﻿50 
  Rydberg: 
  Notes 
  on 
  Rosaceae 
  

  

  Leaves 
  decidedly 
  pubescent 
  beneath. 
  I5- 
  -R- 
  Lyonu 
  

  

  Leaves 
  glabrous 
  or 
  pubescent 
  only 
  on 
  the 
  veins 
  

  

  beneath. 
  

   Leaflets 
  shining 
  above; 
  flowers 
  usually 
  corym- 
  

   bose; 
  prickles 
  as 
  a 
  rule 
  stout 
  and 
  flattened 
  

   at 
  the 
  base. 
  I4- 
  -K- 
  virginiana. 
  

  

  Leaflets 
  not 
  shining 
  above; 
  flowers 
  usually 
  

  

  solitary, 
  rarely 
  a 
  few 
  together; 
  prickles 
  

   slender, 
  terete. 
  

   Leaflets 
  not 
  glandular-dentate. 
  

  

  Plant 
  low, 
  usually 
  less 
  than 
  2 
  dm. 
  high; 
  

   leaflets 
  less 
  than 
  2 
  cm. 
  long, 
  dark 
  

   green; 
  prickles 
  3-5 
  nim. 
  long, 
  stout, 
  

   often 
  curved- 
  I3- 
  ^- 
  nanella. 
  

  

  Plant 
  usually 
  more 
  than 
  2 
  dm. 
  high; 
  

   leaflets 
  usually 
  more 
  than 
  2 
  cm. 
  

   long; 
  prickles 
  more 
  than 
  5 
  mm. 
  

   long, 
  slender. 
  

   Leaflets 
  broadly 
  oval. 
  16. 
  R, 
  obovata. 
  

  

  Leaflets 
  elliptic 
  or 
  narrowly 
  oval. 
  17. 
  R, 
  Carolina, 
  

  

  Leaflets 
  glandular-dentate 
  and 
  rachis 
  usu- 
  

   ally 
  glandular. 
  18. 
  R. 
  serrulata. 
  

  

  Leaflets 
  narrowly 
  elliptic 
  or 
  lance-oblong, 
  usually 
  9- 
  

  

  ■ 
  F 
  

  

  II, 
  at 
  least 
  on 
  the 
  new 
  shoots. 
  19. 
  R- 
  nitida. 
  

  

  Hypanthium 
  decidedly 
  pyriform. 
  or 
  ellipsoid, 
  long-tapering 
  

  

  at 
  the 
  base. 
  

  

  20. 
  R, 
  Bickyiellii* 
  

  

  10, 
  Rosa 
  palustrts 
  Marsh. 
  

  

  This 
  has 
  usually 
  been 
  known 
  as 
  Rosa 
  Carolina 
  L. 
  It 
  is 
  the 
  

   R. 
  Carolina 
  of 
  the 
  second 
  edition 
  of 
  Linnaeus's 
  Species 
  Plantarum, 
  

   but 
  not 
  that 
  of 
  tiie 
  first. 
  It 
  is 
  evident 
  that 
  Linnaeus, 
  when 
  he 
  

   prepared 
  the 
  manuscript 
  for 
  the 
  first 
  edition, 
  did 
  not 
  have 
  any 
  

   specimens, 
  but 
  based 
  his 
  R. 
  Carolina 
  wholly 
  on 
  the 
  plate 
  and 
  

   description 
  in 
  Dillenius's 
  Hortus 
  Elthalmensis, 
  pL 
  245^ 
  /- 
  316* 
  

   This 
  plate 
  does 
  not 
  represent 
  R. 
  Carolina 
  as 
  usually 
  understood, 
  

   but 
  the 
  species 
  commonly 
  known 
  as 
  R. 
  humilis 
  Marsh. 
  In 
  

   discussing 
  R, 
  Carolina, 
  Cr^pin 
  states: 
  "The 
  latter 
  description, 
  

   i.e., 
  that 
  of 
  Hortus 
  Elthalmensis, 
  and 
  the 
  figure 
  can 
  apply 
  either 
  

   to 
  -R. 
  humilis 
  Marsh, 
  or 
  to 
  R. 
  hicida 
  Ehrh., 
  but 
  not 
  to 
  R. 
  Carolina 
  

   as 
  we 
  know 
  it 
  today. 
  Besides 
  the 
  branch 
  which 
  I 
  have 
  seen 
  in 
  

   Sherard's 
  herbarium 
  with 
  Dillenius's 
  name 
  on, 
  belongs 
  to 
  -R- 
  

   humilis 
  Marsh 
  {— 
  R. 
  parviflora 
  Ehrh.). 
  The 
  result 
  is 
  that 
  the 
  

   first 
  name 
  for 
  R. 
  humilis 
  Marsh, 
  is 
  R. 
  Carolina 
  L. 
  Sp., 
  ed.i. 
  

   Cr6pin 
  therefore 
  arrived 
  at 
  the 
  same 
  conclusion 
  as 
  I. 
  The 
  oldest 
  

  

  