498 FOOD REQUIREMENTS 



population growth is correlated more or less directly with the size of the 

 initial population. A number of investigators, however, either have noted 

 no significant correlation, or else have found that the growth rate varies 

 inversely with the initial density of population (e.g., Woodruff, 1911). 

 W. H. Johnson (1936, 1937) has pointed out that these results must 

 be interpreted in relation to the concentration of bacteria in the cultures. 

 Jahn (1929), who partially eliminated bacteria by growing Euglena sp. 

 in inorganic media and washing the flagellates before inoculation, found 

 that the growth rate varied inversely with the initial density of the popu- 

 lation. Most of the literature on bacterized cultures has been reviewed by 

 Jahn (1929),W. H.Johnson (1933, 1936, 1937), and Petersen (1929). 



In contrast to the studies on bacterized cultures, very little work has 

 yet been done with bacteria-free strains. Evidence bearing on the growth 

 of such populations has been presented by Phelps (1935), Mast and 

 Pace (1938), Reich (1938), and Hall and Schoenborn (1939b). 

 Phelps concluded, for Glaucoma piriformis, that the density of popula- 

 tion at the end of the logarithmic phase is, within wide limits, inde- 

 pendent of the initial density of population. This may mean that the 

 logarithmic phase is prolonged in the cultures with small inocula; or 

 that the growth rate is higher in the cultures with low initial density; 

 or perhaps that both the growth rate and the length of the logarithmic 

 phase are increased. Phelps did not consider this question in detail, but 

 some of his data (ser. II, Figs. 4, 5) indicate, at the end of the loga- 

 rithmic phase, an average generation time of about four hours in the low- 

 initial-density cultures and approximately five hours in the cultures 

 started with larger inocula. Such data suggest an inverse relationship be- 

 tween the initial density of population and the rate of population growth. 



In Chilomonas parameciu7n (Mast and Pace, 1938) the rate of repro- 

 duction varies directly with the initial density of population under some 

 conditions, but inversely under others. Some sort of optimal relationship 

 is indicated by the findings of Mast and Pace, since the growth rate in- 

 creased to a maximum with decreasing volumes of medium per flagellate 

 and then decreased to zero. 



Reich (1938) observed that in Mayorella palestinensis the division 

 rate varies directly with the initial density of population in cultures 

 started with less than 3,000 amoebae per cc, although the population 

 "eventually attained is largely independent of the quantity of inoculum." 



