Am.  Jour.  Pharm. 
July,  1890. 
Caffeine  Tri-Iodide. 
349 
want  of  visible  connection  between  the  two  statements,  The  Chemist 
and  Druggist  commentator  of  May  3  gave  Tilden's  corrected 
formula  as  belonging  to  a  closely-related  but  different  compound  to 
that  denoted  by  the  provisional  and  incorrect  formula  given  on  the 
previous  page ;  and  the  printing  of  the  latter  in  the  paragraph  of 
February  15  (copied  into  the  Pharmaceutische  Zeitung)  has  given 
an  opportunity  to  Dr.  Scholvien  to  add  to  the  confusion  by  proving 
at  considerable  length  that  no  body  with  such  a  formula  could 
possibly  exist. 
When  I  was  first  asked  by  Dr.  Mortimer  Granville  to  prepare  a 
quantity  of  the  salt,  Tilden's  second  formula  was  overlooked  alto- 
gether, and  his  original  directions  were  implicitly  followed  ;  but  it 
soon  became  evident  that  the  rapidity  with  which  the  crystals 
formed  was  directly  proportional  to  the  liberation  of  iodine  from  the 
excess  of  hydriodic  acid  present,  an  old  and  decomposed  acid  being 
far  better  for  the  purpose  than  a  freshly-prepared  and  colorless  acid. 
The  next  experiment  was  to  use  an  acid  containing  a  quantity  of 
added  iodine,  when  a  crop  of  crystals  was  quickly  obtained,  more 
than  double  the  weight  of  the  added  iodine,  the  resulting  inference 
being  that  two-thirds  of  the  total  iodine  was  practically  in  the 
"  free  "  state,  and  the  compound,  instead  of  being  a  tri-iodide,  was 
really  (as  indicated  by  Tilden)  the  normal  hydriodide  (hydriodate) 
of  caffeine,  with  two  loosely-combined  atoms  of  iodine,  whence  its 
usefulness  in  therapeutics.  Analytical  results  obtained  by  my 
assistant,  Mr.  Robert  Thomson,  completely  confirm  this  view,  and 
corroborate  those  obtained  by  Professor  Tilden,  except  in  regard  to 
the  water  of  crystallization,  which  we  think  in  no  case  exceeds 
iH20  in  the  molecular  compound. 
The  following  are  the  percentage  results  obtained  from  some 
well-formed  crystals : 
Experiment.    C8Hi0N4O2.HI.  I2.HaO 
Caffeine  (anhydrous),   32*4  32*66 
Hydriodic  acid,   21*59  21*55 
Iodine,   42*67  4276 
Water,   3*03 
99*76  100*00 
Total  iodine,   64*09  64*14 
The  constitution  of  the  body  may  be  shown  by  either  of  the  three 
following  methods:  (1)  By  treatment  with  cold  water  iodine  is 
liberated  to  the  extent  of  42  per  cent,  (two-thirds  of  the  whole), 
