564  rochleder's  proximate  analysis. 
When  suflBcient  fluid  has  been  forced  in  this  way  into  the  funnel,  this  is 
closed  with  the  glass  rod,  and  the  glass  tube  opened.  The  portion  of  the 
fluid  iu  the  funnel  required  for  examination  may  be  removed  by  means  of 
a  pipette. 
I  must  also  mention  a  small  arrangement  which  renders  easy  the  dis- 
placement of  the  hydrogen  gas  by  the  carbonic  acid  gas,  without  permit- 
ting the  entrance  of  air  into  the  apparatus.  This  arrangement  consists  of  a 
spur  made  of  vulcanized  caoutchouc.  This  forked 
7  tube  at  a  is  to  be  connected  with  the  glass  tube,  bent 
U  n — 0 — ^  at  a  right  angle,  which  conducts  the  hydrogen  gas  into 
v:^)  C  the  flask.  The  forked  tube  is  attached  at  h  to  the 
apparatus  for  evolving  hydrogen,  and  is  at  c,  con- 
nected with  a  gasometer  containing  carbonic  acid.  While  the  tube  at  h  is 
closed  with  a  clamp,  carbonic  acid  is  allowed  to  stream  out  of  the  gasometer 
until  all  the  air  is  driven  out  of  the  conducting  tube.  The  carbonic  acid 
passes  at  a  freely,  as  the  tube  at  h  is  closed  by  a  cock.  The  tube  at  c  is 
now  shut  with  a  clamp,  and  the  cock  of  the  gasometer  is  also  closed.  Then 
a  is  attached  to  the  tube  in  the  flask,  and  6  with  the  hydrogen  gas  ap- 
paratus, and  the  air  expelled  from  the  flask  by  the  hydrogen  gas.  After 
the  baryta  has  acted  sufficiently  long,  and  it  is  wished  to  conduct  carbonic 
acid  into  the  flask,  the  tube  is  closed  at  b,  which  in  the  mean  while  was 
open,  and  the  tube  is  opened  at  c,  and  carbonic  acid  allowed  to  pass  in 
from  the  gasometer. 
When  a  clear  fluid  is  obtained  by  the  treatment  with  baryta,  it  is  ascer- 
tained whether,  by  the  addition  of  alcohol,  a  baryta  compound  is  precipi- 
tated, while  the  carbohydrate  remains  dissolved  in  the  spirituous  fluid. 
Often  in  this  way  a  separation  can  be  accomplished. 
If  a  separation  is  not  efiected  in  this  way,  carbonic  acid  is  conducted 
into  the  fluid  until  all  excess  of  baryta  is  saturated  with  carbonic  acid.  In 
many  cases  the  baryta  compound  of  the  decomposition  product  is  thereby 
decomposed.  The  contents  of  the  flask  are  heated  now  to  the  boiling  point, 
while  hydrogen  gas  instead  of  carbonic  acid  is  passed  through  the  fluid. 
By  boiling,  the  dissolved  bi-carbonate  of  baryta  is  converted  into  the  car- 
bonate. The  contents  of  the  flask  are  then  filtered.  On  the  filter,  car- 
bonate of  baryta  and  a  decomposition  product  remain  when  the  latter  is  in- 
soluble in  water,  and  was  dissolved  only  as  a  baryta  salt  in  the  water.  The 
carbohydrate  is  contained  in  the  fluid  frequently  as  a  baryta  compound  (con- 
sequently not  precipitated  from  the  baryta  by  carbonic  acid),  either  alone  or 
together  with  the  second  decomposition  product,  which  may  be  dissolved 
either  free  from  baryta  or  as  a  baryta  salt.  When  the  substance,  as  ononine 
or  populine,  &c.,  is  decomposed  by  baryta  into  a  baryta  salt  and  a  copulated 
carbohydrate,  a  copulated  carbohydrate  and  not  a  separated  carbohydrate 
is  obtained  in  the  filtrate.  On  this  account,  we  have  always  to  ascertain 
whether  a  carbohydrate,  or  such  a  copulated  compound,  is  obtained  by  the 
