,6  Observations  Regarding  Kola  Nuts.   {  ^iS^flSS!11- 
Heckel  and  Schlagdenhauffen3  obtained  as  much  as  2  3  per  cent, 
caffeine,  and,  besides  this,  theobromine,  0  023  per  cent;  fats,  0-5  per 
cent.;  tannic  acid,  1-59  per  cent.;  starch,  337  per  cent.  Schlotter- 
beck  and  Knox  obtained  practically  the  same  figures  as  these,  but 
none  of  these  investigators  state  whether  they  used  West  Indian 
or  African  kola  nuts.  Good  typical  samples  of  both  these  varieties 
of  kola  nuts  in  the  dry  state  were  procured  from  reliable  sources. 
The  African  nuts  were  not  prepossessing  in  appearance,  being  smaller 
and  darker  than  the  Jamaica  nuts,  and  more  shrivelled  and  less  per- 
fectly cured  than  these.  Two  methods  of  assay  were  tried  in  case 
of  each  kind  of  nut,  using  chloroform  in  the  one  and  33^  per  cent, 
alcohol  in  the  other,  as  boiling  with  pure  water  was  impracticable, 
on  account  of  the  large  amount  of  starch  contained  in  the  nuts. 
Method  I. — This  was  similar  to  that  employed  by  Schlotterbeck 
and  Knox,  and  consisted  merely  in  extracting  the  powdered  nuts  in 
aSoxhlet  apparatus  until  the  chloroform  runnings  no  longer  yielded 
a  residue,  evaporating  off  the  chloroform  and  evaporating  the  resi- 
due to  dryness  on  a  water-bath  with  calcined  magnesia  and  sand,  in 
a  flat  porcelain  dish.  This  dry  powder  was  then  placed  in  an  Erlen- 
meyer  flask,  and  boiled  with  chloroform  on  a  water- bath.  It  was  nec- 
essary to  put  a  cork  fitted  with  a  small  condenser  or  a  long  glass  tube 
on  the  flask,  so  as  to  avoid  loss  of  chloroform.  After  heating  to  boil- 
ing for  half  an  hour,  the  contents  of  the  flask  were  allowed  to  cool, 
and  then  filtered  into  a  tared  flask.  On  distilling  off  all  the  chloro- 
form, and  heating  the  flask  for  half  an  hour  on  a  water-bath  at  ioo° 
C,  it  was  weighed  and  the  amount  of  caffeine  obtained  thus  deter- 
mined. The  caffeine  obtained  was  not  quite  white,  but  had  a  light 
brownish  tint. 
Method  II. — The  powdered  nuts  were  boiled  in  an  Erlenmeyer  flask 
with  an  inverted  condenser  or  long  tube  attached,  on  a  water-bath 
for  three  hours,  with  a  mixture  of  two  parts  of  water  and  one  part  of 
alcohol  by  volume.  Part  of  the  starch  will,  of  course,  be  hydro- 
lyzed,  but  not  sufficient  to  render  filtration  impossible.  The  con- 
tents of  the  flask,  after  cooling, were  filtered,  and  the  filtrate  evaporated 
nearly  to  dryness  on  a  water-bath  in  a  porcelain  dish.  When  nearly 
to  dryness,  calcined  magnesia  and  sand  were  added,  and  all  evaporated 
to  complete  dryness,  stirring  carefully  and  frequently.    The  residue 
ZH.  and  S.,  "  Des  Kolas  Africains,"  Paris,  1884,  Masson. 
