599 
cases of regeneration from outside of the breaking-joint, that I ob- 
tained, was regrettably small owing to the fact that after the operation !) 
the leg is usually thrown off at the breaking-joint. This fact itself 
makes those cases in which, after a successful operation, the stump 
of the leg regenerated at its outer end all the more important, since 
it shows how rarely the stump of the leg could remain attached under 
natural conditions, and yet despite this fact the power of regeneration 
exists at all levels. I wished to repeat and extend this experiment 
not only because I had previously obtained only four successful cases, 
but also because the regenerating parts were all from nearly the same 
region. There did not seem to me to be much doubt that the leg could 
regenerate at any level if a successful operation could be performed. 
Owing to the discovery of a new method of operating I have obtained 
a number of new instances of regeneration distal to the breaking- 
joint. The number is sufficient for present purposes, and would have 
been, beyond doubt, much greater had I hit upon the device earlier 
in the summer, so that more time could have elapsed before the ex- 
periments were brought to a close. 
The new method of operating consisted in first cutting the muscles 
proximal to the breaking-joint, and then cutting off the leg outside 
of the breaking-joint. If only the muscles are cut (either from the 
ventral or from the dorsal side of the leg) previous to the second 
operation, the crab makes many efforts to get rid of the leg by 
pulling at it with the claws of the first pair of legs, but generally 
without success, since autotomy can only take place when the muscles 
are intact. If the nerve as well as the muscles is cut in two proximal 
to the breaking-joint, the crab makes no effort to catch hold of the 
stump of the leg, when the leg is cut off distal to the breaking-joint. 
It could not be told beforehand whether by cutting the nerve the 
regeneration at the distal end of the stump would be delayed as 
we should expect to be the case if the regeneration at the outer 
end of the stump is dependent upon the presence of the intact nerve. 
In my experiments I did not attempt to keep apart the animals in 
which both the muscles and the nerve were cut from those in which 
only the muscles were injured. I am not, therefore, able to state what 
is the effect of cutting the nerve. In future experiments I shall hope 
to work out further the effect of this kind of operation. 
In order to carry out the experiments that are here described 
1) The crabs were previously anaesthized to prevent the immediate 
throwing off of the leg. The claws of the first legs were also clipped. 
