Sulphuric  Acid , in  the  Rain-Water  collected  at  Rothamsted.  323 
yielded  an  average  of  0’254  of  nitrogen  as  ammonia  per  million 
of  rain.  Determinations  in  fresh  monthly  mixtures,  extending 
over  two  years,  have  given  an  average  of  0‘316  ; and  determina- 
tions in  fifty  monthly  mixtures,  rather  more  than  half  con- 
sisting of  old  collections,  an  average  of  0'340  of  nitrogen  as 
ammonia  per  million  of  rain. 
It  is  quite  evident  that  the  results  now  obtained,  while  they 
agree  well  with  those  given  by  Frankland’s  analyses,  are  much 
lower  than  those  furnished  by  the  earlier  analyses  made  at 
Rothamsted,  or  by  Professor  Way.  Can  there  be  any  error  in 
the  recent  determinations  tending  to  deficiency  ? Professor  J.  W. 
Mallet  has  lately  called  attention  to  the  necessity  of  supplying 
the  condenser  with  ice-cold  water  when  distilling  ammonia,  if 
loss  is  to  be  avoided.  Thus,  when  distilling  1 milligram  of 
ammonia  from  500  cubic  centimeters  of  water  in  the  ordinary 
way,  he  found  a loss  of  14'5  per  cent.,  and  when  distilling 
0-5  milligram,  a loss  of  7 per  cent.,  the  temperature  of  the 
distillates  being  27°  to  28°  C.  The  temperature  of  our  dis- 
tillates may  in  summer  time  be  as  high  as  the  point  just 
named  ; there  seem,  however,  no  grounds  for  supposing  that  an 
appreciable  loss  of  ammonia  has  occurred.  In  the  first  [place, 
the  quantity  of  ammonia  in  the  retort  has  not  exceeded  0T 
milligram,  and  has  therefore  been  much  below  that  mentioned 
by  Professor  Mallet.  Further,  in  a determination  made  in  June 
with  a supply  of  iced  water,  the  distillate  having  a temperature 
of  11°  C.,  there  was  no  appreciable  difference  in  the  amount  of 
ammonia  found  from  that  obtained  in  a determination  made  in 
the  ordinary  way  with  a distillate  at  25°  C. 
The  evidence  thus  points  to  the  conclusion  that  the  early 
determinations  of  ammonia  in  rain  erred  on  the  side  of  excess. 
The  method  employed  at  Rothamsted  in  1853—4  is  given  in 
detail  in  the  Report  of  the  British  Association  for  1854.  The 
method  used  by  Way  in  1855—6  is  described  in  the  volume  of 
this  ‘ Journal  ’ for  1856,  p.  159.  To  enter  upon  a full  discussion 
of  these  methods  would  lead  us  into  chemical  details  unsuited 
to  the  present  paper.  It  may  be  stated,  however,  that  in  both 
cases  the  ammonia  was  determined  alkalimetrically,  a method 
admittedly  less  delicate  than  the  more  modern  Nessler  process. 
Again,  it  has  been  stated  by  Schloesing,  that  by  the  use  of  a 
glass  condenser  the  distillate  may  acquire  alkali  from  the  glass, 
and  it  is  obvious  that,  so  far  as  this  may  have  happened,  the 
effect  would  be  to  give  high  results  by  the  alkalimetric  method. 
There  is  also  the  further  question  whether  a larger  proportion 
of  the  nitrogenous  organic  matter  was  not  converted  into 
ammonia,  and  determined  as  such. 
Assuming,  then,  the  new  determinations  to  be  substantially 
