Wol. 51.] POSITION OP THE TRILOBITES. 355 



plication ultimately depend than a number of pairs of jaws grouped 

 around, that is, necessarily in less close association with, the mouth. 

 Another advantage, however, might be found in that the atrophy of 

 the jaw-pieces of the second pair of appendages would release their 

 •dorsal branches to become specialized as sensory feelers. Although 

 this latter specialization can hardly be said to have taken place in 

 Apus, we know, from the great development of the second antennae 

 in the higher Crustacea, that this pair of appendages, rejected as 

 jaws, have become most efficient sensory organs. 



Triarthrus represents the primitive type of mouth-formula, which 

 is handed on somewhat specialized to the other merostomata. Apus, 

 however, branched off and by developing the jaw-pieces of the third 

 pair of appendages as specialized mandibles started the type of 

 mouth-parts belonging to the Crustacea proper. That the mouth- 

 formula of Apus is the origin of the mouth-formulae of the higher 

 -Crustacea admits of little doubt when the other primitive features 

 in its organization are taken into account. 



The Appendages of the Trunk and Pygidium. — Burmeister, 1 in his 

 •restoration of the trilobite, figured the under-surface as provided 

 -with two complete series of many-lobed phyllopodan appendages, 

 which, however, did not extend to the rudimentary pygidial seg- 

 ments. Actual discovery of portions of filamentous limbs appa- 

 rently finally disposed of Burmeister's restoration. It was obvious, 

 however, that if the trilobites were related to Apus, the limbs of 

 the fused rudimentary pygidial segments might have been phyllo- 

 podan, as they are in Limulus. On the other hand, Walcott's 

 restoration, partly based upon Hall's discovery of the underside of 

 Asaphus, showed them to be filamentous. On the whole, then, 

 while firmly believing that the trilobites must have at one time 

 had phyllopodan appendages, I was inclined to believe that the 

 limbs in the trilobites, being filamentous in the developed segments, 

 might also appear as such from the first, it not being necessary to 

 assume that they passed through any phyllopodan stage in their 

 ontogenetic development. After all, Burmeister's restoration has 

 been so far justified that one trilobite 2 is now known to have had 

 phyllopodan limbs, not in the thorax, where he placed them, but in 

 the pygidial segments, where he left them out. 



This discovery fully justifies the explanation which I have offered 

 of the morphology of the posterior region of Apus, and the homo- 

 logizing of that region with the pygidial region in the trilobites. 

 In both cases we have a fixation of larval conditions. We now 

 know, then, that the common racial form of both Apus and the 

 trilobites possessed phyllopodan appendages, which in the former 



1 ' Die Organisation der Trilobiten aus ibren lebenden Verwandten ent- 

 wickelt,' Berlin, 1843 ; Engl, transl. edited by T. Bell & Edw. Forbes, Bay Soc. 

 1846. _ J 



2 Since this was written Dr. Beecher has discovered that the pygidial limbs 

 .of Trinuclma show essentially the same characters ; see Am. Journ. Sci. ser. 3, 

 ■yol. xlix. (1895) p. 307, 



