194 BOTANICAL GAZETTE [MARCH 
Heidelberg. Unless otherwise stated, the species used was Spiro- 
gyra longata (Vauch.) Kg. The material and general methods 
were the same as those described in a previous paper (8). The 
original culture of Spirogyra longata was brought from Algiers, but 
the alga had been growing for nearly three years in the laboratory, 
in a stock culture which received small additions of tap water from 
time to time. The filaments of this stock culture were apparently 
healthy and of the usual appearance. The colloidal solutions of 
silver, gold, and platinum here employed were made with non-toxic 
water, according to the method previously described, of distillation 
in glass with animal charcoal present in the retort. They were 
kindly prepared for the author by Dr. W. FRAENKEL in the labora- 
tory of Professor G. BREDIG at the University of Heidelberg, by 
atomizing the metals with an electric arc under water. This 
method has been described by BREDIG (2, 3,4). The solutions were 
kept in flasks of Jena glass. Those of gold and platinum were 
determined by Dr. FRAENKEL and found to contain go ppm. (parts 
per million) of gold and 96 ppm. of platinum, respectively. The 
sample of the silver solution was unfortunately lost before being 
determined, but, as all three solutions were originally prepared so 
as to contain the same amounts of metal (with an error not greater 
than 10 ppm.), the concentration of the silver solution here employed 
may be considered as approximately go ppm. The solution of gold 
was purple in color, that of platinum was yellow-brown, while the 
solution of silver was grayish brown. The gold solution, on stand- 
ing, formed a deposit, probably of the larger particles, on the walls 
both of the flask containing the stock solution and of the culture 
dishes, but the intensity of its color was not perceptibly diminished 
by this deposition. The solutions of silver and platinum gave 2° 
observable deposits. In considering the results obtained with these 
solutions, the possibility is not to be forgotten that small amounts 
of metal oxides may have been formed in the process of preparation, 
and that the solutions are not to be considered as necessarily am 
entirely free from metal in the ionic condition. 
Colloidal solutions of silver, gold, and platinum prepared by 
Brevi, similar to those used by the author except that they were 
weaker and were not made with carbon-distilled water, wet 
