398  Botany  and  Chemistry  of  Essential  Oils.  {Am-lS^;J^xm' 
sl  simple  way,  well  exemplified  in  Pinus.  An  intercellular  space 
already  existing  or  produced  by  division  of  a  mother  cell  into  four 
daughter  cells  becomes  very  much  enlarged,  and  the  cells  sur- 
rounding it  become  modified,  for  by  radial  and  tangential  division 
they  form  an  epithelium  surrounding  the  vessel,  and  these  in  turn 
are  surrounded  by  a  tapetal  or  protective  layer,  and  this  often  by 
a  ring  of  sclerenchyma. 
The  epithelium  secretes  the  oil  as  drops  in  the  vessel,  and  the 
two  protective  layers  prevent  transfusion  of  the  oil  to  the  sur- 
rounding tissue. 
The  drops  of  oil  thus  formed  coalesce  until  gradually  we  have 
the  vessels  filled  with  the  essential  oil.  These  passages  gener- 
ally occur  in  the  cortical  parenchyma  of  the  plant,  which  is  the 
great  metabolic  tissue,  but  they  may  occur  either  in  the  pith  or 
primary  xylem,  as  in  Philodendron,  whilst  in  the  Umbelliferce  the 
passages  are  produced  from  intercellular  spaces  produced  by  the 
splitting  of  the  walls  of  certain  cells  abutting  on  the  pericycle. 
(2)  Cavities  are  generally  formed  lysigenously,  and  occur  in 
Rutacece,  Labiates,  Myrtacece,  and  are  generally  produced  in  the 
following  way  :  Two  cells  becoming  meristematic  give  rise,  by 
cell  division,  to  a  number  of  smaller  cells,  which  secrete  drops  of 
oil,  and  by  the  coalescence  of  these  drops  the  space  formerly 
occupied  by  the  cell  is  filled  with  the  oil,  the  cell  wall  in  the 
meantime  having  been  dissolved,  perhaps  by  an  enzyme. 
In  Myrta  communis,  according  to  Frank  {Beitr.,  125),  a  cell 
divides  into  eight  octants,  and  these  form  an  intercellular  space, 
which,  gradually  becoming  filled  with  oil,  crushes  the  cells  to  the 
side,  and  they  thus  become  flattened,  form  an  epithelium.  This 
is  a  case  of  schizogenous  formation,  but  the  details  are  doubtful^ 
and  fresh  observations  must  be  made  before  fully  accepting  this 
as  a  mode  of  formation  of  the  oil  cavity. 
The  spots  in  Hyperica  and  cells  in  Eucalyptus  are  formed  thus, 
and  it  may  be  taken  as  a  general  rule  that  vessels  are  formed 
schizogenously  and  cavities  lysigenously.  We  now  examine  the 
other  divisions  of  oil  receptacles,  those  of  hysterogenetic  origin. 
(3)  They  exist  both  as  vessels  and  cavities,  but  are  not  nearly  so 
common  as  those  I  have  just  described.  In  Copaifera  and  Dryo- 
balanops  the  oil  receptacles  are  of  enormous  size,  and  are  formed 
by  disorganization  and  solution  of  the  heart  wood,  and  they  are  also 
