Variability of Body Size 45 



clone usually flourishes, to the almost complete ex- 

 clusion of other clones. On the other hand, Jennings 

 ('08b) showed that some wild cultures were very heter- 

 ogeneous in size of individuals. 



It is significant that breadth is more variable than 

 length, at least in Paramecium, quite apart from the 

 larger percentage error involved in its measurement. 



The relatively low variabilities recorded in the last 

 two rows of table 4 indicate that not only homogeneity 

 of the above factors, but also homogeneity of physio- 

 logical condition prevailed. Particularly among con- 

 jugating individuals have low variabilities of body size 

 been found (Pearl, '07; Jennings and Lashley, '13b). 

 But even so, some controllable factors have undoubt- 

 edly escaped attention, for it may be recalled that when 

 trypanosomes lived in the "milieu interieure" of a rat 

 their variability in length (figure 1) was often only 

 half of the lowest variabilities ever observed in Para- 

 mecia. 



The variability of body volume is, of course, much 

 greater than that of any linear dimension. In general 

 it is something greater than the sum of the coefficients 

 for each of three dimensions ; though of course it would 

 be possible for individuals to vary so that shortness 

 tended to be compensated by fatness. This possibility 

 has not been found in any clone of protista which has 

 been measured. 



Some other species. Many other species of unicel- 

 lular organisms than Paramecium have been studied 

 with respect to variability of size. Some representa- 

 tive data are given in table 5. Nearly all, but not all 

 (figure 20) of the distributions are for linear dimen- 

 sions, and for the most part the coefficients of varia- 

 tion are high. This is, obviously, partly due to the 

 heterogeneity of the material. Particularly have the 

 cultural conditions been neglected. In a few cases, as 



