POLYCH.'ETA. BENHAM. 22? 



In a recent article Cros.sland (1904) has tentatively sug- 

 gested that E. hassensis is a synonym of E. antennata, Savigny 

 (1820). Bnt if my identification be correct, E. bassensis 

 differs from that species as described and figured by the 

 authors, Savigny, Gravier (1900), and Crossland, in the fol- 

 lowing features : — 



(1) E. antennata appears to be a small species. Savigny i 

 gives its length as 3 inches. Crossland states that those 

 from Zanzibar are 100 by 5 mm. (including the feet), those 

 from the Maldives are SO x 4 mm., and less. 



E. flaccida, Grube,^ which Crossland identifies with this, 

 was fragmentary ; the 67 segments measure 35 x 2 mm ; 

 while Grube's variety gracilis measured 78 x 2.5 mm. for a 

 worm with 85 segments. Compare this with our first indi- 

 vidual, where the 84 segments measure 60 by 5 mm. In 

 other words, this ''I'asmanian species is evidently a larger 

 worm . 



(2) Crossland^, in his drawing of the head (i'l. xxii., fig. 1) 

 from above, shows that the upper edge of the lower lip is 

 prominent, and is visible from above, as in E. pycnobranchiata, 

 while there is no excavation of the anterior margin of the 

 peristomium. 



(3) The gills are shown as nearly meeting dorsally, whicn 

 is far from being the case in any of our individuals. 



(4) The form of the ventral cirrus, as given by him, is 

 represented (text fig. 60, p. 317) in the 10th segment as long 

 and narrow, without the swollen base which is so evident in 

 ours, and it projects further beyond the lip of the chsetophoraJ 

 sac than in ours. 



(5) The dorsal cirrus is apparently much more slender. 



(6) The nuchal cirrus is shorter. 



(7) According to Crossland's figure of the acicular chaeta 

 the main tooth forms nearly a right angle with the axis, and 

 it may be noted that Gravier's^ figure of E. fiaccida (p. 256) 

 is less than a right angle, whereas in the present sjDecies this 

 angle is much more o])en. 



(8) The jaws of E. antennata show a considerable range of 

 variation ; and if I understand Crossland's formulae on p. 316 

 "G— 7 : 10-f9— 8" to mean '"loft dental plate (II) has six 



1. Savigny — System. Annelidcs, 1820, p. 50, pi. v., fig. 1. 



2. Grube — Annel. d. rothen Meeres, Monatsber. k. preuss. Akad. Berlin, 

 1889, p. 491 (separate copv, p. 11). 



3. Crossland— Proc. Zool. Soc, 1904, I., p. 312. 



4. Gravier — Nouv. Arch. ]\[us. Paris, 1900, p. 2.")5. 



