MR FRANK E. BEDDARD ON THE ANATOMY OF OCNERODRILUS. 565 



latter part of October. This statement implies that, as in the Lumbriculidse and other 

 families of aquatic Oligochseta, Ocnerodrilus passes the greater part of the year in a 

 state of reproductive inactivity. My own specimens must have been collected in June 

 or July, and I think it very probable that, as in Earthworms generally, some individuals 

 are mature during every month of the year. In any case the clitellum, when developed, is 

 very extensive, and easily recognisable in the worm without a microscopical examination. 

 It commences at the XHIth segment, and extends back as far as the XVIIIth, thus 

 occupying altogether six segments. It is absolutely limited to these segments, and 

 does not, as is so often the case with this organ, trespass upon a portion of the adjoining 

 segments at either end. 



The extent and position of the clitellum alone, without going into any details of 

 structure, is sufficient to show how much Ocnerodrilus differs from any of the aquatic 

 groups which are known.* In most of these the clitellum consists of a small number of 

 segments — not more than four, and generally less — which immediately surround the 

 genital pores. Ocnerodrilus is, however, like all the " Limicolce" and the majority of 

 Earthworms, intraclitellian in Permer's sense, i.e., the oviducts and sperm ducts 

 open on to this modified region of the integument. The intraclitellian condition is 

 regarded by Benham [9] as the more primitive ; but it is difficult to see any reason for it 

 shifting back to a point far behind the generative apertures, as is the case with the Lum- 

 bricidse [sensor stricto). Supposing this to be explained, it is then equally difficult to 

 account for the fact that in Perichceta the male apertures usually lie behind the clitellum. 



In Ocnerodrilus and other forms it looks very much as if the clitellum had extended 

 to keep pace with the changed position of the male pores. 



I formerly advocated the same view as Mr Benham, for the reason that the glandular 

 lining of the atria has a certain relation to the structure of the clitellum. In Earthworms, 

 where the clitellum consists of two distinct layers of cells, the atrium is lined with two 

 layers of cells, on the whole similar to those of the clitellum. On the other hand, the 

 aquatic genera, with a clitellum consisting of only a single layer of cells, have atria which 

 are lined by a single layer of cells only. I shall, however, point out later that Ocnero- 

 drilus forms an exception to this rule, at present the only one known. Unless this can 

 be explained on other grounds, it is necessary to abandon this view of the relations of 

 the atria to the clitellum. I am still, however, disposed to believe that the intraclitellian 

 position of the generative opening is the primitive one. It seems to be the most 

 natural position for the development of the clitellum. 



In its minute structure the clitellum of Ocnerodrilus differs from all the aquatic 

 families, and agrees with Earthworms. In all of the former the clitellum consists of a 

 single layer of cells only, while in Earthworms there are two distinct layers, and Ocnero- 

 drilus (see fig. 14) is in this respect an Earthworm, and differs from the Lumbriculidse. 



* With regard to the Lumbriculidse, the number of segments occupied by the clitellum has not yet been accurately 

 described in many types. In Rhynchelmis, Vejdovsky [15] implies that the clitellum occupies segments VIII- 

 XVI. 



