238 THE APODIDyE ' part ii 



In the Eurypteridse, however, even this primitive 

 head-shield seems wholly or almost wholly to have 

 disappeared, and the flattened Annelidan segments 

 relied almost entirely upon the stronger development 

 of the exoskeleton for protection. Like the Xiphos- 

 uridse, they developed comparatively few segments, 

 ending in a caudal spine or plate. In this limited 

 number of segments they show considerable special- 

 isation. The whole structure of the animal is clearly 

 adapted for a free-swimming life, the first trunk limbs 

 forming powerful oars. 



The limbs develop as Crustacean limbs only on the 

 head and first trunk segment ; on the other trunk seg- 

 ments they remain leaf-shaped, i.e. more like the 

 original Annelidan parapodia. The gill portions of 

 these limbs may have had their surfaces increased by 

 means of numerous integumental folds like the leaves 

 of a book, as in the Xiphosuridse. 



We feel some confidence in the following homology 

 of the head limbs, because we have learned, from all 

 the groups hitherto discussed, that the large rowing 

 limb is probably the first trunk limb ; we need not 

 here repeat the reasons already given for this conclu- 

 sion. All that lies in front of these large rowing 

 limbs therefore represents the head. 



Before, however, attempting to examine the parts 

 in detail, we are at once struck by the difference 

 between the heads of these animals and those of the 

 Apodida^, Limulus and the Trilobites. The mouth 

 parts are in fact so specialised that it is not easy to 

 compare them with those of the above-named groups. 



