2i8 THE APODIDvE part ii 



The variation in the number of trunk segments is 

 also a point of no small interest. While some authors 

 have tried to classify the Trilobites according to the 

 number of the trunk segments, Barrande has shown 

 that even within the same genus the number is quite 

 inconstant, the different species varying greatly in this 

 respect, in Olenus 9-15, Cyphaspis 10-17, &c. This 

 is exactly what we find in the Apodidae, where the 

 number of segments varies greatly : from 60-65 ^^ 

 A. cancriformis, to 40 in L. glacialis. We have already 

 discussed the importance of this inconstant number of 

 the segments in our argument that the Apodidae stand 

 half way between the Crustacea, with their small con- 

 stant number of segments, and the Annelida with 

 their large inconstant number. But the argument has 

 not the same weight here as it had in our endeavour 

 to show that the Apodidae were very primitive 

 Crustacea, because in the case of the Trilobites the 

 fact is already apparent from their geological 

 position ; still it is an important characteristic which 

 they have in common with the Apodidae, and as 

 such is so much positive evidence in favour of our 

 argument that both are derived from the bent 

 Annelid. 



(III.) T/ie Pygidmvi is a more or less constant 

 characteristic of the Trilobites. It is the posterior 

 region of the body, composed of a varying number of 

 segments fused together, so that the whole region 

 forms a stiff plate, a sort of tail-shield answering to 

 the anterior head-shield. The morphology of this 

 pygidium has been as little understood as that of the 



