Frederic T. Lewis 119 
(dorsum) is directed anteriorly. The hand is not rotated with the arm, 
but ordinarily in the reverse direction, so that its extensor surface is 
directed anteriorly like that of the foot. Since the arm and hand are 
rotated in opposite directions, a crossing of the bones of the forearm is 
produced. In man the hand may, in later development, be rotated with 
the arm so that its dorsum looks posteriorly and the bones of the forearm 
are not crossed. In this position the inverse symmetry of the arm and 
leg is complete. 
The embryonic rotation of the limbs is not to be compared with their 
voluntary rotation in the adult, for the former is a complex shifting of 
tissues involving modifications in the shapes of the bones. These changes 
in the human leg are clearly shown by Bardeen’s reconstructions in Vol. 
4 of this journal. (Compare Figs. 3, 5, 9, 12, and 13, following p. 302.) 
The external appearances during rotation may be observed in the rabbit 
embryos figured by Minot and Taylor for Keibel’s Normentafeln. From 
these it will be seen that rotation does not occur with mathematical pre- 
cision. 
Interpreted according to the rotation theory, the fundamental veins of 
the arm correspond with those of the leg. Their homologies are shown in 
the accompanying diagram which is based upon the reconstructions pre- 
viously described. The diagram presents throughout anterior views of 
the left limbs, the veins being drawn as they would appear if the hmbs 
were transparent. In Stage 1, at a and A respectively, the arm and leg 
are shown before rotation. Serial homology between the primitive ulnar 
and fibular veins is complete. Then rotation occurs, whereby the fibular 
vein is carried from the posterior to the outer border of the leg, as shown 
at B. The arm, on the contrary, turns so that the ulnar vein is carried 
from the posterior to the inner border, as in b. The forearm rotates in 
the opposite direction from the upper arm, so that the ulnar vein crosses 
from the inner side above to the outer side below. Were the forearm in 
supination, the ulnar border would be internal throughout. After rota- 
tion the ulnar border is no longer homologous with the fibular, but cor- 
responds with the tibial border. 
In Stage 2, veins are established along the inversely homologous ex- 
ternal borders of the limbs, the radial and fibular respectively. As shown 
in the diagram, the cephalic vein must be a new formation throughout, 
but the course of the sciatic vein is already partially occupied by the 
primitive fibular vein. Consequently the sciatic may incorporate a 
portion of the fibular vein. Thus it appears that a real homology exists 
between the cephalic and sciatic veins, although, as Hochstetter pointed 
out, they differ in their relations to the primitive veins of the limbs. 
