^"^'octiyiSs.^^'^  }     Impurities  in  Commercial  Alcohols.  509> 
In  all  such  testing,  the  urine  must  be  free  from  proteid,  as  the  con- 
cluding portions  of  the  paper  show  that  furfural dehyde  is  one  decom- 
position product  of  proteid;  this  fact  is  considered  to  be  the  first 
well-established  chemical  relationship  between  proteids  and  carbohy- 
drates, although  the  physiological  connection  between  the  two  classes 
of  substances  has  long  been  recognized  (Bernard,  Seegen,  &c.).  It  also 
affords  an  explanation  of  the  color  reactions  which  are  caused  by  treat- 
ing proteid  with  acid,  such,  for  example,  as  the  Adamkiewicz  reaction. 
The  amido-acids  which  result  from  the  decomposition  of  proteids  yield 
no  furfuraldehyde. 
DETECTION  OF  IMPURITIES  IN  COMMERCIAL  AL- 
COHOLS.i 
By  L.  Godefeoy. 
6  or  7  cc.  of  the  alcohol  is  agitated  with  one  drop  of  perfectly  pure 
benzene,  mixed  with  6  or  7  cc.  of  pure  sulphuric  acid  of  66°,  and 
again  agitated.  If  reducing  "  head  products  are  present,  the  liquid 
immediately  acquires  a  coloration  which  gradually  darkens  for  a  few 
minutes,  and  varies  from  pale  brownish-yellow  to  black.  Pure  ethyl 
alcohol  gives  no  immediate  coloration,  but  after  8  or  10  minutes  the 
liquid  acquires  a  slight  rose  tint.  This  test  will  detect  1  cc.  of  "  head  " 
products  in  1000  litres  of  alcohol,  or  1  part  per  1,000,000.  The 
quantity  may  be  estimated  by  comparing  the  color  with  that  given  by 
alcohol  containing  a  known  quantity  of  acetaldehyde,  and  expressing 
the  results  in  terms  of  the  latter. 
If  the  liquid  remains  colorless  after  several  minutes,  no  "  head  ^' 
products  are  present.  In  order  to  detect  "  tail  products,  care  is 
taken  that  the  acid  and  alcohol  are  thoroughly  mixed,  in  order  to 
avoid  polymerisation,  and  the  liquid  is  boiled  for  a  short  time  and 
then  allowed  to  remain  for  several  minutes.  Under  these  conditions, 
pure  ethyl  alcohol  gives  an  ochre-yellow  coloration  ;  but  in  presence  of 
"  tail  products  the  liquid  acquires  a  brown  color  with  a  green  fluores- 
cence, the  depth  of  tint  increasing  with  the  quantity  of  impurity.  This 
test  is  not  so  sensitive  as  the  first,  but  will  detect  1  part  in  100,000. 
Neither  of  these  tests  is  directly  applicable  to  wines,  spirits,  etc. 
These  liquids  should  be  distilled,  and  the  first  fraction  tested  for 
"  head    products,  and  the  last  fraction  for  "  tail  products. 
1  Compt.  rend,  cvi.,  1018—1020 ;  Jour.  Chem.  Soc,  1888,  p.  875. 
