256       Crystalline  P)-inciple  of  Barbadoes  Aloes.  {AMjuneT, 
Pharm. 
1872, 
warm  rectified  spirit.  The  tufts  of  bright  yellow  prisms  which  were 
deposited  in  a  few  hours  were  collected  and  dried  by  exposure  to  dry 
air.  They  bear,  without  change  of  color  or  general  appearance,  a 
temperature  of  120°  C,  and  even  much  higher.  At  120°  they  lost 
weight  in  one  experiment  to  the  extent  of  10*86,  in  another  10*04  per 
cent. 
•237  gram  gave  by  boiling  with  nitric  acid  and  nitrate  of  silver 
•216  of  chloride  of  silver,  corresponding  to  22*52  per  cent,  of  chlo- 
rine. 
The  formula  C17B15C13073H20  requires  10*99  per  cent  of  water,  and 
21*66  per  cent,  of  chlorine. 
The  proportion  of  chlorine  found  being  thus  a  little  too  high,  the 
substance  was  recrystallized  and  carefully  washed.  This  time  it  was 
dried  at  120°  previous  to  analysis. 
I.  -247  gram  gave  *251  chloride  of  silver. 
II.  -1835  gram,  by  combustion  with  a  mixture  of  lead  chromate  and 
potassic  dichromate,  -062  H20  and  -304  C02. 
Theory.  Experiment. 
f  A  N  f  A  --\ 
I.  II. 
C17    204  46*62  —  45*17 
H15    15  3-42  —  3*70 
Cl3    106*5  24-34  25-13  — 
07.  .  .  112  —  — 
Again,  therefore,  the  chlorine  is  rather  above,  and  the  carbon  be- 
low the  theoretical  numbers,  although  they  are  sufficiently  near  to 
leave  no  doubt  as  to  the  identity  of  the  body.  I  think  it  probable, 
therefore,  that  notwithstanding  that  the  crystals  are  to  all  appear- 
ance clean,  and  when  dissolved  in  water  give  no  trace  of  turbidity 
with  nitrate  of  silver,  they  are  contaminated  with  a  small  quantity 
of  another  similar  body,  containing  a  higher  percentage  of  chlorine. 
This  chloraloi'n  is  more  soluble  in  water  than  the  corresponding 
compound  containing  bromine,  and  differs  from  the  original  aloin  in 
its  comparative  stability.  Thus,  although  very  soluble  in  aqueous 
ammonia,  it  will  crystallize  out  but  little  altered  when  the  ammonia 
is  allowed  to  evaporate,  and  it  may  be  dissolved  in  ordinary  nitric 
acid  (sp.  gr.  1*37),  without  change  of  color. 
The  aloin  from  which  this  body  is  derived,  when  acted  upon  by 
nitric  acid,  yields,  besides  oxalic  and  picric  acids,  rather  more  than 
