154 THE PHYSIOLOGY OF INSECT SENSES 



information derived from these kinds of experiments is suggestive 

 rather than conclusive ; nevertheless, considering all of the available 

 evidence, it is highly probable that insects can do better than dis- 

 criminating simply between acceptable and unacceptable. 



Bauer (1938) trained Dytiscus marginalis and Hydrous piceus to res- 

 pond positively to sucrose; these animals could not be trained simul- 

 taneously to avoid glucose, and also reacted positively in the majority 

 of cases to some fifteen or twenty other sugars and sugar derivatives. 

 As in the experiments of Schaller (1926) and Ritter (1936) with the 

 same or related species, the beetles learned readily to distinguish be- 

 tween pairs chosen from sucrose, sodium chloride, acids (hydrochloric 

 and acetic), and quinine, and could even be trained to accept quinine 

 and avoid sodium chloride. But specimens that had learned to avoid 

 hydrochloric acid also avoided acetic and could not be taught to 

 respond differently to the two; the same result was obtained when 

 quinine was matched against salicin or aloin. Bauer concluded that the 

 sweet substances (with the possible exception of mannose, which was 

 avoided by some individuals) constitute a single homogeneous 

 grouping; salts, acids, and bitter substances, normally avoided, are 

 distinguished from sweet and from each other, so that taste sub- 

 stances can be classified into the same four quahties for these beetles 

 as for man. 



As mentioned above in the discussion of methods, von Frisch (1935) 

 found additive the stimulatory effects of all sugars acceptable to the 

 bee. Summation was noted also between sodium chloride and lithium 

 bromide, ammonium bromide, or hydrochloric acid, but the repellent 

 effect of quinine was lessened rather than enhanced by the addition of 

 acids (hydrochloric, acetic, sulphuric, citric, lactic). When sucrose 

 solutions, one containing sodium chloride and another quinine hydro- 

 chloride, were prepared and were accepted by equal proportions of 

 the bees they were drunk in different amounts. Almost as much of the 

 quinine was taken as of the control, but considerably less of the 

 solution containing salt. The results would indicate three taste 

 qualities: sweet, acid-salt, and bitter. But in other experiments von 

 Frisch found that starved bees show no better acceptance of sodium 

 chloride than those fully fed, although the threshold for rejection of 

 quinine rises eight times and that for hydrochloric acid by a factor of 

 five. He concluded therefore that salt represents a quality different 

 from either acid or bitter. Furthermore, although there is no summa- 

 tion of repellency between quinine and acids, other bitter substances, 

 such as aloin, arbutin, colocynthin, and salicin, are rendered more 



