Picture a man who has a waywith the ladies, and a characternot unlike James Bond may

spring to mind. He’s clever, classy, fear-less and flashy—characteristics that arealmost universally appealing to the op-posite sex. Throw in the powerful sportscar, and you have a nearly irresistiblecombination.

That females often flock to the mostostentatious males is not a phenome-non unique to humans. In many differ-ent species, successful males—those thatsire the most offspring—are often largeror more brightly colored or “show off”with more vigorous courtship displays.

Females tend to be the choosier sexwhen it comes to selecting a mate, part-ly because males can produce millionsof sperm, whereas females’ eggs are fewand far between. Thus, females may bemore selective because they have moreinvested in each gamete and in the re-sulting offspring. And because the avail-ability of eggs is a limiting factor in re-productive success, males tend to com-pete for female attention and not viceversa.

Charles Darwin was the first to pro-pose that competition for mates playsan important role in reproductive suc-cess—a process he dubbed sexual selec-tion. In The Descent of Man, and Selec-tion in Relation to Sex, published in1871, Darwin hypothesized that anytrait that gives a male mating and fertil-ization advantages will evolve in a pop-ulation because males with such traitswill produce more offspring than theircompetitors. Assuming the trait is heri-table, offspring expressing the beneficialtrait will, in turn, achieve greater repro-

ductive success than their competitors,and so on, through future generations.Further, Darwin proposed that some ofthese traits may have evolved becausethey attract the attention of females.

The idea that females are discriminat-ing and can actively choose with whomto mate was controversial from its in-ception—perhaps because male-malebattles can be quite spectacular. Malesmay fight amongst themselves, occasion-ally in dramatic battles to the death, togain mating privileges with females. Incomparison, female choice is generallymuch more subtle.

Finding Mr. Right

Over the past 25 years, a consider-able body of scientific evidence in

support of female choice has accumulat-ed. Females actively choose their matesin a large variety of species—particular-ly ones in which males are less aggres-sive and display individual differences insecondary sexual characteristics, suchas ornamental plumage or courtshipdisplays. Nevertheless, how and whyfemales select their partners and howmating preferences have evolved remainhotly debated issues among evolution-ary biologists.

A choosy female faces two general

tasks in selecting a mate. First, shemust search for and locate a male.This task can be difficult if the pop-ulation is sparse or if the danger ofpredators prevents her fromspending a good deal oftime searching for asuitable mate. Onceshe has encountereda male, the femalemust then decidewhether to accept orreject him as a mate.The decision ofteninvolves some shop-ping around. In cer-

56 Scientific American April 1998

How Females Choose Their Mates

Females often prefer to mate with the most flamboyant males. Their choice may be based on

a complex interaction between instinct and imitation

by Lee Alan Dugatkin and Jean-Guy J. Godin

FEMALE TRINIDADIAN GUPPIES do the choosingwhen it comes to selecting a mate. Generally speaking,female guppies prefer males that are brighter or moreorange in color (upper right). But even guppies areprone to social pressure. If, for example, an older femaleappears to fancy a drabber male, a young female mayignore her instincts and choose to copy her elder’s mateselection (lower left).

Copyright 1998 Scientific American, Inc.

tain mating systems, females may en-counter a group of available males andcan compare them on the spot. For ex-ample, in early spring, male sage grouse(Centrocercus urophasianus) aggregate“cheek-to-jowl” in temporary commu-nal mating arenas called leks, wherethey strut their stuff for the females. Afemale typically observes the displays ofa number of males, apparently compar-ing them before mating with one luckysuitor. She then leaves the lek to nest

and raise her brood elsewhere. Of allthe potential mates on a lek, a few pre-ferred males receive the bulk of the fe-male attention.

But males are not always convenient-ly displayed like chocolates in a sam-pler box. More commonly, females en-counter males one at a time. Comparingmales in this case is presumably a morechallenging cognitive task, as it involves

remembering the characteristics of anindividual that is no longer in sight.Studies have shown that females canrank the characteristics of sequentiallypresented males. Theo C. M. Bakkerand Manfred Milinski of the Universityof Bern in Switzerland found that fe-male three-spined sticklebacks (Gaster-

ROBE

RTO

OST

I

MALE

MALE FEMALE

YOUNGERFEMALE

OLDER FEMALE MALE

MALE

Copyright 1998 Scientific American, Inc.

osteus aculeatus) will tailor their matechoice to the relative attractiveness ofthe present and previously encounteredmales. Females were more likely to showinterest in a male if his red nuptial col-oring was brighter than the previousmale’s and more likely to reject a suitorwhose coloring was less bright than hispredecessor’s.

Whether a female chooses her matefrom among a dozen dancing grouse orbetween a pair of crimson fish, she gen-erally selects the most conspicuous con-tender. Empirical evidence indicates thatfemales commonly prefer male traitsthat most strongly stimulate their sens-es. (This evidence has recently been re-viewed by Malte Andersson of the Uni-versity of Göteborg in Sweden and byMichael J. Ryan of the University ofTexas at Austin and Anne C. Keddy-Hector of Austin Community College.)For example, when given a choice, fe-male green tree frogs (Hyla cinerea) arepreferentially attracted to males thatcall the loudest and most frequently; fe-male guppies (Poecilia reticulata) to themost brightly colored males; and femalemallards (Anas platyrhynchos) to malesthat court them most frequently. Becauseof such preferences, males have typical-ly evolved exaggerated secondary sexu-al traits to attract the opposite sex.

Why Be Choosy?

Even though evidence indicates thatfemales can actively choose their

mates, the question of why females dis-criminate, rather than mate at random,remains largely unresolved. How did fe-

male choice originate and evolve? Whatare its benefits and costs to individualfemales?

In some cases, females may favor mat-ing with a male that is loud or brightlycolored simply because he is easy to lo-cate. Reducing the amount of time ittakes to find a mate may reduce a fe-male’s risk of being killed by a predator.But for many species, mate choice isprobably more complex. For many birdsand mammals, natural selection appearsto favor females who choose mates thatprovide them with some direct benefitthat will increase their fecundity, theirsurvival or the survival of their offspring.Such benefits might include food, a safehaven or even the prospect of fewerparasites.

In a long-term study of the barn swal-low (Hirundo rustica), Anders P. Møl-ler of the CNRS in Paris observed thatfemales prefer to mate with males pos-sessing elongated tail feathers. As it turnsout, the long-tailed males are infectedwith fewer bloodsucking mites thantheir short-tailed counterparts. Becausethese parasites can jump from bird tobird, females that mate with long-tailedmales benefit by avoiding infection andby producing greater numbers of health-ier chicks than females that mate withshorter-tailed males. Unfortunately, be-cause selecting a mate that offers directbenefits seems so obvious, few studieshave tested this evolutionary model in arigorous way.

When males provide no obvious re-sources, such as food or protection, fe-males may choose to mate with themales that appear to have the best genes.

How do they know which males havegood genes? And why don’t males justcheat by faking the traits associatedwith such genes? In 1975 Amotz Zahaviof the University of Tel Aviv in Israelsuggested that females assess only thosetraits that are honest indicators of malefitness—a hypothesis known as the han-dicap principle. Honest indicators, whichare “costly” to produce and maintain,should be associated with the most vig-orous males.

While studying antipredator behaviorin the Trinidadian guppy, we recentlyobtained some evidence that is consis-tent with the handicap principle. Whena predatory fish nears a school of gup-pies, males, often in pairs, cautiously ap-proach the potential threat to “inspect”it. Such risky behavior has been observedin many species, and behavioral ecolo-gists have suggested that bold males mayswim close to a predator to advertisetheir vigor to nearby females. In fact,laboratory studies have shown that whenno females are present, no male guppyplays the hero by approaching the pred-ator more often than his counterpart.

We hypothesized that boldness exhib-ited during predator inspection might beattractive to females because it shouldbe a reliable indicator of fitness. Lessvigorous guppies who tried to “fake”competence in predator inspectionwould likely be eaten. By using small,custom-built containers that allowed usto position males at different distancesfrom a predator fish, we found that fe-males indeed preferred the most intrep-id males. Such courage appears to cor-relate with color: the males that swim

How Females Choose Their Mates58 Scientific American April 1998

MALE GUPPIES inspect predators; female guppies inspect themales. When a predator—such as the cichlid pictured here—ap-proaches a school of guppies, a pair of males often swims overto inspect the potential threat. Such bold behavior may be at-tractive to females, which tend to choose as a mate the suitor

that swims closest to the predator (left). Although the bravestmales are often the most colorful, females will choose a lessflashy contender if he appears to be more courageous than hisinspection partner (right). In the laboratory, custom-made con-tainers allow the authors to position the males.

ROBE

RTO

OST

I

MALE

MALE

FEMALE

PREDATORMALE

MALE

FEMALE

PREDATOR

Copyright 1998 Scientific American, Inc.

closest to the predator are usually themost colorful. Thus, in the wild, femalesmay have evolved a preference for theflashier males because color is a proxyfor boldness and fitness.

Once females have expressed a pref-erence for a certain trait, a process calledrunaway selection can occur. The mod-el, first brought to the attention of evo-lutionary biologists by Ronald Fisher in1958, suggests that a male trait and thefemale preference for that trait coevolve.For example, females that prefer to matewith large males should produce largesons as well as daughters that show apreference for large males. Under cer-tain conditions, this process can esca-late, producing increasingly exaggerat-ed male traits and stronger female pref-erence for those traits.

A number of behavioral ecologistshave found some evidence for runawaycoevolution of orange body colorationin male guppies and of female prefer-ence for this male trait. But a more con-vincing example of runaway selectionhas recently been presented by GeraldS. Wilkinson and Paul Reillo of the Uni-versity of Maryland in their study of thestalk-eyed fly (Cyrtodiopsis dal-manni). In this species, femalesgenerally prefer to mate withmales possessing widely spacedeyes. By selectively breeding theflies for 13 generations, Wilkin-son and Reillo generated oneline of flies in which the maleshad large eyestalks and anotherline of shorter-stalked males.They found that females in eachline preferred the male trait se-lected for in that line—that is,females from the large-stalk linepreferred males with the longest

stalks, and females from the short-stalkline preferred shorter-stalked males. Fe-male preference thus coevolved withthe selected male trait.

How do preferences about mate choiceoriginate? In some cases, females mayhave a preexisting sensory bias for a cer-tain trait, not because it represents any-thing but because it attracts attention—a hypothesis championed most promi-nently by Ryan and by John Endler ofJames Cook University in Australia. Forexample, female swordtails (Xipho-phorus helleri) prefer males with long“swords” on their tail fins. And al-though males of a related species—theplatyfish Xiphophorus maculatus—lackswords completely, Alexandra L. Baso-lo of the University of Nebraska foundthat when she attached artificial, plasticswords onto these naturally swordlessmales, female platyfish showed an im-mediate, strong and consistent prefer-ence for the males with the counterfeitswords. In other words, platyfish fe-males harbored a preexisting bias forlong swords, even though swords re-veal nothing about the fitness of platy-fish males.

These evolutionary models may beoperating separately or in conjunction;it is difficult to untangle them experi-mentally. Female guppies, for instance,may be partial to orange males becausebright coloring is a proxy for boldnessor for good health (males with thebrightest pigments are probably eatingwell). But the preference could haveoriginated because females are more at-tuned to colors of a particular wave-length and then further evolved througha runaway mechanism.

All these models assume that femalepreference is genetically determined. Re-cent studies indicate, however, that so-cial factors, such as imitation, also influ-ence mate choice.

Copycat Birds and Fish

Some guys get all the girls. On acrowded grouse lek, for example,

the top male may receive 80 percent ofthe mating opportunities. Is he simplyirresistible? Or do females take one an-other’s choices into account when se-lecting a mate? In the early 1990s agroup of Scandinavian researchers, led

How Females Choose Their Mates Scientific American April 1998 59

RUNAWAY SELECTION mayshape mate preferences in stalk-eyed flies. Females of the speciesnormally choose to mate withmales that sport the longest stalks(top center). But when research-ers used selective breeding tech-niques to generate two lines offlies—one in which males havelong stalks (left), the other inwhich males’ stalks are short(right)—they found that femalepreferences evolved along withmale stalk length. Females fromthe long-stalk line were partial tomales with longer stalks (bottomleft), and females from the short-stalk line preferred shorter-stalkedmales (bottom right). DRA

WIN

GS

BY R

OBE

RTO

OST

I; D

IAG

RAM

BY

JEN

NIF

ER C

. CH

RIST

IAN

SEN

LONG-STALKED MALES PREFERRED

LONG-STALKED MALES PREFERRED SHORT-STALKED MALES PREFERRED

Copyright 1998 Scientific American, Inc.

by Jacob Höglund and Arne Lundbergof Uppsala University and Rauno Alat-alo of Jyväskylä University, initiated adetailed study of mate-choice copyingin the black grouse (Tetrao tetrix). Us-ing stuffed dummies to represent inter-ested females, the researchers showedthat female grouse mated preferentiallywith the male that appeared to haveother females in his territory.

Why copy? Perhaps imitation teachesfemales what to look for in a male. Inan extensive series of experiments onmate-choice copying in guppies, we de-termined that young females are morelikely to copy the mate choice of older,more experienced females than vice ver-sa. Further, copying may save time. Re-lying on the judgment of others may al-low a female to assess a potential matequickly and efficiently, leaving her moretime to forage for food or hide frompredators.

For species in which females copy, afascinating question emerges: Howmuch of female mate choice is based on

instinct and how much on imitation?To tease apart the relative contributionsof genetic and social factors involved inmate choice in guppies from the PariaRiver in Trinidad, one of us (Dugatkin)carried out a behavioral “titration” ex-periment. First, a female guppy was al-lowed to choose between two malesthat differed in the amount of orangethat covered their bodies. As expected,females virtually always chose the moreorange of a pair of males. Then a copy-ing opportunity was staged, in whichthe test female was allowed to observeanother female apparently choosing theless orange male as her putative mate.

Which male did she then choose forherself? Remember that the female’s ge-netic predisposition is “pulling” her to-ward the more orange male, but socialcues and the potential to copy are tug-ging her toward the drabber male. Inthe end, her choice depended on howmuch the males differed in coloration.When the paired males differed by small(12 percent) or moderate (25 percent)amounts of orange, the female consis-tently chose the less orange of the two.In this case, the female succumbed topeer pressure, her tendency to copyoverriding her genetic preference for or-ange males. If, however, the males dif-

How Females Choose Their Mates60 Scientific American April 1998

LARA

JO

REG

AN

Gam

ma

Liai

son

Net

wor

kD

AVID

C. F

RITT

S An

imal

s Ani

mal

s

SEXUAL SELECTION comes in manystyles. During mating season, male muledeer challenge their rivals to battle for theright to claim the available females. Malekangaroos also fight for their mates. Butin species in which female choice is themain mode of sexual selection, males haveevolved bold traits designed to attract theattention of females. Male peacocks showoff by being spectacularly colored and me-ticulously groomed. Human males employstrategies that are somewhat less obvious.

M. A

UST

ERM

AN

Ani

mal

s Ani

mal

s

ERW

IN A

ND

PEG

GY

BAU

ER B

ruce

Col

eman

, Inc

.

Copyright 1998 Scientific American, Inc.

fered by a large amount (40percent) of orange, the fe-male ignored the seeminglybad advice and chose themore orange male, her ge-netic predisposition maskingany copying effects.

It appears as if there existsin guppies a color threshold,below which social cuessteer female mate choice andabove which genetic factorspredominate. Dugatkin isperforming further experi-ments to assess whether copy-ing behavior in guppies is it-self heritable. Although imi-tation appears to be basedon social cues, perhaps genesgovern the likelihood that afemale guppy will engage incopying behavior.

Sadie Hawkins Day

Although people are more complex than guppies

and grouse, some of the samemate-choice rules may applyto human dating games. Ac-cording to popular wisdom,it is human females who arethe choosier sex when itcomes to selecting a mate.As a species, humans meet the criteriafor female choice: men, for the mostpart, will avoid fighting to the death forthe hand of a young maiden. And fe-males can distinguish between variousmales on the basis of differences in theircharacteristics: some men are brasher,some are brighter and some have biggerbank accounts.

Women may even engage in mate-choice copying. After all, imitation isimportant in many types of human

learning. To determine whether copyingplays a role in how women rate a man’sattractiveness, Dugatkin is currentlycollaborating with social psychologistsMichael Cunningham and Duane Lundyof the University of Louisville. Althoughtheir results are preliminary, they findthat women are more likely to expressan interest in going out with a man ifthey are told that other women also findhim attractive.

Of course, evolutionary theory will

never be able to explain fully singles bars,personal ads or cyber-romance. Evenfor animals, it appears that the benefitsand costs of being choosy when select-ing a mate differ for different species, indifferent environments and sometimesat different times of day. In any case, ifanimals as simple as guppies can con-sider the opinions of their peers whenchoosing a mate, imagine how complexthe cues must be that guide humans intheir search for the perfect mate.

How Females Choose Their Mates Scientific American April 1998 61

The Authors

LEE ALAN DUGATKIN and JEAN-GUY J. GODIN first joinedforces in Trinidad, where they became fascinated by the mating be-havior of guppies. An evolutionary biologist, Dugatkin has beenan assistant professor of biology at the University of Louisvillesince 1995. He received his Ph.D. in biology from the State Uni-versity of New York at Binghamton in 1991. His research interestsinclude the evolution of cooperation and altruism and the interac-tion of genetic and social factors in shaping behavior. Godin, a be-havioral ecologist, is professor of biology at Mount Allison Uni-versity in New Brunswick, Canada, where he has been on the fac-ulty since 1981. He received his doctorate in zoology from theUniversity of British Columbia and has been a visiting fellow at theUniversity of Oxford. His research focuses on the behavioral ecol-ogy of antipredator, foraging and mating decisions in animals.

Further Reading

Sexual Selection. M. Andersson. Princeton University Press, 1994.Interface between Culturally Based Preferences and Genet-ic Preferences: Female Mate Choice In POECILIA RETICULATA.L. A. Dugatkin in Proceedings of the National Academy of SciencesUSA, Vol. 93, No. 7, pages 2770–2773; April 2, 1996.

Female Mating Preference for Bold Males in the Guppy, POE-CILIA RETICULATA. J.-G. J. Godin and L. A. Dugatkin in Proceedingsof the National Academy of Sciences USA, Vol. 93, No. 19, pages10262–10267; September 17, 1996.

Sex, Color and Mate Choice in Guppies. Anne E. Houde. Prince-ton University Press, 1997.

Sexual Selection and Mate Choice. M. Ryan in BehaviouralEcology: An Evolutionary Approach. Fourth edition. Edited by J. R.Krebs and N. B. Davies. Blackwell Science, 1997.

SA

Call (song) Greater intensityGreater frequencyLonger durationGreater complexityLarger repertoire

Meadow katydidAmerican toadGreen tree frogTungara frogSong sparrow

Courtship display Greater frequency Sage grouse

Body size Larger size Convict-cichlid fish

Tail Longer tailGreater tail heightGreater number of “eyespots”

Barn swallowCrested newtPeacock

Comb Larger comb Red jungle fowl

Bower More decorated bowers Satin bowerbird

Breast stripe

Body color

Larger stripe size

Greater brightnessGreater area of orange

Great titmouse

House finchGuppy

MALE TRAIT FEMALE PREFERENCE SPECIES

What Females Want

JEN

NIF

ER C

. CH

RIST

IAN

SEN

Copyright 1998 Scientific American, Inc.