484 TRANSFORMATION OF ENERGY 



plants that an excretion of cane sugar takes place from the ovule and that the 

 pollen-tube reacts to this substance. There can be no doubt therefore that the 

 ovules of one plant must attract the pollen of entirely different plants, a fact 

 which has indeed been definitely proved in many cases by experiment. Since, 

 however, in nature entry of foreign pollen is prevented, other conditions must 

 co-operate, especially the conditions necessary for the germination of the pollen, 

 where chemical stimuli, often of a highly specific nature must play a special 

 part (pp. 317 and 372). Further, we must not imagine that aU pollen-tubes react 

 only to sugars as stimulants. In this relation Lidfors (1899) has recorded 

 observations which have extended and completed those of Miyoshi. He was 

 able to show that in Narcissus tazetta the attractive substance was not a carbo- 

 hydrate at all, and, after various attempts, he succeeded in proving that it was 

 a proteid that induced the chemotropic reaction. The decomposition-products of 

 the proteid, however, were quite inactive. [As to the chemotropism of roots 

 compare Sammet (1905), Lilienfeld (1905), Newcombe and Rhodes (1904).] 



Hitherto we have considered only liquid or soluble bodies in relation to 

 chemotropic activity ; but it is obvious that gases may also have this effect, for 

 they also diffuse and might affect different sides of plants if in different degrees of 

 concentration. Chemotropic curvatures due to gases have, as a matter of fact, 



been observed by Molisch (1884) in 

 roots and later in pollen-tubes (1893), 

 phenomena to which he has given the 

 name of aerotropism. Molisch' s method 

 of experimenting was as follows : — he 

 separated two chambers from each other 

 by a vertical plate, and placed different 

 gases in each. The plate was pierced by 

 a narrow slit, in front of which, at the 

 smallest possible distance off he placed 

 the radicle of a seedling ; the opposite 

 sides of this radicle were thus in different 

 atmospheres. When the root was placed 

 at the boundary between ordinary air 

 Fig. .S2. Hydrotropism in the root. After SACHS and air poor in oxygcu, a curvature took 



(from Detmer's Smaller Practical Physiology). plaCC tOWards the atmOSphcrC which WaS 



richer in oxygen, and this capacity must 

 obviously prevent the root from penetrating too deeply into the lower layers of 

 the soil ; in other words aerotropism is a factor in determining the depth to 

 which roots penetrate the soil. Negative aerotropism to oxygen has also to be 

 taken into account. This phenomenon makes itself evident when the root has 

 to choose between an atmosphere of ordinary air and one composed of pure 

 oxygen ; the root bends towards that which is poorer in oxygen. By altering 

 the oxygen concentration on both sides a condition is at last reached when 

 neither positive nor negative curvature takes place, a condition of indifference 

 in short so far as this gas is concerned. Molisch found that the root re- 

 sponded only negatively to all the other gases he investigated, viz. carbon-dioxide, 

 chlorine, hydrochloric acid, coal gas, ammonia, and chloroform. When the con- 

 centration of these gases was increased, a positive curvature certainly often 

 appeared, but that was merely due to injury to the concave side of the root, and 

 it was no more a genuine stimulus reaction than the positive curvature akeady de- 

 scribed as occurring in galvanotropic reactions. [Molisch' s results have recently 

 been called in question by Bennett (1904) ; compare also Sammet (1905).] 



Since, according to Molisch, the aerotropic movements took place after 

 removal of the root apex, we must conclude that the act of perception of the 

 stimulus takes place in the growing zone, and this constitutes a difference 

 between aerotropic and hydrotropic curvatures, which have also been observed 



