264 Sou then I ( ^ross. 



AiUaivlic, ItuL it undriiialily (ilisfiuvs the liinit.s oi' llic genera 

 If(fnii(>t/mc aiul JA(ijisi((. Jolliers (18'.*7) lias also expressed hiiiiseli' 

 in (lonbt as to the relations of these two genera in Antarctic waters 

 (see below, \>. 2GG). 



1. IIau.mothoe SriNOSA. 

 (^I'l. XLl., tigs. 1-1, ai.a ri. XUri., f^^s. 1-2 and 4-8.) 



Ilarmothoe spinosa, Kinboru, •'. 'i. II., IHy"). Ofv. Ak. Fork., Stocklioliii, 

 |.. ;5S(i. J,i., KS57. 'Eugenics Jtosa,' p. LM, Taf. VI., lig. ol. Khleis, K., 

 It-'.tT, rolychaeten, Ilajiibuiy. Aiti(jitlhiteiis. iSdiitinchrcisc, p. 12. 



This s}iecies seems to be the representative, in the Antarctic 

 rej;ious. ot" the northern species //. imhricata (L.), the principal 

 diflerence lielween them, so far as I can ascertain by comparison of 

 specimens, relating to the situs ocAdorum. In the Antarctic species 

 the four eyes are visible from above, while II. imhricata, as pointed 

 out by j\[clntosh,^ shows only a single pair of eyes in dorsal view, 

 the anterior pair being concealed below the frontal cones. Both 

 species present a very great range of variation in the iinibriation 

 and pigmentation of the elytra, which has been alluded to by Ehlers 

 (1897) and Mcintosh (1900). 



The number of segments, including the buccal segment, frequently 

 adils up to thirty-seven, but this number possibly represents a 

 particular condition in which the worms are commonly taken, and 

 may not have the signihcance of a fixed (piantity. In one case 

 twenty normal segments are followed abruptly by twelve to thii-teen 

 small regenerated segments ; another specimen of 40 mm. has thirty- 

 eight segments ; in a third specimen of 33 mm. the segments, thirty- 

 four in number, taper to a point behind, but no anal cirri are present.'- 



Judging from the material at my disposal, the variations would 

 seem to culminate in three principal allotypic modifications, of which 

 a brief description follows. One of these modifications is for the 

 jireseut retained as a distinct species (see p. 266). 



' Mclritosli, W. C, IIKX): ' British Aunelids,' Bay Sor. Monograph, \\ 31(>. 



* Killers (lHi»7) finds thirty-seven sognients witii length of body increasing 

 from 12 to 4") imii. In this connection it is interesting to note that the 

 niiiiilKjr of segments in Lcujisca vesiculosa is given both by Grube (1877) and 

 Khlers (lS<t7) as forty-two. Grube's s))ecinien was 21 nun. in length: Klilers 

 given no dimensions. The larger ' C/tal/niger' specimens (named Z. autardica, 

 McInt.X measviring vip to ;i(> mm. in liody-lengtli, have no moic than thirty-nine 

 seumenls. The uuiiiber forty-two may very well be final, in which case it will be 

 inlerenting to learn how it happens that some individuals i.ave less than the normal 

 numlKT «'f .wc.gnient,'5, while others nearly half the size have the full complement. 



