508  Dr.  F.  Horton  on  the  Electrical 
suitable  compounds,  especially  as  the  emission  of  corpuscles 
from  a  large  number  of  metallic  oxides  bad  been  studied  by 
Wehnelt. 
The  oxides  were  used  in  the  form  of  slabs  about  1  sq.  cm. 
in  area,  and  from  1  to  2  mms.  in  thickness.  A  slab  of  oxide 
was  placed  between  two  stout  platinum  plates  which  were 
pressed  firmly  together  by  means  of  two  porcelain  rods 
1  cm.  in  diameter.  These  rods  had  a  small  hole  along  their 
axis,  and  platinum  wires  from  the  electrodes  passed  through 
them  and  served  to  connect  to  the  "Wheat stone  bridge  used 
for  measuring  the  resistance.  The  surfaces  of  the  electrodes 
and  of  the  oxide  in  contact  with  them  were  ground  or  filed 
plane  so  as  to  make  good  electrical  contact  all  over. 
Two  types  of  electric  heating-furnace  were  used  in 
these  experiments.  "With  the  more  easily  fusible  oxides  it 
was  necessary  to  watch  the  oxide  to  see  that  the  temperature 
was  not  raised  above  the  melting-point.  The  furnace  used 
to  heat  these  oxides  consisted  of  an  iron  box  lined  with 
asbestos  cardboard.  The  porcelain  rods  passed  vertically 
through  the  box,  the  slab  of  oxide  being  in  the  middle.  A 
strip  of  platinum  foil,  5  cms.  wide,  passed  above  and  below 
the  oxide,  and  this  was  heated  by  an  alternating  current 
from  a  transformer.  The  oxide  could  be  seen  by  opening  a 
small  door  in  the  side  of  the  furnace.  Oxides  such  as  lime, 
magnesia,  and  quartz,  of  high  melting-point,  were  heated  in 
an  electric  furnace  consisting  of  a  coil  of  No.  20  B.W.Gr. 
platinum  wire  wrapped  round  a  porcelain  tube  2*5  cms.  in 
diameter  and  13  cms.  long.  This  was  placed  inside  a  larger 
earthenware  tube  and  the  space  between  filled  with  magnesia. 
This  furnace  is  shown  in  section  in  fig.  1.  With  both  forms 
of  furnace  the  temperature  of  the  oxide  was  obtained  by 
means  of  a  thermocouple  of  wires  of  platinum  and  an  alloy 
of  platinum  with  10  per  cent,  of  rhodium,  the  junction  being- 
close  to  the  oxide. 
In  order  to  standardize  the  thermocouple  the  melting-point 
of  potassium  sulphate  wTas  taken  as  the  fixed  point,  and  the 
galvanometer  deflexion  corresponding  to  this  temperature 
(1066°  C.)  was  found.  The  platinum  temperatures  given  by 
the  galvanometer-readings  were  corrected  to  degrees  centi- 
grade by  means  of  the  curve  given  by  Professor  Callendar*. 
The  galvanometer  deflexion  corresponding  to  the  melting- 
point  of  platinum  v~as  also  found  and  used  for  standardizing 
readings  up  to  1700°  C. 
*  Callendar,  Ph.il.  Mag.  vol.  xlviii.  p.  519. 
