Sexual selection and parasites

What causes sexual ornaments and behavior ("secondary sexual characteristics": "primary" are the gonads and reproductive equipment themselves)? Peacock's tail, bird song repertoires, conspicuous color patterns, sexual displays

Conspicuous problem: Darwinian question

The two possibilities for sexual selection are (1) male-male competition (horns etc.) or (2) female choice. (Competition is usually among males, and choice usually by females, because females usually invest more in offspring.) We will focus on female choice.

FISHERIAN RUNAWAY model

trait initially advantageous
preference and trait become genetically correlated (trait expressed in males, preference in females)
Requirements: heritability of trait and preference. trait not too expensive (there is certainly no requirement that the trait should be expensive)
Tests/predictions:
- negative or zero correlation between male quality and trait
- genetic correlation between trait and preference (Houde and Endler compared guppy populations in different streams)
- high variability among populations in ornaments?

ZAHAVI's HANDICAP PRINCIPLE

The handicap principle says that traits are not chosen despite their being costly, but because they are costly. "Good genes" model. Ornaments are a signal of good genes: they must be costly, otherwise you can't rely on the honesty of the signal.

revealing traits: indicate male condition/quality
conditional traits: only expressed by high-quality males
viability ("Zahavi") traits: expressed by everyone, but tend to kill off low-quality males

The handicap principle seems paradoxical, but models and observation have suggested that it actually works (in part because traits are only expressed in males)

Problem with handicap principle: what maintains genetic variation for "goodness"? Under normal conditions (directional selection), one would expect that soon everyone would have "good" genes and there would be no variability (nothing for females to choose from except luck). Constant mutation could lead to a mutation-selection balance, or a variable environment could keep changing what genes are "good". Of course (as we know), parasites represent a strong driver of variability in the biotic environment ...

HAMILTON-ZUK hypothesis

(the Red Queen returns)

a subset of the handicap principle
requirements:
- females prefer resistant (showy) males (also required by Fisher)
- correlation between quality (resistance) and trait (not required by Fisher), or negative correlation between parasite load and trait
- (genetic) heritable variation in resistance (the hardest one to prove)
- Parasite load decreases host viability (otherwise there is no point in selecting resistant males)
- (genetic) heritable variation in trait: not quite as important, since H-Z works with a conditional trait

Evidence for H-Z: between-species

If H-Z is operating, we expect a positivecorrelation between parasite load and showiness across species. This correlation has been found, for example, in Hamilton and Zuk's original study which correlated human judgements of species brightness with information on ectoparasite load. Other studies have been more equivocal.

There are some problems with between-species comparisons, though:

reverse causality: showiness could cause parasitism (e.g. brighter species could attract more ectoparasites, or species that invest more in showiness could have less to invest in parasite resistance)
ecological correlates: showiness and parasitism could both be driven by other factors (hole-nesting, polygyny)
falsifiability: when a correlation is not found, how do you decide whether the data are really good enough to reject?
phylogenetic controls
dynamics: how do we know that parasite-driven sexual selection couldn't be so effective that it would drive parasite loads to lower levels in showy species?

Evidence for H-Z: within species

Guppies

Gyrodactylus turbulli infestation lowers brightness of parasitized males
it also reduces the number of sexual displays
females prefer brighter males
we don't know whether resistance is heritable
we can't rule out transmission avoidance (see below)

Barn swallows

females prefer males with longer tails (manipulative experiments, shortening and lengthening male tails)
males with longer tails tend to have fewer mites (field observation)
chicks with heavy mite loads are smaller at fledging, which leads to lower survivorship/fecundity/fitness
heritable variation in resistance: cross-fostering offspring of long-tailed males in the nests of short-tailed males (and vice versa) shows that offspring of long-tailed fathers inherit their fathers' low mite loads. This rules out the hypotheses that
- long-tailed males have just avoided parasitism by luck
- females are selecting males for paternal care (direct benefits) [and offspring are less parasitized, e.g., because they're better nourished]
- offspring benefit by avoiding direct transmission of mites from fathers

Meta-analysis

negative relationship between parasite load & expression of secondary sex characters (more in experiments)
no difference with/without parental care
no difference between behavioral and morphological traits
direct measures of immune function give stronger results than measures of parasite load
stronger for ectoparasites (transmission avoidance?)

Alternatives

Alternatives to H-Z: both can be tested (with some difficulty) by manipulative cross-fostering experiments

selection for parental care (direct benefits) [can be ruled out in species without parental care]
selection for parasite avoidance. e.g.: sage grouse & red ink "haemotomas" [can be ruled out for non-directly transmitted parasites; stronger evidence for ectoparasites suggests parasite avoidance]