112 NATURAL SCIENCE. February. 



sufficiently near the arthropod type to shew us the possibihty of 

 convergent evolution acting so as to produce jointed legs along 

 different lines. The remaining classes of the Arthropoda are, further, 

 by no means easily arranged. The Crustacea must be regarded as 

 standing on their own base down to the lowest forms, and possibly, if 

 the nauplius has any phylogenetic significance, down to much lower 

 forms than exist at present. The whole series of Crustacea, however, 

 give us no point from which the other Arthropoda can be derived. 

 The characteristic specialization of the crustacean head resulting in 

 two pairs of preoral appendages and three pairs of masticatory ones, 

 would have to be undone to arrive at any of the other Arthropoda. 

 It is of course possible here as elsewhere to imagine a hypothetical 

 ancestral form with jointed legs from which the various classes can 

 be derived, but is it necessary ? Is the jointed leg a structure of such 

 profound significance that we should let it trample on all other 

 considerations ? The gulf between a parapodium and such a leg is 

 not so wide and deep as it appears at first sight, and must have been 

 crossed more than once whichever position we assign to Peripatiis. 

 The change is one of obvious advantage in many ways, and given the 

 formation of an exoskeleton, is almost inevitable. Further, the legs 

 of the different classes have no points in common, except the one of 

 being jointed. The legs of the Crustacea are traceable throughout 

 to modifications of the one well-marked biramous type persisting 

 through all kinds of functional changes, but no trace of this type, 

 except a few extreme variations which might be used to prove 

 anything, is seen in the other classes. Similarly, the legs of Arachnida 

 and Insecta have very few points of resemblance. The number of 

 joints, the relation of the anterior appendages to the mouth, the 

 presence of antennae (if these are to be considered as homologous with 

 the other appendages) in the Insecta and Myriopoda, all afford 

 marked points of difference. 



One important point of morphological resemblance remains 

 however, i.e., the coelom. The universal disappearance of this 

 structure in Arthropoda wants explaining, and, if one is prepared to 

 drop Peripatns, is an argument of no little weight for the phylogenetic 

 unity of the group. If we retain Peripatns, its importance vanishes, 

 since, as I have said, it is impossible to consider Pevipatits as ancestral 

 to the whole group, and it follows that the coelom was suppressed 

 independently in the different classes. It seems possible that this 

 reduction of the coelom may be directly associated with the acquisition 

 of jointed legs, and of a body formed of a series of inflexible rings, 

 the consequent enormous muscular development tending to encroach 

 upon it. The excretory function usually associated with the coelom 

 being partly taken over by the formation of the cuticle, would permit 

 of this encroachment on the coelom, and would account for the 

 reduced number of nephridia. The whole question of the coelom in 

 the Arthropoda is, however, wrapped in profound obscurity. 



