clined to ascribe this to the different anatomical
arrangement of the two joints, particularly to the fact that the head of
the femur is included in a bony cup, into the hollow of which it is
accurately fixed by the resilient cotyloid fibro-cartilage. The latter
by its firm grasp of the head allows of little play in the joint; hence
vibrations are conveyed directly to the acetabulum in continuous waves,
and rocking of the articular surfaces is prevented. Beyond this no doubt
the difficulty of detecting small effusions in this joint is an element
which must be taken into consideration.

I do not think that wrenches of the knee-joint in the act of falling
can be suggested as an explanation of the frequency of effusions into
that articulation, since the fractures of the femur were not always
received while the erect position was maintained, and effusion was most
marked when the diaphysis was the part affected, the latter point
illustrating the greater resistance offered by compact bone. Again, when
fracture had taken place, the solution of continuity rendered the
directly injured point the most mobile, and tended to prevent lateral
strain from falling on the joints.

Effusion into the knee or ankle, or sometimes both joints, was common in
fractures of the shaft of the tibia.

In the articulations of the upper extremity the condition was also
common, but somewhat less marked than in the lower limb. Effusions into
the shoulder or elbow occurred. In the former these were less striking;
again, perhaps, as a result of the difficulty of detecting small
effusions in this situation. The elbow was to a certain extent protected
by the possession of a degree of fixity somewhat resembling that already
mentioned in the case of the hip-joint, although here depending on the
conformation of the bones alone. I think this explained the absence of
free effusion in many cases of fracture of the humeral shaft, but when
the latter affected the lower third effusion into the elbow was usually
abundant.

The lighter weight and greater mobility of the upper extremity as a
whole, as decreasing the resistance to the bullet, were also probably an
element in the fact that these effusions were less severe than those in
the joints of the lower limb.

The nature of the effusions was apparently simple, since they were
rapidly reabsorbed, and little thickening of the synovial membrane
remained to suggest either a marked degree of inflammation, or the