REGENERATION IN ANNELIDS
181
neurons to subsist, the axons of which are likely to grow, and of
reparation cells capable of reconstructing neurons so that the conditions for cephalic regeneration can be met in spite of the break in the
posterior nerve cord.
Is it likely that when the nerve chain cannot reach the healing zone,
anterior or posterior regeneration is impossible? The results of Avel
(1938) and Bailey (1939) do not seem to support this view. In both
groups of experiments growth of the nerve chain towards the healing zone
was avoided by bending the cord backwards, yet cephalic regeneration
was obtained in several cases. We feel that these results can be explained
if the histological constitution of the annelid nervous system is taken into
account, together with its great regerative power. The structure of the
ventral ganglia has been well studied by Smallwood (1930) in Eisenia
foetida and by Ogawa (1939) in Pheretima communissima. These ganglia
enclose motor and association neurons ; the sensory system is peripheral.
Nerves containing both sensory and motor fibres leave each ganglion :
there are three pairs in the Oligochaetae. These non-medullated nerves
are accompanied by neurons and undifferentiated cells. They form
numerous ramifications in the body wall. After transection, they
regenerate very rapidly, in a centrifugal as well as centripetal direction, by the axonal lengthening of the sensory neurons on the one
hand and the motor neurons on the other. Axonal growth is accompanied by reserve cell multiplication, and these cells differentiate into
neurons. In the course of our experiments on removal of the nervous
system, positive taxis of the nerves towards the wounded area was observed. If, in the light of these facts, we compare Bailey's experiments
with those of Morgan, an essential difference is apparent. When the
nerve chain is completely destroyed, the very source of the centrifugal motor fibres is suppressed, but if the nerve chain is merely
shifted the stumps of the transected or extirpated nerves remain
attached to the cord and are capable of regenerating ; the incision made
in the body wall so that the nerve chain can be shifted acts as a stimulus
to the formation of a healing zone capable of attracting the regenerating
nerves. A number of these could have reached the level of the incision
and could be damaged by this. This would not only explain the possibility of obtaining regeneration, but also the great variety of results.
Depending on the experimental procedure, it is possible that none of the
nerves reach the blastema fast enough, in which case no regeneration
occurs. If, on the other hand, a significant number of nerves reach the
blastema and activate it, then more or less complete regeneration will
take place.
This interpretation hardly explains the different results obtained
with the two varieties of Eisenia: in this case it would have to be
181
neurons to subsist, the axons of which are likely to grow, and of
reparation cells capable of reconstructing neurons so that the conditions for cephalic regeneration can be met in spite of the break in the
posterior nerve cord.
Is it likely that when the nerve chain cannot reach the healing zone,
anterior or posterior regeneration is impossible? The results of Avel
(1938) and Bailey (1939) do not seem to support this view. In both
groups of experiments growth of the nerve chain towards the healing zone
was avoided by bending the cord backwards, yet cephalic regeneration
was obtained in several cases. We feel that these results can be explained
if the histological constitution of the annelid nervous system is taken into
account, together with its great regerative power. The structure of the
ventral ganglia has been well studied by Smallwood (1930) in Eisenia
foetida and by Ogawa (1939) in Pheretima communissima. These ganglia
enclose motor and association neurons ; the sensory system is peripheral.
Nerves containing both sensory and motor fibres leave each ganglion :
there are three pairs in the Oligochaetae. These non-medullated nerves
are accompanied by neurons and undifferentiated cells. They form
numerous ramifications in the body wall. After transection, they
regenerate very rapidly, in a centrifugal as well as centripetal direction, by the axonal lengthening of the sensory neurons on the one
hand and the motor neurons on the other. Axonal growth is accompanied by reserve cell multiplication, and these cells differentiate into
neurons. In the course of our experiments on removal of the nervous
system, positive taxis of the nerves towards the wounded area was observed. If, in the light of these facts, we compare Bailey's experiments
with those of Morgan, an essential difference is apparent. When the
nerve chain is completely destroyed, the very source of the centrifugal motor fibres is suppressed, but if the nerve chain is merely
shifted the stumps of the transected or extirpated nerves remain
attached to the cord and are capable of regenerating ; the incision made
in the body wall so that the nerve chain can be shifted acts as a stimulus
to the formation of a healing zone capable of attracting the regenerating
nerves. A number of these could have reached the level of the incision
and could be damaged by this. This would not only explain the possibility of obtaining regeneration, but also the great variety of results.
Depending on the experimental procedure, it is possible that none of the
nerves reach the blastema fast enough, in which case no regeneration
occurs. If, on the other hand, a significant number of nerves reach the
blastema and activate it, then more or less complete regeneration will
take place.
This interpretation hardly explains the different results obtained
with the two varieties of Eisenia: in this case it would have to be