INSECT POLLINATION 311 



the stigma of another flower. In that event the 

 quantity of pollen insufficient for anemophilly would be 

 ample for entomophilly, and, moreover, the next genera- 

 tion would inherit the colour-producing tendency, and 

 in succeeding generations those plants which developed 

 the most colour would tend to survive simply because 

 they were more attractive to insects. The secretion of 

 a small quantity of honey at first would be adventitious ; 

 it would not be a deliberate insect lure, but it would 

 happen that the first flower which chanced to secrete 

 honey would attract insects, and these visitors would 

 incidentally become pollinating agents. Here, again, 

 a small quantity of pollen would be effective where even 

 a large quantity might fail in anemophilly. And the 

 honey-producing tendency would be transmitted to 

 posterity, being stimulated by the insects themselves, 

 and become a most important factor in the success of 

 the species. The development of scent would also prove 

 of great value in entomophilly. The production of 

 flowers which would provide shelter for insects may not 

 have been an unimportant development in the direction 

 of successful pollination. It is advisable to note that 

 if, on the one hand, entomophilly has had a powerful 

 influence in the evolution of flowers, on the other hand, 

 it must also have had a considerable bearing on insect- 

 life. Insects well adapted to benefit by the good things 

 provided by flowers have survived.' Thus, only long- 

 tongued insects can secure nectar from certain flowers 

 with long-tubed carollas, and so well adapted are insects 

 to flowers and flowers to insects in these cases that flower 

 and insect stand or fall together. 



Colour, scent, and food form a trinity of irresistible 



