STIMULI AND THEIR ACTIONS 



i51 



conceive this, for, since the assumption of an axial direction is 

 possible only when differences exist at two different points of the 

 surface of the body, it is wholly mystical how the direction of the 

 rays, which is the same upon all sides of the body, can produce 

 such an effect. In nature, under ordinary conditions, the decrease 

 in intensity coincides with the direction of the rays, and hence we 

 always see the phototactic movement take place in this direction. 

 But the decrease in intensity can very easily be experimentally 

 separated from the direction of transmission of the rays. Olt- 

 manns ('92), making use of an idea already employed by Stras- 

 burger, devised a very excellent contrivance for this purpose. He 

 made a wedge of two glass plates, which were inclined toward one 



Fia. 225. — Spoctra of various media ; 1, of a red glass ; 2, of a cobalt glass ; 3, of a green glass ; 4, 

 of a solution of potassium bichromate ; 5, of an ammoniacal solution of a cupric salt. 



another at an angle of 2°, and filled the space between the plates 

 with gelatine clouded with India ink. This wedge let through 

 nearly all the light at its thin end, while at its thick end, where 

 the gelatine was darkest, it absorbed much light. If, therefore, 

 the light fell perpendicularly upon the surface of the plates, the 

 greatest decrease of intensity for objects within a dark box 

 behind the wedge lay at right angles to the direction of the 

 incident rays. By means of these plates and the employment of 

 the proper intensities of light it may actually be proved experi- 

 mentally that it is not direction of ray, but solely difference in 

 intensity upon different portions of the surface of the body, that 

 produces phototactic phenomena. 



4. Thermotaxis 



Heat, like light, can be employed very easily for unilateral 

 stimulation, since, whether transmitted by conduction or radiation, 

 it always decreases with the distance from its source, and hence 



G G 2 



