Anim. Behav., 1998, 55, 33–39
Female swordtail fish prefer symmetrical sexual signal
MOLLY R. MORRIS* & KENNETH CASEY†
*Department of Biology, Montgomery College, Rockville, MD 20850, U.S.A.
†Department of Zoology, University of Texas at Austin
(Received 14 November 1996; initial acceptance 24 January 1997;
final acceptance 5 March 1997; MS. number: 7754)
Abstract. It has been proposed that females use the symmetry of secondary sexual traits to differentiate
between potential mates. The vertical bars on male swordtail fish function as a signal that attracts
females and deters rival males in one swordtail species. In addition, male courtship behaviour of most
Xiphophorus species incorporates serial lateral presentations, which provide females with a clear
opportunity to assess males for bilateral symmetry. We tested the hypothesis that X. cortezi females
prefer males with a symmetrical bar number by determining whether females switched their preference
between two males when we switched which male had a symmetrical number of bars. The ability to
manipulate the bars without influencing other male traits allowed us to control for male characters
correlated with bar symmetry that females might prefer. The degree of asymmetry in bar number we
used was within the degree of asymmetry found for this trait in nature. Females switched their
preference between a pair of males when we switched which male was symmetrical for bar number. We
discuss the possibility that females prefer a symmetrical bar number as well as an alternative hypothesis.
?
mates based on ‘genetic quality’ (Møller 1990;
Møller & Pomiankowski 1993).
Symmetry of sexually selected traits contains
information about male genetic quality, and symmetrical males have greater mating success
(Watson & Thornhill 1994). It has not been
demonstrated, however, that females prefer symmetry in naturally occurring sexual signals. In
some cases, it has been difficult to separate preference for the symmetrical trait from other correlated characters; for example, tail symmetry in
barn swallows, Hirundo rustica (Møller 1992)
affects flight performance (Møller 1991). Other
studies have demonstrated preference for symmetry of unnatural traits, for example, coloured
leg-bands on zebra finches, Taeniopygia guttata
(Swaddle & Cuthill 1994a), and painted tail feathers in barn swallows (Møller 1993). Møller (1993)
painted the swallow tail feathers in an attempt
to control for the affect of asymmetry on flight
performance and assumed that females perceived
the white part of the tail feather as ‘not there’.
Preference for symmetry has also been demonstrated for chest barring on finches (Swaddle &
Cuthill 1994b), but it is not known whether this
trait functions as a sexual signal.
Fluctuating asymmetries are random deviations
from perfect bilateral symmetry in a morphological trait for which differences between the right
and left sides have a mean of zero and are
normally distributed (Van Valen 1962). Fluctuating asymmetries result from the inability of individuals to undergo identical development on both
sides of their bodies (Leary & Allendorf 1989).
Interest in fluctuating asymmetries has recently
spread to studies of sexual selection because of
their potential to link a component of male
viability with criteria used by females in discriminating between mates (Watson & Thornhill 1994).
Some secondary sexual traits show greater fluctuating asymmetries than other morphological traits
(e.g. Møller & Pomiankowski 1993; Møller &
Höglund 1991), and differences between conspecifics in their ability to develop symmetrically
are heritable (Thornhill & Sauer 1992; Møller &
Pomiankowski 1993). It has been proposed that
females could use the symmetry of sexually
selected traits to discriminate between potential
Correspondence: M. R. Morris, Department of Biological, Sciences, Ohio University, Athens, OH 45701,
U.S.A. (email: morrism@oak.cats.ohiou.edu). K. Casey
is at 313 Rawland Dr., Austin, TX 78745, U.S.A.
0003–3472/98/010033+07 $25.00/0/ar970580
1998 The Association for the Study of Animal Behaviour
? 1998 The Association for the Study of Animal Behaviour
33
Animal Behaviour, 55, 1
34
Figure 1. The left (a) and right (b) side of a Xiphophorus
cortezi male, Arroyo La Conchita population, manipulated by freeze-branding to produce an asymmetry of
two (8/6). The male naturally had eight bars on the left
and seven on the right. We removed the posterior-most
bar on the right side.
The signal that we examined is a pigment
pattern of dark vertical bars (Fig. 1). Vertical bars
are widespread throughout the genus Xiphophorus
as well as in other genera of poeciliid fishes (e.g.
Heterandria, Phallichthys). Examination of F1
and back-crossed hybrids among several of the
Xiphophorus species, including X. cortezi, suggests
that the bars have a polygenic basis (Atz 1964).
The bars are composed of hundreds of closely
associated melanophores, and males can vary the
intensity of their bars by dispersing or concentrating the pigment within the melanophores (Gordon
1931). In several species, including X. cortezi, the
bars are a secondary sexual trait found on sexually
mature males and they intensify in relation to
social interactions; the bars become darker on
aggressive males and males courting females
(Franck 1964; Zimmerer & Kallman 1988; Morris
et al. 1995). The bars function in both deterring
rival males and attracting females in the closely
related species X. multilineatus (Morris et al.
1995).
We tested X. cortezi females to determine
whether they preferred males with symmetrical
bar numbers over males with asymmetrical bar
numbers. We controlled differences between pairs
of males other than bar symmetry by testing a
female twice with the same pair of males; we
determined whether females switched the time
they spent associating with one male over another
when we switched which male had symmetrical
bar numbers. We also examined the degree of
bilateral asymmetry in bar numbers for one population of X. cortezi to determine whether the
asymmetry we used in our experiment was within
the natural range of asymmetry for this trait.
MATERIALS AND METHODS
We collected Xiphophorus cortezi from two sites in
Mexico (the Rı́o Axtla, San Luis Potosı́ and the
Arroyo La Conchita at Xiltila, San Luis Potosı́).
We recorded male size (standard length) and the
number of bars on both sides for the males
collected from the Arroyo La Conchita (Table I).
Males were anaesthetized with MS222 before
measurements were taken, which ensured that the
bars were fully expressed when counted. Because
Table I. Bar symmetry (number of bars on left"number of bars on right side) for
X. cortezi males
Symmetrical (0)
Asymmetrical (&1)
Asymmetrical (&2)
Asymmetrical (&3)
Males without bars
Percentage (N)
of males with bars
Average number of bars
right+left ()
35 (8)
39 (9)
17 (4)
9 (2)
N=11
12.5 (4.00)
13.4 (3.13)
12.0 (3.65)
15.0 (5.66)
—
Average size,
mm ()
36.8
39.8
37.5
29.8
37.3
(6.47)
(7.20)
(1.52)
(8.13)
(5.90)
Morris & Casey: Female preference for symmetry
35
Table II. Sizes and number of bars (right side/left side) for pairs of males used in choice
tests as well as the time females spent with each male
Test 1
Test 2
Male size
(mm)
Bars
(R/L)
Time
(s)
Bars
(R/L)
Time
(s)
45.1
43.3
6/6
7/5
1374
424
4/6
5/5
723
1091
38.1
39.2
3/3
2/4
1162
654
3/1
2/2
659
1405
37.2
35.3
8/8
9/7
1398
685
8/6
7/7
684
1089
31.0
31.0
5/5
6/4
1465
562
5/3
4/4
733
1129
32.0
31.5
6/6
7/5
910
737
4/6
5/5
823
1371
34.1
35.0
4/4
3/5
1386
504
4/2
3/3
701
1299
The male that had the symmetrical number of bars in test 1 was freeze-branded to have
an asymmetrical number of bars in test 2.
the bars are a discrete trait, the problems associated with measuring the fluctuating asymmetry of
quantitative traits (Palmer & Strobeck 1986) can
be avoided. We kept males and females in separate 2.5-litre tanks throughout the experiment to
increase sexual interactions during testing. Stimulus males were pairs of males of similar sizes
(within 2.0 mm; Table II). We tested females with
pairs of stimulus males from the same site from
which they were collected, and each female was
tested with the same pair of males twice.
Bar Manipulations
Before the first round of tests, the stimulus
males from a pair were freeze-branded so that
both males had the same total number of bars.
One male, however, received an asymmetrical
number of bars (two more bars on one side
compared to the other), and one male a symmetrical number of bars. The freeze-branding technique
(Raleigh et al. 1973) has been used to examine
the function of the bars as a sexually selected
signal in related swordtail species (Morris et al.
1995; Morris & Ryan 1996) and is an effective
method of removing the bars without altering the
behaviour of the males. We randomly determined
which side of a male would have more bars for the
asymmetrical treatments. Males from each pair
were freeze-branded an equal number of times
during each manipulation, branding between the
bars in cases where fewer bars were removed from
one male compared to the other. When bars
were removed, they were the posterior most bars
(Fig. 1). Although the behaviour of the males did
not appear to be influenced by freeze-branding,
we controlled for any possible influences of this
procedure, because males within a pair always
received the same number of freeze brands in the
same general locations. Before the second round
of tests, we removed two bars from one side of
each male, which reversed which male was symmetrical within a pair while keeping the total
number of bars on the two males equal.
Preference Tests
Simultaneous choice tests were used to test for
female preference. Results from this type of choice
test in the closely related species X. nigrensis are
highly related to other indicators of mate choice
(Morris et al. 1992; Morris & Ryan 1993), in
addition to male reproductive success in the field
(Ryan et al. 1990b). Tests were conducted in an
aquarium (45#41#80 cm) divided into five
equal sections (Fig. 2). The two end compartments were divided from the centre three by glass
plates, which ensured that females received only
Animal Behaviour, 55, 1
36
Figure 2. Design of female choice test tank. Solid lines
indicate solid, transparent partitions; dotted lines indicate lines drawn on the aquarium to visually divide the
middle compartment into three parts.
visual cues. The middle three compartments were
divided by lines marked on the aquarium. We
placed a test female in an opaque tube in the
centre section, and the test males in the two end
compartments. All three fish were allowed to
acclimate for 10 min. When the tube was raised,
we measured the time that the female spent in the
two sections adjacent to the males for 20 min. The
centre section was considered neutral and gave
the females the opportunity to choose not to
associate with either male. Trials in which both
males did not actively interact with the female by
swimming up and down the divider were eliminated. In all cases where the males interacted with
the female, the bars were expressed. After the first
observation period, the female was returned to the
opaque tube, the males were switched end to end,
and all fish were allowed to acclimate for 10 min.
The test was then repeated to control for any
position preference. The total time that a female
spent with a male across both observation periods
was used as an indication of relative preference.
We compared the total time females spent
with the symmetrical males compared to the
asymmetrical males in both rounds of tests with
Wilcoxon matched-pairs signed-rank tests. We
also compared the preference of each female in the
first round of tests to her preference in the second
round of tests to determine whether female preferences switched when we switched which male
had a symmetrical bar number. These data were
analysed with a binomial test.
RESULTS
Bar Symmetries
Not all X. cortezi males from the Arroyo La
Conchita population have bars. Excluding males
without bars (N=11), the total number of
bars was positively correlated with male size
(r222 =0.2, P=0.03). The average& bar symmetry on males with bars was "0.13&1.39
(range="3–+3, N=23), and was not significantly
different from zero (t= "0.449, P=0.66). Twentysix per cent of the males with bars had a degree of
asymmetry in bar number equal to or greater than
the degree of asymmetry we used in this study
(Table I).
Female Preference
Female X. cortezi preferred males with a symmetrical bar number over asymmetrical bar
number in simultaneous choice tests. Females
spent significantly more time with the symmetrical
males compared to the asymmetrical males in
both tests (test 1: mean& time with symmetrical
males=1282.5&85.44 s, time with asymmetrical
males=594.3&48.43 s, Wilcoxon signed-rank
test, Z= "2.2, N=6, P=0.01; test 2: time with
symmetrical males=1230.7&59.07 s, time with
asymmetrical males=720.5&23.21 s, Z= "2.2,
N=6, P=0.01 (Table II). We also compared the
number of females that switched and that maintained their preference when we switched which
male was symmetrical between rounds of tests. All
six females reversed their preference between
the two tests, associating with whichever male
received the symmetrical number of vertical bars
(Binomial test, N=6, r=6, P=0.036; Table II).
These results indicate that female X. cortezi fish
preferred to associate with males that were symmetrical for bar number over males that were
asymmetrical for this trait when the total bar
number was held constant.
DISCUSSION
Mounting evidence suggests that symmetry of
male secondary sexual traits may provide females
with information about male viability, and that
symmetrical males have greater mating success
than asymmetrical males (Watson & Thornhill
1994). Demonstrating that females prefer males
with symmetrical sexual signals and that this
preference has affected the design of sexually
selected male traits has been difficult. Two problems faced in previous studies were identified by
Swaddle & Cuthill (1994b). First, it must be
Morris & Casey: Female preference for symmetry
possible to manipulate the symmetry of the male
trait without changing other correlated traits that
females might prefer. Vertical body bars are a
visual signal that we could manipulate with freezebranding, a technique that did not influence male
behaviour or other male traits. We were able to
show a complete reversal of each female’s preference between a pair of males by changing only the
symmetry of the signal, which rules out preference
for traits correlated with symmetry of the signal.
The potential strength of selection for symmetry
appears to be strong in X. cortezi. Even though
pairs of males may have differed to some degree in
several other characters that could potentially
influence female choice (body size: Ryan et al.
1990b; sword length: Basolo 1990; other pigment
patterns), a change in bar symmetry alone could
override these differences and cause a female to
change her preference from one male to another.
Second, the traits tested must be natural traits,
and preferably within the range of asymmetry
found in nature, to suggest that female preference
for symmetry has affected the design. Vertical bars
are found on males throughout all of the populations of X. cortezi, and the degree of asymmetry
that we used in our choice tests was within the
range of asymmetry in bar number for one of the
populations of X. cortezi tested.
A third type of problem for studies designed to
examine female preference for symmetrical male
traits involves separating preference for symmetry
from preferences for trait size assessed by minimum or maximum expression of the trait on one
side. Because we controlled the total number of
bars, the asymmetrical male had both the greater
and the lesser number of bars on one side. Therefore, females that associated with symmetrical
males may have been expressing a preference for
males with the largest minimum number of bars,
rather than for males with a symmetrical number
of bars. This possibility applies not only to our
study, but also to all studies in which females
showed a preference for symmetrical males when
the average expression of the trait was held constant. This problem was identified by Oakes &
Barnard (1994) when they demonstrated that
female paradise whydahs, Vidua paradisaea, preferred males with asymmetrical tails when average
tail length was held constant. Oakes & Barnard
pointed out that females may actually have preferred the male with the longest single tail feather
rather than asymmetrical males. Studies that
37
show preference for symmetry cannot avoid this
problem either, however. Although preference for
trait size can be tested in an independent experiment (Brookes & Pomiankowski 1994) asymmetry of a trait and size of the trait can not be
manipulated independently. By holding the average size of the trait constant, the asymmetrical
male will have both the maximum and minimum
expression of the trait.
Growing evidence shows that mate choice decisions are based on multiple cues (Burley 1981;
Baker et al. 1986; Brooks & Caithness 1995).
Preference for symmetry and preference for trait
size may be two of several preferences that interact to result in female choice of mates. If preference for exaggeration of the male trait is stronger
than preference for symmetry of the male trait,
it will be impossible to separate the effects of
preference for symmetry and preference for largest
minimum trait in the type of study we have
presented; the necessary test would require that
females prefer a male with symmetrical bar
number over an asymmetrical male that has more
bars on average, so that the minimum number of
bars on one side would be the same for both
males. Alternatively, single stimuli preference
tests, in which the responses of females with
individual males are scored and correlated with
the quantitative differences between males for the
trait, may provide the type of data that can be
used to separate statistically preference for symmetry and preference for largest minimum trait
expression.
Two of the courtship displays that have been
described for swordtail fish provide females with
an opportunity to assess bilateral symmetry. At
least one of the courtship displays found in
X. cortezi and one of the displays found in other
poeciliid fishes incorporates lateral displays to the
female in which the male alternates frequently
from side to side. The ‘figure-eight’ display (Ryan
& Causey 1989), in which males swim back and
forth in front of the female in a figure-eight
pattern, is used by almost all of the northern
swordtails including X. cortezi. Males of the
southern swordtail X. helleri use a courtship display in which they alternate lateral presentations
by swimming back and forth in front of the female
(Clark et al. 1954; Franck 1964). Even the platyfish within the genus Xiphophorus use a behaviour
in which females could assess bilateral symmetry:
the display ‘side change over head’ (Haas 1993), in
38
Animal Behaviour, 55, 1
which the male flips his tail back and forth over
the female’s head, presenting the female with
serial lateral views. It had been suggested that the
serial lateral presentations of the courtship displays of swordtails might accentuate the sword
(Rosen 1960) or limit the ability of the females to
flee (Franck 1964). The results of this study suggest that the function of these components of the
courtship displays should also be examined in
relation to bilateral assessment by females and
their preference for symmetry.
Assuming that the preference we detected was
for symmetry, preference for symmetrically barred
males could evolve because symmetry reflects
individual male quality in relation to the ability of
an individual to cope with genetic and/or environmental stress. As such, the results of this study
would be consistent with ‘good genes’ models of
sexual selection, which predict that females
choose mates based on traits that reflect male
quality (Møller 1990; Møller 1992; Møller &
Pomiankowski 1993). These results would not
exclude other models of mate choice, such as
direct selection (Kirkpatrick & Ryan 1991), sensory exploitation (Ryan 1990; Ryan et al. 1990a)
and selection for mate recognition (Johnstone
1994). For example, the vertical bars are more
variable in the hybrids of several species of Xiphophorus, including offspring from X. cortezi and
X. montezumae crosses (Atz 1962); bar asymmetry
therefore may provide information about the
probability that an individual is a hybrid. Further
studies of the relationship between bar symmetry
and male quality, genetic correlations between
female preference and male trait, and the evolutionary history of preference and trait will make it
possible to consider which of the various models
for the evolution of female preference can explain
preference for males with a symmetrical bar
number.
The vertical bars appear to be ideally designed
as a signal to provide information about symmetry. The number of bars is a discrete character,
so bar number from side to side would provide a
particularly unambiguous measure of symmetry.
Further study is needed, however, to determine
the exact nature of female preference in relation to
the bars in this species. The preference we
detected, whether for symmetrical males or males
with the largest minimum number of bars, would
select for symmetrical males when the total
number of bars is equal. Female preference for
males with the largest minimum number of bars
would select against the addition of bars if not
added symmetrically. In either case, X. cortezi
females are assessing bars on both sides of a male,
and bilateral assessment could have influenced the
design of the vertical bars as a sexually selected
signal as well as the serial lateral presentations of
the courtship display that accentuates this signal.
ACKNOWLEDGMENTS
We thank A. Basolo, K. de Queiroz, I. Schlupp
and W. E. Wagner, Jr for comments, S. J. Siddiqi,
P. Crites and M. Ries for assistance in data
collecting, K. de Queiroz for X. cortezi photos,
and the government of the Republic of Mexico
for permission to collect the fish used in this
study.
REFERENCES
Atz, J. W. 1962. Effects of hybridization on pigmentation in fishes of the genus Xiphophorus. Zoologica, 47,
153–181.
Baker, M. C., Bjerke, T. K., Lampe, H. & Espmark, Y.
1986. Sexual response of female great tits to variation
in the size of male’s repertoires. Am. Nat., 128,
491–498.
Basolo, A. L. 1990. Female preference predates the
evolution of the sword in swordtail fish. Science, 250,
808–810.
Brookes, M. & Pomiankowski, A. 1994. Reply from M.
Brookes and A. Pomiankowski. Trends. Ecol. Evol., 9,
440.
Brooks, R. & Caithness, N. 1995. Female choice in a
feral guppy population: are there multiple cues? Anim.
Behav., 50, 301–307.
Burley, N. 1981. Mate choice by multiple criteria in a
monogamous species. Am. Nat., 117, 515–528.
Clark, E., Aronson, L.R. & Gordon, M. 1954. Mating
behaviour patterns in two sympatric species of Xiphophorin fishes: their inheritance and significance in
sexual isolation. Bull. Am. Mus. nat. Hist., 103, 135–
226.
Franck, D. 1964. Vergleichende Verhaltensstudien und
Lebendegebarenden Zahnkarpfen der Gattung Xiphophorus. Zool. Jahrb. (Physiol. Bd.), 71, 117–170.
Gordon, M. 1931. Morphology of the heritable color
patterns in Mexican killifish, Platypoecilus. A. J.
Cancer, 15, 732–787.
Haas, V. 1993. Xiphophorus phylogeny, reviewed on the
basis of the courtship behaviour. In: Trends in Ichthyology (Ed. by J. H. Schröder, J. Bauer & M. Schartl),
pp. 279–288. Oxford: Blackwell.
Johnstone, R. A. 1994. Female preference for symmetrical males as a by-product of selection for mate
recognition. Nature, Lond., 372, 172–175.
Morris & Casey: Female preference for symmetry
Kirkpatrick, M. & Ryan, M. J. 1991. The evolution of
mating preferences and the paradox of the lek.
Nature, Lond., 350, 33–38.
Leary, R. F. & Allendorf, F. W. 1989. Fluctuating
asymmetry as an indicator of stress: implications for
conservation biology. Trends Ecol. Evol., 4, 214–217.
Møller, A. P. 1990. Fluctuating asymmetry in male
sexual ornaments may reliably reveal male quality.
Anim. Behav., 40, 1185–1187.
Møller, A. P. 1991. Sexual ornament size and the cost of
fluctuating asymmetry. Proc. R. Soc. Lond. Ser. B,
243, 59–62.
Møller, A. P. 1992. Female swallow preference for
symmetrical male sexual ornaments. Nature, Lond.,
357, 238–240.
Møller, A. P. 1993. Female choice for apparently symmetrical sexual ornaments in the barn swallow
Hirundo rustica. Behav. Ecol. Sociobiol., 32, 371–376.
Møller, A. P. & Höglund, J. 1991. Patterns of fluctuating asymmetry in avian feather ornaments: implications for models of sexual selection. Proc. R. Soc.
Lond. Ser. B, 245, 1–5.
Møller, A. P. & Pomiankowski, A. 1993. Fluctuating
asymmetry and sexual selection. Genetica, 89, 276–
279.
Morris, M. R. & Ryan, M. J. 1993. Female mating
preferences for large size in pygmy swordtails, Xiphophorus pygmaeus. Am. Soc. Ich. Herp. Abs., 1993, 229.
Morris, M. R. & Ryan, M. J. 1996. Sexual difference in
signal-receiver coevolution. Anim. Behav., 52, 1017–
1024.
Morris, M. R., Batra, P. & Ryan, M. J. 1992. Male–
male competition and access to females in Xiphophorus nigrensis. Copeia, 1992, 980–986.
Morris, M. R., Mussel, M. & Ryan, M. J. 1995. Vertical
bars on male Xiphophorus multilineatus: a signal that
deters rival males and attracts females. Behav. Ecol.,
4, 274–279.
Oakes, E. J. & Barnard, P. 1994. Fluctuating asymmetry
and mate choice in paradise whydahs, Vidua paradisaea: an experimental manipulation. Anim. Behav.,
48, 937–943.
Palmer, R. A. & Strobeck, C. 1986. Fluctuating asymmetry: measurement, analysis, patterns. A. Rev. Ecol.
Syst., 17, 391–421.
39
Raleigh, R. F., McLearen, J. B. & Groff, D. G. 1973.
Effects of topical location, branding techniques, and
changes in hue on recognition of cold brands in
centrarchid and salmonid fish. Trans. Am. Fish. Soc.,
102, 637–641.
Rosen, D. E. 1960. Middle-American poeciliid fishes of
the genus Xiphophorus. Bull. Fla. State Mus. Biol. Sci.,
5, 57–242.
Ryan, M. J. 1990. Sensory systems, sexual selection, and
sensory exploitation. Oxford Sur. evol. Biol., 7, 157–
195.
Ryan, M. J. & Causey, B. A. 1989. ‘Alternative’ mating
behavior in the swordtails Xiphophorus nigrensis and
Xiphophorus pygmaeus. Behav. Ecol. Sociobiol., 24,
341–348.
Ryan, M. J., Fox, J. H., Wilczynski, W. & Rand, A. S.
1990a. Sexual selection for sensory exploitation in the
frog Physalaemus pustuosus. Nature, Lond., 343,
66–67.
Ryan, M. J., Hews, D. K. & Wagner, W. E. Jr. 1990b.
Sexual selection on alleles that determine body size in
the swordtail Xiphophorus nigrensis. Behav. Ecol.
Sociobiol., 26, 231–237.
Swaddle, J. P. & Cuthill, I. C. 1994a. Female zebra
finches prefer symmetric males. Nature, Lond., 367,
165–166.
Swaddle, J. P. & Cuthill, I. C. 1994b. Female zebra
finches prefer males with symmetric chest plumage.
Proc. R. Soc. Lond. Ser. B, 258, 267–271.
Thornhill, R. & Sauer, P. 1992. Genetic sire effects on
the fighting ability of sons and daughters and mating
success of sons in a scorpionfly. Anim. Behav., 43,
255–264.
Van Valen, L. 1962. A study of fluctuating asymmetry.
Evolution, 16, 125–142.
Watson, P. J. & Thornhill, R. 1994. Fluctuating asymmetry and sexual selection. Trends Ecol. Evol., 9,
21–25.
Zimmerer, E. J. & Kallman, K. D. 1988. The inheritance
of vertical barring (aggression and appeasement signals) in pygmy swordtail Xiphophorus nigrensis
(Poeciliidae, Teleostei). Copeia, 1988, 299–307.