448 University of California Publications in Botany [Vol. 5 



dioxide was not a toxic constituent of tobacco smoke and Miss Doubt 

 {loc. cit.) states that carbon monoxide had no effect in bringing about 

 leaf -fall. Carbon dioxide, in Fitting's (1911) experiments, caused 

 shedding of the corolla prematurely. Brown and Escombe (1902) 

 found for a number of species that in 11.4 parts in 10,000 inflorescence 

 was almost totally inhibited, while in Nicotiana a/finis and N. sylvestris 

 the small flower buds which began to form were all shed long before 

 anthesis. Hannig (1913) found that carbon dioxide was not effective 

 in causing premature floral abscission. 



As will be noted in the case of experiments 3 and 4 in table 1. the 

 flowers fell at the end of forty-eight hours in a dry atmosphere con- 

 taining 10 vol. % of illuminating gas and after approximately the 

 same length of time in the case of the control, whereas in a moist 

 atmosphere containing the same per cent of illuminating gas the 

 flowers fell in twenty-four hours. This situation emphasizes the fact 

 that there is a definite moisture requirement for the abscission process. 

 It has been apparent throughout our field experience with Nicotiana 

 that severe floral abscission often follows sudden changes in tempera- 

 ture, especially when such changes are accompanied by rain or great 

 atmospheric humidity. The role and significance of turgor in the 

 abscission process has been fully dealt with elsewhere (cf. Goodspeed 

 and Kendall, 1916 and Kendall, 1918). The conditions obtaining in 

 experiments 6 and 7 in table 2 were the same as in the experiments last 

 referred to, except that 5 vol. % ethylene in place of 10 vol. % illumi- 

 nating gas was employed. Here, however, the reaction time in the 

 dry and the moist atmospheres was approximately the same, while the 

 first flower in the control fell after a considerably longer period. 

 This may simply be another illustration of the extreme toxicity of 

 ethylene or may possibly be due to the fact that this experiment was 

 performed late in the normal growing season of the plant, with a 

 consequent greater instability of many of the vital physiological 

 processes. 



Our data are in general too fragmentary to permit the drawing of 

 any conclusions as to the basic cause of abscissional responses to 

 poisonous gases. It seems probable that these phenomena will finally 

 be proved to be due to modifications in the normal water relations of 

 the plants concerned. There is abundant evidence that such modifica- 

 tions caused by natural or artificial reduction of the absorbing sur- 

 faces or by increased transpiration under conditions of high tempera- 

 ture and low humidity will bring about abscission. Poisonous gases 



