of  -Dielectric  Constants  of  Non-conducting  Liquids.         11 
variations  in  readings  amounted  to  1  per  cent.,  the  correction 
to  be  applied  would  be  of  the  second  order. 
Liquid  Resistances. 
In  earlier  experiments  these  consisted  of  electrolytic  cells 
of  the  H-bottle  shape  (Kohlrausch  type)  filled  with  redistilled 
water ;  the  electrodes  were  platinized  in  the  usual  way. 
One  cell  was  put  into  each  of  the  two  arms  of  the  bridge,  the 
two  cells  being  immersed  side  by  side  in  a  beaker  of  cold 
water  to  eliminate  effects  of  temperature.  The  resistance 
introduced  into  each  arm  was  approximately  9000  ohms, 
which  was  found  sufficient -for  air  or  liquids  of  low  dielectric 
constant.  In  later  experiments,  liquid  resistance-tubes  of  the 
Nernst  type  were  used,  filled  with  Manganinf's  (mannite- 
boracic  acid)  solution  ;  these  were  found  to  be  more  suitable 
for  liquids  of  relatively  higher  dielectric  constant. 
Lnductoria. 
Two  forms  of  inductorium  were  used,  namely,  that  of 
Ostwald-Luther  and  the  wire  form  (Saiten-unterbrecher). 
My  experience  was  to  the  effect  that  the  latter  gave  the 
sharper  note,  and  consequently  a  more  distinct  minimum  in 
the  telephone,  but  was  more  liable  to  break  down  in  the 
course  of  work,  but  this  might  be  due  to  some  slight 
mechanical  defect  in  the  particular  instrument. 
My  observations  were  in  complete  accordance  with  those 
of  N  ernst  *,  that  a  more  decided  minimum  is  produced  in  the 
telephone  when  the  inductorium  gives  a  buzzing,  rather  than 
a  humming  or  singing  sound.  The  inductoria  were  actuated 
by  a  small  chromic-acid  battery  or  accumulator. 
Telephone. 
The  form  of  telephone  used  was  that  of  Mix  and  Genest 
(resistance  about  130  ohms)  with  an  insulating  handle  ;  it 
was  found  preferable  to  hang  up  the  telephone  on  a  glass 
rod  ;  when  the  wire  inductorium  was  used,  the  note  of  the 
telephone,  except  at  or  about  the  minimum  point,  could  be 
heard  some  feet  away,  so  that  the  initial  adjustments  of  the 
micrometer-screws  could  be  made  with  comfort  and  con- 
venience. The  whole  apparatus  (namely,  liquid  resistances, 
condensers,  connecting  wires)  was  disposed  symmetrically  as 
far  as  possible,  and  each  of  the  two  former  pairs  was  placed 
together. 
*  Nernst  (loc.  cit.  supra)  writes:  ■ ' Ich  liabe  oft  beobachtet  dass  daa 
Minimum  leidet,  -wenn  die  Saite  summt  odor  singt,  anstatt  rasselt ;  ich 
vermajr  dies  niclit  sicher  zu  erldaren." 
