182 CAREY: GROSS STRUCTURE OF AN AGAR GEL 
due to the pressure on the air in the pores of the gel together with 
surface tension changes. The water enters the gel quickly at 
first. While this goes on explosions and splittings take place in 
the gel and air bubbles are formed at little cracks. As the bursting 
of the gel and the evolution of the air proceeds, the volume of 
the air spaces becomes smaller and finally disappears. He does 
not feel that there is any evidence of explosions with water. 
Zsigmondy was, of course, working with a non-elastic gel which 
has rather different properties of hydration and dehydration than 
elastic gels have. 
Zsigmondy also writes that Biitschli and later work with the 
ultra-microscope have shown that with silicic acid gels the 
dehydration does not proceed from the outside of the gel inward, 
but that holes are formed in the interior of the gel. These holes 
contain no liquid. This does not seem to be the case with agar 
gels, when sections of the partly dehydrated gel are viewed under 
the microscope, as the first slits visible are near the surface of the 
gel. 
It appears from this work that the structure of the gels is 
probably due to the rapidity of the drying, which in this case 
varied with the temperature used. All the gels showed structure 
except those dried at room temperature, the majority of which 
showed none; a few, however, showed one or two slits. The 
structure is not due to the strain of stretching on silk, as pieces 
dried suspended or laid on trays in the oven showed the same 
appearance. This varied somewhat in different gels, those above 
5 per cent being finer. Gels of lower concentrations when dried 
at temperatures of 43° C. and 70° C. did not show any great 
difference instructure. The percent by weight of water and agar 
just before and at the beginning of structure formation is not 
absolutely accurate, as the gel contains a higher water concentra- 
tion inside than out, owing to the fact that the drying takes place 
from the outside inward. 
I wish to express my appreciation and thanks to Dr. Herbert 
M. Richards for his most helpful advice and interest. 
CotumBiIA UNIVERSITY 
