Developmental Instability in Japanese Quail Genetically Selected
for Contrasting Adrenocortical Responsiveness1
D. G. Satterlee,*,2G. G. Cadd,* and R. B. Jones†
*Applied Animal Biotechnology Laboratories, Department of Poultry Science, Louisiana Agricultural Experiment Station,
Louisiana State University Agricultural Center, Louisiana State University, Baton Rouge, Louisana 70803; and
†Welfare Biology Group, Roslin Institute (Edinburgh), Roslin, Midlothian EH25 9PS, Scotland
were assessed with Japanese quail of two lines that had
been genetically selected over several generations for re-
duced (low stress, LS) or exaggerated (high stress, HS)
plasma corticosterone response to brief mechanical re-
straint. At 32 wk of age,three bilateral traits were selected
for study in each quail line. The characteristics chosen
ter of the shank (SHD) perpendicular to the spur, and
distance between the auditory canal and the nares (face
length, FL). Significantly greater bilateral trait size vari-
ances were associated with the measurement of SHL (P
< 0.0088) and FL (P < 0.0016) in the HS line than in the
LS line. SHD variances did not differ (P = 0.22) in quail
Differences in developmental instability
(Key words: quail, fluctuating asymmetry, corticosterone, stress)
2000 Poultry Science 79:1710–1714
Developmental instability is attracting growing inter-
est because it is thought to reflect the inability of individ-
ual organisms to produce a stable phenotype under
given environmental conditions (Parsons, 1990; Moller
and Swaddle, 1997; Moller et al., 1999). Unlike genetic
preprogrammingthat canlead toantisymmetry ordirec-
tional asymmetry, developmental instability produces
fluctuating asymmetry(FA) or small,random deviations
from symmetry in otherwise bilaterally symmetrical
Thomson, 1999). In other words, differences between the
size of left and right bilateral characters are normally
distributed with a mean of zero and a variance governed
by the amount of developmental instability (Van Valen,
1962; Thomson, 1999).
It has been proposed that FA is caused by genetic and
environmental stress, and its occurrence has been noted
Received for publication October 29, 1999.
Accepted for publication August 17, 2000.
1Approved for publication by the Director of the Louisiana Agricul-
tural Experiment Station as manuscript Number 99-46-0567.
2To whom correspondence should be addressed: dsatterlee@agctr.
of the HS and LS lines. These findings suggest that devel-
opmental instability (i.e., fluctuating asymmetry, FA) is
more pronounced in HS quail than in LS quail. Previous
studies have shown that not only do quail of the HS
rangeofstressors butthattheyarealso moreeasilyfright-
ened than LS birds. Therefore, the line differences in FA
found here may reflect the birds’ differential respon-
siveness to chronic social and physical environmental
stressors.The presentfindings alsosupport previoussug-
gestions that measuring asymmetries in bilateral traits
could be an additional and valid method of assessing
stress and of comparing phenotypic stability in selected
in a variety of mammalian and avian species (Palmer
and Strobeck, 1986; Jones, 1987; Parsons, 1990; Moller
and Swaddle, 1997; Thomson, 1999). Inbreeding, the in-
troduction of novel mutants into the genome, and hy-
bridization are among the genetic factors that increase
FA, whereas the environmental causes include parasitic
infestation, pathologies, feed deprivation, forced expo-
sure to novelty, loud sounds, and social stress (Parsons,
1990; Moller and Swaddle, 1997). More specifically, FA
was more pronounced in chickens with tibial dyschon-
droplasia and in those reared under high stocking densi-
ties or under permanent light (Moller et al., 1995, 1999).
Strong directional selection, such as that imposed by
animal breeders, may also be particularly influential in
increasing developmental instability. For example,
asymmetry was greater in White Leghorn lines that had
been selected for contrasting antibody response to sheep
red blood cells (Siegel and Gross, 1980) than in their
reciprocal crosses (Yang et al., 1997; Yang and Siegel,
Abbreviation Key: FA = fluctuating asymmetry; FL = face length;
G = generation; HS = high stress; LS = low stress; RA = relative asymme-
try; SHL = shank length; SHD = shank diameter.
DEVELOPMENTAL INSTABILITY AND ADRENOCORTICAL RESPONSIVENESS
The potential value of using asymmetries of bilateral
traitsastools inthestudyofevolution andanimalbreed-
This approach also has considerable strategic relevance
for farm animals. For example, the magnitude of asym-
metries might be used as a valid measure of stress, as a
way of comparing developmental instability between
populations, and as a means of identifying the optimal
rearing conditions for domestic animals (Yang et al.,
1997; Moller et al., 1995, 1999). Such information could
have important implications for well-being and produc-
tivity, both of which can be seriously compromised if
Intuitively, genetic lines that have been selected over
several generations for differences in their physiological
and behavioral responsiveness to stressful stimulation
would be particularly powerful tools for studying devel-
opmental instability. Therefore, in the present study the
occurrence and magnitude of asymmetries were studied
in divergent lines of Japanese quail that had been se-
lected for low (LS, low stress) or high (HS, high stress)
plasma corticosterone responses to brief mechanical re-
straint (Satterlee and Johnson, 1988). These genetic lines
are particularly suitable for exploring the above issues
ently exerted a nonspecific effect on stress susceptibility
to a wide variety of stressors, e.g., cold, crating, food and
water deprivation, social tension, and manual restraint,
than did HS birds (Satterlee and Johnson, 1988; Jones et
al., 1994; Jones, 1996). Second, selection has also influ-
enced underlying fearfulness (propensity to be easily
frightened by diverse events). Fear responses were con-
sistently less pronounced in LS than HS quail when they
were exposed to human beings, exposed areas, unfamil-
iar objects and places, or mechanical restraint (Jones et
al., 1992a,b, 1994, 1999; Satterlee and Jones, 1995; Jones
and Satterlee, 1996). Third, not only is the Japanese quail
an important agricultural species in many countries
(Baumgartner, 1994), but it is also thought to be a useful
model for other more commercially important species,
suchasthe domesticfowl(MillsandFaure, 1992;Aggrey
and Cheng, 1994).
In the first of a series of projected studies, attempts
were made to illuminate the potential role of environ-
mental stress in determining developmental instability.
More specifically, the hypothesis was tested that quail
from genetic lines selected for reduced or exaggerated
sensitivity to stressful stimulation would show de-
creased or elevated levels of FA, respectively. Three bi-
lateral traits were measured in adult male quail from
the LS and HS lines that had been housed in same-line
breeding groups for 26 wk. It was reasoned that they
would likely have been exposed to a variety of environ-
mental stressors (e.g., human traffic, extraneous noises,
3Petersime Incubator Co., Gettysburg, OH 45328.
and interbird aggression) during this time. Fluctuating
asymmetries were then compared between the two lines
by calculating intra-individual variances in trait size
MATERIALS AND METHODS
Genetic Stocks and Husbandry
ica) were studied from two genetic lines that were se-
lected over several generations for low (LS) or high (HS)
plasma corticosterone response to brief mechanical im-
mobilization (Satterlee and Johnson, 1988). After pedi-
sterone levels postimmobilization were 156 and 54% of
the values measured in nonselected controls in the HS
and LS lines, respectively (Satterlee and Johnson, 1988).
erations, after which the lines were maintained for six
generations without selection. Nonselection was accom-
plished by colony breeding of family crosses within a
line, avoiding only full-sib matings. Despite relaxation
of the selection pressure during generation 15 (G15) to
G19, examination of the lines at G20showed that diver-
gencehad beenmaintained; means± SEMofthe circulat-
ing corticosterone concentrations in HS and LS quail
exposed to the immobilization stressor were 19.4 ± 0.7
and 10.0 ± 0.5 ng/mL, respectively. Selection pressure
was reimposed to produce G21wherein immobilization
resulted in corticosterone concentrations of 29.0 ± 5.4
(Jones and Satterlee, 1996). The lines reproduced for an
additional three generations without selection before
their use in the present study (G24).
The quail studied were taken from a larger population
of an 800-bird hatch. Egg incubation and chick brooding,
feeding, and lighting procedures were similar to those
described elsewhere (Jones and Satterlee, 1996) with the
sex, mixed-line groups of 50 within each of 16 compart-
ments of two Model 2SD-12 Petersime brooder batteries3
modified for quail. To maintain the line identity of each
bird, leg bands (placed on chicks at hatching) were re-
placed with permanent wing bands at 14 d of age.
of 10 females and 5 males were placed into 24 colony
cages for breeding (i.e., 12 cages of LS quail and 12 cages
of HS quail). Throughout a laying cycle of 26 wk, and
when extra birds were available, breeders that had natu-
rally succumbed, escaped, or were removed for mechan-
ical reasons were replaced by same-line, same-sex, full-
sibling quail from the original hatch. This practice in-
sured that a high number of experimental units would
be available in each line at the end of the experiment as
the intent was to take measurements on aged quail (32
wk). We reasoned that if exposure to stressful stimula-
tion is positively related to age, then environmentally
SATTERLEE ET AL.
TABLE 1. Bilateral trait means ± SEM for sides (left, L; right, R) and side differences (L minus R)
at 32 wk of age by quail line1
TraitLRL − RLRL − R
35.43 ± 0.12
2.71 ± 0.02
18.79 ± 0.06
35.44 ± 0.11
2.67 ± 0.02
18.66 ± 0.06
−0.01 ± 0.04
0.04 ± 0.01
0.13 ± 0.06
35.93 ± 0.14
2.66 ± 0.02
19.07 ± 0.07
35.90 ± 0.13
2.64 ± 0.02
19.00 ± 0.10
0.03 ± 0.06
0.02 ± 0.01
0.07 ± 0.09
1LS = low stress; HS = high stress.
2Distance between nares and auditory canal.
induced line differences of developmental instability
should be more pronounced in older quail.
Breeders received a laying diet (21% CP and 2,750 kcal
ME/kg) and water ad libitum. A 16 h light:8 h darkness
photoperiod was provided with lights-on occurring at
0500h. Dailymaintenanceand feedingchores weredone
at the same time each day (0800 h).
At 32 wk of age, 53 LS and 60 HS quail were killed
by cervical dislocation. Data were obtained for the fol-
lowing bilateral traits: length of the metatarsus (shank)
(SHL), diameter of the shank (SHD) perpendicular to
the spur, and distance between the auditory canal and
the nares (face length, FL). The same person (blinded
to the experimental protocol) made all quail measure-
ments to the nearest 0.1 mm.
Having calculated absolute FA as the unsigned left-
right character sizes, many researchers (e.g., Yang et al.,
1997; Moller et al., 1999) have then divided this value
(RA) for each bilateral trait. Once summed, these mea-
sures were averaged to produce an overall RA score for
each animal. However, the use of unsigned asymmetries
is thought to be contentious (Swaddle et al., 1994; Gan-
gestad and Thornhill, 1998; Thomson, 1999) because it
violates the assumptions of normally distributed errors
as well as that of homogeneous variance inherent in
techniques like ANOVA, t-tests, and multiple linear re-
gressions. Therefore, in line with Thomson (1999), these
problems were avoided when comparing FA in the two
lines by focusing on intra-individual variance for SHL,
eral variance in trait size, Vind, was calculated from
Vind= (L2+ R2) − ([L + R]2/2).
After calculating Vindfor each individual (quail) within
a given trait and line, trait variance data were fitted to
a general linearized model with gamma errors and a
log link function using PROC GENMOD (SAS Institute,
1985). The dependent variable in each model was quail
line, and the independent variable was Vindof a given
68, 95, and 99.7% of the values in a normal population
are within one, two, or three standard deviations of the
population mean, respectively (SAS Institute, 1985), the
signed asymmetries of each trait were found to be nor-
mally distributed within the population mean for the
of directional asymmetry or antisymmetry. This justified
the use of Thomson’s (1999) procedures for the assess-
ment of intra-individual variance in trait size and subse-
quent analysis for line differences in developmental in-
Table 1 gives bilateral trait line means by side and
mean trait differences (left side minus right side) by
line. Table 2 shows the influence of quail line on intra-
individual, interlateral variances in bilateral trait sizes
(VLSvs. VHS) and the likelihood-ratio test statistic (X21)
associated with each line comparison. Significantly
greater bilateral trait size variances were associated with
the measurement of SHL (P < 0.0088) and FL (P < 0.0016)
in the HS line than in the LS line; however, quail of
the HS line did not show a significant tendency toward
greater (P = 0.22) SHD variance when compared with
quail of the LS line.
than LS quail in two (SHL and FL) of the three bilateral
TABLE 2. Influence of quail line (low stress, LS; high stress, HS)
on intra-individual, interlateral variances in bilateral trait sizes
(VLSvs. VHS)1and the likelihood-ratio test statistic (X21)
associated with each line comparison
Probability > X
1After measuring traits on each side of the body (L = left; R = right),
from Vind= (L2+ R2) − ([L + R]2/2) (see Thomson, 1999).
DEVELOPMENTAL INSTABILITY AND ADRENOCORTICAL RESPONSIVENESS
traits measured in the present study. The interlateral
variance for SHD was also numerically larger in HS than
LS birds, but this difference was not significant. These
findings suggest that developmental instability is more
pronounced in quail from a genetic line selected for
high plasma corticosterone responses rather than low
responses to brief immobilization (Satterlee and John-
Not only do quail of the HS line show greater adreno-
cortical responsiveness to a wide range of stressors, but
also they are also more easily frightened than LS birds
(Satterlee and Johnson, 1988; Jones et al., 1992a, 1999;
Given such contrasting levels of stress susceptibility and
underlying fearfulness, we might tentatively suggest
that the line differences in FA found here reflected the
birds’ differential responsiveness to social and physical
environmental stressors, such as those associated with
mating behavior, interbird aggression, human traffic,
cage cleaning, and extraneous noises. However, al-
though both lines have been subjected to a similar
amount of inbreeding, it cannot be guaranteed that the
resultant effects of such genetic stress (Parsons, 1990;
Moller and Swaddle, 1997) on FA would have been iden-
tical in the LS and HS birds. The inclusion of truly ran-
dombred, nonselected lines or reciprocal crosses in fu-
ture comparisons will allow meaningful assessment of
the role of genetic (directional selection) stress. Quail of
the nonselected Louisiana State University line were not
included in this study because they have been subjected
to inbreeding (genetic stress) over 24 generations.
The possible exploitation of individual variation in
fearfulness and in susceptibility to stress has important
implications for poultry breeding, welfare, and perfor-
mance. For example, in addition to the reduced stress-
responsiveness and fearfulness accompanying selection
of theLS line, thesequail also showlower stress-induced
osteoporosis and greater body weight than HS quail
(Satterlee and Roberts, 1990; Jones, 1996). Furthermore,
the present findings suggest that such genetic selection
stable phenotypes. The present results also support pre-
vious suggestions that measuring asymmetries in bilat-
eral traits could be an additional and valid method of
assessing stress and the suitability of housing systems
(Yang et al., 1997; Moller et al., 1999). Divergent lines,
such as the HS and LS quail, represent valuable models
for resolving some of these issues.
The contribution of R. B. Jones was supported by the
Roslin Institute and the Biotechnology and Biological
Sciences Research Council. The authors are also grateful
Science, Louisiana State University, Baton Rouge, LA
70803, for their technical assistance and to P. B. Siegel
and A. Yang of the Department of Poultry Science, Vir-
ginia Tech, Blacksburgh, VA 24061, for their helpful
Aggrey, S., and K. M. Cheng, 1994. Animal model analysis of
genetic (Co) variances for growth traits in Japanese quail.
Poultry Sci. 73:1822–1828.
genetics. World’s Poult. Sci. J. 50:227–235.
Gangestad, S. W., and R. Thornhill, 1998. The analysis of fluc-
tuating asymmetry redux: The robustness of parametric sta-
tistics. Anim. Behav. 55:497–501.
Jones, J. S., 1987. An asymmetrical view of fitness. Nature
Jones, R. B., 1996. Fear and adaptability in poultry: Insights,
implications and imperatives. World’s Poult. Sci. J.
Jones, R. B., 1997. Fear and distress. Pages 75–87 in: Animal
Welfare. M. C. Appleby and B. O. Hughes, ed. CAB Interna-
tional, Wallingford, UK.
Jones, R. B., and D. G. Satterlee, 1996. Threat-induced behav-
ioural inhibition in Japanese quail genetically selected for
contrasting adrenocortical response to mechanical restraint.
Br. Poult. Sci. 37:465–470.
Jones, R. B., D. G. Satterlee, and G. G. Cadd, 1999. Timidity in
Japanese quail: Effects of vitamin C and divergent selection
for adrenocortical response. Physiol. Behav. 67:117–120.
Jones, R. B., D. G. Satterlee, and F. H. Ryder, 1992a. Fear and
distress in Japanese quail chicks of two lines genetically
selected for low or high adrenocortical response to immobi-
lization stress. Horm. Behav. 26:385–393.
Jones, R. B., D. G. Satterlee, and F. H. Ryder, 1992b. Open-
field behavior of Japanese quail genetically selected for low
or high plasma corticosterone response to immobilization
stress. Poultry Sci. 71:1403–1407.
Jones, R. B., D. G. Satterlee, and F. H. Ryder, 1994. Fear of
humans in Japanese quail selected for low or high adreno-
cortical response. Physiol. Behav. 56:379–383.
Mills, A. D., and J. M. Faure, 1992. The behaviour of domestic
quail. Pages 1–16 in: Nutztierethologie. M. Nichelmann, ed.
Gustav Fisher Verlag, Berlin, Germany.
Moller, A. P., G. S. Sanotra, and K. S. Vestergaard, 1995. Devel-
opmental stability in relation to population density and
breed of chickens Gallus gallus. Poultry Sci. 74:1761–1771.
Moller, A. P., G. S. Sanotra, and K. S. Vestergaard, 1999. Devel-
opmental instability and light regime in chickens (Gallus
gallus). Appl. Anim. Behav. Sci. 62:57–71.
Moller, A. P., and J. P. Swaddle, 1997. Asymmetry, Develop-
mental Stability and Evolution. Oxford University Press,
Palmer, A. R., and C. Strobeck, 1986. Fluctuating asymmetry:
Measurement, analysis, patterns. Annu. Rev. Ecol. Syst.
Parsons, P. A., 1990. Fluctuating asymmetry: An epigenetic
measure of stress. Biol. Rev. 65:131–145.
SAS Institute, 1985. SAS? User’s Guide. Statistics. Version 5
Edition. SAS Institute Inc., Cary, NC.
Satterlee, D. G., and W. A. Johnson, 1988. Selection of Japanese
quail for contrasting blood corticosterone response to im-
mobilization. Poultry Sci. 67:25–32.
Satterlee, D. G., and R. B. Jones, 1994. Fear and adrenocortical
responses to human contact in Japanese quail selected for
low or high plasma corticosterone response to immobiliza-
tion stress. Poult. Sci. 73(Suppl. 1):160. (Abstr.).
Satterlee, D. G., and R. B. Jones, 1995. Timidity in Japanese
quail divergently selected for low or high adrenocortical
response to immobilization stress. Poultry Sci. 66(Suppl.
SATTERLEE ET AL.
Satterlee, D. G., and E. D. Roberts, 1990. The influence of stress
treatment on femur cortical bone porosity and medullary
bone status in Japanese quail selected for high and low
blood corticosterone response to stress. Comp. Biochem.
Siegel, P. B., and W. B. Gross, 1980. Production and persistence
of antibodies in chickens to sheep erythrocytes. 1. Direc-
tional selection. Poultry Sci. 59:1–5.
Swaddle, J. P., M. S. Witter, and I. C. Cuthill, 1994. The analysis
of fluctuating asymmetry. Anim. Behav. 48:986–989.
Thomson, D. L., 1999. Intraindividual variance in trait size
and the analysis of developmental instability. Anim. Behav.
Van Valen, L., 1962. A study of fluctuating asymmetry. Evolu-
Yang, A., E. A. Dunnington, and P. B. Siegel, 1997. Develop-
mental stability in stocks of White Leghorn chickens. Poul-
try Sci. 76:1632–1636.
Yang, A., and P. B. Siegel, 1998. Asymmetries and heterosis of
bilateral traits in parental lines of chickens and their F1
crosses. J. Anim. Breed. Genet. 115:105–111.