Instructions  for  Selecting  and  Sending  Samples  for  Analysis. 
GENERAL  RULES. — (1.)  A sample  taken  for  analysis  should  be  fairly  representative 
of  the  bulk  from  which  it  has  been  drawn. — (2.)  The  sample  should  reach  the  Analyst 
in  the  same  condition  that  it  was  in  at  the  time  when  drawn. 
When  Fertilisers  are  delivered  in  bags,  select  four  or  five  of  these  from  the  bulk, 
and  either  turn  them  out  on  a fioor  and  rapidly  mix  their  contents,  or  else  drive  a 
shovel  into  each  hag  and  draw  out  from  as  near  the  centre  as  possible  a couple  of 
shovelfuls  of  the  manure,  and  mix  these  quickly  on  a floor. 
Halve  the  heap  obtained  in  either  of  these  ways,  take  one  half  (rejecting  the  other) 
and  mix  again  rapidly,  flattening  down  with  the  shovel  any  lumps  that  appear. 
Repeat  this  operation  until  at  last  only  some  three  or  four  pounds  are  left. 
From  this  flU  three  tins,  holding  from  Jib.  to  llh.  each,  mark,  fasten  up  and  seal 
each  of  these.  Send  one  for  analysis,  and  retain  the  others  for  reference. 
Or, — the  manure  may  be  put  into  glass  bottles  provided  with  well-fitting  corks ; 
the  bottles  should  he  labelled  and  the  corks  sealed  down.  The  sample  sent  for 
analysis  can  he  packed  in  a wooden  box  and  sent  by  post  or  rail. 
WTien  manures  are  delivered  in  bulk,  portions  should  he  successively  drawn  from 
dWerent  parts  of  the  bulk,  the  heap  being  turned  over  now  and  again.  The  portions 
drawn  should  he  thoroughly  mixed,  subdivided,  and,  finally,  samples  should  be  taken 
as  before,  except  that  when  the  manure  is  coarse  and  biilky  it  is  advisable  to  send 
larger  samples  than  when  it  is  in  a finely  divided  condition. 
Linseed,  Cotton,  and  other  Feeding  Cakes.— If  a single  cake  betaken,  three  strips 
should  be  broken  off  right  across  the  cake,  and  from  the  middle  portion  of  it,  one  piece 
to  be  sent  for  analysis,  and  the  other  two  retained  for  reference.  Each  of  the  three 
pieces  should  be  marked,  wrapped  in  paper,  fastened  up,  and  sealed.  The  piece 
forwarded  for  analysis  can  he  sent  by  post  or  rail 
A more  satisfactory  plan  is  to  select  lour  to  six  cakes  from  different  parts  of  the 
delivery,  then  break  off  a piece  about  four  inches  wide  from  the  middle  of  each  cake, 
and  pass  these  pieces  through  a cake-breaker.  The  broken  cake  should  then  be  well 
mixed  and  three  samples  of  about  1 lb.  each  should  be  taken  and  kept  in  tins  or  bags, 
duly  marked,  fastened,  and  sealed  as  before.  One  of  these  lots  should  he  sent  for 
analysis,  the  remaining  two  being  kept  for  reference.  It  is  advisable  also  with  the 
broken  pieces  to  send  a small  strip  from  an  unbroken  cake.  ' 
Feeding  Meals,  Grain,  &c. — Handfuls  should  he  drawn  from  the  centre  of  half  a 
dozen  different  bags  of  the  delivery ; these  lots  should  then  be  well  mixed,  and  three 
J-lb.tins  or  hags  filled  from  the  heap,  each  being  marked,  fastened  up,  and  sealed.  One 
sample  is  to  he  forwarded  for  analysis  and  the  others  retained  for  reference. 
Soils. — Have  a wooden  box  made  6 inches  in  length  and  width,  and  from  9 to  12 
inches  deep,  according  to  the  depth  of  soil  and  subsoil  of  the  field.  Mark  out  in  the  field 
a space  of  about  12  inches  square ; dig  round  in  a slanting  direction  a trench,  so  as  to 
leave  undisturbed  a block  of  soU  and  its  subsoil  9 to  12  inches  deep ; trim  this  block  to 
make  it  to  fit  into  the  wooden  box,  invert  the  open  box  over  it,  press  down  firmly,  then 
pass  a spade  under  the  box  and  lift  it  up,  gently  turn  over  the  box,  nail  on  the  lid, 
and  send  by  rail.  The  soil  will  then  be  received  in  the  position  in  which  it  is  found  in 
the  field. 
In  the  case  of  very  light,  sandy,  and  porous  soils,  the  wooden  box  imay  be  at  once 
inverted  over  the  soil,  forced  down  by  pressure,  and  then  dug  out. 
Waters. — Samples  of  water  are  best  sent  in  glass-stoppered  Winchester  bottles, 
holding  half  a gallon.  One  such  bottle  is  suflicient  for  a single  sample.  Care  should 
be  taken  to  have  these  scrupulously  clean.  In  taking  a sample  of  water  for  analysis 
it  is  advisable  to  reject  the  first  portion  drawn  or  pumped,  so  as  to  obtain  a sample  of 
the  water  when  in  ordinary  flow.  The  bottle  should  be  rinsed  out  with  the  water  that 
is  to  he  analysed,  and  it  should  be  flUed  nearly  to  the  top.  The  stopper  should  be 
secured  with  string,  or  be  tied  over  with  linen  or  soft  leather.  The  sample  can  then  be 
sent  carefully  packed  either  in  a wooden  box  with  sawdust,  &c.,  or  in  a hamper  with 
straw. 
Milk.— A pint  bottle  should  be  sent  in  a wooden  box. 
GENERAL  INSTRUCTIONS.  Time  for  Taking  Samples.— AU  samples,  both  of 
fertilisers  and  feeding  stuffs,  should  be  taken  as  soon  after  their  delivery  as  possible, 
and  should  reach  the  Analyst  within  ten  days  after  delivery  of  the  article.  In  every 
case  it  is  advisable  that  the  Analyst’s  certificate  be  received  before  a fertiliser  is  sown 
or  a feeding  stuff  is  given  to  stock. 
Procedure  in  the  Event  of  the  Vendor  wishing  Fresh  Samples  to  be  Drawn. — 
Should  a purchaser  find  that  the  Analyst’s  certificate  shows  a fertiliser  or  feeding  stuff 
not  to  come  up  to  the  guarantee  given  him,  he  may  inform  the  vendor  of  the  result 
and  complain  accordingly.  He  should  then  send  to  the  vendor  one  of  the  two  samples 
which  he  has  kept  for  reference.  If,  however,  the  vendor  should  demand  that  a fresh 
sample  be  drawn,  the  purchaser  must  allow  this,  and  also  give  the  vendor  an  oppor- 
tunity of  being  present,  either  in  person  or  through  a representative  whom  he  may 
appoint.  In  that  case  three  samples  should  be  taken  in  the  presence  of  both  parties 
with  the  same  precautions  as  before  described,  each  of  which  shotdd  he  duly  packed 
up,  labelled  and  sealed  by  both  parties.  One  of  these  is  to  be  given  to  the  vendor,  one 
is  to  be  sent  to  the  Analyst,  and  the  third  is  to  be  kept  by  the  purchaser  for  reference 
or  future  analysis  if  necessary. 
Ill 
