AmAu°gu"'^77Tm  1        The  Tannic  Acid  °f  Guar  ana.  389 
The  chemical  investigations  of  Theodore  Martius,  in  1826,  proved 
that  it  was  not  a  gum-resin,  as  had  been  supposed,  but  a  mixture  of  the 
seeds  of  the  Paullinia  sorbilis  with  starch,  and  that  it  contained  a  crys- 
talline principle,  which  he  styled  guaranin,  under  the  supposition  that  it 
was  a  new  principle.  The  true  nature  of  guaranin  was  not  detected 
until  1840,  when  Berthemot  and  Dechastelus  submitted  it  to  an  ulti- 
mate analysis,  which  established  the  fact  that  it  was  identical  in  com- 
position to  cafFeina.  Their  very  valuable  researches  on  the  subject 
("Jour,  de  Pharm.,"  1840,  p.  578)  were,  however,  confined  to  the 
extraction  and  determination  of  the  characters  of  the  cafFeina,  and  they 
merely  refer  incidentally  to  the  gum,  starch,  tannin  and  oleaginous 
matters  which  exist  in  the  guarana.  The  "Journal  de  Pharmacie  et  de 
Chimie  "  (vol.  xxxix,  1861,  p.  291)  gives  an  extract  from  a  note  of  M. 
Fournier  in  regard  to  an  analysis  he  had  made  of  guarana  ;  but  as  it 
is  only  stated  that  he  had  found  gum,  starch,  a  fixed  green  oil,  three 
volatile  oils,  a  peculiar  principle  not  fully  determined,  tannate  of  caf- 
feina,  and  free  tannic  acid,  without  any  reference  to  the  relative  pro- 
portions of  the  different  substances,  or  the  methods  that  had  been 
employed  to  separate  them,  it  can  hardly  be  said  that  the  analysis  added 
anything  of  importance  to  what  was  already  known  in  regard  to  the 
chemical  composition  of  guarana.  It  is  to  be  regretted  that  M.  Four- 
nier did  not  communicate  his  methods  of  isolating  the  different  consti- 
tuents, as  an  investigation  of  the  physiological  action  of  the  component 
parts  in  a  separate  state  might  have  afforded  much  information  in  regard 
to  the  value  of  the  preparation  as  a  curative  agent.  As  far  as  I  have 
been  able  to  ascertain,  no  later  chemical  investigations  of  guarana  have 
been  made.  It  is  possible,  however,  that  Dr.  Peckolt,  who  has  exam- 
ined a  great  number  of  the  medicinal  plants  of  Brazil,  may  have  re-in- 
vestigated the  subject  and  given  the  results  in  his  work,  "Analyses  de 
Materia  Medica  Brazileira." 
During  an  investigation  lately  made,  with  a  view  to  the  extraction  of 
the  caffeina  from  guarana,  several  of  the  reactions  of  the  accompanying 
tannic  acid  were  so  strikingly  dissimilar  from  those  of  the  tannic  acids 
in  general,  that  I  determined  to  isolate  it  and  examine  it  carefully.  For 
this  purpose  a  quantity  of  guarana  in  fine  powder  was  treated  with 
successive  portions  of  boiling  alcohol  (75  per  cent.),  the  alcoholic  solu- 
tions filtered  when  cold,  and  the  alcohol  driven  off  on  a  water  bath. 
The  aqueous  solution  was  then  diluted  with  distilled  water,  and  a  slight 
