60 MAMMALIAN DESCENT. [r^BCT. IIL 



— Frogs and Toads — should show a tendency to become 

 members of the Amniota, the higher gill-less types 

 which develop an amnion and an allantois, two of the 

 three membranes that enclothe the embryo. 



No other explanation of the curious pouch -developed 

 in front of the lower part of the great intestine of the 

 Frog, when it is passing into an air-breathing type, 

 can be given, than that it is a rudiment of the 

 allantois. Yet this is all ; there is no amnion corre- 

 lated with it, and only in Eeptiles, Birds, and Mammals^ 

 have we all three of the foetal membranes or bags — 

 yolk-sac, amnion, and allantois. But there is every 

 reason to suppose that other temporary gUl-bearers, 

 forms on a level with the existing Amphibia, did, 

 during transformation, develop the rudiment of the 

 amnion as well as of the allantois. Such Proto- 

 amniota, under the quasi-magical influence of new 

 surroundings, may, very probably, have grown into 

 forms, of which some kinds had a huge development of 

 all the three bags, while in others they aU appeared, 

 but were aU arrested. 



In none of the Marsupials or of the higher mammals 

 is there more than a slight development of the yolk-sac ; 

 none have any bulk of food-yolk, so that the embryos of 

 both these groups are dependent upon some other source 

 of food. In all cases the mother supplies this pabulum, 

 but whilst in Birds and Eeptiles the food is prepared 

 in the ovary and its duct, in the Metatheria or 



