148 JOHN H. LOVELL 



lands where they are or were endemic, will not be questioned by 

 any orthodox floroecologist. But manifestly when they are cul- 

 tivated in widely separated stations, under the most diverse 

 conditions, there is a strong probability that in many localities 

 their normal pollinators will be entirely absent or extremely 

 rare; while the flowers themselves modified both in form and 

 function by artificial selection and hybridization may cease to 

 remain equally attractive, e.g., double flowers may be devoid of 

 both nectar and pollen. On the other hand why should we expect 

 common Hymenoptera and Diptera frequently to visit flowers 

 from which they can not legitimately obtain nectar, and to which 

 they are not beneficial ; or why should we look for diurnal insects 

 as common visitors to crepuscular flowers ? One of the advan- 

 tages of reciprocal adaptation between flowers and their polli- 

 nators is the exclusion of injurious and useless forms. 



But Plateau assumes that all bright-hued flowers, according to 

 the theories of M tiller and Knuth, no matter what their manner 

 of pollination, should frequently be visited by diurnal insects. 

 The rarity of insect visitors to many beautiful flowers with very 

 showy colors, he remarks, places the biologists of the school of 

 Hermann Muller in a singularly embarrassing position. « 



He summarizes his views as follows : 



' My observations establish the truth of the fact, well-known 

 though not sufficiently insisted upon, of the existence of many 

 plants with flowers formed on the entomophilous type and pre- 

 senting large dimensions as well as brilliant colors, which attract 

 almost no -diurnal insects. 



It follows that the attractive role, or, as it is often called, 

 vexillary role of the forms and colors of floral envelopes is either 

 nul or of very little importance. 



Causes of attraction other than colored surfaces are necessary 

 to bring pollinators to flowers and to lead them to return again 

 after a first visit; they are an odor, which is agreeable to the 

 insects, and a sweet liquid, which permits them to appease their 

 hunger and provide food for their young. * 



Unfortunately for the general acceptance of Plateau's con- 

 clusions, they are not of universal application, but are controverted 



6 Plateau, F., " Recherches experimentales sur les fleurs entomophiles peu visiters 

 par les insectes rendues attractives au moyen de liquides sucres odorants," Mem. de 

 V 'Acad. roy. de Belgique, 2me ser., 2:5, 1910. 



• hoc. cit., pp. 51-2. 



