﻿The 
  British 
  Species 
  of 
  Andrena 
  and 
  Nomada. 
  231 
  

  

  The 
  most 
  remarkable 
  divergence 
  found 
  in 
  our 
  British 
  

   species 
  is 
  in 
  the 
  case 
  of 
  N.flavipes 
  {solidaginis). 
  Obviously 
  

   this 
  parasite 
  belongs 
  to 
  a 
  most 
  distinct 
  group, 
  including 
  

   N. 
  tormentillae 
  {roberjeotiana) 
  and 
  obtusifrons, 
  which 
  attack 
  

   A. 
  tarsata 
  and 
  coitana, 
  but 
  itself 
  is 
  especially 
  attached 
  to 
  

   A.fuscipes, 
  belonging 
  to 
  quite 
  a 
  different 
  group 
  of 
  Andrena. 
  

  

  In 
  the 
  present 
  state 
  of 
  our 
  knowledge 
  I 
  think 
  it 
  a 
  mis- 
  

   take 
  to 
  accept 
  numerous 
  and 
  diverse 
  species 
  of 
  Andrena 
  

   as 
  the 
  host 
  of 
  a 
  single 
  Noynada 
  on 
  casual 
  observations. 
  It 
  

   is 
  quite 
  certain 
  that 
  parasitic 
  bees, 
  either 
  for 
  shelter 
  or 
  

   in 
  search 
  of 
  their 
  proper 
  host, 
  do 
  at 
  times 
  enter 
  burrows 
  

   of 
  species 
  on 
  which 
  they 
  are 
  not 
  parasitic, 
  and 
  certainly 
  

   non-parasitic 
  bees 
  that 
  burrow 
  in 
  the 
  ground 
  do 
  some- 
  

   times 
  make 
  use 
  of 
  a 
  burrow 
  not 
  formed 
  by 
  themselves, 
  

   but 
  by 
  some 
  other 
  non-parasitic 
  species, 
  just 
  as 
  we 
  know 
  

   that 
  wood-boring 
  bees 
  constantly 
  make 
  use 
  of 
  burrows 
  

   formed 
  by 
  other 
  Hymenoptera 
  or 
  by 
  Coleoptera. 
  

  

  When 
  Saunders 
  tells 
  us 
  that 
  iV^, 
  marshamella 
  parasitises 
  

   A. 
  nigroaenea 
  and 
  atriceps 
  and 
  Alfken 
  supposes 
  or 
  con- 
  

   jectures 
  that 
  nitida 
  is 
  its 
  host, 
  one 
  may 
  not 
  be 
  able 
  to 
  

   prove 
  that 
  such 
  is 
  not 
  the 
  case; 
  but 
  when 
  after 
  very 
  

   close 
  attention 
  to 
  the 
  habits 
  of 
  this 
  Nomada 
  in 
  widely 
  

   different 
  localities 
  one 
  has 
  found 
  it 
  peculiarly 
  attached 
  

   either 
  to 
  A. 
  trimmerana 
  Auct., 
  or 
  more 
  rarely 
  to 
  its 
  close 
  

   ally 
  A. 
  spinigera, 
  neither 
  of 
  which 
  are 
  mentioned, 
  one 
  

   would 
  like 
  to 
  know 
  on 
  exactly 
  what 
  evidence 
  it 
  is 
  assigned 
  

   to 
  species 
  representing 
  two 
  quite 
  different 
  groups 
  from 
  

   that 
  of 
  its 
  normal 
  host. 
  Both 
  English 
  and 
  Continental 
  

   writings 
  are 
  full 
  of 
  these 
  abnormal 
  associations 
  of 
  host 
  

   and 
  parasite, 
  and 
  since, 
  as 
  I 
  have 
  shown, 
  we 
  can 
  be 
  certain 
  

   that 
  many 
  of 
  these 
  were 
  mere 
  guesses, 
  we 
  shall 
  do 
  well 
  

   to 
  look 
  with 
  suspicion 
  on 
  all 
  such, 
  mitil 
  they 
  have 
  been 
  

   thoroughly 
  investigated. 
  

  

  In 
  spite 
  of 
  their 
  parasitic 
  habits 
  the 
  Nomadas 
  are 
  freely 
  

   attracted 
  by 
  flowers, 
  but, 
  as 
  might 
  be 
  expected, 
  in 
  general 
  

   the 
  males 
  visit 
  these 
  more 
  frequently 
  than 
  the 
  females. 
  

   It 
  is 
  noteworthy 
  that 
  some 
  species 
  are 
  particularly 
  attracted 
  

   by 
  the 
  same 
  flower 
  that 
  is 
  most' 
  attractive 
  to 
  their 
  hosts. 
  

   Thus, 
  as 
  is 
  well 
  known, 
  N. 
  armata 
  and 
  atrata 
  frequent 
  the 
  

   flowers 
  of 
  scabious 
  with 
  A. 
  hattorfiana 
  and 
  marginata, 
  

   N. 
  tormentillae 
  those 
  of 
  Potentilla 
  with 
  A. 
  tarsata, 
  and 
  so 
  

   with 
  others. 
  This 
  seems 
  particularly 
  the 
  case, 
  where 
  the 
  

   Andrena 
  affects 
  a 
  very 
  limited 
  number 
  of 
  plants^ 
  — 
  a 
  fact 
  

   in 
  itself 
  of 
  considerable 
  interest. 
  

  

  