Are body fluctuating asymmetry and the ratio of 2nd to 4th digit length reliable predictors of semen quality?

Renée C. Firman1, Leigh W. Simmons1,4, James M. Cummins2 and Phillip L. Matson3

1 Department of Zoology, The University of Western Australia, Nedlands, WA 6009, 2 Division of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, WA 6150 and 3 Hollywood Fertility Centre, Hollywood Private Hospital, Nedlands, WA 6009, Australia

4 To whom correspondence should be addressed. e-mail: Lsimmons{at}cyllene.uwa.edu.au


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Recent attention has been paid to patterns of fluctuating asymmetry (FA) in paired bilateral traits and the extent to which they reflect phenotypic and genetic quality. The FA–fertility hypothesis proposes that FA may be a reliable indicator of ejaculate quality in humans and other animals. The common control by the Hox genes of the differentiation of both the urogenital system and the appendicular skeleton in vertebrates has been proposed as an explanation for the recent finding that FA, and the second to fourth digit ratios (2D:4D) are both associated with semen quality in men. METHODS: A group of 50 men was evaluated for FA, calculated by the sum of three different body FAs, 2D:4D ratios, and seminal parameters of masturbatory semen samples. RESULTS: Composite FA had a significant effect on semen parameters; the 2D:4D ratios did not predict semen quality. CONCLUSIONS: Comparison of our data with previous studies suggests that the putative relationship between semen quality and 2D:4D may have been driven by the inclusion of severely oligozoospermic men within the original subject group. Our sample included men with equally high 2D:4D ratios but who had normal semen. Thus, the 2D:4D ratio may not reliably indicate poor semen quality although FA might.

Key words: 2D:4D ratio/fertility/fluctuating asymmetry/semen quality


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The developmental norm is for bilaterally paired morphological traits to show perfect symmetry. However, random deviations from symmetry can arise during development when factors interfere with the ability of an organism to execute its developmental plan equally on both sides of the body, generating what is known as fluctuating asymmetry or FA (Livshits and Kobyliansky, 1991Go; Markow, 1992Go; Polak, 2003Go).

Over the last decade, studies of FA in a number of organisms have suggested a link between developmental stability and an organism’s general health and reproductive fitness (see reviews in Møller and Swaddle, 1997Go; Polak, 2003Go). For example, in human populations developmental stability has been linked with a number of diseases and genetic conditions (Thornhill and Møller, 1997Go). Recently, Manning et al. (1998aGo) reported an association between developmental stability and semen quality in men. Levels of FA in digits two to five were found to be negatively related to total sperm numbers per ejaculate and to sperm motility, with azoospermic subjects exhibiting the largest measure of FA (Manning et al., 1998aGo). Relationships between FA and semen quality have also been suggested from studies of antelopes (Roldan et al., 1998Go; Gomendio et al., 2000Go) and insects (Farmer and Barnard, 2000Go).

While this relationship may seem obscure, there could be a proximate link between FA of the digits and gonadal function in mammals. A group of homeobox genes, the Hox genes, which are organized into four clusters (Hoxa, Hoxb, Hoxc and Hoxd), control the differentiation of the digits and the urogenital system during vertebrate development. Evidence in support of a link between digit development and genital function has accumulated through a number of studies, including Peichel (1997Go) who reported that deregulation of Hoxd alters the relative lengths of the digits and affects growth of the genital bud. Furthermore, in humans, the hand–foot–genital syndrome, characterized by defects in the fingers, toes and gonads, results from a mutation within Hoxa (Mortlock and Innis, 1997Go). Manning et al. (1998bGo) thus further explored the relationship between digit lengths and semen quality by measuring the ratio of the second to fourth digit (2D:4D), which present a pattern of approximate symmetry around the central axis of the third. They found that low 2D:4D values in the right hand were associated with high concentrations of testosterone and high sperm counts, and suggest that the negative relationship arises because of the dual control of hox genes on the development of the digits and gonads. The 2D:4D ratio has thus been advocated as a tool for diagnosis, prognosis, and early life-style and/or detection of infertility, as well as for a number of other medical conditions (Manning and Bundred, 2000Go).

More generally, the finding that FA may be linked to fertility has broader implications for the evolution of human sexual behaviour. Gangestad et al. (1994Go) reported that non-facial body FA was correlated negatively with ratings of their facial attractiveness, while Grammer and Thornhill (1994Go) reported that opposite-sex attractiveness ratings of facial photographs of men correlated positively with the measured bilateral symmetry of each face. Moreover, Thornhill and Gangestad (1994Go) reported a relationship between body asymmetry and reproductive success in a group of 60 men, where more symmetric men reported a greater number of lifetime sexual partners. A preference amongst women for symmetric males could be adaptive if symmetry were to be correlated with a fitness trait such as fertility.

The subjects in the studies by Manning et al. (1998aGo,b) were recruited from an infertility clinic and thus were likely to be involved in relationships with fertility problems. By assessing a group of men randomly recruited from the general population, our study had two aims: first, the relationship between general body FA and ejaculate quality in men was investigated; second, the potential of the 2D:4D ratio for predicting male fertility was assessed.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Measurement considerations
While many authors have used FA extensively as a tool for measuring developmental stability, its assessment is not as straightforward as might first appear. Because FA is characterized by a normal distribution around a mean of zero it is likely to be indistinguishable from measurement error, which shows similar properties (Palmer and Strobeck, 1986Go). This aspect is exacerbated by FAs usually being small in relation to the trait being measured (Møller and Pomiankowski, 1993Go). Therefore, repeated measures must be taken from the same individual to ascertain the relative influence of measurement error on the estimates of asymmetry. Swaddle et al. (1994Go) showed that the accuracy, and hence the reliability, of asymmetry data will increase with increasing numbers of repeats. Tomkins and Simmons (2003Go) have performed a meta analysis of the FA literature suggesting that experimenter expectation and measurement error may compound to produce patterns of significance that have little biological basis (see also Simmons et al., 1999Go). It is therefore recommended that trait measurements be conducted at least twice by a single investigator and should be performed blindly with a time interval to avoid bias (Palmer, 1994Go). Few studies of human FA have conformed to this strict experimental protocol (Tomkins and Simmons, 2003Go).

Subjects
Our subjects were 50 men recruited randomly from advertisements displayed at the University of Western Australia (UWA) and in the Post Community Newspaper. A target age group of 18–35 years was specified on the recruitment advertisements to encapsulate the peak ‘fertile’ years of a man’s life as there is some suggestion that sperm productivity may decrease with age after 35 years (Rolf and Nieschlag, 1997Go, 2001). Of the 50 participants, seven were known to be fertile (i.e. having achieved at least one conception during their lifetime), the other 43 reported no known infertility problems.

Body measurements
The initial measurement session was conducted at the Zoology Department, UWA. The height and weight of each subject was measured and eight bilateral traits (ear length, wrist diameter, elbow diameter, ankle diameter, foot length, foot width and the lengths of digits two and four on the hand) were assessed. These traits were chosen because preliminary studies suggested that they might exhibit FA (Livshits and Kobyliansky, 1991Go). The right and left sides of each trait were measured independently to the nearest 0.1 mm using precision digital plastic callipers (700–103B, Mitutoyo, Japan). Each trait was measured twice, with a lapse of 5–15 min between the first and second measurement.

FA analysis
For each trait, FA was calculated by subtracting the left from the right side value. The repeatability of the two independent FA measurements was tested using a one-factor, repeated-measures ANOVA (Palmer and Strobeck, 1986Go). Measures were accepted as being repeatable if there was a significantly greater variance between individuals than between the repeated FA measurements of the same individual. Repeatability estimates, based on the intraclass correlation of the variance component of the repeated measures ANOVA, were also determined (Winer, 1971Go). Repeatability estimates vary from zero to one, with higher values indicative of greater repeatability. Traits that showed significant repeatability can only be accepted as showing FA if the signed asymmetries are normally distributed (as determined by a Shapiro-Wilk test) around a mean of zero (determined by a one-sample t-test with the null hypothesis set to a mean of zero) (Palmer and Strobeck, 1986Go). Absolute mean FAs were scaled according to trait size by dividing each one by the average value of the trait measurements. Composite FA was then calculated as the sum of the individual scaled FAs.

Semen analysis
Each subject was given clear verbal and written instructions concerning the collection and transport of the semen sample. Participants were required to collect semen by masturbation into a sterile vial after a sexual abstinence minimum of 48 h, but no longer than 7 days. Samples were delivered to the Department of Zoology (UWA) for analysis. To minimize degeneration of motility with time, a period of 2.5 h was set as the maximum time interval between collection and analysis (on one occasion a sample was returned after 2.5 h and consequently was omitted from motility assessment). Furthermore, subjects were asked to surround the collection vial with a strip of aluminium foil to retain the heat of the sample during transport to the laboratory. A regression analysis relating the time between ejaculation to sperm motility assessment and the percentage of motile sperm confirmed that there was no significant relationship, indicating that the method adopted was acceptable.

Sample examination began after liquefaction had occurred. Semen analyses were conducted according to the World Health Organization (WHO) (World Health Organization, 1999Go) protocol. Sperm motility assessment was conducted at the Zoology Department (UWA) by a simple grading system in which 10 microscopic fields (at 400x magnification) were assessed in a systematic way, and all free sperm were graded according to the WHO (1999Go) classifications. In brief, our estimate of motility consisted of the proportion of sperm exhibiting class A or B behaviour, where A = proportion with rapid progressive motility (>20 µm/s at 20°C, with 20 µm being estimated as five head lengths) and B = proportion with slow or sluggish progressive motility (World Health Organization, 1999Go). Sperm counts and preparation of permanent mounts for sperm morphometry were conducted at the Hollywood Fertility Centre. The number of sperm in the ejaculate was determined by volumetric dilution and counting using a Neubauer haemocytometer. Sperm counts were significantly repeatable: the numbers of sperm in the ejaculates of seven men were counted four times, twice on two consecutive days. Repeated measures ANOVA showed significantly greater variance between subjects than between repeated counts of their semen samples (F(6, 21) = 261.8, P < 0.001, repeatability estimate 0.996). Permanent slides were prepared by semen smears, which were fixed and stained with the Diff-Quik (Allegiance Healthcare Corp., McGraw Park, Illinois, USA) stain set and used for sperm morphometry. Sperm lengths (head and tail) were assessed by measuring 10 cells per sample. Images were imported into the Optimas 6.2 Image Analysis software package (Media Cybernetics, Silver Springs, MD, USA) using an Hitachi HV-C20 camera mounted on a Leica DMLS compound microscope. Sperm were viewed under light field at 400x magnification. Averages of the parameters were calculated for each sample. It should be noted that absolute measures of sperm morphology obtained from these fixed samples are likely to be lower than those obtained from fresh samples due to shrinkage. Nevertheless, they allow direct comparisons among individuals within the context of our study because samples from all individuals were treated in an identical manner.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Statistical properties of body measurements
Repeated measures ANOVA confirmed that, with the exception of elbow diameter, the measurements of asymmetry were significantly repeatable. That is, repeatability estimates were generally high (range 0.564–0.933; F-ratios ranged from 2.24–14.84 with all P-values < 0.01, df 49 and 50 for all traits). Our measures of asymmetry in elbow diameter were not repeatable (F(49,50) = 0.39, NS) and were therefore not considered further in our analyses.

Population mean values of asymmetry differed significantly from zero for ankle diameter (the mean ± SEM left-right value was –1.25 ± 0.24, t = 5.16, P < 0.001), foot width (–1.32 ± 0.33, t = 4.01, P < 0.001), and the length of digit two (–0.44 ± 0.16, t = 2.70, P < 0.01). These traits were larger on the right side of the body. Such directional asymmetry arises from a developmental bias toward one side of the body and, unlike FA, may not be indicative of underlying variation in developmental stability (Palmer and Strobeck, 1986Go). Therefore they cannot be used to estimate FA. Wrist diameter also tended to be larger on the right (–0.46 ± 0.22, t = 1.91, P = 0.06) and was not normally distributed (Shapiro-Wilks W = 0.943, P = 0.032) but rather skewed toward larger trait sizes on the right, also indicative of directional asymmetry rather than FA. Wrist diameter was thus not considered further in our estimation of body FA. This left three traits (ear length, the length of digit four and foot length) that had statistical properties indicative of FA. The absolute values of these traits were scaled for trait size and summed to provide our estimate of overall body FA.

Our estimates of the 2D:4D ratio were high and significantly repeatable for both hands (right hand, 0.972, F(49, 50) = 35.89, P < 0.001; left hand, 0.979, F(49, 50) = 48.28, P < 0.001). The population mean (± SEM) values for the 2D:4D ratio in the right hand were 0.964 ± 0.004 (range 0.913–1.029) and the left hand 0.962 ± 0.004 (range 0.901–1.041). These values do not differ from the sample reported in Manning et al. (1998bGo) for which they also had measures of semen quality.

Ejaculate parameters
The mean (± SEM) ejaculate traits were: total number of sperm 105 ± 11(x106); motility 36.39 ± 2.78%; sperm head length 4.99 ± 0.07 µm; sperm tail length 34.87 ± 0.63 µm. A repeated-measures ANOVA of the 10 independently measured sperm per individual revealed a significantly greater variation in sperm morphometry between males than between sperm ejaculated by individual males (head length, F(49, 450) = 6.96, P < 0.001; tail length, F(49, 450) = 3.97, P < 0.001). Thus men differed significantly in sperm dimensions. Interestingly, there was significant within-male variability in the length of sperm tails (within subject F(9, 441) = 1.92, P = 0.048) but sperm head length was far more consistent (within subject F(9, 441) = 0.51, P = NS).

FA, 2D:4D ratios and ejaculate quality
We first performed a single multivariate analysis to examine the influence of the three male morphological variables (FA, left and right hand digit ratios) on semen quality. There was a significant effect of morphology on the multivariate measure of semen quality (whole model Wilks’ Lambda = 0.801, F(3, 45) = 3.74, P = 0.018). Univariate tests showed that there were significant relationships between body FA and total sperm number, sperm motility, and sperm head length (Table I). However, there were no significant relationships between 2D:4D ratios and semen parameters, either for right or left hands (Table I). We also included a number of potentially confounding variables into our analysis; body weight, height, age and period of abstinence prior to ejaculate collection. Including these variables in multiple regression analyses had no influence on the conclusions drawn from Table I (partial slopes of FA on sperm numbers, –14.10 ± 6.59, t44 = 2.15, P = 0.037; on motility, –14.69 ± 6.72, t43 = 2.18, P = 0.034; on head size, 1.41 ± 0.54, t44 = 2.64, P = 0.01).


View this table:
[in this window]
[in a new window]
 
Table I. The results of regression analyses of ejaculate parameters on body fluctuating asymmetry and the 2D:4D ratio of left and right hands (all data were log (1 + x) transformed prior to analyses; df = 48 for number of sperm and sperm morphologies, and df = 47 for motility)
 
To facilitate inspection of the variance in sperm numbers and for comparison with the results of Manning (1998a,b), the numbers of sperm per ejaculate are plotted against body FA in Figure 1 and right hand 2D:4D ratio in Figure 2.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 1. The relationship between composite body FA and the total number of sperm in masturbatory ejaculates for a sample of 50 men.

 


View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. The relationship between right hand 2D:4D ratio and the total number of sperm in masturbatory ejaculates for a sample of 50 men.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Like Manning et al. (1998aGo) we found significant negative relationships between FA and total numbers of sperm, and the motility of sperm contained in masturbatory samples of semen. Unlike Manning et al. (1998bGo) we found no relationships between the 2D:4D ratio and measures of semen quality.

Manning et al. (1998aGo) found a significant negative relationship between digit FA and sperm number and sperm velocity for 52 men. However, the subject pool from which men were drawn was a non-random sample of the male population; being drawn from a group attending an infertility clinic. Manning et al. (1998aGo) suggested that a similar study conducted by Baker (1997Go), who reported a comparable correlation between body FA and sperm numbers, and which sampled a seemingly random population in regard to their fertility, was based on a better representation of the general population. The subject group used by Baker (1997Go) consisted of 34 undergraduate students from Manchester University, UK, ranging in age from 19–22 years. However, it is difficult to evaluate the biological significance of Baker’s (1997) data because of his failure to adequately assess FA. The study measured asymmetry in four traits, index finger length, ear length, wrist width and ankle width. We found that asymmetry in three of these traits—index finger length, wrists and ankles—were not characteristic of FA and cannot therefore be used as an indicator of developmental instability. Our study correctly identified traits that were revealing of developmental instability through FA, and showed that developmental instability may indeed be associated with poor semen quality in the general population.

In a second study, Manning et al. (1998bGo) found negative relationships between the 2D:4D ratio in the right hand and measures of total sperm number per ejaculate, sperm motility, and two measures of sperm swimming speed. Our study, which has a comparable sample size, found no significant associations between 2D:4D ratio and ejaculate quality. Comparison of our Figure 2 with Figure 2 of Manning et al. (1998bGo), shows that the digit ratios between the two subject groups are comparable, but the total sperm numbers are not. Manning et al. (1998bGo) had a subset of oligozoospermic males, which reduced the overall mean sperm numbers and may account for the significant relationship between sperm number and 2D:4D ratio in their study, and the absence of one in our study. Manning et al. (1998bGo) had 12 subjects with germ cell failure (i.e., they were azoospermic or severely oligoasthenozoospermic) and these men had significantly higher 2D:4D ratios in their right hand than did men with active sperm. Our data are certainly a better representation of the general population. Although infertile men may have high 2D:4D ratios (Manning et al. 1998bGo), our data show that men with high 2D:4D ratios do not necessarily have abnormal semen. Thus, the 2D:4D ratio is unlikely to be of general predictive value for a male’s ejaculate quality (Manning and Bundred, 2000Go).

Our finding of consistent differences between men in the morphology of their sperm is interesting, and replicates a recent find by Morrow and Gage (2001Go) who found consistent between-male variation in sperm morphologies across a range of animals, including humans. They suggested that this variation could have important implications if sperm morphology influences motility and/or fertility. A recent comparative study across primates suggests that sperm competition has been important in the evolution of sperm morphology and motility (Anderson and Dixon, 2002Go). Although we know that reduced sperm motility or asthenozoospermia may be caused by functional abnormalities in the morphology of the midpiece (Hargreave and Nillson, 1983Go), little is known about how natural variation in sperm morphology influences characteristics such as motility or longevity. Our finding that sperm head length increased with increasing FA might suggest that large head size is associated with poor sperm performance. In IVF trials, the average length of sperm heads was found to be significantly greater in samples with general poor morphology and low rates of zona pellucida binding and fertilization (Liu and Baker, 1992Go), suggesting that this may indeed be the case. Large sperm head size may be associated with abnormal chromosome constitution (Seuanez et al., 1977Go). These aspects of sperm biology warrant further study.


    Acknowledgements
 
We thank Emily Zuvela, whose expertise and guidance in semen analysis were invaluable, and Lyn Beazley for comments on the thesis draft from which this paper was derived. R.F. thanks Matthew Love whose support was continual and unconditional. This research was supported by the Department of Zoology, UWA, and an Australian Research Council grant to L.W.S. and was conducted with the approval of the University of Western Australia Human Research Ethics Committee, Project No. 0306.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Anderson, M.J. and Dixon, A.F. (2002) Motility and the mid-piece. Nature, 416, 496.

Baker, R.R. (1997) Copulation, masturbation and infidelity. In Schmitt, A., Atzwanger, K., Grammer, K. and Schäfer, K. (eds), New Aspects of Human Ethology. Plenum Press, New York, USA. pp. 163–188.

Farmer, D.C. and Barnard, C.J. (2000) Fluctuating asymmetry and sperm transfer in male decorated field crickets (Gryllodes sigillatus). Behav. Ecol. Sociobiol., 47, 287–292.[CrossRef][ISI]

Gangestad, S.W., Thornhill, R. and Yeo, R.A. (1994) Facial attractiveness, developmental stability and fluctuating asymmetry. Ethol. Sociobiol., 15, 73–85.[CrossRef][ISI]

Gomendio, M., Cassinello, J. and Roldan, E.R. S. (2000) A comparative study of ejaculate traits in three endangered ungulates with different levels of inbreeding: fluctuating asymmetry as an indicator of reproductive and genetic stress. Proc. R. Soc. Lond. B, 267, 875–882.[CrossRef][ISI][Medline]

Grammer, K. and Thornhill, R. (1994) Human (Homo sapiens) facial attractiveness and sexual selection: the role of symmetry and averageness. J. Comp. Psychol., 108, 233–242.[CrossRef][ISI][Medline]

Hargreave, T.B. and Nillson, S. (1983) Seminology. In Hargreave, T.B. (ed) Male Infertility. Springer-Verlag, Berlin, pp. 56–74.

Liu, D.Y. and Baker, H.W.G. (1992) Morphology of spermatozoa bound to the zona pellucida of human oocytes that failed to fertilise in vitro. J. Reprod. Fert., 94, 71–84.[Abstract]

Livshits, G. and Kobyliansky, E. (1991) Fluctuating asymmetry as a possible measure of developmental homeostasis in humans: a review. Hum. Biol., 63, 441–466.[ISI][Medline]

Manning, J.T. and Bundred, P.E. (2000) The ratio of 2nd to 4th digit length: A new predictor of disease predisposition? Med. Hypoth., 54, 855–857.[CrossRef][ISI][Medline]

Manning, J.T., Scutt, D. and Lewis-Jones, D.I. (1998a) Developmental stability, ejaculate size and sperm quality in men. Evol. Hum. Behav., 19, 273–282.[CrossRef][ISI]

Manning, J.T., Scutt, D., Wilson, J. and Lewis-Jones, D.I. (1998b) The ratio of 2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Hum. Reprod., 13, 3000–3004.[Abstract]

Markow, T.A. (1992) Human handedness and the concept of developmental stability. Genetica, 87, 87–94.[CrossRef][ISI][Medline]

Møller, A.P. and Pomiankowski, A. (1993) Fluctuating asymmetry and sexual selection. Genetica, 89, 267–279.[ISI]

Møller, A.P. and Swaddle, J.P. (1997) Asymmetry, Developmental Stability and Evolution. Oxford University Press, Oxford, UK.

Morrow, E.H. and Gage, M.J. G. (2001) Consistent significant variation between individual males in spermatozoal morphometry. J. Zool., 254, 147–153.[CrossRef][ISI]

Mortlock, D.P. and Innis, J.W. (1997) Mutation of hoxA13 in hand-foot-genital syndrome. Nature Genet., 15, 179–180.[ISI][Medline]

Palmer, A.C. and Strobeck, C. (1986) Fluctuating asymmetry: measurement, analysis, pattern. Ann. Rev. Ecol. Syst., 17, 391–421.[CrossRef][ISI]

Palmer, A.R. (1994) Fluctuating asymmetry analysis: a primer. In Markow, T.A. (ed), Developmental instability: its origins and evolutionary implications. Kluwer Academic Publications, London, UK. pp. 335–364.

Peichel, C.L., Prabhakaran, B. and Vogt, F. (1997) The mouse ulnaless mutation deregulates posterior hoxd gene expression and alters appendicular patterning. Development, 124, 3481–3492.[Abstract/Free Full Text]

Polak, M. (2003) Developmental Instability: Causes and Consequences. Oxford University Press, New York, USA. in press.

Roldan, E.R. S., Cassinello, J., Abaigar, T. and Gomendio, M. (1998) Inbreeding, fluctuating asymmetry and ejaculate quality in an endangered ungulate. Proc. R. Soc. Lond. B, 265, 243–248.[CrossRef][ISI][Medline]

Rolf, C. and Nieschlag, E. (1997) Senescence. In Nieschlag, E. and Behre, H.M. (ed) Andrology. Springer, Berlin, Germany. pp 397–407.

Rolf, C. and Nieschlag, E. (2001) Reproductive functions, fertility and genetic risks of ageing men. Exp. Clin. Endocrinol. Diabetes, 109, 68–74.

Seuanez, H.N., Carothers, A.D., Martin, D.E. and Short, R.V. (1977) Morphological abnormalities in spermatozoa of man and great apes. Nature, 270, 345–347.[ISI][Medline]

Simmons, L.W., Tomkins, J.L., Kotiaho, J.S. and Hunt, J. (1999) Fluctuating paradigm. Proc. R. Soc. Lond. B., 266, 593–595.[CrossRef][ISI]

Swaddle, J.P., Witter, M.S. and Cuthill, I.C. (1994) The analysis of fluctuating asymmetry. Anim. Behav., 48, 986–989.[CrossRef][ISI]

Thornhill, R. and Gangestad, S.W. (1994) Human fluctuating asymmetry and sexual behavior. Psychol. Sci., 5, 297–302.[ISI]

Thornhill, R. and Møller, A.P. (1997) Developmental stability, disease and medicine. Biol. Rev., 72, 497–548.[CrossRef][ISI][Medline]

Tomkins, J.L. and Simmons, L.W. (2003) Fluctuating asymmetry and sexual selection: Paradigm shifts, publication bias and observer expectation. In M. Polak (ed) Developmental Instability: Causes and Consequences. Oxford University Press, New York, USA. in press.

Winer, B.J. (1971) Statistical Principles in Experimental Design. McGraw-Hill, New York, USA.

World Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen and Sperm-Cervical Mucus Interaction. Cambridge University press, Cambridge, UK.

Submitted on June 27, 2002; resubmitted on December 9, 2002; accepted on January 6, 2003.