s TROCHILID A. 147 
fact we have personally ascertained, having repeatedly 
discovered minute flies in the stomachs of those we dis- 
sected. In speaking of these charming birds, the 
naturalist is almost tempted to abandon that didactic 
style best suited to his subject, and to clothe his inform- 
ation in the language of poetry ; yet both must fail in 
conveying to the mind an adequate idea of their sur- 
passing beauty. The rainbow colours of the most 
resplendent gems are here superadded to a living form, 
which in itself is exquisitely graceful and animated in 
all its movements: the flight of these pigmy birds is so 
rapid as to elude the eye ; for a few moments they may 
be seen hovering over a flower, but so soon as they have 
sipped its sweetness they vanish in an instant ; they may, 
‘in truth, be said to “ come like shadows, so depart.” 
(166.) Between the humming-birds and the last genus 
there is obviously a strong affinity, although we are un- 
acquainted with the precise link which connects the two. 
The old authors not unfrequently confounded the two 
‘races ; but they are too distinct to be mistaken even by 
a modern student. The Cinnyride have full-sized legs, 
and their wings moderate and rounded: the humming- 
birds, on the contrary, have the feet excessively short 
and remarkably small; while the wings, for the size of 
their body, are frequently longer than those of the 
‘swallows. As the sunbirds are restricted to the tro- 
‘pical latitudes of Africa and India, so are the humming- 
birds confined to America; both groups are rich in 
species; and of this, in particular, the variety of 
secondary forms is almost innumerable. We have 
endeavoured to determine the five principal genera, but 
the subgenera can only be correctly ascertained by a 
much more rigid analysis than we have yet been able to 
make. In the genus 7'rochilus, as now restricted, we 
have all those whose bills are perfectly straight, the tail 
being either even or slightly divaricated. Cynanthus 
comprehends such species as have the bill slightly bent, 
with a tail very long, and deeply forked. If we look to 
the sunbirds on the one hand, and to the hoopoes on 
“2 
