540 
HISTORY  OP  PETROLEUM  OR  ROCK  OIL. 
In  the  above  table  we  see  the  transition  from  peat  and  brown 
coal  to  lignite,  and  thence  to  bituminous  coal.  Professor  Johns- 
ton, from  his  experiments  on  various  coals,  including  cannel 
from  Wigan,  splint  coal  from  Workington,  and  caking  coal  from 
Newcastle,  deduced  the  composition  given  in  14,  in  which  with 
H9  the  oxygen  varies  from  two  to  four  equivalents.  It  will 
be  seen  from  a  comparison  of  the  infusible  Albert  coal  with  the 
bitumens  16,  17,  and  18,  how  gradual  is  the  transition  to  the 
true  petroleums  and  naphthas,  from  which  oxygen  is  absent. 
The  asphalts  also,  as  will  be  observed,  differ  very  much  in  their 
composition,  and  though  generally  much  richer  in  hydrogen 
than  the  bituminous  coals,  the  variety  from  Naples  (17)  which 
is  completely  fusible  at  140°  C,  contains  less  hydrogen  and 
more  oxygen  than  the  Albert  coal  analysed  by  Wetherell  ; 
while  the  idrialine  or  bitumen  found  with  the  mercury  ores  of 
Idria,  approaches  very  nearly  in  composition  to  the  bituminous 
coals,  11,  12,  and  13,  with  which  many  asphalts  may  be  said  to 
be  isomeric.  It  is,  however,  probable  that  those  oxygenised 
bitumens,  unlike  the  coals,  are  products  of  the  oxidation  of 
naptha  or  petroleum,  by  a  process  similar  to  that  by  which 
resins  are  derived  from  vegetable  hydrocarbons.  These  for- 
mulae must  be  taken  as  representing  not  the  true  equivalents, 
but  only  the  proportions  of  the  elements  in  the  bodies  in  ques- 
tion, which  are  in  most  cases  mixtures  of  various  substances. 
This  is  especially  true  of  naphtha,  which  may  be  taken  as  the 
representative  of  pure  unoxidised  petroleum,  and  which  is  sepa- 
rated by  distillation  into  oils  of  very  different  boiling  points. 
The  late  analyses  by  Uelsrnann  of  the  rectified  rock  oil  from 
Sehnde,  near  Hanover,  gave  the  formula  C18H20  and,  according 
to  De  la  Rue  and  Muller,  the  greater  part  of  the  Rangoon  pe- 
troleum consists  of  hydrocarbons,  in  which  the  number  of  equiv- 
alents of  hydrogen  is  a  little  greater  than  the  carbon  ;  one  gave 
C26H28.  Associated  with  these  are,  however,  portions  of  bodies 
containing  a  less  proportion  of  hydrogen,  so  that  we  may  con- 
ceive the  mean  composition  of  petroleum  to  be  represented,  as 
in  the  preceding  table,  by  equal  equivalents  of  hydrogen  and 
carbon  ;  many  forms  of  solid  bitumen  also,  as  ozokerite  and 
hatchetine,  have  the  same  general  composition. 
By  referring  to  what  has  been  said  above,  it  will  be  seen  that 
