288 THE ZOOLOGIST. 



under trial. The connection between sensation and response 

 was immediate. There was no delay. The accidental success 

 of the first trial did not establish a new road to the nest. In 

 other respects, such as the slow elimination of useless move- 

 ments, especially in the beginning of each reaction, the results 

 do not favour the presence of images nor ideas, nor even of the 

 sensory thought described by Hunter."'^ There was evidently 

 sensation, but not perception. Hence the word retentionf is to 

 be preferred to the word memory as a term for the mechanism 

 of the more or less permanent modifications of behaviour in the 

 Wasp. The conclusion is reached that the learning of the Wasp 

 did not transcend the sensori-motor level, and that images or 

 ideas were not elements essential to an explanation of the 

 observed reactions. 



From the absence of locality-studies prior to the start of the 

 experiments, the discontinuance of the studies after the plug 

 was left out of the tunnel, and their cessation after the sixteenth 

 trial excepting when the plug had to be renewed or remodelled, 

 from the similarity of the behaviour in front of the plug on 

 arrival from the feeding grounds, and the locality studies made 

 before the tunnel on departure from the nest, it may be surmised 

 that the locality study is a compulsory reaction to more or less 

 novel environmental stimuli, having a biological significance to 

 the Wasp, rather than a means directed towards securing an 

 accurate return to the point of departure. The locality-study is 

 certainly a feature of Aculeate behaviour deserving of a far more 

 critical examination than any the study has hitherto received. 



* Hunter, " The Delayed Eeaction in Animals and Children," ' Behaviour 

 Monographs,' vol. 2, No. 1, 1913. 



f Watson, ' Behaviour,' p. 241, 1914. 



