llarily.  I have  not  been  able  to  discover  any  reason  for  this  differ- 
ence. It  cannot  be  due  to  a difference  of  speci(^s,  nor  to  the  elevation 
at  which  ihe  trees  are  growing,  nor  to  the  teinperaturedimits  of  the 
respective  districts,  as  in  other  districts,  at  both  higher  and  lower 
elevations,  and  higher  and  lower  temperature  limits,  e.^.,  Mexico 
on  the  one  part,  and  Ecuador  on  the  other,  the  Castilloa  trees  pro- 
duce a perfectly  fluid  latex.  It  is  not  impossible  that  the  condition 
of  the  soil,  and  th(‘  annual  rainfall  may  have  some  influence  upon 
this  point,  but  in  the  absence  of  any  positive  proof  I prefer  to  leave- 
this  matter  for  the  present  undecided. 
d he  latex  of  Castilloa  at  the  moment  of  issuing  from  the  cuts 
forms  an  almost  pure  white,  thick  creamy  mass,  which,  however, 
almost  immediately  begins  to  discolour,  assuming  at  first  a pale 
drab  colouration,  which,  in  the  course  of  a very  short  time  darkens 
into  a brownish  black.  This  phenomenon,  which  is  at  least  one  of 
the  causes  ol  the  very  bad  colour  of  all  the  Central  American  rub- 
bers of  the  present  day,  in  fact,  of  all  the  rubbers  obtained  from 
Castilloa  elastica,  I found  to  be  due  to  the  presence  in  the  latex  of 
an  oxidising  ferment  (oxydase)  and  it  is,  therefore  obvious  that  in 
attempting  to  produce  a high-class,  pure  rubber  from  Castilloa 
latex  the  presence  of  this  ferment  has  to  be  taken  into  consi- 
deration. 
The  taste  of  the  Castilloa  latex  is  intensely  bitter.  This  appears 
to  be  due  to  the  presence  in  it  of  a substance  of  the  class  of  bodies 
chemically  described  as  glucosides.  It  is  this  same  body  wdiich  is 
the  cause  of  the  intense  dark  green  colouration  produced  by  the 
addition  to  the  latex,  or  better  to  its  aqueous  vehicle,  of  a few^  drops 
of  a solution  of  ferric  chloride.  I presume  that  this  reaction  has 
been  observed  before,  and  led  the  observers  to  the  altogether  er- 
roneous assumption  that  the  Ca‘^tilloa  latex  contains  tannic  acid, 
which  latter  as  is  well  known,  produces  much  the  same  colouration 
with  ferric-chloride.  As  a matter  of  fact,  there  is  not  the  slightest 
trace  of  tannic  acid  to  be  found  in  this  latex,  and  I doubt  wTether 
it  occurs  in  the  latex  of  any  other  rubber  tree.  It  is  really  only 
necessary  to  state  that  the  latex  of  Castilloa  elastica,  beside  the 
bodies  already  named,  contains  a very  large  proportion  of  albumen, 
and  to  remind  the  reader  that  albumen  may  be  quantitatively  pre- 
cipitated with  tannic  acid,  in  order  to  prove  that  the  presence  of 
tannic,  acid  in  the  latex  is  an  impossibility.  Indeed,  on  adding  to 
solution  of  the  aqueous  vehicle  of  the  latex  of  Castilloa  a few 
drops  of  a dilute  solution  of  tannic  acid,  a most  coj)ious  precipitate 
of  albumen dannate  is  at  once  obtained.  Considering  that  all  the 
different  specimens  of  rubber  latex  I have  so  far  an  opportunity 
of  examining  contain  albumen  in  varying  quantities,  though  none 
as  much  as  the  latex  of  Castilloa  elastica,  is  at  the  same  time  suffi- 
cient proof  of  the  absence  of  tannic  acid  in  every  case. 
A quantitative  determination  cf  the  amount  of  albumen  and 
albuminous  matters  in  general  in  the  latex  yielded  the  ratln^^r  sur- 
prising result  that  there  is  as  much  as  1 1 per  cent,  of  these  bodies 
present.  This,  I believe,  is  the  cause  of  the  extreme  ease  with 
which  the  latex  of  Castilloa  elastica  can  be  coagidated.  I am  quite 
