f one or more portions of bone that would have
contributed to the repair. Sometimes the delay cannot be so explained;
Bier suggested that it is due to the escape of blood at the wound,
whereas in simple fractures the blood is retained and assists in
repair.

If sepsis gains the upper hand in a compound fracture there is,
firstly, the risk of infection of the marrow--osteomyelitis--which in
former times was liable to result in pyaemia; in the second place, not
only do loose fragments tend to die and be thrown off as sequestra,
but the ends of the fragments themselves may undergo necrosis;
involving as this does the dense cortical bone of the shaft, the dead
bone is slow in being separated, and until it is separated and thrown
off, no actual repair can take place. The sepsis stimulates the
bone-forming tissues and new bone is formed in considerable amount,
especially on the surface of the shaft in the vicinity of the
fracture; in macerated specimens it presents a porous, crumbling
texture. Sometimes the new bone--which corresponds to the involucrum
of an osteomyelitis--imprisons a sequestrum and prevents its
extrusion, in which case one or more sinuses may persist indefinitely.
Cases are met with where such sinuses have existed for the best part
of a long life and have ultimately become the seat of epithelioma.

It should be noted that all the above changes can be followed in
skiagrams.

_Treatment._--The leading indication is to ensure asepsis. Even in the
case of a small punctured wound caused by a pointed fragment coming
through the skin it is never wise to assume that the wound is not
infected. It is much safer to enlarge such a wound, pare away the
bruised edges, and disinfect the raw surfaces.

In cases of extensive laceration of the soft parts, all soiled,
bruised, or torn portions of tissue should be clipped away with
scissors, blood-clots removed, and the bleeding arrested by
forci-pressure or ligature. If there is any reason to believe that
the wound is infected, any fragments of bone completely separated from
the periosteum should be removed. In comminuted fractures, extension
applied by strips of plaster or by means of ice-tong callipers or
Steinmann's apparatus (p. 150) often facilitates replacement of the
fragments and their retention in position. Plates and screws are not
recommended for comminuted fractures, owing to the mechanical
difficulty of fixing a number of small fragments and the risks of
inf