[Vol. 2 



480 ANNALS OF THE MISSOURI BOTANICAL GARDEN 



plants can survive in a medium without nutrient materials. 

 That these plants could not survive for that length of time in 

 the other media, however, shows that in those cases a real 

 toxicity enters into consideration. 



In addition to the actual time limits for recovery just tabu- 

 lated, as well as the method of recovery and delayed maturity 

 mentioned in the preceding section, another interesting point, 

 which was very noticeable in the cultures and which can also 

 be seen in the plates, is the character of growth of the root- 

 lets in the boundary cultures, by which is meant those cultures 

 which have remained in the inimical media nearly as long as 

 their endurance would permit, and whose recovery in full 

 nutrient solution is slower or more difficult than the normal 

 unaffected plants. In the latter case the roots are short and 

 compact and usually extend down only to about one-half the 

 distance to the bottom of the tumbler. In the case of the first 

 mentioned cultures, however, when transferred to full nutrient 

 solution the rootlets develop a long, slender growth easily 

 extending to the bottom of the tumbler. 



S 



VI. Effect of Sterilizing the Water During Growth of 



Plants 



The foregoing series pointed, therefore, to factors other 

 than extraction or loss of solute from the plant tissue as bein 

 responsible for the deteriorating phenomenon observed when 

 growing plants are placed in distilled water. In the unre- 

 newed water cultures in the previous series a brownish colora- 

 tion developed and the roots appeared, in their gelatinized 

 condition, to be covered by bacterial and fungous growths. 

 Suspecting that these organisms played an important role, it 

 was decided to grow additional cultures to test this point. 

 Four cultures, each containing ten plants of Pisum sativum,. 

 were set up in distilled water: in one the medium was not 

 renewed ; in a second the water was renewed every four days ; 

 and in the remaining two the medium was sterilized every four 

 days by boiling in a return condenser one-half hour. The re- 



sults are given in table vii (series 5) and the cultures are 



shown in pi. 16 fig. 1. The full nutrient solution cultures, 



