432 JOHN B. WATSON 



is found it folds its fins, assumes a characteristic position (ganz 

 schrag auf seine Breitscite gelagert) and remains in this atti- 

 tude from one-half to two hours. Only the regular but slow 

 movements of the gill cover show that life is present. Romeis 

 doubts whether the above described form of sleep is very pro- 

 found, since the near approach of any object, or noises, causes the 

 fish to take flight immediately. The author sees in this activity 

 a mechanism which is economical with respect to metabolism. 

 It should be designated generally by "Ausruhstellung." 



At night the behavior is different ; the fish ordinarily sinks to 

 the bottom of the container and rests on the ventral surface. 

 Fin movements cease. It seems to be genuinely asleep, since 

 light in moderate intensities and noises do not disturb it. 



Amphibians. Mulder (18) finds that the speed of the rota- 

 tion reflex is directly proportional to the amount of the stimulus. 

 The relation is not logarithmic as is the case with sensations. 



Birds. Breed (4) finds that the drinking instinct does not 

 have to be supplemented by imitation, accident, intelligence, 

 instruction, etc., in order to act. The pecking response as a 

 whole is very imperfect at birth. For purposes of observation 

 the response is subdivided into three separate reactions: strik- 

 ing, seizing and swallowing. The amount of daily improvement 

 of the act as a whole and of its separate parts was measured up 

 to the 25th day. Pecking improved in accuracy rapidly during 

 the first two days. On the third day efficiency is placed at 

 29.29 on a scale of 50, 40.10 on the eleventh day and never 

 higher than 42.57 for the rest of the time of observation. Social 

 influence exerted little effect in improving accuracy. 



On page 40 Breed raises the very interesting question of the 

 relation of habit to instinct. Although in pecking there is 

 improvement with practice, the question remains whether the 

 increase in accuracy is dependent upon practice, or whether it 

 is the natural functional correlate of structural maturation. 

 The question might be answered experimentally, it would 

 appear, by forced feeding" of a group of chicks for the first week 

 and then making a comparison of the accuracy of pecking with 

 that of a group brought up in the normal way. Since the nervous 

 system resists starvation well, it is doubtful if a few days of 

 slight underfeeding would check the normal rate of the develop- 

 ment of the sensory -motor arcs. 



