454 
CONTRIBUTIONS    TO    ECONOMIC    GEOLOGY,  1905. 
limestone  than  through   sand,   is  troublesome  because   it    renders  the  hatch  less  fusibl  I 
[f  the  sand  is  derived  from  indurated  sandstone  the  latter  should  be  friable  or  easily  crushe 
In  examining  sand,  in  order  to  ascertain  its  value  Tor  glass-making  purposes,  inspecticl 
with  a  magnifying  glass  is  the   besl    preliminary  test.     The  following  points  should  l| 
observed:  The  sand  should  be  nearly   white  in  color:  it   should   be  of  medium   linene  I 
(passing  a  20  to  50  mesh  horizontal  sieve);  the  grains  should  be  uniform  in  size,  even 
and  angular,  or,  less  preferably,  they  may  be  rounded.     A  simple  chemical  test  may  I 
employed  by  heating  the  sand  in  a  dilute  acid.      Effervescence  indicates  the  presence  ( 
lime:  loss  of  color  shows  the  presence  of  clay  impurities.     Iron  in  the  most  minute  qua! 
tity  may  be  detected  by  dissolving  sand  in  hydrofluoric  acid  and  adding  potassium  fei 
rocyanide,  which  produces  a  blue  precipitate  if  iron  is  present.     Complete  quantitatif 
analyses  as  well  as  a  furnace  test  should  be  made  as  a  final  determination  of  the  characfj 
of  a  prospective  ^la-s  sand.     The  impression  ha-  long  prevailed,  especially  among  writJ 
on  the  subject    of  glass  making,  that  round-grained  sands  are  at  a  decided  disadvantage  i. 
comparison  with  the  sharp,  angular  variety.     Practice  seems  to  disagree  with  this  idea 
at  least  in  the  case  of  the  Mississippi  Valley  plants,  where  smool  h,  rounded  sand  is  success 
fully  used  for  all  ordinary  varieties  of  glass  and  for  some  line  Hint  ware.  As  a  rule,  in 
very  close  check  i-  kepi  on  materials  by  manufacturers  of  the  ordinary  commercial  grade; 
of  glass.  Results  are  watched  with  care,  however,  and  experience  with  certain  material: 
usually  determines  their  use  <>r  rejection.  Sand  uniformly  liner  than  one-sixtieth  incl 
is  said  to  "burn  out  "  in  the  batch  and  not  to  produce  as  much  glass  per  unit  of  weight  ■ 
doc-  coarser  sand.  In  a  mixture  of  coarse  and  line  sand  the  liner  sand  is  liable  to  settl 
to  the  bottom  of  the  batch,  thus  preventing  an  even  mixture  of  the  materials  and  pio- 
dueing  in  consequence  a  glass  uneven  in  texture. 
ANALYSES   OF   GLASS    SAND. 
Window  ( rlass  Company 
The  following  analyses  submitted  l>\   the  America 
different  qualities  of  sand  used  in  its  glass  work-: 
the 
Analyses  <>/  glass  sand  used  /"/  American  Window  Glass  Company. 
il  iiriit . 
Silica  (Si02) 
Alumina  |  AI2O3) 
1  ron  oxidi     I  ■  .<  • 
Lime  and  magnesia  (Ca<  >  and  Mgi  1 
No.  1. 
No.  2. 
No.  3. 
No.  1. 
1 
99.714 
99.659 
99.571 
.008 
.  280 
.310 
.:;.-,() 
Slighl 
t  race. 
.006 
.011 
.021 
.002 
.020 
.020 
.050 
100.000 
100.020 
100.000 
100.000 
No.  1  is  suitable  for  the  very  highest  grade-  of  glassware  and  Hint  glass.  Nos.  2  and  3 
are  suitable  for  tableware,  plate  glass,  chimneys,  prescription  ware,  etc.,  and  No.  -1  is  used 
for  window  glass. 
The  following  analyses  are  submitted  by  the  Pittsburg  Plate  (ilass  Company  as  samples 
of  sand  used  by  its  plants: 
