484 Journal of Comparative Neurology and Psychology. 



seen. With the tip of such a rod it is possible to stimulate 

 Paramecium locally, without jarring the animal as a whole. It 

 is then found that a mechanical stimulus back of about the an- 

 terior one-third causes a movement forward. It is notable that 

 at the anterior end the lightest touch produces a strong avoid- 

 ing reaction, whereas an equally light stimulus elsewhere pro- 

 duces no reaction whatever. I was not able to confirm with 

 the rod Roesle's view that the region about the mouth is espe- 

 cially sensitive, but this seems highly probable on general prin- 

 ciples, as well as in view of Roesle's results ; the technical 

 difficulties of reaching precisely the region about the mouth 

 with the rod are very considerable. 



A very powerful stimulus even on the posterior part of the 

 body induces the avoiding reaction. But this may be due to 

 the mechanical transmission of the shock to the anterior end. 



Apparently a very light, unlocalized stimulus likewise pro- 

 duces forward swimming, as I noted in a previous paper (1899, 

 a, p. 104). This is true of a slight jarring of the vessel con- 

 taining resting individuals. Roesle (1902) states that an in- 

 duction shock sometimes has the same effect, though as Stat- 

 KEwrrscH (1903) shows, this stimulus usually produces the 

 avoiding reaction. 



2. Reaction to Electricity. 



Part Played by the Aetion System. — The reaction to the 

 electric current presents certain features not found in the reac- 

 tions to other stimuli. According to the account of this reac- 

 tion in the foundational paper of Ludloff (1895), the cilia on 

 the cathode half of the body of Paramecium strike forward, 

 those on the anode half backward. The inevitable result is 

 that any specimen not in line with the current will be turned 

 directly around, until the anterior Qwd is toward the cathode. 

 The reaction seems, according to this account, to be much sim- 

 pler and more schematic in character than the reactions to other 

 stimuli, the characteristic "action system" seeming to play no 

 differential part. But the recent papers of Pearl (1900), Put- 

 ter (1900) and Wallengren (1902, 1903) show that the reac- 



