128 
Life-History  of  a  Doctrine. 
Am.  Jonr.  Pharnj. 
March,  1903. 
number  of  atoms  or  groups.  Whether  that  which  holds  together 
the  atoms  that  make  up  the  core  is  what  in  similar  compounds 
manifests  itself  as  valence  remains  to  be  seen.  At  all  events,  if  the 
views  of  Werner  should  prove  to  be  correct,  we  shall  have  two  kinds 
of  valence  to  deal  with — that  of  the  inner  sphere  and  that  of  the 
outer  sphere,  or  that  of  the  core  and  that  of  the  shell.  In  a  recent 
article  Werner  extends  his  views  and  introduces  the  conception  of 
secondary  valences.  Thus  he  holds  that  in  ammonia  the  three 
valences  that  enable  the  nitrogen  auom  to  hold  the  three  hydrogen 
atoms  in  the  molecule  of  ammonia  are  different  from  that  which 
enables  ammonia  to  combine  with  a  molecule  of  hydrochloric  acid. 
The  former  he  calls  the  "  primary  valences  "  (Hauptvalenzen),  the 
latter  a  "  secondary  valence  "  (Nebenvalenz).  He  does  not  think 
that  the  two  differ  fundamentally.  So  Thiele  in  his  study  of  the 
phenomena  of  saturation  among  organic  compounds  is  obliged  to 
assume  the  existence  of  "  partial  valences  (Partialvalenzen),  and  the 
facts  described  by  him  are  singularly  in  accord  with  the  assumption. 
This  applies  up  to  the  present  only  to  the  compounds  of  carbon. 
Thiele's  "  partial  valences  "  are,  however,  not  to  be  confounded  with 
the  secondary  valences  of  Werner  or  the  other  earlier  "  residual 
valences  "  of  Armstrong.  A  discussion  of  this  subject  might  be 
made  interesting  and  profitable,  but  I  cannot  go  into  it  here.  So 
many  curious  valence  phenomena  have  been  observed  of  late  that 
one  cannot  help  feeling  that  we  are  about  to  have  a  revelation  that 
will  make  the  old  as  well  as  the  new  phenomena  appear  clear.  Car- 
bon is  bivalent  and  quadrivalent.  That  has  always  been  clear, 
though  Nef  has  made  it  clearer  than  it  used  to  be.  But  now  comes 
trivalent  carbon  that  Gomberg  has  shown  us,  and  we  may  be  pre- 
pared for  almost  anything.  And  oxygen  that  has  been  regarded  as 
a  very  model  of  bivalency  these  many  years  is  getting  restless,  and 
is  beginning  to  show  that  it  too  can  do  the  unexpected.  It  seems 
clear  that  it  can  act  as  a  quadrivalent  element,  but,  according  to 
Walden,  it  has  even  higher  powers. 
Whatever  may  come  of  all  this,  it  is  clear  that  we  cannot  enlarge 
our  conception  of  the  atom.  It  not  only  has  the  power  to  combine 
with  other  atoms,  but  under  given  conditions  it  has  a  definite  num- 
ber of  such  powers.  If  we  attempt  to  represent  these  powers  to  our 
minds  we  can  only  use  the  grossest  methods.  The  union  of  two 
univalent  atoms  does  not  necessitate  the  conception  of  direction. 
