474  Rhizome  and  Roots  of  Phlox  Carolina.  {AoctobeJ;  mo™' 
and  frequently  three  areas — an  outer  layer  composed  of  compact, 
strongly  lignified  cells,  which  are  present  in  more  lignified  stems 
but  not  here  illustrated;  a  middle  layer,  as  in  Fig.  B,  in  which 
there  are  radial  rows  of  parenchyma  cells  separating  the  thick- 
walled,  lignified  cells;  and  a  continuous  zone  of  lignified  cells,  as 
in  Fig.  C.  The  tracheae  are  marked  with  bordered  pores,  and 
those  near  the  centre  of  the  rhizome  have  annular  and  spiral  mark- 
ings, the  number  of  tracheae  thus  marked  being  rather  striking  and 
characteristic.  Most  of  the  lignified  cells  of  the  xylem  are  in  the 
nature  of  tracheids,  which  are  narrower  and  longer  than  those  in 
Spigelia,  being  usually  not  more  than  20  microns  in  diameter  and 
about  500  microns  long  (Fig.  F).  The  parenchyma  cells  in  the 
outer  layers  of  the  wood  not  infrequently  show  the  presence  of 
a  number  of  small  starch  grains  similar  to  those  already  described. 
Underlying  the  xylem  tissue  somewhat  tabular  cells,  resembling 
those  of  the  sieve,  sometimes  occur,  but  for  the  most  part  all 
of  the  cells  beneath  the  xylem,  constituting  the  pith,  are  made  up 
of  somewhat  thick-walled  parenchyma  cells  resembling  those  of 
the  cortex.  The  walls  of  these  cells  are  non-lignified,  and  are 
wanting  in  simple  pores.  The  cells  are  rich  in  protoplasmic 
contents,  and  may  contain  a  large  number  of  starch  grains. 
The  most  characteristic  features  of  the  rhizome  of  Phlox  ovata 
may  be  enumerated  as  follows  :  ( 1 )  The  upright  or  vertical  position 
of  the  rhizome;  (2)  the  few  rather  long  and  comparatively  thick 
roots  extending  from  the  nodes;  (3)  the  comparatively  thick 
xylem;  (4)  the  absence  of  an  internal  phloem,  the  fibrovascular 
bundles  being  of  the  collateral  type,  and  (5)  the  presence  of  starch 
in  at  least  the  rhizomes  of  the  fruiting  plants.  The  fact  that  neither 
Stockberger  nor  Holm  found  starch  in  this  rhizome  is  probably  due 
to-  their  having  worked  with  material  in  which  it  was  present  in 
rather  small  quantities,  the  amount  varying  unquestionably  with 
the  season  of  the  year.  I  especially  mention  this  point,  in  view  of 
the  fact  that  Stockberger  places  Phlox  in  a  group  in  which  starch 
is  wanting,  and  calls  attention  to  its  supposed  absence  as  a  differen- 
tiating character. 
The  tendency  of  this  plant  to  produce  mechanical  cells  is  further 
shown  by  the  fact  that  some  of  the  cells  of  the  pith  as  well  as 
of  the  cortex  are  thick-walled,  strongly  lignified,  the  walls  being 
marked  by  rather  fine  simple  pores  (Figs.  B,  D).  These  cells 
as  they  occur  in  the  pith  are  either  cubical,  or  elongated  and 
with  square  ends,  while  those  in  the  cortex  are  narrow,  with 
