396 
IMPROVED  METHOD  OF  ULTIMATE  ANALYSIS. 
heat,  namely,  the  center,  is  occupied  by  a  bad  conductor,  to  the 
exclusion  of  any  portion  of  the  gases  to  be  burned. 
The  oxidized  copper  surfaces  fill  the  tube  more  perfectly  than 
the  oxide  of  copper  by  the  usual  method,  and  are  so  easily  per- 
meable that  the  gases  more  readily  and  more  slowly  percolate 
the  roll  under  circumstances  more  favorable  to  perfect  com- 
bustion. Then  as  portions  of  these  surfaces  are  reduced  during 
combustion,  the  reduced  metal  does  not  protect  or  shield  the  ox- 
ide not  yet  reduced,  from  contact  with  the  gases,  but  from  being 
distributed  with  or  behind  the  line  of  deoxidation  is  in  the  most 
active  condition,  and  the  best  position  for  decomposing  the  oxides 
of  nitrogen  that  may  form.  Hence,  probably,  the  circumstances, 
that  the  reduced  roll  placed  in  the  anterior  end  of  the  tube  com- 
monly remains  quite  free  from  visible  oxidation  till  the  combustion 
is  nearly  finished,  and  may  only  be  necessary  as  a  precautionary 
measure. 
The  rolls,  by  the  readiness  with  which  they  conduct  and  dis- 
tribute the  heat,  admit  of  perfect  combustion  with  less  intense 
heating,  and  thus  measurably  preserve  the  tube  from  being  soft- 
ened, whilst  they  strengthen  and  keep  it  in  shape  if  it  softens, 
so  that  the  distribution  and  flow  of  the  gases  can  never  be  ob- 
structed or  rendered  irregular  by  this  not  uncommon  accident  to 
thin  and  good  tubes. 
They  effect  a  very  considerable  economy  in  time  and  expense, 
and  require  much  less  skill  and  experience  in  order  to  attain 
to  the  same,  or  a  greater  degree  of  accuracy,  than  by  the  common 
method. 
On  the  point  of  accuracy,  however,  it  is  necessary  to  state  that 
the  writer  has  never  used  the  rolls  for  furnishing  oxygen  in  any 
delicate  or  scientific  analyses,  neither  has  he  tested  the  method 
upon  any  substance  the  composition  of  which  was  accurately  known 
beforehand.  The  method  is  well  adapted  to  the  large  class  of  tech- 
nical analyses  daily  required,  and  is  suggested  that  it  may  be 
improved  in  its  design  and  application,  for  the  benefit  of  the 
daily  increasing  number  of  chemists  and  geologists,  upon  whom 
such  analyses  devolve.  At  the  same  time,  in  practice  thus  far, 
it  appears  well  adapted  to  make  analyses  of  the  greatest  ac- 
curacy. 
U.  S.  Naval  Laboratory,  New  York,  May  1856. 
