Am.  Jour.  Pharm.  ) 
February,  1914.  j 
Colloids  and  Crystals. 
75 
III. 
On  the  subject  of  the  classification  of  colloid  systems  we  must 
be  very  brief.  One  proposal  subdivides  them  into  suspensoids,  such  as 
the  sols  3  of  gold  and  arsenious  sulphide,  in  which  the  dispersed 
phase  is  solid,  and  emulsoids,  in  which  the  dispersed  phase  is  liquid. 
This  classification  would  appear  to  be  an  attempt  to  extend  the 
familiar  distinction  between  liquid  and  solid  to  a  domain  in  which 
that  distinction  has  little  if  any  meaning.  To  assert  that  a  thing  is 
solid  is  to  say  that  it  has  a  definite  shape,  which  it  retains  with  some 
persistence.  There  is  not  the  slightest  reason  to  think  that  the 
particles  in  a  gold  sol  are  solid.  It  is  usual  to  assume  that  they  are 
spherical,  but  this  is  done  merely  because  it  is  the  simplest  assumption 
to  make.  There  are  faint  indications  that  they  really  have  the  form 
of  leaflets  or  of  little  rods,  but  they  appear  in  the  ultra-microscope 
simply  as  brilliant  dancing  points,  and  in  reality  we  know  nothing 
whatever  about  their  shape.  In  connection  with  this  it  is  interesting 
to  recall  the  fact  that  the  formation  of  a  crystal  begins  with  the 
appearance  of  minute  liquid  spheres  (globulites)  ,4  which  pass 
through  several  stages  (margarites,  longulites,  etc.)  before  the  crys- 
tal is  formed.  It  seems  possible  that,  under  such  enormous  sub- 
division, cohesion  retires  into  the  background  and  surface  tension 
assumes  the  chief  role,  so  that  the  gold  particles  are  rather  to  be 
compared  to  minute  drops  than  to  little  crystals. 
Enough  has  been  said  to  make  clear  the  uncertainty  which  attaches 
to  the  attempt  to  classify  colloid  solutions  according  to  the  state  of 
aggregation  of  the  particles.  A  better  classification  is  into  reversible 
and  irreversible  colloids,  according  to  the  way  in  which  the  dissolved 
substance  behaves  when  separated  from  the  solution.  Thus,  when  a 
gelatin  solution  is  evaporated  until  it  "  sets  "  it  is  only  necessary 
to  warm  the  jelly  with  water  to  obtain  it  again  in  colloid  solution. 
Gelatin  is  a  typical  reversible  colloid.  But  when  the  gold  is  caused 
to  separate  from  a  gold  sol — which  can  easily  be  brought  about  by 
adding  any  electrolyte  to  the  sol — the  gold  will  not  again  enter  into 
colloidal  solution.    Shaking  or  warming  with  water  gives  a  mere 
3  Thomas  Graham  introduced  the  term  sol  as  an  abbreviation  for  colloidal 
solution. 
4  Fink,  "  PoggendorfFs  Annalen,"  vol.  46,  p.  258  (1839)  ;  Schmidt,  "  Liebig's 
Annalen  der  Chemie,"  vol.  53,  p.  171  (1845);  Frankenheim,  "  PoggendorfFs 
Annalen,"  vol.  in,  p.  i  (i860). 
