SUMMER FLOWERS 



other colour, that they frequent reds with a tinge of purple, 

 but leave cinnabar reds alone. In such observations it is 

 necessary to be very careful. A flower is a complex of 

 stimuli, and the fact that blue-violet blossoms are most 

 visited does not, in itself, prove that they are preferred 

 because of their colour. 



Sir John Lubbock, now Lord Avebury, tried to get away 

 from the risk of fallacy by baiting different slips of paper 

 of different colours with the same sugar, and the interesting 

 and valuable result of his experiment was that the bees 

 " preferred " the blues and violets to yellows, and so on. 

 Even here, however, one must tread warily, since it seems 

 likely that the bee has established associations partly 

 instinctive and partly based on previous experience, so that 

 the stimulus of certain colours may act merely as the 

 memorandum of previous good feeding. And, again, there 

 seems good sense in Plateau's objection, that observations 

 and experiments on this subject have not discriminated 

 adequately between colour and absolute intensity of illumi- 

 nation. Those insects that love light choose the brightest 

 surfaces; but brightness is one thing, and colour is 

 another. 



Plateau worked a good deal with dahlias, which are 

 visited by humble-bees, butterflies, and other insects. His 

 method was to hide the colour and form of the inflorescence 

 by means of pieces of cardboard, which were variously 

 coloured (white, black, green, etc.), or by means of green 

 leaves. The results of his experiments led him to a heretical 

 conclusion : " The form and the colour do not seem to have 

 any attractive role ; the insects are evidently guided to the 

 capitula of the composites by some other sense than sight 

 probably by smell." Now, we do not believe that these and 

 similar experiments have upset the theory that floral colour 

 is one of the attractive stimuli which draws insects to the 



