1904.] 



MARIXE FAUNA OF ZAXZIBAE. 



317 



considerably bent (text-fig. 59, p. 316). In the posterior feet this 

 tendency to the addition of a third smaU hook often makes itself 

 appai-ent in the oi-dinary compound setse also, and even the acicula 

 are sometimes bifid at their extremity, though the form of the 

 projections they beai- is not such that they could rightly be 

 described as hooked, PL XXII. figs. 5 & 6. (Compare the species 

 described below, and also E. elseyi, Baird, ' Ohallengei' ' Repoi-t. 

 xii. pi. XX A. figs. 14, 15, 16.) 



There ai-e two acicula throughout, and only one or two acicular 

 setse postei'ioi-ly (cf. E. incUca, where there may be as many as 

 four). 



Tlie gills are Avell developed, very regular in foi-m, and of stiff 

 consistency, the name ^'■flaccida" being equally a misnomer 

 whether applied to them or to the body of any of these specimens, 

 large or small. They vary in length and complication to some 

 extent independently of the size of the worm, but never quite 

 cover the back. The first appears with great regularity on the 

 6th foot, only in 16 per cent, being on the 5th, and only one case 

 each on the 3rd, 4th, and 7th. The largest gills, which may be 

 composed of 15 filaments, are only found anterioily fi-om the 7th 

 to -the 20th foot in small specimens, or as far as the 40th in large. 



Text-fie-. 60. 



W^-'^F^ 



Three feet of lH. antennata (a small specimen of vav. viridis), shov^ing the pi'o- 

 portions of the gills in anterior, middle, and posterior parts of the body. 



From this point to one approximately the same distance fi-om the 

 tail a variable amount of reduction takes place, after which, in the 

 hinder thuxl of the bod}', the gills again become lai-ge (see figure 

 of hind end of body, PI. XXII. fig. 7 and text-fig. 60), though 

 never equal to those of the anterior end *. The reduction in the 

 middle region may be sufiicient to cause a striking difTerence in the 

 appeai'ance of the animal, or, in large specimens, be not apparent 

 until a close examination is made. In a few specimens the gills 

 are reduced to single filaments, or even disappear altogether in the 



* A similar phenomenon is noted in the ca.-e of IS. iorresiensis by Mcintosh 

 ' Challenger,' xii. p. 271. 



