130 THE APODID^ part i 



pancreas of the higher Crustacea. But this is 

 certainly not the case. Neither in position nor in 

 structure do the glands remind one of Annulatan 

 nephridia, but, on the other hand, they correspond in 

 a most remarkable manner with the acicular glands 

 of the Annelidan parapodia. 



(5) We further assume that the habit of browsing 

 of the bent Annelid was originally acquired by the 

 ad^ilt animals, in which the nephridia in the anterior 

 segments have generally disappeared in the course 

 of development ; so that Apus, which represents 

 such a browsing Annelid, would probably have no 

 nephridia in the anterior or head segments. In the 

 trunk segments, on the other hand, we shall find 

 abundant traces of nephridia. 



These considerations, which, taken singly, do not 

 possess much weight, taken all together make the 

 ncphridial origin of these glands very improbable 

 compared with that which we here attribute to them. 



Setiparous glands on the trunk segments are hardly 

 to be expected ; the dorsal parapodia are developed 

 into complicated limbs covered with setae, and the 

 ventral parapodia are also thickly beset with setse of 

 different kinds. We have succeeded, however, in 

 finding two such glands on the same limb in one 

 specimen of Apus cancriformis (see Fig. 31). We 

 could find no similar glands on the gnathobase or 

 ventral parapodium of the corresponding limb of the 

 same segment, and only on one other limb. Perhaps 

 further search would reveal more, but it is certain 

 that these glands occur quite irregularly. We are 



