362  Some  Recent  Sugar  Reactions.  {^j^SJ'iwf^ 
From  the  above  results,  the  deduction  must  be  made  that  such 
urines,  presumably  glycuronic,  affect  the  safranine  as  well  as  the 
copper  test,  acting  like  a  small  amount  of  glucose. 
But  the  statement  that  any  discharge  of  color  by  a  sample  of 
urine  freed  from  albumin  may  be  set  down  as  being  due  to  the 
presence  of  sugar,  is  not  allowable  in  the  case  of  such  urines. 
A  suspicion  of  diabetes  can,  however,  be  entertained,  if  succes- 
sive additions  of  the  safranine  solution  are  decolorized  by  the 
urine. 
The  question  may  be  raised  in  these  cases  whether  the  reducing 
substance  is  a  glycuronic  compound  or  a  small  percentage  of  glu- 
cose. 
In  all  doubtful  cases,  such  urine  was  precipitated  by  basic  lead 
acetate,  the  excess  of  lead  removed  by  diluted  sulphuric  acid. 
If  glucose  was  present,  the  application  to  the  filtrate  of  either  the 
alkaline  copper  or  the  safranine  test  readily  revealed  its  pres- 
ence. 
It  may  also  be  questioned  if  small  quantities  of  glucose  are  not 
precipitated  and  removed  by  this  treatment. 
Various  trials,  made  by  the  writer,  showed  that  urine  containing 
o-i  per  cent,  of  glucose  can  be  subjected  to  the  lead  treatment  and 
still  reveal  the  presence  in  the  filtrate  to  the  copper  or  safranine 
tests. 
Further  dilutions,  however,  responded  but  feebly  or  entirely  in 
the  negative. 
Milk  sugar,  substituted  for  glucose  in  these  determinations,  gave 
the  same  results. 
For  the  quantitative  estimation  of  sugar  in  diabetic  urine,  the 
safranine  test  does  not  give  accurate  results. 
It  has  been  stated  that  the  discharge  of  color  of  each  2  c.c.  of  the 
safranine  solution,  I  part  to  1,000,  corresponds  roughly  to  01  per 
cent,  of  sugar. 
In  working  on  this  basis,  the  difficulty  lies  in  determining  the 
end  of  the  reaction,  when  the  mixture  is  required  to  assume  the 
original  red  color  before  boiling. 
In  urine  containing  a  large  sugar  percentage,  dilution  is  neces- 
sary. 
The  writer  has  followed  the  practice  of  retaining  a  portion  of  the 
original  mixture,  and  comparing  with  the  boiled  portion. 
