164 W, C. ALLEE 



starvation, all of wliich depress the rate of animal metabolism, 

 also decrease the positive rheotactic response of these isopods. 

 Heightened oxygen tension, caffein, and a gradual increase in 

 temperature have the opposite effect ('12, '13). ^ 



Changes in the percentage of the positive rheotactic reaction 

 are correlated with resistance to potassium cyanide ('14) and 

 with changes in carbon dioxide production (Allee and Tashiro 

 '14), At the present time their resistance to cyanide and their 

 carbon dioxide production furnish the best known indices of 

 the rate at which metabohsm is carried on in these small 

 organisms. 



Essentially the same method of testing the rheotactic tendency 

 has been used throughout this series of isopod experiments. 

 The animals to be tested, usually three in number, are placed 

 in a waxbottomed enamel ware pan 25 cm. in diameter and 6 

 cm. deep in 2 cm. of water to which they are accustomed. The 

 pan is placed in diffuse light and after standing for about fif- 

 teen minutes a current is set up by stirring with the bulb end 

 of an ordinary pipette. Rheotactic reactions given in a circu- 

 lar current have been found comparable with those given in a 

 straight current and because of the greater speed with which 

 tests may be made the circular current method has been exclu- 

 sively used in these experiments. Great care is taken to create 

 an approximately uniform rate of current. The reaction of 

 each isopod is taken for one minute and these are separately 

 recorded. As a rule this is repeated ten times but with highly 

 toxic chemicals the preliminary' interval was shortened and 

 only five trials were made. 



The chemicals used were Kahlbaum's c. f. a. in all cases 

 except caesium chloride which was Baker's 'Standard Purity.' 

 The solutions were made with distilled water using the molec- 

 ular weights in Chemiker Kalender for 1913. Allowance was 

 made for water of crystallization but no quantitative tests were 

 run even in the deliquescent salts. However, the experiments 

 on calcium chloride and magnesium chloride were repeated each 



' Numerals standing alone refer to my earlier papers. 



