114 DR. J. STEPHENSON ON THE MORPHOLOGY, CLASSIFICATION, 



with the substitution of a number, sometimes a very large number, 

 of small micronephridia for the single pair of meganephridia in 

 each segment ; and the single central canal of the prostate may 

 branch, with the consequence that the organ is no longer tubular 

 and cylindrical in form, but racemose. In a, small group of 

 genera there is a development of two or more gizzards, instead of 

 the single gizzard of Phtlellus. 



Plutellus, then, has meganephridia, eight setae per segment 

 arranged in four pairs (the lumbricine arrangement), and a pair 

 of tubular prostates opening on segment xviii. in common with 

 the vasa deferentia. From Plutellus is derived Megascolides, in 

 which the nephridia are breaking up or have broken up ; this 

 apparently does not always take place in the same way : in one 

 group of forms there are three or four nephridia on each side of 

 each segment, all about the same size, while in other cases there 

 is one large one and a number of quite small ones ; however, all 

 stages of the process are united in this genus, so long as the 

 prostates and setae retain their original condition. The next 

 stage is Notoscolex : the prostates now become branched ; in a 

 number of cases the branches of the central canal are so insig- 

 nificant tlmt they have no effect on the form of the gland, and 

 can only be demonstrated in sections — the genus, however, is 

 defined as including all forms in which there is any branching at 

 all. Following this we come to Megascolex, where the seta? take 

 on the perichsetine aiTangement ; here again there are a number 

 of intermediate stages ; in a number of species the anterior 

 segments retain the lumbricine arrangement, and the increase in 

 the number of seta? takes place gradually as we move backwards ; 

 in others the a.nterior segments show an increase too, but the 

 paired arrangement still holds — there are six pairs, or eight pairs, 

 instead of four; and so on. The last genus along this line is 

 Pheretima • the essential characters are those of Megascolex , but 

 the gizzard is further back, the testes and male funnels are 

 enclosed in testis sacs instead of being free in the segments (this 

 occurs occasionally in Megascolex), and on the whole the ring of 

 setse is more closed up — has smaller gaps in the dorsal and 

 ventral lines than is usual in Megascolex. 



But there are other lines starting from Plutellus. In the line 

 just considered the first change was the breaking up of the 

 nephridia; in another line the multiplication of the setse comes 

 first. This change, occurring in the basal genus Plutellus, gives 

 Dlporoehfxta, the generic characters of which are therefore 

 tubular prostates, meganephridia, and perichsetine setae. It is, 

 of course, impossible to derive this form from any of the first 

 line, since those all have micronephridia; the meganephridial 

 condition is the primitive one, and a meganephridial cannot be 

 derived from a, micronephridia] form. From Dijjoroclixeta is 

 derived Perionyx, in which the prostates have branched ; this 

 genus therefore possesses meganephridia, perichaetine setae, and 

 racemose prostates. As in the case of Megascolides and Notoscolex, 



