290 DISCOVERY REPORTS 



in 14 from septum 14/15, in 15 from 15/16, in 16 from 15/16, in 17 from 16/17 and in 18 

 from 17/18; the last empty sacs without contents; an ovary and oviduct in segment 18. 

 Left side: three spermathecae at intersegments 15/16, 17/18 and 18/19; testes and 

 spermiducal funnels, free, in 16, 18 and 19; seminal vesicles apparently absent; ovary 

 and oviduct apparently absent. 



It is impossible to refer this specimen to any familiar species. If the right side alone 

 is considered and the anterior testes and funnels assumed arbitrarily to be in their 

 normal position in segment 10 the gizzard would lie in segment 24 and the clitellum on 

 about segments 34-42. If the gizzard is assumed to be in a normal position in segment 

 17 the clitellum would be situated on about segments 27-35. Neither of these positions, 

 when taken in conjunction with other characters, fit any known species, and in any case 

 the interval between the anterior testes and the gizzard is exactly twice the normal 

 interval (fourteen segments instead of seven). On account of the dorsal position of the 

 spermathecae and the restriction of the gizzard to one segment, this specimen might at 

 first sight be referred to the genus Eiseniella, especially as E. tetraedra occurs in the same 

 collection. Unfortunately the position of the male pores lateral to setal line h precludes 

 this identification. In specimens of Dendrohaena siibriibicunda from the Falkland Islands, 

 described in the present communication (p. 288), the gizzard was found to be confined 

 frequently to one segment ; unfortunately the dorsal position of the spermathecae make 

 it improbable that the abnormal specimen under consideration is referable to this 

 species ; moreover D. submbicunda has not as yet been recorded from Tristan da Cunha. 

 The dorsal position of the spermathecae suggests that it belongs to the genus Eisenia 

 (since it is not either Eiseniella or, on account of the free testes, Octolasium). Michaelsen^ 

 observes that in some species of Eisenia the gizzard occupies only a part of segment 18 in 

 addition to segment 17. In E. veneta, the only commonly peregrine species of this genus 

 in which the setae are not closely paired, the clitellar bands are in the form of two pairs 

 of tubercles and not a continuous wall as in the present specimen. In any case, on 

 account of the intense pigmentation, this specimen cannot be referred to E. rosea, the 

 only Eisenia recorded from the island, and it is equally certain that it cannot be 

 identified with any of the other lumbricid species so far known to occur on Tristan da 

 Cunha {Lumbriais rubellus, Allolobophora caliginosa and Bimastiis constrict us). 



The cause of the extreme abnormality of this specimen is quite obscure. Hyper- 

 regeneration of segments has not been recorded in the Lumbricidae which, on the 

 contrary, usually regenerate fewer than the normal number of segments when the 

 anterior end is removed. This phenomenon is, however, known in the allied form 

 Criodrilus. There are no obvious signs of injury in the present specimen and, in the 

 absence of further evidence, the abnormality would seem to be more probably of a 

 developmental nature. It may be compared with the abnormal specimen of Micro- 

 scolex georgianiis described previously (p. 277), in which the prostatic pores are situated 

 very far back and abnormally reduplicated. 



1 Michaelsen, W., Ann. Mus. Zoo]. Acad. Imp. Sci. St PStersbourg, xv, 1910, p. 9. 



