The cost of sexual selection: a study in pheasants

We’ve known for a long time that sexual selection—ultimately caused by differences in gamete size—can produce marked differences in the appearance and behavior of males versus females within a species. Often males are more ornamented than females, with bright colors and long feathers or ornaments on the head.  We also know that colors and ornamentation of males puts them at a disadvantage in certain respects, as they are more easily detected by predators than are the females, or have difficulty flying because of exaggerated feather displays. This disadvantage also applies to sexually-selected “weapons” like deer horns and moose antlers, which are shed and have to be regrown, at great metabolic expense, each year.

Perhaps the most famous of these features is the tail of the peacock, in which males have long, decorated, and spreadable tails that females lack.  We are pretty sure that this difference is due to sexual selection because experiments show that the “eyespots” on the male tails attract females: the more eyespots you have, the higher chance you have of reproducing. Thus the genes for exaggerated tails accumulate via sexual selection by females.

Of course female preference plays a key role here, as that preference has to exist to give more elaborate males a reproductive advantage.  We don’t fully understand, however, exactly why females prefer many exaggerated male traits. In some cases, like the orange-red color of the male house finch, we have an answer.  As I said, there are also costs of sexually-selected male traits like big bodies (elephant seals) or antlers (moose), who use them to directly fight for access to females. (Darwin called this the “law of combat”.)

But in most cases we don’t understand why females prefer certain bright colors or long tails, though we have theories that are largely untested. This difference in patterning and color was called “the law of beauty” by Darwin, who was the first person to suggest the idea of sexual selection (1871).

Both forms of sexual selection show that this type of selection—really a subset of natural selection—involves tradeoffs.  Males sacrifice flight ability, become more obvious to predators, and have to re-grow antlers and horns each year, which are considerable disadvantages. But those have to be more than compensated for by either the success in combat or the increased attractiveness to females of males with those traits—otherwise the exaggerated traits would not have evolved.

A new paper in Biology Letters (click title screenshot below) shows a novel form of tradeoff in pheasants, and the first such tradeoff known in any animal. In two species of pheasants, males have evolved “capes” around their neck that, when expanded, occlude the male’s visual field (but not the female’s), as well as head feathers that also appear to block the male’s vision.  These are sexually selected traits.  Noticing them, the five authors hypothesized the tradeoff: in the two species of pheasant with head and neck ornamentation (the Golden and Lady Amherst pheasant), they tested whether the male’s head feathers blocked part of his visual field compared to females in the same species.  As a control, they used two pheasant species (Silver pheasants and Green pheasants), in which males don’t have head ornamentation that would block the visual field.

The authors then measured the visual field of males and females of all four species, and, lo and behold, males of the Golden and Lady Amherst’s pheasants did have a considerable blockage of the vertical field of vision compared to conspecific females, while there was little or no difference between the sexes in the two control species.

Click the title below to read the original paper for free, or find the pdf here. There is also a brief précis piece in Science if you want the abridged version.  The quotes and figures below come from the original paper, while the six full-bodied photos of the pheasants come from Wikipedia (credits shown).

First, the birds.

A male Golden pheasant, Chrysolophus pictus:

. . . and a female Golden pheasant. The sexual dimorphism is bloody obvious.

The heads of males (l) vs. females (r) of the Golden Pheasant, taken from the paper itself. You can see how the male’s head feathers could occlude its vision.

The one other “experimental” species with male vision-occluding feathers.

Male Lady Amherst’s pheasant, (Chrysolophus amherstiae):

A female Lady Amherst’s pheasant:

And one of the two control species, the Green Pheasant, (Phasianus versicolor). First, a male, with vision not impeded by a crown. (The other control species, the Silver pheasant, Lophura nycthemera, isn’t shown.) Both of the control species show sexual dimorphism of color and plumage in the expected direction, but there are no feathers on the male’s head that could block his vision.

And a female:

How did they measure the visual field of males and females? They simply put the pheasants in a padded box and fixed their heads firmly so that they could not move. (No pheasants were harmed in this study, which is excellent.) Then, to measure whether an eye could see at a certain angle, they shined a light on the eye. If there was a reflection from the retina at the back of the eye, that meant the bird could see the light from that angle. By performing many tests at various angles around the head, the researchers were able to judge the field of vision of each bird. They could also do this in pheasants whose heads were tilted up or down (see below).

The differences were most pronounced in the vertical line of sight. For example, as shown below, when the head is horizontal or looking down,  the male of the Golden pheasant sees 30° less above his head than does the female.  This would be a problem because, as the authors say, “Sexually selected traits such as feather ornamentation of male birds can act as an impediment to movement and predator detection.”  When you’re a male pheasant busily foraging on the ground, which is how they eat, you may not see an approaching predator. That is the cost of the sexual selection that produced head and neck feathers. (The figure says this is a Lady Amherst’s pheasant but it is apparently a Golden pheasant.)

Here are all four species.  The Lady Amherst’s pheasant has an even more severe impediment of vision in the male: he can see vertically a full 40° less than do conspecific females.  In contrast, the sex difference in the control species is much less: a mere 5° reduction in males in the Silver pheasant and no difference in the green pheasant.

The figure below in the paper gives a three-dimensional depiction of a bird’s view, with males on the left and females on the right. You can see that the males are effectively blind (black area) over a much larger space than are the females, and that space is mostly above the bird’s head. Since pheasants are ground foragers, blacking-out of “down” vision would be a very serious impediment, making males unable to locate food. Blocking “up” vision would surely have a smaller cost.

The upshot is that the authors’ hypothesis is supported: males but not females in the pheasants having feathers around their eyes appear to have occluded vision, mostly above their heads.  Now we don’t know whether this occluded vision translates into a loss of fitness at all, much less a loss that is outweighed by the gain in fitness caused by the head and neck ornamentation.  Trying to answer questions about fitness is nearly impossible, as you’d have to measure survival and offspring production of males who have bigger and smaller feathers within a species (would you have to give the birds a haircut?). But there is a period of moulting in which males lose their head and neck feathers, and at least researchers could measure the field of vision, and perhaps foraging efficiency, during that period.  Nevertheless, I do suspect that occluded vision reduces fitness, and that the head ornamentation more than compensates for it.

Besides these results, the paper does show how natural selection and adaptation involves tradeoffs.  There are usually no mutations that are “universally” adaptive in that they convey a benefit without any cost. As I said, natural selection will favor the increase in frequency of mutations that produce net reproductive benefits to the individual that outweigh the costs.