280 Effects of Electrolytes on Gelatin. I 
accurate enough for the purpose. Two determinations were 
made for each point and the average was taken as the true value. 
The test-tubes containing the samples to be titrated with alco- 
hol were kept at constant temperature for some hours before us- 
ing in order to reach equilibrium. The time allowed for this and 
the temperature chosen varied considerably in the different ex- 
periments but, in plotting, only curves which are comparable 
in all essential respects are placed together, unless mention is 
made to the contrary. In mixing, the gelatin was dissolved in 
water at 90°C. and 5 ce. of the resulting solution were mixed 
with 5 cc. of the solution of the electrolyte to be investigated. 
The two halves, of 5 ec. each, of these mixtures were used for the 
two determinations of each point. Three grades of commercial 
gelatin were used throughout. Gelatin A, which was used in all — 
except the earlier experiments, set at a lower concentration and 
gave a clearer solution than gelatin B or C. The concentration 
of the gelatin was so chosen (2 or 3 per cent) that it would not 
set at the temperature of the experiment. Throughout this 
paper, the number of cc. of 95 per cent alcohol necessary to pro- 
duce an opaque precipitate in 5 cc: of the gelatin-electrolyte 
mixture will be spoken of as the “alcohol number’ for the sake 
of brevity. In interpreting the results, it should be remembered 
that in titrating, both the gelatin and the electrolyte are being 
diluted, and the solution is being cooled by the addition of the 
alcohol. The alcohol number is ordinarily increased by a rise 
in temperature and by dilution of the gelatin. The cooling effect 
of the alcohol could be avoided by allowing the gelatin to reach 
equilibrium at room temperature, but sharper end-points are 
obtained by using stronger gelatin. This necessitates a higher 
temperature to prevent setting. If, in titrating, the alcohol 
is run in too rapidly, the value obtained for the alcohol number 
will be too high as the precipitate takes a certain short time to 
form. This error is important only in the case of NaOH and 
HCl and similar electrolytes. The gelatin is stirred after each 
addition of alcohol. The end-point is defined as the point where 
a pencil held behind the test-tube can no longer be seen. 
The alcohol number is ordinarily increased by a rise in tem- 
perature but in the presence of NagSO, in concentrations greater 
than 0.2 m and NaCl o or stronger, the reverse is true. 
