472  Action  of  Water  on  Glass.  {Ams^uS^vn"' 
before,  except  that  it  is  unnecessary  to  ascertain  the  oxygen  lost  by 
ignition.  The  result  presents  about  two-thirds  of  the  total  volatile 
organic  matter  present ;  further  small  quantities  can  be  recovered 
from  the  distillate  by  repeating  the  process. 
By  the  employment  of  sodium  carbonate  the  whole  of  the  com- 
pounds of  calcium,  magnesium,  and  iron  are  precipitated,  and  any 
combined  ammonia  in  the  water  is  volatilized.  There  only  remains 
sodium  chloride,  alkali  nitrates,  and  uncombined  silica  to  interfere 
with  the  loss  on  ignition  being  accepted  as  a  measure  of  the  organic 
matter  present,  and  neither  of  these  compounds  being  present  in 
estimating  volatile  organic  matter,  the  results  may  be  accepted  as 
free  from  objection  on  that  account.  As  regards  sodium  chloride, 
the  burning  of  the  organic  matter  is  so  rapid  (a  few  seconds  suffices), 
and  the  temperature  so  low,  that  none  is  volatilized,  or  if  a  little  is 
lost  through  excessive  heating,  the  loss  can  be  ascertained  and  due 
correction  made.  As  regards  alkali  nitrates,  provision  is  made  in 
the  process  for  ascertaining  and  correcting  for  the  loss  of  oxygen  by 
reduction  of  nitrates,  but  it  is  seldom  of  any  great  importance,  and 
has  never  exceeded,  in  ordinary  drinking  waters,  the  equivalent  of 
O  07  grain  per  gallon.  The  presence  of  nitrates  assists  the  burning 
of  the  organic  matter  very  materially,  and  in  the  case  of  very  foul 
waters,  such  as  sewage  effluents  or  seriously  polluted  waters,  which 
rarely  contain  any,  the  author  finds  it  advisable  to  add  a  drop  or  two 
of  solution  of  potassium  nitrate  before  the  final  evaporation.  With 
regard  to  the  uncombined  silica,  the  author  has  never  found  it  pres- 
ent, and  if  it  should  be,  he  does  not  think  heat  required  to  burn  oft 
the  organic  matter  is  sufficently  great  to  cause  it  to  decompose  the 
sodium  carbonate. 
THE  ACTION  OF  WATER  UPON  GLASS. 
By  F.  Mylius  and  F.  Forrester. 
The  authors  summarize  their  rough  researches  in  the  following 
propositions,  which  they  consider  proved  by  their  own  observations 
and  those  of  Pfeiffer  and  Kohlrausch  : 
(1)  The  solution  of  glass  in  water  depends  on  a  decomposition 
in  which,  in  the  first  place,  free  alkali  appears. 
(2)  The  silica  of  the  glass  is  secondarily  dissolved  by  the  free 
alkali. 
