FEEDING EXPERIMENTS ON TADPOLES 467 



It is thus clear that, whenever thyroid forms a part of the diet, 

 a precocious differentiation sets in; when thymus is added (thy- 

 mus-thyroid, thymus-hypophysis, thymus-plants), the opposite 

 influence is noticed. 



One group (9 m) of Bufo tadpoles was starved to test again 

 whether hunger might cause differentiation, but no differentia- 

 tion was noted even 5 weeks after the start of the experiment. 



Finally another group was fed on testicle. The testicle diet 

 had not been fully tested before. The development of these tad- 

 poles did not differ from the control and this group can therefore 

 serve for a further comparison with the groups given a mixed 

 diet. 



Experiments were also begun on Triton alpestris and on Bufo 

 \n.ilgaris tadpoles which had a part of their tails amputated, but 

 the time was too short to carry them out completely. 



Feedings were started with Triton alpestris larvae on June 24. 

 On July 12 the outer gills of the thyroid-fed ones had disappeared 

 and the fore legs were well differentiated. The thymus-fed speci- 

 mens still possessed their outer gills and had only small buds of 

 their fore legs. The animals treated with other foods showed 

 intermediate conditions. 



Bufo vulgaris tadpoles of 19 to 23 mm in lengthy tails 10 to 14 

 mm. long, had 5 mm of their tails amputated on June 24. The 

 next day separate groups were started on thyroid, thymus, mus- 

 cle and vegetable diets. The thyroid group developed hind legs 

 on July 3 (8th day) and fore legs on July 12 (17th day) . The thy- 

 mus and muscle groups developed hind legs on July 9 (14th day). 

 Their measurements are given in table 15, page 468. 



The thymus-fed ones had grown considerably by July 14 and 

 regained their former average length, since they had almost 

 completely regenerated their tails. The other groups had grown 

 and regenerated much less, the thyroid least of all, while in turn 

 it showed the fastest differentiation. These observations cor- 

 respond with those of 1911. Romeis has studied extensively the 



