378  The  Volatility  of  Glycerin.  { *m^?7h8arm' 
containing  20  per  cent,  by  weight  of  glycerin,  only  16  per  cent,  of 
residue  was  obtained,  showing  that  20  per  cent,  of  the  glycerin  in  the 
mixture  had  been  carried  off  with  the  vapor  of  the  alcohol  and  water. 
This  result  seemed  so  at  variance  with  the  recognized  non-volatile 
character  of  glycerin  that  the  following  experiments  were  made  to  test 
it,  the  results  of  which  do  not  at  all  accord  with  Dr.  Squibb's  theory. 
A  mixture  of  glycerin,  water  and  alcohol  was  made  in  the  same 
proportions  used  by  Dr.  Squibb.  The  glycerin  was  taken  from  a  five 
pound  can  opened  expressly  for  the  purpose,  and  tested  by  exposing  a 
carefully  weighed  quantity  to  the  heat  of  a  water  bath,  for  30  minutes, 
•during  which  time  it  lost  no  appreciable  weight,  showing  that  it  lost  no 
water  at  this  temperature.  In  this,  as  in  the  other  experiments,  the 
liquid  was  weighed  in  a  tared  watch  glass  set  in  a  second  one,  the  two 
being  used  to  avoid  loss  from  "  creeping  "  of  the  liquid  over  the  edge 
of  the  glass.  They  were  then  set,  together,  in  a  large  porcelain  dish, 
heated  over  boiling  water  and  covered.  The  temperature  of  this  "  air 
bath  "  was  shown  to  be  1700  to  i8o°F.,  by  a  chemical  thermometer. 
Each  evaporation  was  carried  to  constant  weight.  The  glycerin 
mixture,  containing  20  per  cent,  of  glycerin,  was  kept  in  a  well 
stopped  bottle  to  avoid  concentrations  by  evaporation  of  the  alcohol. 
Four  experiments  gave  residues  weighing  in  per  cents,  as  follows  : 
20'  1,  20*5,  20*3,  20*6  ;  mean  result,  20*4  per  cent.  The  o!4  per 
cent,  excess  is  due  to  the  rapid  evaporation  of  the  alcohol  during 
mixing  and  weighing,  thus  concentrating  the  solution  of  glycerin. 
Two  experiments  were  made  with  a  mixture  of  strong  alcohol  and 
glycerin,  with  30*8  per  cent,  of  the  latter  ;  the  residue  weighed  31*0, 
31*5  ;  mean,  31*3  per  cent.     Excess  slightly  greater  than  before. 
Glycerin  167  per  cent.,  water  83*3  per  cent,  gave,  under  the  same 
treatment,  residues  weighing  16*5  and  16*5  per  cent. 
These  results -show  no  loss  of  glycerin  whatever;  the  results 
obtained  by  Dr.  Squibb  must  have  been  due  to  some  cause  other  than 
the  volatility  of  glycerin  in  the  presence  of  alcohol  and  atfatmospheric 
pressure. 
Nat.  W.  Lord,  E.M., 
Analytical  Chemist. 
