ABSORPTION OF MOISTURE BY GELATIN IN A SATURATED ATMOSPHERE 323 
The generalized formula is 3^ = a logio {hx + i) + c, in which y is the 
percentage of total intake, and x the time, with a, h, and c constants. 
ISO' 
UO- 
$0' 
70- 
50 
30 
1 0 
/O 10 30 40 days. 
Fig. I. Curve of moisture intake of gelatin in a saturated atmosphere. Lower curve, 
from von Schroder's data. Upper curve, from data presented in Table 4. 
In the case of seeds we found it necessary to use two or even three curves 
to approximate the experimental data (4). Similarly it is necessary to use 
two equations for curves joining in a common tangent value to express 
the gelatin absorption curve. 
The first part of the curve, with the values of a, b, and c substituted 
in the equation, is as follows: y = 93.4 logio (0.032:^; + i) + 10.413; while 
the later part is expressed thus: y = 141. 9 logio (0.0064X + i) + 40.82. 
These two curves have equal tangents at x — 209.47 hrs., at which time 
the two values for y are, yi = 93.22723, and 3^2 = 93.22764, showing a 
break in the curve of only .00041 per cent. In other words, at the end of 
about six days, when the gelatin has taken in something less than its 
own weight of water (93 percent) it requires different values for the con- 
stants in the general formula to keep a calculated curve running close to 
the data. With these two successive curves there is fairly close agreement 
between the calculated and the observed data as shown in table 5. 
The agreement is very good with the exception of the first and the next to 
the last readings. The data at 960 hours may be in error, although there 
is nothing in the records to indicate it. It seems scarcely likely that the 
