he membrane, _B_, separates two liquids, one
containing the dissolved food outside the villus, and the other
containing blood inside the villus. Here are proper conditions for
osmosis, and this process of dialysis will take place whenever the
intestinal contents holds more dialyzable material than the blood.
Under these conditions, which will always occur after food has been
digested by the digestive juices, the food will begin to pass through
this membranous wall of the intestine into the blood under the influence
of the physical force of osmosis. Thus the primary factor in food
absorption is a physical one.

We must notice, however, that the physical force of osmosis is not the
only factor concerned in absorption. In the first place, it is found
that the food during its passage through the intestinal wall, or shortly
afterwards, undergoes a further change, so that by the time it has
fairly reached the blood it has again changed its chemical nature. These
changes are, however, of a chemical nature, and, while we do not yet
know very much about them, they are of the same sort as those of
digestion, and involve probably nothing more than chemical processes.

[Illustration: FIG. 3--Diagram of the intestinal walls. _A_, lumen of
intestine filled with digested food. _B_, villi, containing blood
vessels. _C_, larger blood vessel, which carries blood with absorbed
food away from the intestine.]

Secondly, we notice that there is one phase of absorption which is still
obscure. Part of the food is composed of fat, and this fat, as the
result of digestion, is mechanically broken up into extremely minute
droplets. Although these droplets are of microscopic size they are not
actually in solution, and therefore not subject to the force of osmosis
which only affects solutions. The osmotic force will not force fat drops
through membranes, and to explain their passage through the walls of the
intestine requires something additional. We are as yet, however, able to
give only a partial explanation of this matter. The inner wall of the
intestine is not an inert, lifeless membrane, but is made of active bits
of living matter. These bits of living matter appear to seize hold of
the droplets of oil by means of little processes which they thrust out,
and then pass them through their own bodies to excrete them on their
inner surface into the blood vessels. Fig. 5 shows a few of these living
bits of the membrane, each containing several su