48 
Thermometric  Measurements. 
/Am.  Jour.  Pharm. 
\      Jan.,  1884. 
depression  of  the  zero  point  may  be  effected  by  heating  a  thermometer 
at  various  temperatures  for  a  prolonged  time,  and  then  leaving  it  to 
cool  in  the  air,  and  these  depressions  will  necessarily  increase  all  the 
constants  of  the  thermometer  when  referred  to  the  zero  point.  After 
such  a  depression  has  been  effected,  the  thermometer,  slowly  at  ordinary 
temperatures,  but  more  quickly  when  warmed  slightly,  tends  to  revert 
to  its  original  readings. 
A  considerable  elevation  of  the  zero  point,  10°  to  26°,  is  produced 
by  heating  the  thermometer  for  a  week  at  355°,  which  is  caused  by  the 
expansion  of  the  glass  bulb  after  it  has  been  blown  out  and  then  sud- 
denly cooled. 
The  elevation  of  zero  in  a  thermometer  maintained  at  ordinary 
temperatures  diminishes  gradually  and  ceases  to  be  appreciable  after 
five  or  ten  years.  Similarly  variations  produced  by  protracted  heat- 
ing tend  towards  constant  limit ;  thus,  for  example,  a  thermometer 
may  be  heated  for  several  days  at  300°,  or  for  several  months  at  100° 
without  causing  a  variation  in  the  effect  produced  by  heating  to  355°. 
From  the  facts  detailed  above,  it  is  necessary  to  heat  a  thermometer 
required  for  ordinary  experiments  for  a  week  in  boiling  mercury,  the 
whole  of  the  stem  being  enclosed  in  the  vessel ;  after  this  treatment, 
the  points  0°  and  100°  will  have  a  permanent  value. 
The  author  further  remarks  that  thermometers  with  a  limited  range, 
from  200°  to  300°  for  example,  cannot  be  graduated  with  the  same 
degree  of  precision,  for  the  determination  of  the  fixed  points  0  and 
100  becomes  impossible,  owing  to  the  falling  of  the  mercury  within 
the  reservoir.  In  order  to  fix  definite  points  above  100°,  the  author 
suggests  the  use  of  naphthalene  and  benzophenone,  substances  which 
can  be  obtained  in  a  state  of  purity;  the  former  boils  at  218°,  and  the 
latter  at  306°  under  normal  pressure;  in  a  table  in  the  original 
memoir,  the  boiling  points  of  these  two  substances  under  various 
pressures  are  given. 
In  a  thermometer  which  has  been  thoroughly  deprived  of  air,  the 
phenomenon  of  volatilization  of  mercury  can  be  observed  at  100°  ;  the 
column  of  mercury  gradually  descends,  and  after  about  15  minutes 
the  variation  is  about  0'01 — 0*02  degree.  If  the  zero  point  is  rede- 
termined after  each  warming,  no  error  is  caused  by  the  descent  of  the 
mercury.  In  all  cases  the  mercury  with  which  the  thermometer  is 
filled  must  not  only  be  purified  but  boiled  for  a  long  time  to  rid  the 
