The extravagant and long tails of many male birds are great examples of ornaments that have evolved through sexual selection. Females in such species prefer males with elaborate tails, and hence those that overcome the costs imposed by bearing fancy tail feathers are often reproductively successful. But what are the actual costs, for a bird, of sporting a large, attractive tail? Given that most birds fly, and that long tails might increase the resistance experienced by a flying animal, decreased locomotor performance seems likely. Christopher Clark and Robert Dudley at UC Berkeley wanted to examine this issue more closely and recently quantified how much a long,ornamental tail affected flight performance in Anna's hummingbirds.

Clark and Dudley captured birds in Berkeley CA, and trained them to fly over a range of speeds in a wind tunnel. To alter tail length they either glued long feathers from a different species (red-billed streamertail) onto the tails of their hummingbirds, or removed some or all of the hummingbirds'tail feathers. They also had two controls, an unmanipulated condition and a sham condition, in which hummingbird tail feathers were first removed and subsequently glued back on. To quantify locomotor performance the researchers examined two parameters: metabolic cost and maximum speed.

In respirometry experiments, birds exhibited the now familiar `U-shaped'curve relating speed and metabolism in which minimum costs are achieved at low to moderate flight speeds and costs increase at slower and faster speeds. Animals with augmented tail length incurred greater metabolic costs, on average, than controls, and this was most evident at high speeds where long tails led to an 11% increase in metabolism. Maximum flight speeds ranged between 13 and 16 m s–1 across all treatments, but animals consistently showed a small decrease in maximum speed after tail elongation(average decrease of 3.4%).

The main conclusion appears to be that the cost of bearing a long tail is pretty minimal in hummingbirds, at least in relation to steady flight performance (maneuverability might be another story). This point is even more striking when you consider that the tails used to augment length in these experiments are among the longest found in any species of hummingbird –more modest tail elongation presumably is even less costly. Moreover, the greatest increases in metabolism were found at the highest flight speeds,which animals use only about 5% of the time in the wild. If these ornamental tails are in some sense `handicapping' the males, they certainly don't seem to be doing so very severely when it comes to steady locomotion. And if females are using variation in tail length as a cue for interpreting male `quality',the signal to noise ratio doesn't seem likely to be very good. Subtle,intraspecific differences in tail length among males are likely meaningless in terms of their effects on flight costs.

More quantitative work evaluating the costs of sexually selected ornaments is required before large, sweeping generalizations can be made. However, this work on the tails of hummingbirds, as with work on the horns of dung beetles,suggests that evolving elaborate structures to attract females or combat other males may not necessarily always come at as much of a cost as we once imagined.

Clark, C. J. and Dudley, R. (
). Flight costs of long, sexually selected tails in hummingbirds.
Pro. R. Soc. B