AN ARTIFICIAL OSMOTIC CELL 49 



s 



is greatest along a line of the long diameter. To produce ver- 

 tical steins in such vessels, in the winter, in a room heated in 

 the ordinary manner, is impossible. The inclination of the 

 stems in such cases is always toward the source of heat in the 

 room. A vessel of the type above described was placed under 

 conditions where the temperature on all sides of it were as nearly 

 equal as I could make them. The room (in the winter) was 

 unheated, and the vessel, well isolated with towels, was placed 

 in the central compartment of the bottom drawer of a desk and 

 left there undisturbed overnight. In the morning it was found 

 that the stems were inclined toward a vertical line passing 

 through the center of the vessel. The temperature of the room 

 had fallen during the night, while the stems were growing, a 

 little over one degree, and the center of the vessel has there- 

 fore become the warmest point. 



The direction of the stems is, however, not continuously 

 toward the warmer side. It is at first toward, then away from 

 it; then again toward, then again away; and so on in a series of 

 upward curves. Stems were grown in a vessel of the type de- 

 scribed above, except that it had a tin chimney passing vertically 

 through its center, in which was placed a small flame. The re- 

 sult was a series of regularly symmetrical curves on each side 

 of the chimney, as will be seen in figure 2, E. So far I have no 

 adequate explanation to offer for the formation of these curves. 



When the stem has reached the top of the solution, and the 

 source of heat is definite, as from a Bunsen burner, the direction 

 of horizontal surface-branches is definitely toward the warmer 

 side of the vessel. 



A similar bending toward the source of strong illumination 

 is also observed, but this may be another instance of heat effect. 



THE PRODUCTION OF BRANCHES 



When the stem has been compelled, b}' the turning of the vessel, 

 to make a number of sharp bends, it ceases to grow. Instead 

 a new stem shoots out either from the osmotic sac or from the 

 side of the old stem. It has already been remarked that the 



