266 NATURAL SCIENCE. April, 



blood-sinus formed by the fusion of the veins whose walls have dis- 

 appeared, leaving their openings into the heart itself as naked ' ostia.' 

 (The theoretical explanation of the derivation of the arthropod heart 

 from a median dorsal vessel fed by paired lateral veins was given in 

 my article on the *' Ccelom and Vascular System of Mollusca and 

 Arthropoda," Quart. Journ. Micr. Sci., vol. xxxiv., 1893). 



The parapodial jaws and the ostiate heart cannot be supposed to 

 have been both developed independently in each group of arthropods, 

 or in any two of them. That is a supposition which seems to me, 

 in our present state of knowledge as to " homoplastic structures," 

 inadmissible. Hence I conclude that they were inherited from a 

 common ancestral stock, in which they arose (since we find them 

 nowhere outside the Arthropoda), and the derivation of all the 

 Arthropoda from this one stock which had already acquired two great 

 distinctive characters of all known Arthropoda, renders it impossible 

 — without some new and unrecognised use of language — to speak of 

 the Arthropoda as constituting more than one natural group. 



On all hands it is admitted that this group comprises (as most 

 large groups do) several diverging lines of descent. 



The conception which I have formed, from the facts now ascer- 

 tained, as to the relationships of those groups is indicated by the 

 subjoined diagram. I do not introduce the pycnogonids or other 

 degenerate groups into consideration. 



I regard the trilobites as a lower ancestral grade of the Arachnida, 

 possessing the parapodial antennae which the higher grade have lost, 

 and I propose to distinguish these grades as Arachnida eucevata and 

 Arachnida lipocerata. The trilobites exhibit the same wide range in 

 the number of body-segments which is observed in the lower 

 Crustacea, whereas the most fully-developed Arachnida (Eurypterines, 

 Limulus, and Scorpio) present eighteen segments arranged in three sets 

 of six. The immediate ancestors of the trilobites and those of the 

 Crustacea were probably more closely allied than were either to the 

 ancestry of hexapods, chilopods, and diplopods, but it must be 

 remembered that the common ancestor of trilobites and Apiis could 

 not be reckoned a crustacean, since it probably had only one and not 

 two pairs of praeoral tactile appendages, 



In any scheme of Arthropod classification we have to admit the 

 occurrence of remarkable instances of homoplastic agreement — that 

 is to say, we cannot conceive of the derivation of the classes in such 

 a way as to assign to certain important and remarkably elaborated 

 organs a single point of origin. The facetted eyes of Insecta and 

 Crustacea are examples — also the tracheae of Peripatus, Diplopoda, 

 Chilopoda, Hexapoda, and certain Arachnida — and again the enteric 

 coecal tubes of Amphipoda and of scorpions, which so much 

 resemble but differ in origin from the proctodoeal malpighian tubes 

 of Hexapoda. 



There are, no doubt, besides the two great characters which I 



