Epilogue: W hat Comes of Studying Vertebrates 787 



The respiratory organs of fishes are gills, with lungs in Dipnoi 

 playing an accessory role. Amphibian larvae are gill-breathing, but 

 most adult amphibians are lung-breathing, the skin in many cases 

 being a more or less important accessory. All Amniota are lung- 

 breathers. The pharyngeal chambers are at the maximum in number 

 and functional importance in fishes, and at the minimum in birds and 

 mammals. 



The heart is two-chambered in lungless fishes; three-chambered 

 in Dipnoi and Amphibia; three-chambered, but with a partial division 

 of the ventricle, in Reptilia excepting the crocodilians. In Crocodilia, 

 Aves, and Mammalia, the heart is four-chambered. A review of the 

 aortic arches (Fig. 73) and the arrangement of the main venous channels 

 (pp. 69-78) reveals in each of these two systems a graded series beginning 

 with fishes and passing through amphibians and reptiles to a duplex 

 culmination in birds and mammals. 



The pronephros is the functional kidney of a few fishes. In most 

 fishes and in amphibians the mesonephros is the adult kidney (see p. 

 79). In Amniota the adult kidney is the metanephros. 



In most fishes reproduction is effected with a minimum of acces- 

 sory provision for protection or other needs of the embryo. A relatively 

 small egg, fertilized after it is discharged into the water, is freely 

 exposed to the hazards of the environment. High mortality is com- 

 pensated for by production of vast numbers of eggs. Reproduction in 

 most amphibians is essentially similar to that in most fishes, but with 

 the long larval stage interpolated into the course of development. The 

 relatively large eggs of reptiles, deposited on land, are enclosed by a 

 protective membrane and a shell. The early embryo produces the 

 protective amnion and the respiratory allantois. Reproduction in birds 

 is essentially of the reptilian type, but with eggs of maximum relative 

 size. In monotreme mammals reproduction is definitely reptilian. In 

 utmost contrast to this, most mammals produce eggs of the minimum 

 size in vertebrates, and development is placental. 



In this series of vertebrate reproductive methods there are two 

 breaks. A very abrupt one occurs between amphibians and reptiles. 

 The amnion and allantois correspond to nothing in the embryo of an 

 amphibian or fish. A second break occurs, not between two classes, 

 but within the Class Mammalia. There is nothing in a monotreme 

 corresponding to a placenta. A simpler ichthyopsidan type of repro- 

 duction is sharply contrasted with a more elaborate sauropsidan 

 type. Monotreme mammals are sauropsidan in their reproduction, and 

 the conditions in modern marsupials do not bridge the gap between 

 monotreme and placental reproduction. 



Classed with reference to the structure of the brain, the serial 



