Yol.  49.]  PROF.  J.  W.  JUDD  ON  COMPOSITE  DYKES  IN  ARRAN.  557 
to  12  feet  (see  fig.  4).  It  here  appears  to  send  off  sheets,  laterally, 
between  the  sandstone-beds,  and  to  be  connected  with  the  great 
north-and-south  dyke  which  crosses  it  on  the  shore. 
Some  distance  southward  a  simple  dyke  of  augite-andesite,  about 
5  feet  wide  (P  of  Jameson),  is  found  crossing  the  shore. 
The  Southern  Transverse  Composite  DyTce  (IV  in  Map). — This 
dyke  is  not  included  in  Jameson’s  plan,  but  it  occurs  only  a  little 
south  of  the  point  where  the  great  north-and-south  dyke  rims  out  to 
sea,  and  appears  to  be  closely  connected  with  the  other  dykes  of  this 
plexus  ;  it  is  about  300  yards  south  of  the  middle  composite  dyke. 
This  southern  dyke  is  about  90  feet  wide,  and  traverses  the  sand¬ 
stone,  which  it  greatly  alters,  in  a  N.NVE.-and-S.S.W.  direction.  The 
point  where  it  runs  into  the  cliff  is  called  on  the  Ordnance  map  An 
Cumhann,  and  the  intrusion  forms  a  great  buttress  across  the  shore. 
In  places  the  strike  of  this  dyke  closely  agrees  with  that  of  the 
sandstone-rocks  into  which  it  is  intruded,  and  the  mass,  like  the 
great  north-and-south  dyke,  which  is  parallel  to  it  at  its  southern 
end,  may  be  regarded  as  being  of  the  nature  of  a  sheet  rather  than 
a  dyke. 
The  rock  forming  the  greater  part  of  this  dyke  is  a  quartz-felsite, 
with  very  large  porphyritic  crystals,  closely  resembling  that  which 
constitutes  the  headland  of  Drumadoon.  On  the  south-eastern  side 
of  the  mass  there  is  a  dyke  of  augite-andesite,  2j  feet  wide,  which 
has  inserted  itself  between  the  quartz-felsite  and  the  sandstone,  in 
the  latter  having  sent  off  a  branch  which  can  be  seen  in  the  cliff. 
On  the  north-western  side  is  a  second  dj7ke  of  augite-andesite,  which 
bifurcates  where  it  rises  into  the  cliff  into  two  bands  each  3  feet 
wide,  including  between  them  a  ‘  horse  ’  of  the  quartz-felsite.  A 
third  dyke  of  augite-andesite,  4  feet  wide,  traverses  the  middle  of 
the  great  quartz-felsite  dyke  in  a  sinuous  line.  It  is  one  of  the 
andesite  d}Tkes  that,  weathering  away  easily,  has  formed  a  ravine 
down  which  the  path  from  the  top  of  the  cliff  to  the  shore  has 
been  formed  (see  fig.  5,  p.  558). 
There  can  be  no  doubt  that  these  three  dykes  of  augite-andesite 
were  injected  at  a  later  date  than  the  mass  of  the  quartz-felsite. 
The  relations  of  the  several  rocks  render  this  point  clear,  and,  if  any 
doubt  existed  on  the  subject,  it  would  be  removed  by  the  presence  of 
numerous  crystals  of  quartz  and  felspar  in  the  augite-andesite,  which 
have,  as  I  shall  presently  show,  been  derived  from  the  quartz-felsite. 
Passing  southward,  past  the  well-known  King’s  Cave,  we  come 
to  the  cliff  in  which  the  masses  of  red  and  brown  pitchstone  and 
quartz-felsite  described  by  Allport1  occur,  and  a  little  farther  on  a 
composite  dyke  occurs  with  augite-andesite  at  its  sides  and  pitchstone 
in  its  centre.  Nearer  Drumadoon  Point  many  dykes  of  different 
composition  are  seen,  one  of  augite-andesite  having  produced  partial 
f  usion  in  a  mass  of  quartz-felsite,  as  shown  by  Prof.  Bonney.2  Other 
augite-andesites  are  remarkable  for  the  number  of  quartz-crystals 
they  have  caught  up  in  traversing  the  quartz-felsite.  As,  however5 
1  Geol.  Mag.  for  1872,  pp.  538-540. 
