1904.] MARIXE FAUXA OF ZAXZIBAR. 325 



all stages occurring between the ' valida ' and siciliensis types. 

 Indeed, this was noted long ago by Grube in the case of the 

 features (1) and (2), and variations in the colour of living examples 

 are given above. In spiiit, colomiess specimens are not common, 

 but shades of light and dark brown are numerous. 



Gravier's figures (pi. 13. figs. 71-82, and text 130 and 134) are 

 excellent illustrations of the species and the extremes of its 

 variation. 



Ehlers's South American species E. leucodon difliersin (1), as the 

 name implies, the bi-oad white band surrounding the plates of the 

 upper jaw and the almost completely calcareous composition of 

 the lower ; (2) the shortness and ringing of the tentacles. The 

 mottling of the body -pigment and other small differences may be 

 found as variations in E. siciliensis ; but I do not find in these 

 collections any specimens the tentacles of which are shorter than 

 one and a half times the length of the prostomium, or the upper 

 jaw-plates of which are not quite black with but a narrow and 

 inconspicuous white edging. Ringing of the tentacles may oi- 

 may not be developed. The points of agreement between the two 

 species so preponderate over these small, but constant difierences, 

 that it seems best to follow out Ehlei's's own suggestion and to 

 I'egard his specimens as belonging to a local vaiiety of the cosmo- 

 politan species JiJ. siciliensis. 



Claparede and Ehlei's (Borstenwiirmer) figui-e compound setse 

 with elongated articulated pieces, whereas Gravier's figures (of 

 both varieties) show shoi't triangular hooks as in Ehlers's figures 

 of the leucodon variety. These seta? vary considerably, those from 

 antei-ior segments differing from those found posteriorly in the way 

 shown by figs. 8, a & h, PI. XXII., which represent the extremes 

 of the forms met with. Those of the hinder feet are much the 

 stouter, as here commonly one compound seta is found alone. 

 Comb and acicular setse are entirely absent fi'om all these speci- 

 mens, even in cases where two acicula occur in a few feet. Theii- 

 absence forms a very definite distinction between this and others 

 which have simple gills, e. g. E. schizohrancMa Clap. *, E. maren- 

 zelleri Gravierf, and E. cirrohranchiata McI. J. These three 

 species indeed show none of the othei- features of specialisation 

 which make E. siciliensis so well marked a foi-m. 



Several specimens, e. g. four from Hulule, Male Atoll, and one 

 from Minikoi i-eef, show remarkable papilla' disti-ibuted over the 

 three middle tentacles, as shoAvn in fig. 9, PI. XXII. Possibly 

 these are connected with the " papilles en forme de domes sur- 



* Oj3. cit., Supplement, 1870, p. 30, and pi. ii. fig. 6. This species has been regarded 

 as sjuonymous with IE. s/ciZieMS?s by later workers. Though Claparede says nothing 

 about its jaw-apparatus and lays emphasis upon probably variable features such as 

 colour and propoi'tions of the buccal segment, the presence of these seta: remains an 

 indubitable distinction of quite specific importance. 



t Op. cit. p. 230. 



J ' Challenger ' Eeports, vol. xii. p. 277. It is remarkable that no specimen of so 

 common and widely distributed a species as E. siciliejisis occurred in the 'Challenger' 

 collection. 



