Mr.  J.  A.  Wanklyn  on  Water -Analysis  and  Water,        527 
quired  to  supply  the  whole  of  its  nitrogen  with  sufficient  hydro- 
gen to  form  ammonia.  By  assimilation  of  water  it  yields 
ammonia  and  carbonic  acid ;  but  by  assimilation  of  oxygen  it 
could  not  possibly  yield  up  more  than  half  of  its  nitrogen  in  the 
shape  of  ammonia,  and,  as  I  have  said,  has  been  found  experi- 
mentally to  yield  up  absolutely  none  as  ammonia. 
In  dealing  with  urea  there  is  no  difficulty  in  making  it  take 
up  the  elements  of  water  and  evolve  all  its  nitrogen  in  the  form 
of  ammonia.  This  may  be  very  conveniently  accomplished  by 
maintaining  it  for  a  short  time  at  a  temperature  of  150°  C.  in 
contact  with  caustic  potash.  Now  this  peculiar  character — to 
yield  up  nitrogen  in  the  form  of  ammonia  on  heating  to  150°  C. 
with  caustic  alkali,  and  not  to  yield  up  nitrogen  as  ammonia 
when  oxidized — appears  to  pertain  exclusively  to  urea.  Not 
only  does/ree  urea  exhibit  this  character,  but  coupled  urea  shows 
it  also;  thus  creatine,  which  is  urea  coupled  with  sarcosine, 
yields  only  the  ammonia  arising  from  the  sarcosine  when  it  is 
oxidized  by  alkaline  permanganate,  yielding  the  ammonia  from 
its  urea  only  on  treatment  with  alkali. 
Reasoning  from  this  behaviour  of  coupled  urea,  I  have  been 
led  to  the  interesting  conclusion  that  albumen,  caseine,  and 
fibrine  are  coupled  ureas,  but  that  gelatine  is  not  a  coupled 
urea. 
Although  all  nitrogenous  organic  compounds,  with  the  ex- 
ception of  nitro- com  pounds,  ferrocyannides,  and  urea,  yield  am- 
monia to  alkaline  permanganate,  yet  many  nitrogenous  organic 
compounds  do  not  yield  their  total  nitrogen  in  the  form  of  am- 
monia. The  natural  alkaloids,  as  for  example  morphia,  often 
give  up  half  of  the  nitrogen  as  ammonia,  as  is  also  the  case  with 
naphthylamine  and  toluidine. 
From  100  grms.  of  albuminous  substances  about  10  grms. 
are  obtained  by  means  of  alkaline  permanganate. 
"When  my  colleagues  and  I  first  directed  our  attention  to  the 
subject  of  water,  we  endeavoured  to  provide  a  test  which  should 
not  fail  to  recognize  germs  by  their  chemical  characters — a  test 
which  could  not  fail  to  distinguish  between  water  that  was  pure 
and  water  that  was  charged  with  germs.  With  this  object  in 
view,  we  selected  egg-albumen  as  the  representative  of  germs, 
and  then  sought  for  the  most  accurate  and  convenient  method 
of  measuring  the  strength  of  excessively  dilute  solutions  of  it. 
How  accurately  the  ammonia  process  effects  this  object  may 
be  judged  of  by  the  following  examples,  which  I  quote  from  a 
paper  of  mine  published  by  the  Chemical  Society  in  the  year 
1867. 
