1053 
theii  sifuated  most  frequeiitly  in  the  ueigbourliood  of  the  central 
oanal  (tig.  15).  If  the  cell  is  at  a somewhat  later  stage  of  develop- 
raent,  an  incision  of  the  nucleus  is  usuallj  accompanied  by  a division 
of  the  suiTOunding  protoplasm,  which  even  at  the  same  stage  shows 
protoplasmic  processes  engaged  in  development  (fig.  10).  Cells  of 
this  last  type  are  situated  farther  away  from  the  centi-al  canal. 
It  is  noteworthy  that  the  structnre  of  the  nucleus  in  the  cell  engaged 
irl  amitotic  division  seems  to  be  relatively  intact  in  comparison  with 
the  corresponding  condition  in  mitotic  cell  division. 
With  regard  to  the  degree  of  the  neurone  formation  1 thinkthat, 
on  the  ground  of  the  reasons  given  above,  I may  go  a step  further 
than  Paladino  when  he  writes : “En  conclusion,  Ie  tissu  nerveux 
ne  fait  pas  exception  a la  loi,  d’après  laquelle  tont  tissn  vit  dans 
l’ensemble  et  se  renouvelle  isolément,  pour  remplacer  les  éléraents 
qui  se  détériorent  et  se  détruisent;  en  d’autres  tenues,  Ie  tissu  ner- 
veux, Ini  aussi,  est  un  siège  de  regénémtion  pour  ainsi  dire  restauratrice.” 
It  seems  to  follow  from  what  has  been  shown  above  that  we  are 
not  dealing  w'ith  merely  a restoration  of,  but  also  with  an  increase 
in  the  mimber  of  neurones. 
In  order  to  complete  this  sui-vey  I shall  add  the  results  of  iny 
investigations  of  the 
Appearances  of  groiotk 
of  the  axons  in  the  dorsal  and  ventral  rools  of  the  spinal  nerves.  As 
has  already  been  shown  above,  the  calculations  of  the  number  of 
the  nerve  fibres  in  cross-sections  of  the  dorsal  and  ventral  roots  of 
the  spinal  nerves  made  at  a,  and  b,  hi  lext  fig.  1 gave  such  valnes 
that  one  might  expect  that  tignies  of  gi'owth  might  also  really  be 
shown  in  longitudinal  sections  of  these  roots.  Silver-impregnated  roots 
from  some  intact  lurnbal  nerves  of  a 17  days  old  dog  were  set  up 
in  series  of  sections  of  suilable  thickness  atid  then  the  preparations 
were  searched  for  tigures  of  growth.  These  investigations  showed 
the  occurrence  of  a large  number  of  figures  of  nerve-fibres  free 
from  medullary  sheaths,  whose  ends  are  situated  between  the  two 
section  surfaces  of  the  preparations;  the  shape  of  these  ends  shows 
that  they  could  scarcely  be  due  to  the  nerve-fibres  having  been  cut 
otf  when  the  sections  were  cut.  Of  the  different  shapes  that  the 
ends  of  these  nerve-fibres  take  1 will  only  mention  one  here,  namely, 
that  which  shows  a swelling  at  the  point;  this  swelling  has  in  most 
cases  a claviform  shape  (figs.  22,  23,  24,  25,  and  26).  The  nerve- 
fibres  in  these  cases  were  very  tine,  and  showed  repeated  convolu- 
tions  during  their  course  (fig.  26).  A large  number  of  such  nerve- 
6b* 
