AND ZOOaEOaRAPHY OP INDIAN OLIGOCH^TA. 117 



I now wish to argue that a number of genera of Megascole- 

 cidye are probably polyphyletic. I take as a text a sentence of 

 Benham's, in his paper on the Oligochfetes of the Subantarctic 

 Islands of New Zealand (4) : — " According to Michaelsen, species 

 of Microsmlex may arise at different times, in different parts of the 

 world, from different species of Notiodrilus. This thesis involves 

 so profound a modification intlie accepted ideas of evolution that 

 space will not permit me to discuss the pi'oblem here." I think 

 we may ta.ke it, then, that the genei'al view with regard to the 

 multiple origin of species, genera, and larger groups is one of 

 scepticism ; the orthodox view is that each group has arisen 

 once and once only. 



But we have to remember that the essential variations in the 

 genera of this family are not innumerable, but limited. It is 

 not the case that the modes of variation are so numerous, and 

 the possible combinations thei'efore so greatly more numerous 

 still, that there is no reasonable chance of the same combination 

 of characters ever being repeated — this seems to be the foundation 

 of the orthodox view. The combinations of charactei's that 

 distinguish the various genera can be obtained in more than one 

 way, and the characters and their combinations are few enough 

 to render it possible that this has happened ; I would even say 

 probable that this has happened, and perhaps often. 



Let us i-emember that the perichtetine arrangement of sette is 

 secondary to the lumbricine, the micronephridial condition 

 secondary to the meganephridial, and the racemose prostate to 

 the tubular, and tha.t these changes have demonstrably taken 

 place more than once ; and let us take such a form as Megascolex, 

 with perichfefcine seta?, micronephridia, and racemose prostates. 

 It may ha.ve arisen from a, foim with lumbricine set?e, micro^ 

 nephridia, and racemose prostates (i.e. Noioscolex) by the multipli- 

 cation of the setae ; or it may have arisen from a form witli 

 periclipetine setse, racemose prostates, and meganephiidia 

 (i. e. Perionyx) by breaking up of the nephridia. Both these 

 modes of origin have, in fact, been suggested; the point seems to 

 be decided — for some species of Megascolex, at any rate — by 

 finding a number of intermediate forms between Noioscolex and 

 Megascolex. There is a third possibility, from a form with 

 perichsetine setse, micronephridia, and tubular prostates, by 

 the change of the tubular into the racemose prostate — this 

 would mean that Spenceriella was the ancestor. 



Or take Perionyx, with perichsBtine setae, racemose prostates, 

 and meganephridia. It might be derived from a form with 

 lumbricine setae, meganephridia, and racemose prostates (i. e. 

 Woodwardia) by multiplication of setae; or from one with peri- 

 chaetine setae, tubular prostates, and meganephridia (i. e. 

 Diporochceta) by the branching of the prostatic lumen. Here the 

 existence of intermediate forms has decided in favour of the 

 latter. 



Sjienceriella is a very small genvis, with one species in India 



