﻿134 
  ^'EW 
  YORK 
  STATE 
  MUSEUM 
  

  

  With 
  enlargement 
  of 
  the 
  underground 
  tunnel 
  the 
  roof 
  tends 
  to 
  

   cave 
  in, 
  at 
  first 
  where 
  thinnest, 
  followed 
  by 
  gradual 
  lengthening. 
  In 
  

   most 
  cases 
  the 
  cover 
  is 
  thinnest 
  toward 
  the 
  stream 
  mouth 
  and 
  cav- 
  

   ing 
  in 
  begins 
  there 
  and 
  works 
  slowly 
  upstream. 
  In 
  the 
  case 
  of 
  the 
  

   creek 
  at 
  Felts 
  ]vlills, 
  just 
  referred 
  to, 
  the 
  map 
  shows 
  Lowwille 
  lime- 
  

   stone 
  in 
  its 
  bed 
  for 
  a 
  half 
  mile 
  above 
  its 
  mouth, 
  beyond 
  which 
  the 
  

   Leray 
  forms 
  the 
  bed 
  rock. 
  In 
  the 
  Lowville 
  for 
  part, 
  of 
  its 
  

   course 
  the 
  stream 
  is 
  above 
  ground, 
  and 
  the 
  point 
  where 
  the 
  forma- 
  

   tional 
  contact 
  crosses 
  the 
  stream 
  marks 
  the 
  point 
  of 
  emergence 
  

   from 
  underground, 
  and 
  the 
  slow 
  upstream 
  working 
  of 
  the 
  roof 
  

   cave 
  in. 
  In 
  plate 
  36 
  is 
  a 
  rather 
  unsatisfactory 
  \-iew 
  of 
  the 
  caved 
  

   roof 
  of 
  a 
  small 
  stream, 
  unsatisfactory 
  because 
  no 
  position 
  of 
  the 
  

   camera 
  which 
  looked 
  upstream 
  could 
  be 
  obtained, 
  and 
  we 
  are 
  here 
  

   merely 
  looking 
  across 
  from 
  one 
  bank 
  to 
  the 
  other, 
  with 
  the 
  nearer 
  

   bank 
  somewhat 
  hiding 
  the 
  view 
  of 
  the 
  opposite 
  one. 
  The 
  stream 
  is 
  

   a 
  small 
  one, 
  fed 
  by 
  the 
  underground 
  waters 
  of 
  a 
  Leray 
  pro-m- 
  

   ontory 
  of 
  no 
  great 
  extent, 
  but 
  its 
  waters 
  emerge 
  from 
  well 
  do^-n 
  

   in 
  the 
  Lowville, 
  (which 
  alone 
  appears 
  in 
  the 
  plate) 
  and 
  can 
  be 
  

   seen 
  in 
  the 
  extreme 
  lower 
  left-hand 
  comer. 
  The 
  caving 
  extends 
  

   many 
  yards 
  upstream 
  and 
  amount^ 
  t:^ 
  -ome 
  20 
  feet 
  in 
  hight 
  at 
  

   the 
  lower 
  end. 
  

  

  Plate 
  37 
  gives 
  an 
  interesting 
  illustration, 
  on 
  a 
  small 
  scale, 
  of 
  

   another 
  feature. 
  The 
  view 
  shows 
  a 
  Lowville 
  platform, 
  surfaced 
  

   by 
  a 
  resistant 
  layer 
  of 
  somewhat 
  less 
  solubiHty-, 
  and, 
  on 
  the 
  right, 
  

   the 
  point 
  of 
  emergence 
  of 
  a 
  small, 
  wet 
  weather 
  stream, 
  flowing 
  in 
  

   a 
  shallow 
  underground 
  channel 
  in 
  the 
  more 
  soluble 
  material 
  un- 
  

   derneath. 
  The 
  stream 
  course 
  then 
  curves 
  across 
  the 
  foreground 
  

   and 
  passes 
  back^-ard 
  and 
  toward 
  the 
  left, 
  its 
  course 
  margined 
  by 
  

   the 
  projecting 
  edge 
  of 
  the 
  hard 
  layer, 
  wliich 
  has 
  otherwise 
  been 
  

   removed 
  from 
  the 
  channel 
  with 
  the 
  exception 
  of 
  the 
  fragment 
  

   left 
  as 
  a 
  tiny 
  "natural 
  bridge 
  " 
  on 
  the 
  left. 
  

  

  As 
  already 
  pointed 
  out 
  by 
  Ruedemann. 
  in 
  his 
  account 
  of 
  the 
  Low- 
  

   ville 
  inliers 
  in 
  the 
  Leray 
  limestone, 
  very 
  interesting 
  under- 
  

   ground 
  features 
  are 
  shown 
  in 
  the 
  Perch 
  river 
  valley 
  about 
  Limerick 
  

   (Ga3rton 
  sheet). 
  The 
  rock 
  structure 
  there 
  seems 
  to 
  us 
  to 
  be 
  anti- 
  

   clinal, 
  with 
  the 
  Leray 
  limestone 
  at 
  Limerick 
  marking 
  the 
  

   site 
  of 
  a 
  sag. 
  and 
  the 
  Lowville 
  inlier, 
  just 
  south, 
  the 
  site 
  of 
  a 
  

   dome 
  of 
  the 
  anticlinal 
  crest. 
  North 
  of 
  Limerick 
  increased 
  south- 
  

   erly 
  dip 
  transfers 
  the 
  stream 
  from 
  the 
  Lowville 
  to 
  the 
  Leray 
  

   horizon, 
  south 
  of 
  it 
  diminished 
  south 
  dip 
  transfers 
  the 
  stream 
  back 
  

   to 
  the 
  Lowville 
  again, 
  the 
  point 
  of 
  transfer 
  being 
  marked 
  by 
  a 
  fall 
  

   [pi. 
  23] 
  as 
  is 
  the 
  rule 
  in 
  the 
  streams 
  of 
  the 
  region 
  when 
  passing 
  

  

  