 

 those from East Africa : — 



There is here seen to be variation between specimens of the same 

 size, though, on the whole, gill complexity is correlated with increase 

 of the size of the body, a conclusion corroborated by a comparison 

 of the earlier descriptions of this species. The second Seychelles 

 specimen is distinctly abnormal in the early commencement of the 

 gills, but this may be connected with the fact that whereas in other 

 specimens these attain their full size three or four segments after 

 their commencement, in this case the increase is more gradual. 



This variation is a great deal wider than is that commented 

 upon by Ehlers when comparmg certain specimens from East 

 Africa with these originally named by Peters *. To explain the 

 fact that some specimens have gills of four filaments while othei-s 

 attain to six, he propounds the theory that the species is sexually 

 dimorphic in respect to its gills. It is easy to imagine a priori 

 reasons for such a dimorphism, but so far as this account is 

 coucei'ued the grounds foi- the assertion are of the slightest, and I. 



Eh 



Nachr. zu Gott. 189/ 



