32  Conversion  of  Starch  into  Glucose.     {Am  ji°nu^887.arm 
(2)  That  an  agreement  between  the  results  afforded  by  the 
optical  and  copper  methods,  while  proving  the  conversion  to  be 
complete,  required  necessarily  to  tally  with  the  real  amount  of 
dextrose  known  to  be  represented  by  the  quantity  of  starch  opera- 
ted upon. 
My  procedure  was  as  follows : — Ten  grammes  of  starch  were 
heated  in  flask  provided  with  condenser,  with  190  cc.  of  the  dilute 
"  HC1 "  for  the  prescribed  time.  The  solution  was  next  cooled,  made 
up  to  200  cc,  and  filtered,  in  some  cases  decolorized.  It  was  subse- 
quently polarized,  using  the  half-shadow  instrument ;  a  portion  was 
also  neutralized,  suitably  diluted,  and  estimated  volumetrically  by 
"  copper." 
Repeated  experiments  were  made  in  both  cases,  and  the  averages 
taken. 
My  results  are  tabulated  below,  and  may  be  summed,  up  thus : — 
(1)  That  immersing  the  flask  containing  the  mixture  in  a  boiling- 
water  bath  was  equally  efficient  with  boiling  by  the  direct  applica- 
tion of  flame. 
(2)  That  in  all  cases  where  the  conversion  was  complete  my  re- 
sults were  seriously  below  the  truth  to  the  extent  of  from  four  to 
seven  per  cent. 
(3)  That  it  was  difficult,  if  not  impossible,  so  to  limit  the  time  of 
heating  as  to  prevent  the  destruction  of  some  of  the  newly-formed 
dextrose. 
(4)  That  prolonged  heating  resulted  in  regularly  diminislnng 
dextrose. 
In  conclusion,  while  regretting  my  inability  to  confirm  some  of  Mr. 
Heisch's  results,  I  am  of  opinion  that  the  main  issues  of  that  gentle- 
man's paper  continue  unaffected,  viz.,  that  the  percentage  of  starch 
in  ash  and  water-free  pepper  is  remarkably  constant,  and  that  the 
estimation  of  the  former  affords  valuable  aid  in  the  analysis  of  the 
latter. 
I  would  also  refer  to  some  researches  bearing  on  this  subject  by  F. 
Allihn,  published  in  the  Journal  of  the  Chemical  Society,  vol.  46  (ab- 
stracts), p.  721,  in  which  the  results  appear  to  confirm  mine. 
I  have  also  pleasure  in  acknowledging  the  assistance  of  my  friend 
and  pupil,  Mr.  John  Howard,  during  my  long  and  somewhat  tedious 
work. 
