METAMORPHOSIS OF NEOTENOUS AMPHIBIANS 401 



due to respiratory difficulties. Several tadpoles were kept alive 

 until tail resorption was nearly complete by placing them in 

 shallow containers and passing a stream of compressed air 

 through the water. 



In a later series of experiments the thyroids were cut into small 

 pieces and each part transplanted separately into immature 

 tadpoles without hind hmbs. One animal received three large 

 colloid-filled vesicles dissected out of the gland; another received 

 four follicles; the remainder received seven large follicles each. 

 The grafts were made July 26th, and on July 30th, when ex- 

 amined, several engrafted individuals showed evidences of 

 hyperthyroidism, such as emaciation and limb development. 

 By August 2nd all of the animals showed a marked reaction to 

 the graft; emaciation was very marked, the eyes protruded, and 

 the legs had greatly increased in length. One animal had the 

 right fore leg through the skin, and the remainder showed autol- 

 ysis of the skin in the region where the fore limb later appears. 

 The control animals remained unchanged (figs. 3 to 6). August 

 7th, when the experiment was discontinued, all engrafted animals 

 were in advanced stages of metamorphosis. During the course of 

 the experiment the animals were fed quantities of Spirogyra. 

 They fed very little after the first three or four days, and not at 

 all following the onset of marked metamorphic change. 



This experiment shows clearly that the thyroid glands of adult 

 Necturus are highly active metamorphosis-inducing agents. 

 It is reasonable to assume that if they are capable of producing 

 marked metamorphic changes when transplanted into immature 

 anuran larvae within eight days, they are potentially competent' 

 of doing likewise in Necturus if this animal still retained any 

 capacity to transform. It will be shown later that the relation 

 of the thyroids to metamorphosis in Necturus is quite different 

 from the situation existing in axolotls and neotenous anurans. 



While writing this paper the writer came upon a statement by 

 Uhlenhuth ('21) that Jensen ('14) had subjected Proteus to the 

 action of thyroid substance, but did not get any demonstrable 

 results. Uhlenhuth's comment upon this experiment is interest- 

 ing. He says: "Many causes may have been responsible for this 



