488  Determination  of  Melting  Points.  {Am,o^il£ari^ 
the  temperature  at  which  the  fat  is  pressed  upward  in  the  tube.  This 
latter  method  does  not.  give  as  accurate  results  as  the  former,  some 
fats  becoming  very  soft  before  they  are  entirely  melted ;  but  it  can 
be  used  to  good  advantage  in  the  determining  of  the  melting  points 
of  some  resins  and  waxes. 
Class  III.  Substances  not  readily  oxidized.  The  method  given 
under  Class  I.  can  be  used  here,  but  only  in  such  cases  where  no  re- 
action takes  place.  Where  this  method  cannot  be  used,  we  must 
resort  to  the  sand-bath  and  a  thin  porcelain  capsule.  The  tempera- 
ture registered  by  the  thermometer,  if  placed  in  the  sand-bath,  will 
be  several  degrees  too  high ;  as  a  correction  we  can  use  the  compari- 
son mentioned  under  Class  I.  Or  the  bulb  of  the  thermometer  is 
immersed  in  the  material,  which  is  slowly  heated  to  liquefaction. 
Class  IV.  Substances  easily  oxidized  in  contact  with  air.  Sul- 
phur, phosphorus,  and  other  bodies  which  easily  burn  or  oxidize,  or 
in  which  there  takes  place  dissociation  or  other  chemical  change  when 
heated  in  air,  must  be  melted  in  some  liquid  in  which  the  article  under 
examination  is  not  at  all  or  only  very  slightly  soluble,  and  with  which 
no  chemical  decomposition  takes  place.  The  best  manner  in  which 
to  apply  this  method,  is  to  suspend  a  piece  of  the  substance  in  the 
liquid,  and  keep  the  bulb  of  the  thermometer  as  close  as  possible  to 
it.  This  .method  has  been  tried  by  Gerardin  (Compt.  rendues,  1862, 
liv,  1082)  with  sulphur  and  phosphorus  in  different  media,  and  he 
found  the  melting  point  to  be  invariably  111.5°  C  and  44.2°  C 
respectively. 
At  various  times  descriptions  of  apparatuses  for  the  determination 
of  melting  points  have  been  published,  of  which  the  following  may 
be  briefly  mentioned.  DragendorfF  (Analyse  der  Pflanzen  12, 13)  does 
not  apply  heat  directly  to  the  beaker  spoken  of  under  Class  I.  He 
allows  the  rim  of  the  beaker  to  rest  on  the  top  of  a  filter-dryer  and 
covers  the  whole  with  a  bottle,  the  bottom  of  wilich  has  been  taken 
out.  The  thermometer  is  suspended  through  the  neck  of  the  bottle, 
and  is  there  fastened  by  a  cork.  The  whole  apparatus  stands  on  an 
iron  plate,  the  filter-dryer  being  of  such  a  height  that  the  beaker  does 
not  rest  on  the  plate.  Heat  is  now  applied  gradually  to  the  iron 
plate,  carefully  regulating  the  temperature. 
Another  form  of  apparatus  has  been  described  by  Gustav  Oldberg 
(Rep.  d.  Anal.  Chein.,  1886,  94).  The  apparatus  consists  of  two 
glass  tubes  of  different  diameters  ;  a  bulb  is  blown  into  one  end  of 
