230         CONTMBtTTlOHS    TO    ECONOMIC    GEOLOGY,  L906,  PART   I. 
AL,0,  +  FeA 
SiO, 
should    be    more    nearly    approximated.     The 
o  r 
alumina   and   iron  oxide  together  should  not   be  greater  than  ■   0  - 
and  it  is  apparent  from  the  above  analyses  that  at  Potosi  Station,  as 
well  as  at  Spechts  Ferry,  their  sum  is  greater  than  this.  It  is  desir- 
able, therefore,  both  on  account  of  the  chemical  composition  and  the 
relative  thinness  of  the  shah4  beds,  that  some  other  supply  of  silica 
and  alumina  should  be  at  hand.  It  is  possible  that  such  a  supply 
might  be  obtained  from  the  residual  clay  and  loess  at  the  top  of  the 
hills  wherever  these  materials  average  rich  in  silica  and  poor  in  lime. 
An  idea  of  the  composition  of  the  clay  and  loess  present  in  the  region 
can  he  had  from  the  accompan}ring  analyses.  While  the  samples  of 
clay  and  loess  were  not  taken  from  the  same  sections  as  the  limestone 
and  shale,  there  is  an  abundance  of  this  unconsolidated  material 
above4  the  Galena  beds  ;it  each  locality,  and  its  composition  is  prob- 
ably such  that  it  may  be  considered  an  important  factor  in  the 
sit  uat  ion. 
Analyses  of  residual  clay  and  loess  from  th  Drifiless  Arm.  " 
[Analyst,  R.  !'•    Riggs.] 
1. 
J. 
3- 
4. 
r>. 
6. 
3iO     
71.  L3 
L2.50 
15 
I.", 
02 
04 
85 
38 
l.M 
l  63 
.  L9 
1'.  59 
L8.64 
17.  I'.t 
.27 
28 
03 
.01 
.7.; 
80 
.93 
L0.46 
-, ,  09 
21.  !• 
8.53 
86 
1., 
03 
i,.; 
.  95 
l  43 
I   45 
L0.79 
.  29 
.22 
l'i.  L3 
20.08 
ll.iil 
.93 
L3 
.04 
.00 
1 .  22 
1.92 
1.33 
L.60 
11.72 
.  39 
L.09 
72.68 
12.03 
:<.  53 
.  96 
.72 
.  23 
.06 
L.59 
1.11 
l.liS 
2.  13 
2.50 
39 
.09 
.53 
04.  01 
U20  : 

i  i  l  i 
10.64 
2.  til 
.  :,i 
I  iOs 
.  40 
!•■<>. 
.06 
MnO 
.05 
CaO 
5.  41 
MgO 
:',.  oo 
\  .i  ■() 
I .  :i.r) 
2.  0(1 

EfjO  (+  1 1  of  organic  matter) 
2.05 
6.31 

c 
L3 
SOj 
.  11 
ih  Aim.  Rept.  i".  s.  Geol.  Survey,  L885,  pp.  2.~>o,  282. 
Nbs.  1  and  J  are  samples  of  clay  from  the  same  vertical  section, 
\o.  1  having  been  taken  11  feet  from  the  surface  and  Xo.  2  a  little 
more  than  si  feet  from  the  surface,  in  contact  with  the  underlying 
limestone.  Nos.  :>  and  1  are  .simples  of  clay  1  hat  are  similarly  related, 
having  been  taken,  respectively.  :!  and  J  I  feet  from  the  surface,  the 
latter  clay  in  contact  with  the  rock.  \'o.  .">  is  loess  from  DubuqueJ 
Iowa;  Xo.  6  is  loess  from  Galena,  111. 
A  review  of  the  character  of  the  materials  available  shows  that  a 
cement  manufactured  in  this  district  would  be  of  the  type  made  from 
a  mixture  of  ordinary  hard  limestone  and  clay  or  shale.  About  36 
per  cent  of  the  Portland  cement  now  made  in  the  United  States  is  of 
this  type.  Magnesium  carbonate,  the  most  objectionable  of  impuri- 
ties in  raw  limestone  materials,  falls  in  these  rocks  well  within  the 
