138 THE APODIDAL PARTI 



in support of our main argument ; they not only 

 remove a difficulty, but bear positive testimony to the 

 truth of our theory. Those who, however, may think 

 this view of the disappearance of the Annelidan nc- 

 phridia (with the exception of their peritoneal cover- 

 ings) far-fetched, should remember that the weight of 

 all the positive evidence brought forward as to the 

 relationship of Apus and the Annelida is only really 

 diminished if we cannot show that a difficulty is sur- 

 mountable. It is by no means necessary for our 

 argument cither to remove all difficulties so long as 

 they are not positive contradictions, or to state exactly 

 Jioiv such and such a transformation came about, but 

 only to show that such transformations arc not incon- 

 ceivable. We believe, however, that in this case we 

 have not only shown this, but more, viz. that the pro- 

 cess of the disappearance of the nephridia was what 

 we have described. 



We have now dealt with the principal glands of the 

 Crustacea and of the Annelida. We have deduced 

 the Crustacean glands from the Annelidan setiparous 

 glands, and followed the Annelidan nephridia in their 

 transformation into Crustacean genital glands. 



Of the typical dermal glands of the Annelida we 

 have found no trace in Apus (except in the dorsal 

 organ, see below and Appendix IV). The hypodermis 

 is very thin, and seems to be entirely taken up in 

 secreting the cuticle in its cr r adual transformation 



o *-> 



into an cxoskeleton. 



There are very numerous glandular cells in the 

 hind-gut, which have already been mentioned in 



