' 
44 BOTANICAL GAZETTE [JULY 
path of the cell, 7. ¢., so that the light should fall always ahead 
of the cell, we can arbitrarily prolong the movement in the same 
direction and hinder the cell from retracing its path. 
These observations showing the dependence of movements 
upon light, their prevalence in the red light which, as known, is 
the most advantageous for assimilation, and the influence of 
the direction of incident light on that of the movements, give 
support to the supposition that assimilation is the real ground 
of movements. 
That being stated, how Y 
shall we conceive the mech- 
anism of these movements? 
Imagine a diatom cell to be pat # 
‘ % \ ‘ , Pik 
at a certain distance from hee a ers 
ra Ny en 
- io ud me ee 4 / y 3 
the apex (s) of the light om t3 
cone, with one of its ends AUN reed 
. re ‘ wy ¥ ra 
directed toward it (see fig- Ne baa 
SOA ae 
ure Sale 
The half directed towards 
the light (which we will name 
the anterior) will receive 
more light (because of the less inclosed direction of rays 
towards it) than the posterior half. Consequently assimilation 
will take place here with more energy, and in the same ratio the 
endosmotic current to the interior of the cell will be more 
energetic, since it is dependent on the quantity of elaborated 
products of assimilation. The result will be that the cell will 
be impelled towards the source as a canoe would be in whose 
bow a suction pump had been placed. 
While the cell is approaching the apex of the light-cone, the 
difference of insolation of both halves of the cell is gradually 
lessened, and thus the acceleration in the given direction, depend- 
ing upon this difference, grows more and more near zero, which 
value it reaches at the moment when the center of the cell cor- 
responds to the plane (+y) passing through the apex of the light- 
cone perpendicular to the direction of the movement at the cell. 
