1903.] MARINE FAUNA OF ZANZIBAR 131 



on feet 14 to 35. The prostomium is of nearly the same size in 

 both specimens, yet the unpaired tentacle, when laid along the 

 back, extends beyond the anterior border of the fourth segment, 

 or only up to that of the second segment. The colour of the two 

 specimens differs markerlly, being much duller in no. 2. 



Similar results were obtained in the case of Diopatra neapoli- 

 tana, in which species such variations are much more conspicuous 

 in accordance with the great development of the organs concei-ned. 

 The resulting differences of facies have caused some synonymy, 

 and have made some authors content to give quite insufficient 

 descriptions. 



The Onuphidin^. 



The genera Rhamphohi^achium (Ehlers), Onuphis (Aud. et Ed.), 

 Hyalinmcia (Malm), and Diojmtra (Aud. et Ed.) form a very 

 well-marked group characterised by modifications for a perma- 

 nently tubicolous mode of life. 



The last three genera are differently defined, however, by the 

 pi-incipal authorities. All authors but Ehlers agree in separating 

 the genus Diopatra because of its very characteristically formed 

 and distributed gills. In other cases Ehlers would be theoretically 

 right in objecting that the gills of Annelids are too variable to 

 be made a basis for generic distinctions, and in urging that the 

 gill-less species are derivable from either Onuphis or Diopatra, if 

 the distinction between these genera is maintained. Practically, 

 however, we find no form with gills intermediate in structure 

 between those of Diopatra and any other Eunicid whatever ; and 

 this distinction is emphasised by the fact that no other member 

 of this group has its largest gills confined to the anterior part of 

 the body. There is no real necessity, therefore, for the confusion 

 introduced by assigning the two gill-less species, /?'«c/osa* and 

 glutinatrix*^ and the forms with pectinate gills p>osteriorly ^ 

 pourtalesii*, magna t, and dorsalisX, to the genus Biopatra 

 instead of to the genera Paronuphis and Onuphis, as would have 

 been done by any other writer. 



The confusion in the definitions of the genera Hyalincecia and 

 Onuphis has its origin simply in the name of the former. The 

 addition of a coating of mud or larger pieces of foreign material 

 to the foundation of the tube secreted by the body of the worm 

 cannot be regarded as a sufficient cause for generic distinction, 

 however striking the difference in appearance of the tube. The 

 name Hyalincecia may be retained when not literally applicable, 

 the transparency of their abode being characteristic of the great 

 majority of the species. 



The character of the gills forms the basis of a distinction 

 between nearly all the species of the twp genera, but, as else- 

 where, all stages are found betweeia the typical comb-like gill of 



* Elilers, " Annelids of tlie U.S. Survey ship ' Blake,' " Cambridge, Mass., 1887. 



t Andrews, Pr. U.S. Nat. Mus. xiv. p. 277. 



J Ehlers, 'Hamburger Magal. Sammelreise,' p. 71 (1897). 



9* 



