ction the outer layers of the mass become still more highly
differentiated and form a distinct spore membrane, and the spore itself
is now highly refractile. It has been suggested, and apparently on good
grounds, that the spore membrane consists of two layers, the exosporium
and the endosporium. Each cell forms one spore only, usually in the
middle, occasionally at one end (some exceptions, however, are recorded;
e. g., B. inflatus). The shape of the parent cell may be unaltered, as
in the anthrax bacillus, or altered, as in the tetanus bacillus, and
these points serve as the basis for a classification of spore-bearing
bacilli, as follows:

(A) Cell body of the parent bacillus unaltered in shape (Fig. 91, a).

(B) Cell of the parent bacillus altered in shape.

1. _Clostridium_ (Fig. 91, b): Rod swollen at the centre and
attenuated at the poles; spindle shape; e. g., B. butyricus.

2. _Cuneate_ (Fig. 91, c): Rods swollen slightly at one pole and more
or less pointed at the other; wedge-shaped.

[Illustration: FIG. 91--Types of spore-bearing bacilli.]

3. _Clavate_ (Fig. 91, d): Rods swollen at one pole and cylindrical
(unaltered) at the other; keyhole-shaped; e. g., B. chauvei.

4. _Capitate_ (Fig. 91, e): Rods with a spherical enlargement at one
pole; drumstick-shaped; e. g., B. tetani.

The endo-spores remain within the parent cell for a variable time (in
one case it is stated that germination of the spore occurs within the
interior of the parent cell--"endo-germination"), but are eventually set
free, as a result of the swelling up and solution of the cell membrane
of the parent bacillus in the surrounding liquid, or of the rupture of
that membrane. They then present the following characteristics:

1. Well-formed, dense cell membranes, which renders them extremely
difficult to stain, but when once stained equally difficult to
decolourise.

2. High refractility, which distinguished them from vacuoles.

3. Higher resistance than the parent organism to such lethal agents as
heat, desiccation, starvation, time, etc., this resistance being due to

(a) Low water contents of plasma of the spore.

(b) Low heat-conducting power } of the spore
(c) Low permeability } membrane.

This resistance varies somewhat with the particular species--e. g.,
some spores may resist boiling for a few minutes--but practically all
are killed if the boiling is continued for ten minutes.

~Germination.~--When tran