20 



INTRODUCTION 



branchs and teleosts, ganoids and dipnoi. None are known in am- 

 phibia, birds or mammals, but in crocodiles and turtles so-called peri- 

 toneal canals occur, usually ending blindly in turtles but emptying 

 into the cloaca in the crocodilians. These may be homologous with 

 the abdominal pores, but this can be decided only by a study of the 

 development. In some fishes the abdominal pores serve for the 

 escape of the eggs and spermatozoa; in other animals their function 

 is uncertain. 



Mesenchyme. — The principal places where the mesenchyme 

 arises have been alluded to incidentally in the foregoing pages. 

 These statements may be brought together here. A part is derived 

 from the splanchnic walls of the mesomere and from the ventral part 

 of the myotomes, each of which is a centre of rapid cell proliferation, 



i^^^yyyyy^^^y'yyyy'^y,^^. 



Fig. 12. — Diagram of possible connexions of ccelom with the exterior, modified 

 from Bles. c, coelom; d, cloaca; g, glomerulus of kidney; i, intestine; n, nephrostome; 

 pa, porus abdominalis. 



the resulting cells passing toward the median plane of the body. 

 From the method of formation from segmented structures, these 

 bodies of mesenchyme are at first metameric, and since in part, they 

 give rise to the axial skeleton, the separate portions are called 

 sclerotomes. This mesenchyme extends dorsally beside the cen- 

 tral nervous system and notochord, and ventrally on either side of 

 the axial blood-vessels and the alimentary canal, thus forming a 

 partition between the two sides of the body (fig. 2,i)' 



A second source of mesenchyme is found in the breaking up of the 

 somatic walls of the myotomes, all of the cells of which are trans- 

 formed into this layer and come to lie immediately beneath the ecto- 

 derm, forming a complete layer of mesenchyme around the whole 

 body. There is also a probability that some mesenchymatous cells 

 are budded from the entoderm of the archenteron, but it is uncertain 

 whether the ectoderm contributes to the layer, farther than by the de- 

 generation and metamorphosis of some transient nervous structures. 



