320 ASA A. SCHAEFFER 



with the feet of the frog. (Former control experiments on other 

 frogs showed that the skin of frogs is very sensitive to sohitions 

 of CaCla). The worm was not entirely swallowed. Sometime 

 between 15 and 20 minutes after the worm was snapped up, 

 the worm broke in two, the anterior third being swallowed and 

 the remainder dropping to the floor of the cage. 



Fourth day. 12 m. An untreated worm was presented to 

 the frog. No reaction occurred in 30 seconds. 



Fourth day. 4 p. m. I presented a worm treated with CaClj. 

 The worm wriggled for two minutes unnoticed. I then presented 

 a mealworm which was eaten in about 15 seconds. 



On the fifth and sixth days no experiments were made. 



Seventh day. A mealworm was eaten in five seconds. A 

 worm painted with CaCl^ was then placed into the cage. It 

 crawled into a very favorable position. The frog bent down 

 the head in about 45 seconds after introduction, apparently 

 to look at the worm more closely. In a few seconds after bend- 

 ing down the head, the frog shot out the tongue, missed the 

 worm, and got instead a few fibres of grass. In ten seconds 

 the frog hopped away. I then gave it a mealworm which was 

 eaten in about 20 seconds. 



Eighth day. The frog snapped at, but missed, an untreated 

 worm two minutes after it was placed into the cage. The worm 

 crawled around the cage for 20 minutes without calling forth 

 any response from the frog. A mealworm which was now in- 

 troduced, was eaten in two seconds. A normal untreated 

 worm was then put into the cage, but no change of behavior 

 resulted. Another mealworm placed 15 cm. to the side of the 

 frog was eaten in three seconds. 



Ninth day. The experiments so far show the effects of dis- 

 agreeable chemical stimuli. The object was now to test the 

 formation of habits by using a disagreeable electrical stimulus. 

 An induced current was employed. One electrode was con- 

 nected with the pan holding wet sand on which the frog sat. 

 The other electrode was attached to a very thin copper wire, 

 the end of which was wrapped around the body of a cockroach 

 in such a way as not to interfere much with the roach's crawling 

 movements. The idea was that as soon as the roach should be 

 snapped up, the tongue of the frog, touching the wire, would close 

 the circuit, and a shock would then be sent into the tongue. The 



