VARIATION AND ADAPTATION 229 



the young are the dorsal pouch in Nototrema and the united 

 and enlarged vocal sacs in the little Rhinoderma darwini. In 

 the former case the pouch is present only in the female and 

 the eggs are probably introduced by the male ; pressure of the 

 eggs into the skin may from the Lamarckian point of view be 

 suggested as the original cause of the formation of the pouch. 

 In Rhinoderma the brood sac is confined to the male, and it is 

 difficult to explain the origin of such a condition except as the 

 result of the distension of the vocal sacs by the eggs. The 

 male apparently developed the habit of taking the eggs into the 

 mouth like the Cichlidas among fishes, and thence they passed 

 into the vocal sacs. The sacs are united, but they still open by 

 two apertures into the mouth as in other Anura. When empty 

 the sac only extends back to the pectoral region, but when filled 

 with eggs it reaches to the groins and upwards to the dorsal 

 surface, a truly extraordinary modification, sacs belonging to 

 the mouth cavity acting as a brood-pouch and extending be- 

 neath the skin over the greater part of the body. The vocal 

 sacs themselves in ordinary cases are at the same time an 

 adaptation and a secondary sexual character ; they serve as 

 resonators to the voice which is produced by the larynx and in 

 the breeding season the males are continually expressing their 

 sexual excitement by their loud croaking, though whether this 

 music has any effect in promoting the satisfaction of the sexual 

 instincts may be doubted. 



Regeneration, or recrescence, of lost parts, occurs in many 

 Amphibia. This process is sometimes regarded as an adap- 

 tation evolved by natural selection, but to the present writer 

 it seems more reasonable to regard it as a primitive pro- 

 perty of the organism which tends to be lost with increasing 

 specialisation of the individual. In Amphibia it is exhibited 

 more strongly in the young stages ; the gills and limbs of 

 axolotls if cut or bitten off grow again, and this is true also of 

 many or all Urodeles both in the larval and adult stages. 

 Proteus reproduces a whole leg in about eighteen months, and 

 in adult Triton the same complete recrescence has been ob- 

 served. In Anura, on the other hand, recrescence of lost limbs 

 is complete in tadpoles, but in the adult stages is slow and im- 

 perfect. 



Animals of different lines of descent may become adapted 



