which in spite of moderate suppuration united well in
the course of six weeks. The case is of particular interest as
illustrating the nature of the fracture to be expected when the velocity
retained by the missile is low.

The above instances show that such peculiarities as belong to wounds
produced by pom-pom shells depend on the comparatively small size and
weight of the fragments, and on the small degree of impetus with which
they are propelled.

[Illustration: FIG. 95.--Boer Segment Shell, or Shrapnel. The large
fragment is a piece of the case, the smaller are two of the pieces of
iron packed within]

Fig. 95 illustrates a form of shrapnel employed by the Boers, the case
of which is of cast metal arranged in definite segments, while the
interior is filled with small fragments of iron so shaped as to pack in
concentric layers. As to the wounds produced by the contained fragments
I have no experience, since I never saw one of the pieces of iron
removed. This no doubt depended in part on the very unsatisfactory
practice made by the Boers with shrapnel generally. Even when they fired
English shrapnel, the shells were, as a rule, exploded far too high to
cause any serious danger to the men beneath. I saw on one occasion a
large number of shrapnel shells exploded over a body of Imperial
Yeomanry, but as a result of the great height at which all the shells
were exploded, not a single casualty resulted.

The segment casing of the shell, however, I several times saw removed
from the body. The fragment shown in fig. 95 was removed from the
buttock of a man after one of Lord Methuen's early battles. It may be
remarked that the buttock is rather a common, and also a favourable,
seat for shell wounds with retention of the fragment. This no doubt
depends on the fact that the buttock is one of the few superficial
regions in which sufficient depth of tissue exists for the retention or
the passage of so large an object as a fragment of shell.

Fig. 96 is of a number of leaden shrapnel bullets from our own shells. A
normal undeformed bullet, such as was the usual cause of wounds, is
shown at the left-hand upper corner. The remainder show common forms of
deformity caused by striking on the ground or against rocks. I attribute
small importance to the deformed bullets, as I never saw one removed,
and it is probable that a ricochet shrapnel bullet would rarely retain
sufficient force to penetrate. The lower fragments are inserted to