44 Regulation of Size in Unicellular Organisms 



To define how much the individuals of a group dif- 

 fer among themselves, it is convenient to use coeffi- 

 cients of variation. A coefficient of variation is the 

 percentage of the mean or average which is formed 

 by the mean squared or standard deviation. 



Paramecium. In table 4 are represented data which 

 were obtained by various surveyors of Paramecia. A 

 number of points are brought to light in this table. 

 There is a general improvement in the degree of simi- 

 larity among the individuals of one group as more and 

 more factors are taken under control. The factor of 

 age after fission has been evaluated in the previous 

 chapter. Homogeneity of hereditary factors, which 

 is obtained by limiting the group to one clone (which 

 consists of descendants of a single individual) is seen 

 to be important from the fact that the various groups 

 measured differ tremendously in mean dimensions. 

 Homogeneity of cultural conditions over a consider- 

 able period of time before measurements were taken, 

 will be shown below to be a prime factor. 



The factor of growth or age was first demonstrated 

 to be of significance by Simpson ('02). Sister individ- 

 uals showed similarity in length such that a correla- 

 tion coefficient of +0.70 was found ( + 1.00 being com- 

 plete similarity). It also turned out that as growth 

 proceeded, sister individuals became more and more 

 diverse; and this requires that it be recognized that 

 growth in mass was partially independent of age. 



Inheritance was shown by Jennings ('08b) to influ- 

 ence size within a unicellular species. It seems neces- 

 sary to remark, however, that cultures which do not 

 consist of a clone are nevertheless often homogeneous 

 in race, but of course not certainly so, because of the 

 fact that all the parent individuals in the culture may 

 have come eventually from some common ancestor; 

 and because under particular conditions some one 



