rochleder's  proximate  analysis. 
183 
the  separation  is  accomplished  by  fractional  precipitation  with  a  suitable 
metallic  salt.  In  this  case  it  is  necessary  to  take  care  that  no  excess  of 
ammonia  is  present.  Solutions  of  acetate  of  copper,  acetate  of  lead,  or  chlo- 
ride of  calcium  are  generally  the  most  suitable  salts  for  fractional  precipi- 
tation. By  a  preliminary  experiment  it  may  be  learnt  which  is  the  best 
to  employ. 
By  the  prescribed  treatment  of  the  extract  whose  preparation  was  ex- 
plained in  page  13,  a  number  of  substances  have  been  prepared,  not  all  of 
which  can  be  regarded  as  isolated  constituents,  although  such  may  be 
the  case  with  many  of  them.  Substances  which  have  been  separated 
from  others,  by  the  methods  described  in  the  2nd  section,  will  frequently 
be  obtained  in  a  state  of  contamination  with  small  quantities  of  other  con- 
stituents. But  as  we  have  prepared  these  bodies  from  different  extracts  of 
the  same  material,  according  to  different  methods,  generally  twice  or  oftener, 
their  peculiarities  may  still  be  recognized,  even  when  the  substance  has  in 
no  case  been  obtained  quite  pure,  because  the  impurities  from  the  different 
methods  of  preparation  must  be  alike.  After  following  the  directions  given, 
and  from  the  evidence  thus  obtained,  it  is  often  impossible  to  know  whether 
we  had  to  do  with  one  individual  body  or  several  bodies.  Thus,  by  the  frac- 
tional precipitation  or  solution  of  precipitates,  it  must  often  be  felt  undecided 
whether  the  individual  solutions  or  precipitations  contain  the  same  or  differ- 
ent bodies  ;  and  when  the  presence  of  several  bodies  has  been  detected  in 
the  different  solutions  or  precipitates,  whether  these  individual  portions  are 
to  be  fractionized  still  further  or  not  by  a  repetition  of  this  kind  of  treat- 
ment, in  accordance  with  our  failure  or  success  in  isolating  a  body  from 
one  of  these  portions.  The  following  section  will  treat  of  the  manner  in 
which  we  can  obtain  information  on  these  points,  by  the  aid  of  reagents,  and 
of  the  kind  of  reagents  which  in  the  majority  of  cases  will  answer  this 
purpose. 
Section  III. 
Reagents  and  their  application. 
It  has  often  been  stated  in  the  second  section,  that  the  further  examina- 
tion of  a  body,  by  the  observation  of  its  behaviour  to  reagents,  is  required, 
not  merely  to  decide  by  its  reactions  the  identity  of  a  body  with  one  already 
known,  but  chiefly  to  prevent  constituents  being  overlooked  by  the  aid  of 
their  reactions. 
If  we  have  obtained  a  substance  by  fractional  solution  of  a  precipitate, 
or  by  fractional  precipitation,  of  which  we  could  not  previously  know 
whether  it  is  only  one  body  or  a  mixture  of  two,  three,  or  several  bodies, 
and  have  divided  it  into  several  portions,  and  allowed  a  number  of  reagents 
to  act  on  each  portion,  we  shall  be  able  to  ascertain  whether  we  have  to  do 
with  one  or  several  bodies  differing  from  one  another. 
I  am  certainly  far  from  believing  that  only  the  reagents-described  in  the 
following  pages  are  useful  means  for  the  attainment  of  this  object,  but  I 
