178 INTRODUCTION 
that it does not discover and utilize. Even the deeply hidden nectar-receptacles of 
Iris are found and sucked by Rhingia. In anthophilous insects, the power to detect 
hidden nectar increases pard passu with the structural adaptations for securing it. 
When Sprengel described flies as stupid insects, incapable of finding nectar which lay 
concealed, that statement applied to the great majority of short-tongued forms, but 
not at all to forms with long proboscides, such as Syrphidae, Bombyliidae, Conopidae, 
and Empidae. 
Even in the Syrphidae, however, only a few species have acquired so highly 
specialized a proboscis as Eristalis: the great majority are in a much lower stage 
of adaptation, as shown in Fig. 75. The labium is much shorter, its extensible 
middle joint is wanting, the terminal flaps are swollen and cushion-shaped, and 
there is a corresponding diminution in intellectual power 
in regard to the spoliation of flowers. This is indicated 
by the large amount of variation in the length of the 
proboscis, as follows :— } 
mm. 
Syrphus-balteatis= 6 6.4. on eae 2 
a) MibesitG, Gi...) Bo ee 
Eristalis arbustorum .».. 5 0.0. °. 4.9 ee 
Helophilus trivittatus. . . . . . . 6-4 
Eristalis tenax +8) J) sone 9. oe 
YVolucella bombylais vy.) Pua 0 Se 4 8 
Rhingia rostrata -4. 0. 00a) 2) ae 
With reference to this dissimilarity in the length 
of the proboscis of species belonging to the same family, 
Loew very rightly remarks (‘Blumenbesuch,’ II, p. 117) 
Fic. 75. Proboscis of Syrphus that it is necessary to test each Syrphid as regards its 
balteatus, Deg.; seen from below stage of adaptation. For this, however, the observations 
(after Herm. Miller). References 4 4 é 
as in Fig. 73. so far recorded are insufficient. Loew thinks that there 
may possibly be a continuous series of transitional 
stages between purely allotropous forms and well-marked hemitropous ones. 
Loew (‘ Blumenbesuch,’ II, p. 116) calls attention to the fact that the more 
or less well-developed feathering of the antennae of some species is of importance 
for the transfer of pollen from flower to flower. It is especially well marked in the 
genus Volucella. The curved pendent dorsal brush of the antennae is more than 
2 mm. long in V. bombylans, while the lateral branches are about 1 mm., and 
numerous pollen-grains are not infrequently found adhering to the dense feathery 
hairs. As Loew further remarks, the dorsal brush forms the most prominent part 
of the head, so that the fly in alighting on flowers will—so far as the structure 
of these permits—in many cases deposit pollen-grains on the stigma. An equally 
well-developed feathering of the dorsal brush also occurs in Sericomyia and Arcto- 
phila, while in Cheilosia and Eristalis there are species with naked dorsal brush 
as well as others where it is feathery. 
Loew further regards the hairs below the eyes of many Syrphidae (Leucozona, 
Volucella, Sericomyia, Arctophila, Eristalis, and species of Cheilosia) as an adaptation 
