﻿130 
  PKOCKKDINGS 
  OF 
  TIIK 
  ]\I 
  ALACOLOGICAL 
  SOCIKTV. 
  

  

  and 
  on 
  both 
  the 
  east 
  and 
  west 
  coasts 
  is 
  found 
  high 
  up 
  in 
  mountain 
  

   streams, 
  while 
  the 
  Anodontas, 
  on 
  the 
  contrary, 
  are 
  essentially 
  

   inhabitants 
  of 
  lakes 
  and 
  the 
  deeper 
  waters 
  of 
  slow-moving 
  rivers, 
  

   a 
  fact 
  which 
  may 
  explain 
  the 
  ' 
  capture 
  ' 
  of 
  the 
  Margaritanas 
  of 
  the 
  

   Missouri, 
  in 
  which 
  the 
  Anodontas 
  were 
  not 
  involved. 
  On 
  the 
  other 
  

   hand, 
  not 
  a 
  single 
  one 
  of 
  the 
  several 
  hundred 
  species 
  of 
  the 
  XJnione 
  

   fauna 
  of 
  the 
  Mississippi 
  and 
  eastern 
  J^orth 
  America, 
  which 
  no 
  doubt 
  

   are 
  descended 
  from 
  the 
  Cretaceous 
  TJnionidae 
  of 
  the 
  western 
  states, 
  

   has 
  ever 
  succeeded 
  in 
  passing 
  over 
  or 
  around 
  the 
  mountains 
  and 
  

   obtaining 
  a 
  foothold 
  on 
  the 
  Pacific 
  coast 
  or 
  in 
  Asia. 
  

  

  If 
  it 
  were 
  true 
  that 
  the 
  Margaritanas 
  of 
  the 
  Pacific 
  coast 
  of 
  ISTorth 
  

   America 
  and 
  Asia 
  were 
  the 
  descendants 
  of 
  an 
  early 
  migration 
  to 
  

   tlie 
  north-west 
  from 
  central 
  North 
  America, 
  it 
  is 
  certainly 
  very 
  

   remarkable 
  that 
  the 
  species 
  was 
  not 
  accompanied 
  by 
  any 
  of 
  the 
  

   species 
  Avhich 
  must 
  have 
  been 
  associated 
  with 
  it 
  in 
  Cretaceous 
  or 
  

   early 
  Tertiary 
  times. 
  But 
  there 
  is 
  not 
  the 
  slightest 
  evidence 
  of 
  

   anything 
  of 
  the 
  kind. 
  And 
  in 
  view 
  of 
  this 
  fact, 
  as 
  well 
  as 
  the 
  

   acknowledged 
  relations 
  of 
  the 
  Pacific 
  coast 
  fauna 
  with 
  that 
  of 
  north- 
  

   eastern 
  Asia, 
  it 
  would 
  seem 
  to 
  be 
  certain 
  that 
  no 
  such 
  migration 
  ever 
  

   did 
  occur, 
  and 
  that 
  the 
  present 
  Margaritanas 
  of 
  the 
  Pacific 
  coast 
  are 
  

   derived 
  from 
  Asiatic 
  sources. 
  

  

  If 
  this 
  be 
  conceded, 
  then, 
  if 
  Margaritana 
  originated 
  in 
  North 
  

   America, 
  its 
  occurrence 
  in 
  the 
  Old 
  World 
  must 
  be 
  accounted 
  for 
  by 
  

   a 
  north-easterly 
  migration 
  from 
  central 
  North 
  America 
  across 
  the 
  

   Oreenland 
  bridge 
  into 
  Europe, 
  and 
  from 
  there 
  entirely 
  across 
  Europe 
  

   and 
  Asia 
  and, 
  via 
  the 
  Behring 
  bridge, 
  into 
  California. 
  The 
  first 
  part 
  

   of 
  the 
  hypothesis 
  would 
  accord 
  with 
  Dr. 
  Scharff's 
  theory, 
  but 
  not 
  the 
  

   necessary 
  extension 
  of 
  the 
  route 
  to 
  account 
  for 
  the 
  occurrence 
  of 
  the 
  

   species 
  in 
  eastern 
  Asia 
  and 
  western 
  North 
  America. 
  This 
  he 
  would 
  

   apparently 
  attribute 
  to 
  a 
  north-western 
  migration 
  from 
  North 
  America. 
  

   But 
  this 
  we 
  have 
  shown 
  is 
  untenable. 
  

  

  There 
  is 
  apparently 
  little 
  doubt 
  but 
  that, 
  following 
  the 
  manifold 
  

   mutations 
  of 
  the 
  Unionidse 
  in 
  the 
  Upper 
  Cretaceous, 
  as 
  Simpson 
  

   (1895, 
  pp. 
  336-7) 
  has 
  already 
  pointed 
  out, 
  there 
  was 
  an 
  extensive 
  

   dispersal 
  of 
  the 
  TJnionidse 
  in 
  two 
  directions 
  — 
  eastward 
  through 
  British 
  

   America 
  and 
  the 
  northern 
  United 
  States 
  (in 
  the 
  early 
  Tertiary, 
  the 
  

   Gulf 
  of 
  Mexico 
  extended 
  north 
  above 
  the 
  present 
  junction 
  of 
  the 
  Ohio 
  

   and 
  Mississippi 
  Rivers) 
  and 
  south 
  into 
  Mexico. 
  It 
  was 
  with 
  the 
  

   northern 
  migration, 
  if 
  at 
  all, 
  that 
  the 
  original 
  Margaritanas 
  reached 
  

   the 
  Atlantic 
  coast 
  and 
  Greenland 
  from 
  the 
  region 
  of 
  the 
  Mississippi. 
  

   It 
  seems 
  to 
  be 
  clear 
  that 
  the 
  migration 
  did 
  extend 
  to 
  the 
  Atlantic 
  

   coast 
  from 
  the 
  close 
  relationship 
  that 
  exists 
  between 
  many 
  of 
  the 
  

   existing 
  species 
  of 
  the 
  Atlantic 
  drainage 
  with 
  those 
  of 
  the 
  Mississippi 
  

   system. 
  

  

  The 
  exact 
  relation, 
  as 
  to 
  specific 
  identity, 
  between 
  the 
  pre-Glacial 
  

   fauna 
  of 
  eastern 
  North 
  America 
  and 
  that 
  which 
  now 
  exists 
  is 
  not 
  

   known. 
  Except 
  as 
  scattering 
  fragments 
  may 
  have 
  survived 
  in 
  

   favourable 
  localities 
  along 
  the 
  coast, 
  it 
  was 
  entirely 
  exterminated 
  

   north 
  of 
  the 
  Ohio 
  by 
  the 
  ice 
  of 
  the 
  Glacial 
  Epoch. 
  From 
  that 
  time 
  

   the 
  Atlantic 
  and 
  Mississippi 
  faunas 
  as 
  a 
  whole 
  have 
  been 
  entirely 
  

  

  