632 
ME.  J.  LISTEE  OX  THE  CUTAXEOL'S 
not  exist  in  fig.  5,  which  represents  the  lower  of  the  two  cells  in  a more  advanced  state 
of  concentration.  The  remote  branches  of  the  processes  were  then  for  the  most  part 
invisible,  and  those  which  did  appear  were  generally  pale,  instead  of  dark,  as  they  had 
been  dming  full  diffusion.  This  difference  does  not  depend  on  contraction  of  the 
branches,  but  on  the  granules  being  absent  from  them,  or  sparsely  scattered  instead  of 
closely  packed  ; and  I have  often  ascertained  from  some  granules  remaining  widely  sepa- 
rated in  a process,  that  it  was  of  large  calibre,  though,  without  careful  searching,  it  would 
have  seemed  invisible.  Even  in  fig.  6 concentration  is  not  represented  absolutely  perfect ; 
for  a few  molecules  are  to  be  observed  near  the  black  mass  in  the  more  cu-cumferential 
parts  of  the  body  of  the  cell.  The  extreme  delicacy  of  the  cell-wall  makes  it  very 
difficult  to  trace  it  among  the  surrounding  tissues,  and  I have  not  attempted  to  give  it 
in  these  figures,  which,  it  must  be  clearly  borne  in  mind,  represent  only  the  coloming 
matter.  The  external  parts  of  the  body  of  the  cell  and  the  principal  processes  may, 
however,  be  sometimes  discovered,  though  perfectly  colourless  in  consequence  of  concen- 
tration : they  are  then  found  to  be  of  the  usual  dimensions  met  with  in  full  difiusion, 
showing  that  they  are  still  full  of  fluid  though  destitute  of  granules.  In  fact  the  only 
change  of  form  to  which  the  cells  appear  liable,  is  a slight  bulging  of  the  central  part 
at  the  seat  of  the  black  mass  in  the  concentrated  state,  which  I have  detected  in  some 
cases  by  camera  lucida  sketching,  and  which  is  consistent  vdth  the  separation  of  the  cell- 
wall  from  the  nucleus,  implied  by  the  ultimate  disappearance  of  the  central  pale  pomts 
of  fig.  4. 
The  movement  of  the  granules  towards  the  centres  of  the  cells  may  be  seen  without 
any  great  difficulty.  The  death  of  a healthy  frog  is  always  followed  by  complete  concen- 
tration of  the  pigment  for  a time,  however  much  diffused  it  may  have  previously  been, 
and  the  process  taking  place  gradually,  its  progress  can  be  observed.  If  a fi’og  with  the 
skin  dark,  and  the  pigment  therefore  diffused,  be  killed  and  the  web  examined  soon 
with  a good  glass  of  high  power,  the  granules  may  be  seen  distinctly  moving  along  the 
offsets  of  each  cell  to  join  the  dark  mass  which  is  becoming  accumulated  in  the  centiul 
part.  If  the  process  is  going  on  languidly,  the  individual  molecules  advance  slowly  with 
slightly  dancing  movements,  indicating  that  they  are  free  in  the  fluid  and  not  confined 
in  any  way  to  the  cell-wall.  If  concentration  is  taldng  place  more  speedily,  the  granules 
rush  along  so  quickly  that  no  time  is  allowed  for  observing  their  molecular  movements, 
and  often  their  motion  is  so  rapid  as  to  elude  the  eye  altogether.  In  one  instance  a large- 
sized offset,  which  at  first  contained  abundance  of  pigment,  became  gradually  cleared  in 
this  way  of  its  colouring  matter  without  any  change  in  its  dimensions,  till  it  was  almost 
invisible  on  account  of  the  very  small  number  of  molecules  remaining  in  it. 
It  is  thus  a matter  of  dkect  observation,  that  the  pigment-granules  move  along  into 
the  bodies  of  the  cells  during  concentration,  and  leave  colourless  fluid  behind  them  in 
the  processes.  It  is  clear  that  their  motion  cannot  be  explained  by  currents  in  the 
fluid ; for  streams  proceeding  towards  the  centre  of  a cell  would  necessarily  be  accom- 
panied by  a returning  flow  in  the  opposite  direction,  which  would  carry  the  pigment  with 
