104 Dli. J. STErHENSON OX THE JIOKPHULOGV, CLASSIFICATION, 



The clatisification of the Megascolecida?, to which both sub- 

 families, the Octochfetinte and Trigastiiii£e, beloug, proceeds on 

 phylogenetic lines ; and a very considerable degree of success has 

 been reached in the filiation, and consequently in the definition 

 and arrangement of genera. The origin of the family, as is now 

 generally recognized, is to be sought in the '• original Acantho- 

 driline,"" a form which is represented at the present day by the 

 genus Notiodrilns as defined by Michaelsen in the Tierreicli 

 volume (1). Its essential characters (for our present purpose) aie 

 as follows : A pair of male pores o\\ xviii., two pairs of prostates 

 opening separatel}^ on xvii. and xix., a single ojsophageal gizzaid, 

 four pairs of setse per segment, one pair of meganephridia per 

 segment, no calciferous glands. 



The important characters of the genera which enter into tlie 

 following discussion may be stated thus : — 



Diplocardia, one remove from the original Acanthodriliue, aii'l 

 the ancestor of the Trigastrinte, differs from the original Acan- 

 thodriline only in having two gizzards. It is found in North 

 and Central America. 



Trigaster, descended from Diplocardia, differs from tjie latter 

 genus in being micronephridial ; the posterior male organs are 

 either of the acanthodriliue type or they may have undergone 

 the " microscolecine reduction " (disappearance of the posterior 

 pair of prostates, and union of the male pores with the anterior 

 prostatic openings on xvii.) ; there are two or three gizzards, but 

 no calciferous glands. Trigaster is found in Mexico and the 

 West Indies. 



Dichogaster is similar to Trigaster, except that there are 

 calciferous glands in two or three of the segments xiv.-xvii. A 

 point to which attention has been called is that while in Trigaster 

 the setal interval cd is greater than ah, in Dichogaster these 

 intervals are approximately equal. Eudichogaster was supposed 

 to resemble Trigaster in this respect, but in a few species ah is 

 equal to cd. Dichogaster is endemic in tropical Africa, and 

 probably in Central America and the West Indies, but has spread 

 widely in the tropics and warmer temperate zones. 



The first development, therefore, in the above line of descent 

 (Diplocardia-Trigaster-Dichogaster), and the primary charac- 

 teristic of the Trigastriufe (I inclnile Dijilocardia in the 

 Trigastrinte, v. post.), is the reduplication of the gizzard. 



The genus Octochcetas belongs to a difi'erent line. It diflers 

 from the original Acanthodriliue in being micronephridial, and 

 in having a pair of calciferous glands in segment xv. or xvi. (a 

 few species, to which more particular refei-ence is made subse- 

 quently, are without calciferous glands); it has the single gizzard 

 and other characters of the ancestral form. It occurs in India 

 and New Zealand. 



The other genera of the subfamih^ need not be mentioned here. 

 The first development in the Octochtetinse, and therefore their 



