556         Root* Structure  of  Spigelia  Marilandica.     { ADecimbe?.i906?' 
as  an  exception,  in  regard  to  the  occurrence  of  cystolithes,  from  all 
the  other  members  of  the  Polemoniacece. 
However,  the  roots  of  true  Phlox  ovata  show  the  structure  as 
follows :  There  are  many  root-hairs,  and  inside  epidermis  is  an  exo- 
dermis,  a  single  layer  of  pentagonal  cells  (when  viewed  in  trans- 
verse section),  which  are  thin-walled  and  about  as  large  as  the 
adjoining  cells  of  the  cortical  parenchyma.  The  cortex  consists  of 
about  twelve  layers,  of  which  the  peripheral  two  or  three  are 
slightly  thick-walled,  the  others  are  thin-walled  and  constitute  quite 
a  compact  tissue,  the  intercellular  spaces  being  narrow. 
A  thin-walled  endodermis  surrounds  the  continuous  pericambium, 
inside  of  which  are  many  groups  of  leptome  and  rays  of  vessels 
with  strata  of  moderately  thickened  conjunctive  tissue.  No  crystals 
occur  in  any  of  these  tissues:  no  "  stone-cells  "  and  no  "cysto- 
lithes." The  structure  agrees  in  all  respects  with  that  of  the  roots 
of  other  species  of  Phlox  examined  for  this  particular  purpose. 
If  Professor  Greenish  had  had  access  to  some  more  complete 
specimens  of  true  Phlox  ovata,  he  would  have  observed  that  their 
root  structure  does  not  differ  much  from  that  of  Spigelia  Marilandica, 
which  may  be  described  as  follows :  Epidermis  and  exodermis  as 
above,  but  the  cells  of  the  latter  are  not  so  wide  as  the  adjoining 
cortical  parenchyma.  Cortex  consists  of  about  fifteen  thin-walled 
and  compact  strata.  Endodermis  is  thin-walled,  and  the  pericam- 
bium shows  numerous  cell-divisions,  besides  that  the  stele  has 
commenced  to  increase  in  thickness;  the  conjunctive  tissue  is  thick- 
walled.  No  crystals  or  cystolithes  were  observed.  The  structure 
is,  thus,  almost  identical  with  that  of  true  Phlox  ovata.  While  thus 
the  roots  of  Phlox  and  Spigelia  look  very  much  like  each  other,  the 
rhizomes  are  very  distinct,  and  no  species  of  Phlox  posesses  a  rhizome 
that  in  any  way  can  be  compared  with  that  of  Spigelia :  with  the 
"  cup-shaped  scars  "  from  the  dead  stems  and  the  somewhat  matted 
roots,  developed  from  the  "  very  short  internodes  of  the  rhizome." 
It  was,  therefore,  no  difficult  matter  to  decide  that  the  plant  de- 
scribed by  Professor  Greenish  was  no  Phlox  and  of  course  no  Spigelia 
either.  But  what  was  it  ?  As  mentioned  above,  the  writer  had 
commenced  a  study  of  the  Acanthacece,  and  it  so  happened  that  one 
of  these,  Ruellia  ciliosa,  showed  a  root-structure  so  characteristic 
that  I  feel  confident  that  this  is  the  plant  which  was  confounded 
with  Phlox  Carolina  and  sold  as  a  substitute  for  Spigelia  in  accord- 
ance with  Professor  Greenish. 
