368 
MESSRS.  A.  HARKER  AND  J.  E.  MARR  ON  THE  [Aug.  1 893, 
in  the  relatively  pure  limestone  now  noted,  the  formation  of  lime- 
silicates,  etc.,  has  gone  on  wherever  impurities  were  collected. 
While  the  greater  part  of  the  rock  consists  of  rather  finely  crystal¬ 
line  calcite  (perhaps  with  dolomite),  the  hand-specimens  show  little 
streaky  patches  which  are  harder  than  the  rest  and  give  no  effer¬ 
vescence  with  acid.  Of  these  little  patches  some  are  of  a  greenish 
tinge  and  are  probably  augite,  others  have  the  characteristic  lustre 
and  brown  colour  of  idocrase,  while  others  again  are  white  and 
apparently  felspathic.  We  also  remark,  scattered  through  the  mass, 
little  granules  of  a  black  iron  ore  and  of  a  yellow  pyrites. 
Another  feature  of  this  rock,  doubtless  connected  with  the  nodular 
structures  so  prevalent  in  the  purer  portions  of  the  Coniston  Lime¬ 
stone  group,  is  the  occurrence  of  distinct  ovoid  nests  of  lime-bearing 
silicates  with  a  definite  arrangement  of  nucleus  and  outer  shell. 
The  nucleus  consists  of  brown  crystals  of  idocrase  [1427,  1428], 
often  1  inch  in  length,  disposed  in  stellate  groups ;  the  outer  part 
of  the  nest,  forming  a  shell  |  to  1  inch  in  thickness,  is  of  a  white 
crystalline  mineral  which  proves  to  be  chiefly  a  plagioclase-felspar 
[1746].  The  mineral  forms  striated  crystal-plates  considerably 
larger  than  any  yet  observed  in  these  limestone-rocks.  Associated 
with  it  are  specks  of  a  brightly-polarizing  pyroxene  and  grains  of 
iron  ore,  the  former  becoming  more  abundant  where  the  felspathic 
nodule  meets  the  matrix  of  calcite.  (See  PL  XVII.  fig.  6.) 
VI.  Some  Concluding  Considerations. 
In  conclusion  we  would  venture  on  one  or  two  remarks  which, 
following  out  ideas  suggested  in  the  foregoing  notes,  may  be  found 
to  have  a  more  general  bearing.  All  that  we  have  seen  in  the  Shap 
district  confirms  the  belief  that  thermometamorphism  is  not  in 
general  accompanied  by  any  change  in  the  chemical  composition  of 
the  rocks  affected.  The  exceptions  we  have  already  noted  :  namely, 
the  (partial)  loss  of  water  and  the  expulsion  (under  proper  conditions 
only)  of  carbonic  acid.1  These  do  not  seriously  modify  the  law  in 
question.  We  have,  however,  gone  farther  in  applying  this  law  not 
onl}T  to  the  metamorphosed  rocks  as  a  whole,  but  also  to  any  small 
portion  of  any  of  these  rocks.  In  the  course  of  our  descriptions  we 
have  seen  innumerable  facts  which  point  to  the  conclusion  that  no 
transference  of  material  has  taken  place  within  the  mass  of  the  rocks 
except  between  closely  adjacent  points.  If  this  be,  as  we  believe,  a 
general  law,  it  follows  that  the  mineral  produced  by  complete 
thermal  metamorphism  at  any  point  of  a  rock  depends  upon  the 
chemical  composition  of  the  rock-mass  within  a  certain  small  distance 
around  that  point.2  The  question  naturally  arises,  what  is  the  radius 
of  this  sphere  of  influence  ?  Can  we  form  any  estimate  of  the  maxi¬ 
mum  range  at  which  interchange  of  material  takes  place  between 
points  in  the  interior  of  a  rock  undergoing  thermal  metamorphism  ? 
The  distance  in  question  will  presumably  not  be  the  same  for 
1  In  some  districts  we  should  have  to  allow  other  exceptions  for  rocks  in  the 
neighbourhood  of  an  acid  intrusion:  namely,  the  introduction  of  boric  and 
hydrofluoric  acids. 
2  Comp.  Harker,  Bull.  Geol.  Soc.  Amer.  vol.  iii.  (1891)  p.  20. 
