246  FORMATION  OF  GUM  TRAGACANTH. 
walls  of  the  cells,  while  the  inner,  by  a  sharp  line  separating  it 
from  the  outer  homogeneous  mass,  still  showed  fine  stratification. 
Finally,  the  last  transformation  into  perfectly  formed  traga- 
canth  takes  place  when  the  cells  lose  their  outer  solid  definition 
[Begrenzung^  and  these  outer  layers  become  merged  into  a 
more  or  less  uniform,  mucilaginous  mass  ;  in  which  condition  (as 
is  also  often  the  case  in  exuded  gum"!  the  inner  layers  may  still 
be  present  in  their  perfect  integrity. 
The  cells  transformed  in  the  manner  described,  present,  at 
least  when  moistened,  a  much  greater  diameter  than  the  thin- 
walled  cells  out  of  which  they  were  formed ;  thus,  one  of  the 
larger,  unaltered  cells  of  the  medullary  rays  of  Astragalus  de- 
nun  at  us  has  a  diameter  of  0/;/.00G4,  whilst  a  transformed,  yet 
sharply  defined  cell  of  the  inner  part  of  the  same  medullary  ray, 
was  0'/r.035,  that  is  about  five  times  tiiesize;  in  Astragalus 
echinus  the  size  of  the  transformed  medullary  cells  had  increased 
to  0/r/.06,  and  thus  attained  about  the  same  size  as  the  cells 
contained  in  exuded  tragacanth. 
The  cells  are  affected  by  the  action  of  iodine,  according  to  the 
degree  of  transformation  they  have  undergone.  The  unchanged 
cells  of  the  pith  and  medullary  rays  are  rendered  deep  violet  in 
twenty-four  hours  by  the  action  of  an  iodized  solution  of  chlo- 
ride of  zinc.  The  same  occurs  in  the  cells  which  have  only  been 
transformed  in  a  slight  degree  and  still  possess  the  form  of  angu- 
lar, but  thick-walled  cells  of  parenchyme.  This  coloring  is,  how- 
ever, not  uniform  throughout  the  whole  thickness  of  the  walls 
and  of  the  cells,  but  it  is  principally  the  outer  and  inner  layer 
that  is  vividly  colored ;  several  thin  layers  among  the  seconda- 
ry strata  exhibit  also  a  violet  color.  Whether  the  [apparently] 
uncolored  strata  situated  between  these  two  colored  ones  were 
entirely  without  color,  or  were  of  very  pale  violet,  I  could  not 
make  out.  A  similar  condition  is  often  observable  in  other  thick 
walled  parenchyme-cells,  which  soften  and  swell  up  in  water,  as 
for  instance,  in  those  of  Schotia. 
The  more  the  breaking  up  of  the  cells,  and  their  transforma- 
tion into  gum  tragacanth  progresses,  the  more  pale  violet  does  the 
general  body  of  them  become  ;  while  uncolored,  or  faintly  col- 
ored strata,  more  and  more  preponderate  over  the  colored  ones, 
and  the  colored  strata,  especially  the  outer,  show  a  fainter  col- 
