SECT, xin THE TRILOBITES 235 



The Alimentary Canal has already been referred to. 

 It has the very pronounced bend on which we lay so 

 much importance (see Fig. 46). Although \ve think 

 our proof is not much weakened by our not finding 

 any traces of the sternal plate, still it would be 

 interesting if it were to be found, as it must without 

 doubt have been there, i.e. if there is any truth of our 

 deduction of these animals from bent Annelids. The 

 habit of rolling up would lead to a strong development 

 of the ventral muscle bands, and consequently of this 

 sinewy mass for their attachment (cf. p. 261). 



We think, then, that we have here made it highly 

 probable that if our deduction of Apus from a bent 

 carnivorous Annelid holds, the Trilobitcs must have 

 had the same origin. This fact, that the most primi- 

 tive Crustacean known to the palaeontologist should 

 show so many points in its organisation directly 

 clcducible from the Annelids, i.e. deducible after the 

 Apodidae have supplied us with the key to their cor- 

 rect interpretation, is one of those confirmations of a 

 theory which we think amounts almost to a demon- 

 stration. 



The Trilobitcs, then, arc nothing but specialised 

 carnivorous Annelids, browsing under cover of the 

 dorsal integument, which, starting from the head- 

 shield, gradually spread out like a flattened jointed 

 roof, covering all the segments. Every imaginable 

 variation in the sculpture of the surface of this roof, 

 and in the thorns for its protection, arc to be met 

 with in the Trilobites. 



