EARTHWORMS 421 



collection made by Mr Perkins I find the following variations in structure, from what 

 may perhaps be regarded as the typical organisation of this species. The number of 

 the papillae in the neighbourhood of the male pores varies ; I have found only one or 

 two, three or four; Rosa says two or three. In one example however I found six of 

 these papillae on each side. 



The number of the papillae therefore does not allow of a separation of hawayanus 

 and bermudensis. 



The setae upon the anterior segments are larger than those which follow ; in one 

 example segments 4 — 7 were furnished with these larger setae. This was originally 

 one of the reasons for separating bermudensis from hawaya)iiis. 



The clitellum was described by Rosa to stop short at the middle or thereabouts of 

 the 1 6th segment. I have observed both this arrangement and that generally found 

 in bermudensis, i.e. that the clitellum does not commence accurately at the beginning of 

 the 14th segment while it stops short as in the typical hawayanus. In one example the 

 clitellum was exactly coincident with segments 14 — 16. 



The clitellum has usually setae upon its last segment, i.e. the i6th of the body. 

 There are ten to fifteen of these setae. In two examples I could see no setae anywhere 

 upon the clitellum. This seems to have been the case with the individuals examined 

 by Rosa. A. bermudensis appears to always have setae upon this segment. 



A feature not yet recognised in the worms which I referred to the species bermu- 

 densis was found in two examples oi hawayanus. In one of them there was a pair of 

 papillae anterior in position lying on the 7th segment near to its posterior end and the 

 orifices of the spermathecae. In a ripe individual there was but one of these papillae, 

 that of the right side. This fact will be seen presently to bear upon the identity of the 

 present species with others hitherto supposed to differ specifically from it. As to 

 internal characters the caecum of the intestine has not always the series of short out- 

 growths on the under surface that has been described for this species and for bermuden- 

 sis. The spermiducal gland is generally long, occupying segments i 7 — 22 about. Some- 

 times the duct is given off at the top when the gland commences in segment 17. In 

 one specimen the gland was much abbreviated and lay only in 18, 19 on one side and 

 17, 18 on the other. This looks like a commencing loss of the gland which is known 

 to occur in some other species, e.g. Aniyntas heterochaeta. The spermathecae seem 

 always to lie in segments 6, 7, 8. In one example they were particularly large ; but, as 

 this individual had no other features which seemed to remove it from the species, the 

 difference in size (the pouches were as large as the gizzard) does not seem to be more 

 than a variation to be neglected for systematic purposes. 



The size of Amyntas hawayanus varies to some extent. The greatest and least 

 lengths which I observed were 150 and 69 mm. The number of segments varied 

 between 97 and "jt,. 



Next as to the identity of Amynias hawayanus with A. barbadensis — the original 



