no 



NATURAL SCIENCE. 



February, 



Omitting all considerations of minor groups like the Tardigrada, 

 Linguatulida and Pycnogonida, I think that recent students have 

 shown that there are three large groups among the forms commonly 

 considered as Arthropoda. The first of these, the Branchiata, 

 includes the Crustacea and Arachnida. The arguments for this 

 association have been so often presented that I need not rehearse them 

 here. For the second group the name Insecta or Antennata may be 

 employed. It includes the Hexapoda and the Chilopoda. Both 

 Pocock and myself have repeatedly pointed out that there is no valid 

 group of Myriopoda. The Chilopoda are decidedly like the Hexapoda 

 in their broader anatomical features, while the Diplopoda' differ so 

 totally from these that they may form our third group. Concerning 

 these Diplopoda, however, so little is known that they may be left out 

 of consideration in what follows. 



For convenience of contrast of the characters of the Branchiata 

 and Antennata I employ parallel columns. 



Branchiata. 



Respiration by gills (or lungs or 

 tracheae derived from gills) arising in 

 connection with appendages. 



Functional nephridia in somites, 

 two, five, or both. 



Genital ducts (modified nephridia) 

 opening near the middle of the body. 



Malpighian tubes (when present) 

 entodermal. 



Stomodaeum long. 



" Head " indefinite, ranging great y 

 in limits even in the same group. 



Antennata. 



Respiration by tracheae (modified 

 glands ?) never connected in origin with 

 the appendages. 



No functional nephridia. 



Genital ducts (modified nephridia) 

 opening near the end of the body. 



Malpighian tubes ectodermal. 



Stomodaeum short. 



Head distinct, consisting of pre- 

 cephalic lobes and five segments (as 

 indicated by the neuromeres). 



These are, it seems to me, points of fundamental importance, and 

 I can hardly see how one can reconcile them without going back to 

 the annelid ancestor. Certain it is that the two groups must have 

 become differentiated at a time when nephridia were present in all 

 segments, when these served as the only excretory organs, when there 

 was no head comprised of coalesced somites, when there were no 

 specialised respiratory organs apart from the general body-surface or 

 possibly portions of the appendages, and when no appendages were 

 specialised as jaws to the exclusion of others. No known form 

 outside the annelids meets these requirements ; and these facts, it 

 seems to me, go far towards showing that the group Arthropoda does 

 not form a true phylum, and that the similarities which are often 

 advanced to support such a view are the results of convergence, many 

 of them explicable as homoplastic. 



J. S, KiNGSLEY. 



Tufts College, Mass., U.S.A. 



1 F. C. Kenyon has recently shown that Pauropus is an aberrimt dipiopod : " The 

 Morphology and Classification of the Pauropoda," Tufts College Studies, No. 4, 1895. 



