228 BOTANICAL GAZETTE [SEPTEMBER 
membranes, but these do not seem to interfere with the work or the 
strength of the membrane. 
In three days an osmometer made in this manner and filled with a 
25 per cent cane sugar solution, under ordinary laboratory conditions, 
will send up a column of liquid 130-197 cm. high. When similar 
osmometers are filled with a ro per cent sodium chloride solution, the 
iquid will rise 138-172 cm. in the same length of time and under the 
same conditions. At the end of this time the water column begins to ” 
sink slowly. 
Attempts were made to render membranes semipermeable by using 
tannic acid and copper ferrocyanide. Osmometers having membranes 
prepared as described above were filled with a 5 per cent tannic acid 
solution and then set in dishes containing the same solution. Enough 
melted ro per cent gelatin solution to form a film was poured in the 
bottom of other osmometers, allowed to harden, and then treated with 
the 5 per cent tannic acid as before. After standing in the tannic acid 
solution for 48 hours, the osmometers were placed in water to soak out 
the excess of tannic acid. The acid leeched out for a long time after- 
ward. In some cases the leeching continued for as long as three weeks, 
even after several changes of water. On account of the difficulty of 
freeing the membranes of the excess of tannic acid, no further experi- 
ments were made with this substance. 
For the formation of the copper ferrocyanide membranes, M/20 
solutions of copper sulphate and potassium nitrate were used. Osmom- 
eters were filled with the copper sulphate solution and immersed in the 
potassium ferrocyanide solution, while others containing potassium 
ferrocyanide were placed in the copper sulphate solution (fig. 3). The 
latter method proved the better, as PrerrEeR® found in his experiments 
with porous clay cups. By the end of the third day the celloidin film 
becomes impregnated with the copper ferrocyanide. There may be 
some exosmosis of the potassium ferrocyanide and a consequent forma- 
tion of copper ferrocyanide on the outside of the membrane, but this 
is easily washed off with water. The precipitation membrane is not 
formed uniformly throughout the celloidin, but has a mottled appear- 
ance at first. Later the membrane becomes uniform in color and texture. 
These membranes need not be used at once, but should be used 
within a month. It was found by BicrLow and GreMBERLING that for 
celloidin membranes standing in water for three months the permeability 
1s considerably decreased. The rapidity of the passage of the liquid 
’ Prerrer, W., Osmotische Untersuchungen. Leipzig. 1877. 

