'^Vay'i-.mr-}        On  Breakage  of  Glass  Vessels.  199 
rod,  are  still  hot  enough  to  warm  the  surface  of  the  glass,  but  not  hot 
enough  to  cause  it  to  crack.  They  form  a  layer  of  moderately  warm 
liquid  on  the  surface  of  the  glass.  The  next  portions  of  water  coming 
'down  are  somewhat  warmer,  and  are  followed,  toward  the  end  of  the 
pouring,  by  liquid  which  is  quite  hot ;  but  the  bottle  has  been  gradu- 
ally warmed  by  the  first  portions  of  the  liquid,  so  that  it  will  not  now 
crack  when  the  hot  liquid  is  poured  in,  as  there  is  no  very  sudden 
change  of  temperature.    This  idea  suggested  to  my  mind 
Experiment  XVI.  Procured  ten  common  pickle  bottles  made  of 
green  glass,  flat  sided  and  tapering  the  entire  length,  so  that  the  bot- 
toms were  4J  inches  wide,  while  the  tops  were  only  I  J  inches,  the 
width  of  the  mouth.  Into  each  of  these  bottles  I  placed  a  rod  in  such 
manner  that  the  lower  end  rested  on  one  end  of  the  bottom,  and  the 
upper  pointed  diagonally  upward.  The  bottles  were  cooled  and  boil- 
ing Avater  w^as  poured  in.  Owing  to  the  position  of  the  rods,  most  of 
the  water  fell  off  them  soon  after  entering  the  bottles,  and  had  not 
time  to  be  cooled  by  passing  all  the  way  down  on  the  cold  metal.  The 
water  fell  on  the  bottoms  at  a  point  two  inches  from  the  ends  of  the 
rods.     Out  of  the  ten  bottles,  eight  were  broken. 
Experiment  XVII.  Took  a  good,  wide  mouthed,  one  pint  bottle, 
•cooled  the  bottle  and  placed  in  it  a  rod  of  brass,  poured  in  boiling 
water  until  full,  and  immediately  plunged  in  a  thermometer.  The 
mercury  in  the  thermometer  rose  only  to  181*^,  showing  that  31°  of 
heat  had  been  lost  b}^  the  entire  mass  of  liquid  in  passing  down  the 
rod.    This  circumstance  suggested 
Experiment  XVIII.  Procured  a  good,  wide  mouthed  bottle,  of  clear 
white  glass.  The  bottle  was  2|  inches  in  diameter  at  the  bottom,  and 
6 J  inches  high  to  the  shoulder.  Placed  in  it  a  thermometer  and  brought 
the  bottle  to  a  uniform  temperature  by  filling  with  water  at  50°  F. 
and  placing  it  in  a  bucket  of  the  same  water.  Allowed  the  thermome- 
ter to  remain  in  the  bottle,  placed  in  the  same  bucket  a  rod  of  brass 
and  allowed  all  to  remain  for  twenty  minutes  ;  so  that  bottle,  rod  and 
thermometer  tube  might  be  of  the  same  temperature.  Heated  water 
to  boiling  in  a  vessel  having  a  good  spout.  Before  placing  the  bottle 
in  the  bucket,  I  had  pasted  on  it  a  strip  of  paper,  graduated  to  show 
the  points  by  which  it  was  filled  by  two,  four,  six,  eight,  ten,  twelve, 
sixteen  and  twenty  fluid  ounces.  After  removing  the  bottle  from  the 
cold  water  and  emptying  it,  I  placed  in  it  the  brass  rod  and  ther- 
mometer.   Poured  boiling  water  down  the  rod  to  the  first  mark,  re- 
