﻿258 
  J. 
  Barrett 
  — 
  Piedmont 
  Terraces. 
  

  

  Figure 
  8 
  gives 
  the 
  projected 
  profile 
  of 
  the 
  Stamford 
  

   (Conn.) 
  and 
  northwestern 
  portion 
  of 
  the 
  Oyster 
  Bay 
  

   (N. 
  Y.) 
  quadrangles 
  where 
  the 
  terraces 
  are 
  narrowest, 
  

   the 
  direction 
  of 
  view 
  being 
  N 
  30° 
  E 
  or 
  nearly 
  parallel 
  to 
  

   the 
  shore. 
  A 
  comparison 
  of 
  this 
  profile 
  with 
  the 
  preced- 
  

   ing 
  one 
  appears 
  to 
  indicate 
  that 
  the 
  terraces 
  there 
  seen 
  

   are 
  present 
  but 
  much 
  less 
  strongly 
  developed. 
  The 
  

   problem 
  of 
  this 
  weaker 
  expression 
  may 
  be 
  left 
  for 
  future 
  

   solution. 
  It 
  may 
  be 
  said 
  that 
  the 
  explanation 
  does 
  not 
  

   seem 
  to 
  be 
  found 
  in 
  the 
  seaward 
  protection 
  afforded 
  by 
  

   Long 
  Island, 
  for 
  the 
  high 
  level 
  terraces 
  show 
  a 
  corre- 
  

   sponding 
  narrowness 
  as 
  compared 
  with 
  their 
  eastern 
  por- 
  

   tions, 
  and 
  it 
  can 
  not 
  well 
  be 
  supposed 
  that 
  a 
  permanent 
  

   barrier 
  has 
  existed 
  since 
  the 
  latter 
  part 
  of 
  the 
  Creta- 
  

   ceous 
  at 
  which 
  time 
  the 
  two 
  highest 
  terraces 
  are 
  assumed 
  

   to 
  have 
  been 
  cut. 
  The 
  main 
  factors 
  appear 
  to 
  be 
  a 
  some- 
  

   what 
  greater 
  resistance 
  of 
  the 
  rock 
  formations 
  to 
  erosion 
  

   and 
  a 
  steeper 
  general 
  surface 
  slope 
  than 
  in 
  the 
  eastern 
  

   part 
  of 
  the 
  state. 
  There 
  was 
  also 
  an 
  initial 
  difference 
  

   in 
  the 
  topography 
  of 
  the 
  two 
  localities 
  though 
  not 
  suffi- 
  

   ciently 
  marked, 
  perhaps, 
  to 
  have 
  had 
  much 
  effect 
  on 
  the 
  

   rate 
  at 
  which 
  the 
  benches 
  may 
  have 
  been 
  cut. 
  

  

  It 
  should 
  not 
  be 
  inferred 
  that 
  the 
  recognition 
  of 
  the 
  

   terraces 
  was 
  based 
  only 
  on 
  the 
  accordance 
  of 
  hilltops 
  

   because 
  the 
  profiles 
  give 
  that 
  impression. 
  There 
  is 
  al- 
  

   ways 
  the 
  possibility 
  of 
  a 
  chance 
  accordance 
  of 
  hilltops 
  

   and 
  mere 
  accordance 
  no 
  doubt 
  has 
  been 
  too 
  loosely 
  used 
  

   as 
  a 
  criterion 
  for 
  the 
  recognition 
  of 
  peneplanes. 
  Profes- 
  

   sor 
  Barrell 
  had 
  this 
  possibility 
  clearly 
  in 
  mind 
  and 
  in 
  

   his 
  earliest 
  note-book 
  certain 
  criteria 
  are 
  given 
  for 
  the 
  

   recognition 
  of 
  true 
  coincidences 
  of 
  level 
  in 
  a 
  partly 
  re- 
  

   duced 
  upland. 
  They 
  are: 
  " 
  (1) 
  There 
  must 
  be 
  some 
  

   ' 
  special' 
  feature 
  (or 
  features) 
  in 
  the 
  level 
  observed, 
  not 
  

   a 
  mere 
  hilltop; 
  (2) 
  similar 
  features 
  elsewhere 
  must 
  be 
  

   accordant 
  and 
  lie 
  in 
  the 
  same 
  plane, 
  (3) 
  the 
  baselevel 
  is 
  

   as 
  low 
  as 
  the 
  lowest 
  features 
  correctly 
  determined 
  as 
  

   belonging 
  to 
  the 
  stage, 
  and 
  (4) 
  levels 
  should 
  be 
  region- 
  

   ally 
  independent 
  of 
  structure, 
  and 
  rocks 
  of 
  similar 
  resist- 
  

   ance, 
  when 
  in 
  similar 
  relation 
  to 
  baselevel, 
  should 
  have 
  a 
  

   corresponding 
  topographic 
  development." 
  An 
  idea 
  of 
  

   what 
  these 
  special 
  features 
  were 
  is 
  gained 
  from 
  this 
  

   article 
  and 
  an 
  inspection 
  of 
  maps 
  and 
  notebooks 
  shows 
  

   that 
  Professor 
  Barrell 
  was 
  constantly 
  on 
  the 
  lookout 
  for 
  

   them. 
  

  

  [To 
  be 
  continued] 
  

  

  