1919] Esterhj: Reactions of Various Plankton Animal* 19 



that were obtained at the surface during the night. Table 1 (p. 20) 

 is a summary of the distribution of four different sets of animals 

 (A-D) in a column of water. The tube containing the animals was 

 marked off into five equal divisions and exposed to light for a short 

 time and the distribution noted one or more times. Then it was 

 covered so as to exclude all light, and after an interval the cover was 

 removed and the distribution noted. The alternation of light and 

 darkness was repeated several times, and all the records for a given 

 set of animals are put together in the four lines of each half of the 

 table. 



The form of tabulation used in table 1 is found in others in this 

 paper, and the following explanations of terms will apply to all tables 

 of this sort. The "whole number of animals observed" is the sum 

 of all records of distribution in all experiments. Each living animal 

 was counted as many times as there were observations made : if there 

 were five individuals in a tube and their distribution was recorded ten 

 times, the "whole number observed" would be fifty. If it is necessary 

 to count quickly it sometimes happens that all the individuals known 

 to be in the container can not be located ; in such cases the ' ' whole 

 number" does not equal the product of "number of individuals" by 

 "number of observations." "Percentage distribution in the sections" 

 is the proportion of the total number of animals observed in each 

 section, to the whole number observed. "Number of observations" 

 means the number of times the distribution was recorded. The "center 

 of distribution" is the average position (as between the ends of the 

 container) of the whole number of animals observed. It is obtained 

 by multiplying the total number of animals recorded in each section 

 by the number of the section, and dividing the sum of the products 

 by the "whole number of animals observed." The centers are set 

 down in units and tenths and the unit is considered as the middle point 

 of the section having the corresponding roman numeral. If one center 

 is compared with another the difference between them, if any, is the 

 magnitude of the shift of the whole population. This useful method 

 of expressing average position was first used, so far as I know, by 

 Yerkes (1899). 



The ascent is accomplished by means of swimming while in descend- 

 ing the animals drop passively, with the head uppermost and the an- 

 terior antennae spread to their full extent. The passive descent con- 

 tinues after the animals are lighted, following a short time in darkness, 

 but in a few seconds they begin to swim up in characteristic jerky 

 fashion. 



