1915.] : J. STEPHENSON : Indian Oligochaeta. 105 
But let us state the case a little more fully. The characters of Notiodrilus which 
areimportant from a phylogenetic and therefore from a systematic point of view are:— 
acanthodriline disposition of the posterior portion of the male organs, holoandric 
Notiodrilus. 
Ges eee | re ba Ed 
H sav 
dE | : | b 
Erythraeodrilus. Octochaetus. 
| 
AR ee 
| 
Hoplochaetella. 
condition of the anterior male organs, lumbricine arrangement of the setae, megane- 
phridia only, and spermathecae opening in 7/8 and 8/9. Howascolex is exactly the 
same in all these features, except that micronephridia coexist with the meganephridia. 
Octochaetus also presents the same features, except that micronephridia exist alone. 
Now we have reason to suppose that a pure micronephridial condition did not arise 
throughout the body at a single bound, but that it was preceded by a mixed mega- 
and micronephridial condition ; —one in which, apparently, the whole of the megane- 
phridium in each segment had not dissolved into micronephridia, but a portion remained 
still as a recognizable meganephridium. Such intermediate forms exist elsewhere at 
the present day (Megascolides, Lampito). Therefore, even if we had not Howascolex 
before us, we should have to postulate a form which by its definition would coincide 
with Howascolex as the immediate ancestor of Octochaetus, and the intermediary 
between Octochaetus and Notiodrilus.! 
I consider therefore that the soiution of the problem of classification is to be 
found in reckoning the Octochaetinae as beginning with Howascolex, and placing 
Erythraeodrilus and Octochaetus as independently evolved descendauts of Howascolex ; 
the line cd in the foregoing chart is therefore to be considered as that which delimits 
the Octochaetinae from the Acanthodrilinae. 
Family GLOSSOSCOLECIDAE. 
Genus PONTOSCOLEX. 
West Haputale Estate, Ceylon, 6200 ft.; dug from earth; Dec, 1913 (S W. Kemp). Four speci- 
mens of a species of Pontoscolex, probably P. corethrurus, but the specimens were not fully 
mature. 
! It is no doubt possible to agree with the above (which is indeed almost self-evident), and still, on 
the principle that the facts of geographical distribution are an essential part of the definition of a group, 
to hold that two (morphologically) indistinguishable genera are nevertheless to be kept separate. The 
point is that it seems premature at present to discuss what must, after all, be an extremely rare 
exception to the general rule, in connection with Howascolex and the putative ancestor of Octochaetus. 
