971 



1885.] ii<»-L [Packard. 



Neocarida, branched oS from a common ancestor, but the more important 

 morphological points show that the terrestrial, air-breathing tracheates 

 were a much later branch of the Arthropod tree than the marine 

 branchiate Palaeocarida and genuine Crustacea. Probably the Palseo- 

 carida (Limulus, Merostomaia and Trilobita) were the earliest arthro- 

 pods lo appear ; after them arose the Crustacea, perhaps at nearly the 

 same time the Arachnida, and finally the Myriopoda and the winged 

 insects. Without much doubt, the earliest branchiate forms were our Pro- 

 tocyclus,* the ancestor of the Palaeocarida ; and a Protonauplius form, the 

 forerunner of the Crustacea ; these were marine, perhaps branchiate 

 organisms-, with a few pairs of simple oar like swimming appendages 

 either around or just behind the mouth, and which were free swimming or 

 creeping forms ; the Protocyclus was, perhaps a solid oval creeping animal 

 living at the bottom on mud or sand. The branchiae probably became 

 first developed on the limbs of the free-swimming Protonauplii, as they 

 needed, owing to their great rapidity of movement, the means of rapid 

 aeration of the blood ; while in the heavily-moulded less oxygen-consum- 

 ing Protocyclus, the evolution of gills was somewhat postponed. The 

 steps from Protocyclus to Agaostus was not a very long one. The oldest 

 arthropods, notwithstanding the recent discovery of a Silurian scorpion, 

 were trilobites. 



The following conclusions are drawn from a study of the stage of Limu- 

 lus here figured. 



The fact that the embryo Limulus has at first no abdominal appendages 

 (uropoda), whereas there are temporary abdominal appendages in the 

 tracheates, shows that Limulus in this important respect has little in 

 common with the Arachnida, Myriopoda or Hexapoda. On the other 

 hand In the embryo Crustacea the cephalic limbs are first indicated ; the 

 nauplian limbs, as well as the zoean appendages being cephalic ; the uro- 

 pods not appearing until after the Crustacea leave the egg. 



These facts indicate that Limulus probably descended from a type 

 in which there were cephalic appendages only, and no abdominal append- 

 ages. The absence of a serous membrane, of an amnion, and of pro- 

 cephalic lobes, of temporary embryonic abdominal appendages (at the 

 stage above described) ; also of protozonites (seen in the early embryo of 

 the scorpion and spider) tend to prove that the embryo of Limulus has 

 little in common with that of Tracheata. 



On the other hand the earlier stages in the embryology of Limulus re- 

 semble those of Crustacea in the absence of the procephalic lobes ; in the 

 primitive development of cephalic appendages alone ; the comparatively 

 early appearance of the branchiae of Limulus in the stage succeeding that 

 figured in this essay, shows that the Limulus probably never had any 

 genetic connection with a tracheate arthropod. 



On the other hand, the tracheate features of mesoblastic somites are also 

 ^een in the worms, in Peripatus and in Annelida. 



*See Development of Limulus, 1S72, p. 



