366 THE BIOLOGY OF FLOWERING PLANTS 



pattern of blue or yellow rays is distinguished from a 

 pattern of four blue or yellow arms ; the former gives the 

 appearance of a composite flower, the latter of a gentian 

 flower. There is thus, in addition to a sense of colour, a 

 very distinct sense of form ; it is limited, however, to the 

 recognition of forms which have some affinity to those of 

 flowers ; purely geometrical and artificial patterns are not 

 distinguished. Von Frisch holds that the form sense, 

 including the power of distinguishing between colour 

 patterns, is chiefly important in enabhng the bee to pick up 

 the particular flower it is visiting from others of similar 

 colour. The habit of visiting only one species is very 

 strongly marked. The honey guide may help the insect 

 to probe the flower more easily, but it is more probable that 

 it forms a colour pattern which makes identification of the 

 flower to be visited easy, and that this is its chief function. 

 It is clear that there is plenty of room for experimental 

 work here. 



Using the results and methods of von Frisch, Knoll 

 (192 1, 1922 a and h) has investigated the colour sense and 

 relation to flowers of two other insects. In the flower of 

 Muscari racemosum the tips of the perianth segments form 

 a white circle surrounding the dark opening, and standing 

 out from the violet-blue ground colour of the perianth. 

 The plant is visited by various insects, of which Knoll chose 

 the humming-bird hawk-moth {Macroglossum stellatarum) 

 to work with. He imitated the colour, but not the form, of 

 the flower with suitably coloured paper ; through a hole in a 

 white circle on this sugar-water was available. The moths 

 were drilled with this artificial flower, and were then pre- 

 sented with the same appliance without the food and covered 

 by a glass plate. They flew to the white ring, and the traces 

 of their tongues on the glass sheet were all grouped round 

 the ring (Fig. 52). They did not visit the other parts of the 

 violet paper, nor did they visit white rings on yellow or grey 

 paper. Violet paper without the white ring had relatively 

 little effect. This is a good case of a honey guide. The 

 experiment does not show that the white ring helps the insect 



