ill  certain  parishes  is  due  to  a  comparatively  limited  number  of  what 
may  be  termed  plague  spots,  where  malaria  is  rife ;  and  this 
information  would  facilitate  very  much  the  application  of  the 
anti-malarial  measures  with  which  I  shall  deal  presently.  But  I  hope 
to  arrive  at  this  by  another  method. 
Before  considering  the  table,  it  is  necessary  for  me  to  point  out 
one  or  two  sources  of  error.  In  the  first  place,  no  census  has  been 
taken  since  1891,  owing  to  financial  reasons,  and,  as  the  Registrar- 
General  rightly  points  out,  the  figures  relating  to  population  must  be 
taken  as  approximate  and  subject  to  correction  at  the  next  census, 
which  I  trust  will  be  taken  in  1911.  I  ought  also  to  mention  that  the 
slight  discrepancy  which  may  be  observed  between  the  Registrar- 
General’s  figures  of  the  total  death-rate  and  those  given  by  myself 
are  due  to  the  fact  that  I  have  calculated  the  death-rate  on  the 
estimated  population  for  the  whole  year,  while  he  has  taken  the  mean 
population  calculated  to  the  middle  of  the  year. 
There  is  another  source  of  error  which  must  be  remembered  so 
far  as  the  malarial  death-rate  is  concerned,  and  that  is  that  a  very 
large  proportion  of  the  deaths,  more  especially  in  the  outlying  parts 
of  a  parish  are  uncertified,  and  consequently  a  number  of  deaths  which 
are  registered  as  being  due  to  ‘  fever  ’  may  be  unconnected  with 
malaria.  Any  illness  associated  with  a  rise  of  temperature,  no  matter 
what  the  cause,  is  invariably  spoken  of  by  the  uneducated  native  as 
‘  fever  ’  and  registered  accordingly. 
So  that  possibly  a  more  accurate  diagnosis  might  m  some  cases 
tend  to  reduce  the  malarial  death-rate.  On  the  other  hand  a  number 
of  complaints  are  complicated  by  malaria,  which  may  be  fatal,  though 
the  death  is  returned  under  the  original  head,  so  that  this  neutralises 
to  some  extent  the  other  factor.  And,  further,  the  error  is  a  constant 
one.  It  will  probably  occur  to  the  same  extent  all  over  the  Island  and 
in  different  years,  so  that  for  practical  purposes  it  will  afford  a  very 
fair  standard  for  comparison. 
The  table  is  one  of  extreme  interest,  and  from  it  I  have  calculated 
the  mean  death-rates  for  the  decennium  1898  to  1907,  which  for 
facility  of  reference  I  give  below,  but  I  would  recommend  the  serious 
study  of  the  large  table  to  those  who  are  interested  in  the  health  and 
sanitary  condition  of  their  respective  parishes. 
