424 tiiE riooioalsft. 



consolidum and of Primula veris ; from a dark purple to a bright 

 yellow variety of Viola tricolor ; and with two species of Papaver, 

 from one variety to another which differed much in colour" 

 (p. 416). 



Having the colour blue specially in mind, I have made some 

 observations on bees visiting lavender flowers. Surely, if they 

 prefer blue, they will, on leaving a bush of lavender, seek out 

 another or some other blue flower. During these observations a 

 good many bees escaped my notice on leaving the lavender, but 

 I have been able to note in the case of the following the flowers 

 they next visited : — 



(1) . From lavender to Geranium Robertianum. 



(2). From lavender to another bed of same. 



(3). From lavender to Geranium Robertianum, and then to 

 Stachys sylvestris. 



(4). From lavender to Leycesteria formosa (white flowers). 



(5). From lavender to Epilobium montanum, and then, after 

 one visit, back to lavender. 



(6). From lavender to snapdragon, of which it tried several 

 flowers unsuccessfully, and then to Stachys sylvestris. 



(7). From lavender to evening primrose (yellow). 



(8). From lavender to Geranium Robertianum. 



Thus, out of these eight cases, seven left the lavender for 

 flowers of another colour, and one only for another patch of 

 lavender. One of the seven returned to the lavender after 

 visiting one pink flower. 



On another occasion I watched some bees which were visiting 

 in large numbers some blue campanulas, which grew near a 

 bush of white Syringa in flower. During the few minutes I 

 watched them six passed from blue Campanula to white Syringa, 

 and probably many passed in the other direction ; but I only 

 noted those going from the Campanula. 



This changing from species to species is important, apart 

 from the question of colour. There is a more or less vaguely 

 received idea that bees are rather constant to the flowers they 

 visit ; that they usually keep to one species during a single 

 journey. My observations lead me to conclude that this is far 

 less the case than is usually supposed, and certainly much less 

 than required to enable them to differentiate a new species by 

 their selective action. 



