mS  Furfuraldehyde  Color  Reaction.        {^"^  ocu  ig^"""- 
F.  Several  of  the  substances  examined,  for  instance  a-naphthol,  give 
the  test  with  greater  delicacy  than  is  the  case  with  cholic  acid. 
The  fact  that  the  test  is  given  by  coniferin  gave  rise  to  a  number  of 
experiments  on  the  color  reactions  produced  by  strips  of  the  wood  of 
the  pine  and  other  trees. 
In  conclusion,  the  delicacy  of  the  reaction  as  applied  to  the  detec- 
tion of  bile  is  discussed  ;  using  a  1  per  cent,  solution  of  furfuraldehyde 
in  the  manner  already  detailed,  it  is  found  possible  to  obtain  a  color 
with  0*000033  gram  of  cholic  acid  ;  a  quantity  of  0*00005  gram  gives 
a  color  sufficiently  intense  to  show  its  spectroscopic  bands ;  in  order 
to  obtain  evidence  of  the  presence  of  bile  acids  in  urine,  it  is  generally 
necessary  first  to  isolate  them  from  that  secretion.  Normal  urine  does 
not  contain  bile  acids. 
The  question  whether  normal  urine  contains  carbohydrates  has 
been  one  on  which  a  large  amount  of  work  has  been  done,  but 
has  never  been  satisfactorily  settled.  The  most  recent  of  these  ob- 
servations are  those  of  Landwehr  (1886),  who  states  that  animal 
gum  may  be  present  in  the  urine,  and  the  author's  own  researches 
on  humous  substances  (1887).  The  fact  that  carbohydrates  yield 
furfuraldehyde  on  treatment  with  acid,  which  can  be  identified  by 
means  of  the  characteristic  color  given  with  bile  acid,  a-naphthol,  and 
many  other  substances,  has  led  to  the  present  reinvestigation  of  the 
question.  The  special  method  adopted  is  that  of  H.  Schiff  {Ber.,  xx., 
540),  in  which  strips  of  filter-paper  are  dipped  in  a  mixture  of 
xylidine,  glacial  acetic  acid,  and  alcohol,  then  dried.  The  substance 
suspected  to  contain  carbohydrate  is  heated  in  a  tube  with  sulphuric 
acid  ;  the  fumes,  which  contain  furfuraldehyde,  strike  a  red  color  with 
the  strips  of  paper  placed  at  the  mouth  of  the  tube.  Using  this 
method  with  quite  small  quantities  (a  few  drops)  of  normal  urine,  the 
color  never  fails  to  appear.  The  conclusion  is  drawn  that  normal 
urine  contains  carbohydrates,  although  of  what  kind  is  doubtful.  A 
reaction  described  by  Molisch  (Amer.  Joue.  Pharm.,  1887,  p.  74), 
in  which  either  thymol  or  a-naphthol  and  sulphuric  acid  give  a  red 
coloration,  may  also  be  used  with  the  same  result. 
In  cases  of  glycosuria  these  reactions  occur  more  readily,  and  by  a 
minutely  described  process  of  appropriate  dilution  of  the  urine,  an 
approximate  quantitative  result  may  be  obtained,  at  least  sufficiently 
near  to  enable  one  to  say  whether  the  secretion  contains  more  than  a 
normal  amount  of  carbohydrate. 
