286 "Journal of Comparative Neurology and Psychology. 



and definiteness. There are rarely premonitory symptoms of auy 

 kind before the reaction takes place, even thongh the reaction time is 

 long. In snmmary, it is evident that the dogfish is sensitive to acids 

 of a solution of n/75, both in the mouth and spiracle in which are 

 found taste buds^ and also on the outer body surface when no such 

 structures are found. 



J^aOII is the only hydroxide given in the table, although experi- 

 ments were made with KOH which indicate that essentially similar 

 reactions would be secured by its use. To NaOH the dogfish reacts 

 quickly and definitely in the stronger solutions, although not quite so 

 quickly as to acids. It will be noted likewise that the animal is not 

 sensitive generally to so dilute solutions. Reactions are secured from 

 the general body surface, however, to a solution of n/70. It is of 

 special importance to note that the mouth and spiracle are almost in- 

 sensitive to alkalis except in very strong concentration. Some slight 

 chemical reactions take place between the hydroxide and the sea 

 water, as was the case with the acids used also. This probably renders 

 the solutions less powerful than they would otlierAvise be. 



To salts the reactions are slower than to acids and alkalis. The 

 responses to both lithium and ammonium chlorides are practically the 

 same. IvTo reactions to sodium chloride could, however, be obtained. 

 This is due, doubtless, to its })resence in such quantities in the sea 

 water. The reactions to the chlorides are usually preceded by pre- 

 monitory symptoms a few seconds before the definite reaction occurs. 

 These consist of a local or general uneasiness. The reactions are also 

 often prolonged, continuing for a few seconds after the stimulus is 

 removed. It will be noted here, as in the case of the alkali, that the 

 general body surface is more sensitive than the mouth. Definite re- 

 actions are secured from the former to a 2n solution, while the fish 

 shows a sensitiveness to a normal solution. The mouth and spiracle, 

 however, react very weakly to solutions as strong as 5n, and almost 

 never to a lesser degree of concentration. The general lack of effect 

 of the solutions of a weaker gi-ade than iioniial is ])r()bably due to 

 the fact that the salts of the sea water make it about a 5/8 normal 

 solution. This inter])retation is supi)orted l)y the results when NaCU 

 was used nnd ;dso l)y the tests made by Parker ('08b) on the fresh 



