STRUCTURE OF THE VERTEBRATES 345 



the Crossopterygii and amphibia; and (6) the marked similar- 

 ity between the positions of the head bones in the two groups. 



In view of the extraordinary similarity between an amphibian 

 tadpole and a fish larva, both structurally and functionally, 

 the embryological transition from fish to amphibian is simpler 

 than that from amphibian to reptile. The most difficult point to 

 understand is the development of feet and toes, and this has 

 been partly bridged by the discovery of crossopterygian fossils 

 {Sauripterus and Eusthenopteron). The most fish-like amphib- 

 ian yet discovered [Eogyrinus) is elongate; covered with bony 

 plates; the head bones have the crossopterygian pattern; and 

 there is a large cleithrum, a typical bone of the fish, connecting 

 the shoulder girdle with the skull. The vertebrae are of the 

 primitive type with two distinct rings to each segment. 



The fossil Stegocephalia (the first amphibia) evolved a num- 

 ber of specializations. The line leading toward the living amphib- 

 ia is soon lost, the legless Apoda most resembling the Stego- 

 cephalia in skull type and in having minute scales beneath the 

 skin. Others of the Stegocephalia became reptilian in body 

 shape, and until larval forms were discovered these were known 

 as Pro-reptilia. 



The fossils supply evidence as to the evolution of the skeleton 

 from amphibia to reptiles, but this gives us no clue as to the 

 evolution of the amniote type of development. The gap is partly 

 filled by the type of egg cleavage, gastrulation, and early de- 

 velopment of the Apoda. The egg is relatively large and yolk- 

 laden and cleavage is partly meroblastic, the dorsal blastoderm 

 being a layer of columnar cells. Gastrulation is almost reptilian, 

 and is an essential link in understanding the process in reptiles 

 and birds. A small area on the edge of the blastoderm folds 

 under and the inturned tissue grows forward as the endodermal 

 layer. It will be recalled that gastrulation in the chick begins 

 as a bit within the blastodermic disc. 



Although this does not account for the origin of the amnion, 

 it is not difficult to understand how the yolk mass became greater 

 and the blastoderm relatively smaller. This being admitted, 

 the overgrowth of amniotic folds is not the enormous transition 

 that might at first be supposed; for, not only is the embryo 

 spread out as a sheet with its archenteron open to the yolk 



