cle-fibre and nerve-fibre was
thus a compound cell. Capillaries, Schwann held, were formed by cells
hollowed out like drain-pipes, and set end to end--a mistaken view soon
corrected by Vogt (_Embryologie des Salmones_, p. 206, 1842).

In this detail part of his book Schwann accumulates material for a
general theory of the cell which he develops in the third and last
section. Taking up the physiological or dynamical standpoint, he points
out that one process is common to all growth and development of tissues
both in animals and plants, namely, the formation of cells, a process
which he conceives to take place in the following manner. There is,
first of all, a structureless substance, the cytoblastem, the matrix in
which all cells originate. The cytoblastem may be either inside the
cells, or, more usually, in the spaces between them. It is not a
substance of definite chemical and physical properties, for the matrix
of cartilage and the plasma of the blood alike come within the
definition. It has largely the significance of food material for the
developing cells. In plants, according to Schleiden, cells are never
formed in the intercellular substance--the cytoblastem is within the
cells; but extracellular cell formation seems to be the general rule in
animals. An intracellular formation of cells occurs only in the ovum, in
cartilage cells and chorda cells and in a few others, and even there it
is not the exclusive method of formation; a formation of cells within
cells never occurs in muscles and nerves, nor in fibrous tissue (p.
204). In the cytoblastem granules appear, which gradually increase in
size and take on the characteristic shape of nuclei; round each of these
a young cell is formed. Sometimes the young cells appear to have no
nuclei, as in the intracellular brood of chorda cells, but, as a rule, a
nucleus is clearly visible. The nucleus is indeed the most
characteristic constituent of the cell. "The most important and most
constant criterion of the existence of a cell is the presence or absence
of the nucleus," writes Schwann near the beginning of his book (p. 43).

As a general rule the nucleolus is formed first, and round it by a sort
of condensation or concretion the nucleus, which is frequently hollow,
and round this again, by a somewhat similar process, the cell. "The
whole process of the formation of a cell consists in the precipitation
round a small previously formed corpuscle (the nucleolus) of first one