1981] 
Tepedino & Parker — Flower-Visiting Insects 
325 
Discussion 
In this study, foraging insects did not appear to react to the 
presence of other insects in choosing flowers. Only 2 of the censuses 
on onions and none of the censuses on sunflowers displayed a 
significant departure from a random distribution (Table 1). Sun- 
flower foragers (Apis, Perdita, Halictus) frequently entered the 
flower by landing on the back of the petals or on the involucral 
bracts and then crawling onto the head. If occupancy by another 
insect were important, this would be an inefficient method of 
choosing a flower. In a similar study Waddington (1976) also 
concluded that halictid bees were foraging independently on bind- 
weed ( Convolvulus arvensis). None of the abundant species 
present appeared to forage other than randomly with respect to 
other flower occupants. This was especially surprising for honeybees 
which have been reported to more readily tolerate, or even form, 
clumped distributions (Kalmus 1954, Benest 1976). However, con- 
tagious distributions of honeybees may occur only under unusual 
circumstances; the data of Kalmus (1954) were gathered from a 
small number of feeding dishes and are quite artificial. Benest’s 
(1976) suggestion that honeybees are more tolerant of joint foraging 
than bumblebees does not stand close examination. Additional 
study is required before such conclusions are warranted. 
Instead of using the presence of insects on inflorescences as cues, 
some flower-visiting insects may make selections based on the 
number of open flowers or the amount of nectar or pollen available. 
Although all heads censused had some open flowers, some had more 
open flowers than others and insects may have been choosing those 
heads with more flowers irrespective of other visitors. Even if heads 
were equivalent in number of flowers, continuous removal of nectar 
and pollen by foragers would cause variation in resource availability 
between heads (e.g., Pleasants and Zimmerman 1979) and insects 
may be responsive to such variation prior to landing on a flower. 
For example, Thorp et al. (1975) have suggested that the fluorescent 
nectar (and perhaps pollen) of many species with open flowers may 
be used as a cue by foraging insects (see also Kevan 1976, Thorp et 
al. 1976); and onion nectar is intensely fluorescent (Thorp et al. 
1975). Recently Heinrich (1979) has shown that bumblebee foragers 
reject many more nectar depleted (recently visited) white clover 
( Trifolium repens) heads than heads with abundant nectar. Rejec- 
