vii ASSIMILATION AND READJUSTMENT 139 



example, that they have, for the most part, little capacity 

 for distinguishing individuals. 1 Very similar facts are 

 established with regard to frogs and toads. Reptiles, like 

 fish, haunt definite places. Crocodiles, for instance, haunt 

 watering places 2 where they pick up cattle or human beings, 

 and a tame yellow adder has been known, when let loose, 

 to return to his cage. 3 After what has been said above, 

 Schneider's description of a turtle persistently chasing a 

 hermit crab will be readily understood as a cognate example 

 of intelligence. Of direct modification by experience, 

 the case of a Texan Snapping Turtle will serve as an 

 example. 4 Having been caught with a bait, and having 

 successfully resisted being dragged ashore, it afterwards 

 avoided both the bait and the place where it was 

 caught. 



To pass to Invertebrates ; there is no doubt that in 

 some form or other the power of learning by experience 

 is to be found among many Insects, Spiders, and, I should 

 say, Crustacea. Of the Hymenoptera it is needless to 

 speak. Among insects other than the Hymenoptera, 

 which may be classed together for our purposes as the 

 Lower Insects, the evidence that I have seen for the power 

 of learning by experience is not extensive. As laboratory 

 evidence we have the case of the mayflies mentioned 

 above, and some experiments in which Szymanski suc- 

 ceeded in teaching cockroaches to conquer their desire for 

 darkness by administering electric shocks. The acquired 

 habit persisted only for a short time, varying from about 

 four to fifty-five minutes, but it could be re-learnt with a 

 large reduction in the number of shocks. 5 Romanes 8 

 quotes a story of a fly that became tame enough to 

 perch on its friend's thumb and eat sugar, but it is signi- 



1 Brehm, VII. p. 25. On the limits to the taming of snakes, see p. 220. 



2 Ib. p. 508. s Ib. p. 297. 



* Ib - PP. 555, 556. 



6 J. A. B., 1912, p. 89 especially. 



6 Pp. 231 and 229. The story of a beetle that fetched a companion to 

 move a cockchafer is beyond me. The apparent co-operation of dung 

 beetles in rolling their balls (ib. pp. 236-238) is probably to be explained 

 by the fact that to the beetle one ball is as good as another. Balls, as 

 such, provoke the beetle to roll even if they are quite of the wrong kind. 

 (Schneider, p. 187.) 



