CORRELATION BETWEEN METABOLIC GRADIENTS. 325 



quate to account for the facts in the majority of cases. The be- 

 havior of the cilia in galvanotaxis has been carefully investigated 

 by Ludloff ('95), Pearl ('oo), Wallengren ('02, '03), Statkewitsch 

 ('05), and others, and the galvanotactic response has been assigned 

 in most cases to the direct action of the current on the cilia. But, 

 according to Statkewitsch ('05), Parameciuin under optimum con- 

 ditions responds to the current with very little change of the cilia* 

 from their normal behavior. Bancroft ('06) found that exposure 

 to various salt solutions renders Paramecium, which is normally 

 cathodic, anodic, and some solutions destroy the galvanotropism of 

 this ciliate. Dale ('01) found that Opalina, usually described as 

 exhibiting anodic galvanotropism, is cathodic in slightly acid media 

 and anodic in slightly alkaline media ; the same dependence of the 

 response on the medium is likewise true of other parasitic ciliates 

 of the frog. The behavior of the cilia is the same in either reac- 

 tion. According to Terry ('06) and Bancroft ('07), Volvox is 

 cathodic in the light and becomes anodic after being kept in the 

 dark. When cathodic, the orientation is brought about by the in- 

 hibition of the flagella on the cathodal side ; when anodic, by their 

 inhibition on the anodal side. It thus appears that in Protozoa 

 the galvanotactic response is determined chiefly by the internal 

 condition, presumably ionic, of the organism, and not by the direct 

 effect of the current on the cilia. Similar reversals have been 

 noted by us in multicellular animals as the result of environmental 

 conditions or prolonged exposure to the current and appear to be 

 due to internal physiological changes and not to the action of the 

 current on the neuro-muscular mechanism. 



Since the inception of the ideas recorded above it has been our 

 purpose to collect further data on the correlation between metabolic 

 differences, electrical differences, and the galvanotactic response in 

 animals. We hope to test in as many forms as possible the meta- 

 bolic gradient along the axis, the potential differences along the 

 axis, and the galvanotactic behavior. For the last three or four 

 years this work has been carried on whenever opportunity has 

 afforded. It seems desirable, owing to the time which has elapsed 

 since the presentation of the preliminary paper (Hyman, '18), to 

 record the principal data which have accumulated since that publi- 

 cation. To present these data and to point out the correlation 



