140 TBOPISMS 



order to get clear results a method must be used which 

 prevents a rapid diffusion of the substance; and, more- 

 over, the current of diffusion must be confined to an 

 almost straight line. It is possible that Pfeffer's method 

 satisfies this condition. 424 . 425 He introduced the sub- 

 stance to be tested for its chemotropic effect into a capil- 

 lary tube, the end of which was then sealed. The other 

 end was pushed into a drop of water containing the sus- 

 pension of the organisms whose chemotropism was under 

 investigation. From this capillary the diffusion was ex- 

 tremely slow. Moreover, the current of diffusion was 

 approximately linear at the orifice. Hence the test for 

 the existence of positive chemotropism was perhaps pos- 

 sible. When an organism, struck sidewise by the line of 

 diffusion near the opening of the capillary tube, turns 

 toward the tube going into it, some probability of positive 

 chemotropism exists ; and when all the organisms coming 

 near the orifice of the tube are thus compelled to go into 

 it, the probability may become certainty, provided that 

 the substance used does not paralyze the organism and 

 therefore act as a trap, allowing the organisms to come 

 in but not to go out. The capillary tubes used were of 

 10 to 15 mm. length and of a width of about 0.1 mm. 

 Pfeffer and his pupils found that the spermatozoa of 

 ferns go in large numbers into a capillary tube containing 

 sodium malate in a concentration of 0.01 per cent, (a solu- 

 tion ten times as diluted is still slightly active). This 

 effect of the malate is specific in this case and this indi- 

 cates that either a definite chemical action of the malate 

 ion or a specific permeability of the organism for it is the 

 source of the chemotropism. Such specific chemotropic 

 effects are not rare, since Pfeffer found that Bacterium 

 termo and Spirillum undula are positively chemotropic 



