HABIT FORMATION IN FROGS 325 



only is necessary. This does not mean that the muscular act 

 of rejection may not sometimes aid in strengthening a habit 

 or in perfecting it more quickly. 



With regard to the rapidity of the formation of a habit my 

 results differ markedly from those obtained by Yerkes in his 

 careful and extensive work on frog reactions. But our results 

 are not contradictory, for Yerkes worked on habit formation 

 by using the labyrinth and other devices, making use of what 

 might be called for convenience the homing instinct ; while 

 my work is based on discrimination involving the feeding in- 

 stinct. The apparatus Yerkes used in his experiments seems 

 to be the simplest that could be devised which insured trust- 

 worthy results. One piece of apparatus was " a small box with 

 an opening 15 cm. by 10 cm. so that the animal could escape 

 from confinement in it through the opening, the lower portion 

 being closed by a plate of glass 10 cm. by 10 cm. leaving a space 

 5 cm. by 10 cm. at the top." The frogs being placed in the box 

 " tended to jump toward the opening because it was light, 

 but they did not learn with 20 or 30 experiences that there 

 was a glass at the bottom to be avoided." (11, p. 582.) Other 

 experiments with a very simple maze or labyrinth showed also 

 that from 50 to 100 trials were necessary to p>erfect the habit 

 of correctly threading the maze. The question now arises, 

 why should Yerkes' frogs require from 20 to 100 trials to form 

 a habit, while my frogs required only from two to seven trials? 



The writer is convinced that the solution lies in the fact that 

 his experiments involved adaptations in a highly developed 

 and plastic instinct (using this word here and throughout the 

 paper in a general sense) that is exercised many times daily 

 in the frog's active natural life, while Yerkes' experiments 

 involved adaptations in a very simple, hard and fast, instinct. 

 Undoubtedly situations frequently arise in a frog's wild life 

 when a disagreeable insect or other food animal is tested and 

 rejected. If the frog did not learn in a few trials to leave the 

 disagreeable object quite alone — if the feeding instinct was 

 as inflexible as it has usually (and erroneously, as I shall show 

 at a later date) been supposed — the frog would be condemned 

 to try the disagreeable food object at least 20 to 100 times, 

 or perhaps indefinitely. Aside from the waste of time and 

 energy which might otherwise be employed in getting food, 



