THE THEORY OF THE SKELETON. 161 



simply the interspaces left between the inferior or 

 posterior wall of the prolongation of the carapace and 

 the originally exposed external faces of these regions of 

 the cephalic integument. 



Fourteen somites having thus been distinguished in 

 the cephalothorax, and six being obvious in the abdomen, 

 it is clear that there is a somite for every pair of append- 

 ages. And, if we suppose the carapace divided into 

 segments answering to these sterna, the whole body will 

 be made uj) of twenty somites, each having a pair of 

 appendages. As the carapace, however, is not actually 

 divided into terga in correspondence with the sterna 

 which it covers, all we can safely conclude from the 

 anatomical facts is that it represents the tergal region of 

 the somites, not that it is formed by the coalescence of 

 primarily distinct terga. In the head, and in the greater 

 part of the thorax, the somites are, as it were, run 

 together, but the last thoracic somite is partly free and 

 to a slight extent moveable, while the abdominal somites 

 are all free, and moveably articulated together. At the 

 anterior end of the body, and, apparently, from the an- 

 tennary somite, the tergal region gives rise to the 

 rostrum, which projects between and be^^ond the eyes. 

 At the opposite extremit}^ the telson is a corresponding 

 median outgrowth of the last somite, which has become 

 moveably articulated therewith. The narrowing of the 

 sternal moieties of the anterior thoracic somites, to- 



