46 AFFINITY OF THE TRILOBITES 



long as in Apus. I presume, therefore, that in this respect the Trilohites resembled 

 Branchipus rather than Apus ; or that those at least which possessed the power of doubling 

 themselves up had the anterior pair of extremities perfectly similar to the rest. And this, 

 indeed, appears also to have been the case, from the equality of the thoracic rings observable 

 in Branch'qms and in many Trilobites capable of doubling themselves up, this equality 

 appearing to correspond with a similar equality in the dimensions of the extremities attached 

 to them. Since also in Branchiptus there are no feet attached to the abdominal rings, 

 in this respect also we might expect a correspondence with the group of Trilobites now 

 under consideration, and this seems the more probable in the case of those genera amongst 

 them which have a short axis, and no lateral ribs on the caudal shield {Asaphm, Illamtn, 

 NiJeus, Aiiqyyx). For similar reasons, however, I assume that this structure obtained also in 

 the other genera {Calymene, Homahnotus, Phacops) capable of doubling themselves up, and 

 also in others {Oc/ngidce, Odonfoplcuruhe) not having this power, but characterized by equal 

 thoracic rings. In these, however, the existence of abdominal feet may perhaps be inferred 

 from the lateral furrows of the caudal shield. 



On the other hand, it appears pi-obable, from the decrease of size observable in com- 

 paring the anterior with the posterior portion of the body, that in the other groups (the 

 OlenidcB and Campylopleiira), in which the animal was not able to roll itself into a ball, the extre- 

 mities were not all of equal size, but diminished towards the posterior part of the body, with the 

 diminution in the size of the rings ; while the thoracic rings passed gradually into abdominal 

 rings. This is the case in Apus and Limuadia, where the rings increase a little at first, but 

 then diminish in size from the centre of the thorax, and, becoming progressively smaller, 

 pass into the rings of the abdomen. From the analogy here presented, we might also expect 

 that the first pair of feet from the thoracic feet of the Oleuidce and Campylopleura were pro- 

 vided with filaments instead of fins, or that at least the antennse of the head were larger and 

 more like those of Limnadia, while in other Trilobites they must certainl}^ have been short 

 and small ; and they could not have projected beyond the margin of the cephalic shield, for 

 the same reason which prevented the development of the lobes of the first pair of feet into 

 filaments. 



SECTION XXII. 



Convinced that the reasons already offered will be deemed sufhciently conclusive to 

 satisfy the unprejudiced reader, I venture now to offer the following deductions and general 

 conclusions : 



The Trilobites zoere a pecidiar fcunili/ of Crustacea, nearly allied to the existing 

 Phyllopoda, approaching this latter family most nearly in its genus Branchipus, and 

 forming a link connecting the Phyllopoda with the P^ecilopoda. 



In order, however, to estimate fairly the affinity of the Trilobites with the Phyllopoda, 

 we must not lose sight of the important fact, that the Trilobites differ not only from the 

 Phyllojjoda, but from all other existing families of Crustacea in the varying numerical pro- 

 portion of their thoracic rings ; a peculiarity neither exhibited at present as a characteristic 

 of any natural family among the Crustacea, nor in any of the heterogeneous Articulata. 



