MR FRANK E. BEDDARD ON THE ANATOMY OF OCNERODRILUS. 579 



I am inclined, however, to agree with Mr Benham, and to retain this group on 

 account of the occurrence of the asexual mode of generation.* 



The second sub-class, Lumbricomorpha, contains all the remaining Oligochseta, which 

 are divided into two series, Microdrili and Megadrili. 



The only character which absolutely distinguishes these two is the presence in 

 Megadrili of a capillary network upon the nephridium, and its absence in Microdrili, and 

 it is admitted that this difference may be due to size. Ocnerodrilus renders this division 

 no longer tenable. 



Mr Benham has, however, not mentioned three points, which he might have used to 

 distinguish the two groups. These are — -(1) large size of ova, (2) clitellum consisting of 

 only a single layer of cells, (3) sexual maturity at a fixed season. Until the publication 

 of the facts contained in the present paper, these points would, so far as I am aware, 

 apply to all Microdrili, and to none of the Megadrili. However, in Ocnerodrilus the ova 

 appear to approximate in size to those of the Microdrili. The question to be now con- 

 sidered is, How far are the two last points sufficient to characterise the Microdrili ? If 

 they are sufficient, it will be tantamount to restoring the old grouping into Limicolse and 

 Terricolse, for the Naidomorpha in these points agree with the Microdrili. The classi- 

 fication would be as follows : — 



I. Clitellum one cell thick. 



Sexual maturity at a fixed period. 



( A, Asexual reproduction occurs, Naidomorpha. 



\ B, No asexual reproduction, sexually mature at fixed periods, Microdrili. 



II. Clitellum, composed of two distinct layers. 

 Sexual maturity more or less continuous. 



Megadrili. 



This mode of division does not appear to me so satisfactory as the one proposed by Mr 

 Benham, and I should be inclined, therefore, not to divide his Lumbricomorpha, except 

 of course into families and genera. 



I do not propose to discuss the limitations of these families, as I have already done so 

 elsewhere [3 J in so far as concerns Earthworms. 



The question is whether Ocnerodrilus is referable to any known family, or whether 

 it should form a distinct family. 



This genus was originally placed by Eisen [1] in the family Lumbriculidse, though 

 the reasons which led to this view are not plainly stated. Indeed, the whole paper is 

 occupied with a description of the differences between Ocnerodrilus and other genera 

 of Lumbriculidse ; nowhere is there any indication of what are regarded as the points of 

 affinity between Ocnerodrilus and other Lumbriculidse. 



* This scheme is practically identical with that proposed by D'Udekem in 1853 [13], and further elaborated in 

 1863 [14]. The division is into "Agemmes" and " Gemmipares" these names implying the principal distinction between 

 the two groups. Another distinction referred to is the persistence of the genital organs in the "Agemmes," and their 

 appearance only at certain epochs in the " Gemmipares." 



VOL. XXXVI. PART II. (NO. 21). 4 S 



