Behavioral Ecology

doi:10.1093/beheco/arh141
Advance Access publication 18 August 2004

Allometry and sexual selection of male

weaponry in Wellington tree weta,
Hemideina crassidens

Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022

Clint D. Kelly
Department of Zoology, University of Toronto at Mississauga, Mississauga, Ontario, Canada L5L 1J7

Both male and female Wellington tree weta, Hemideina crassidens, use cavities in trees as diurnal shelters. That these galleries are
often limiting in nature offers males the opportunity to increase their reproductive success by monopolizing galleries and the
females residing in them. Male H. crassidens, can mature at either the 8th, 9th, or 10th instar, whereas females mature at the 10th
instar only, and male head (and mandible) size positively covaries with ultimate instar number. It has been suggested that males
fight for control of galleries by using their enlarged mandibles as weapons, and males with larger mandibles control galleries with
more females. In the present study, I present a statistical examination of sexual dimorphism, showing that traits related to head
size are on average significantly larger in males, whereas traits related to body size are on average significantly larger in females. I
tested three predictions addressing the hypothesis that sexual selection is driving megacephaly in male H. crassidens. First, as
predicted, traits related to head size show a positive allometric relationship with body size in males but not in females. Second,
adapting a novel statistical technique based on maximum likelihood and bootstrapping revealed that males, but not females,
exhibit a multimodal distribution in head and body size traits. This is likely a consequence of males maturing at one of three
instars, which results in positive covariance between the ultimate instar number and morphological traits. Third, as predicted,
single adult males with larger heads reside in galleries housing larger groups of adult females. Key words: allometry, harem success,
sexual selection, tree weta, weaponry. [Behav Ecol]

ales in many animal species control resources required
M for breeding by sexually receptive females or by compet-
ing directly for control of harems (Emlen and Oring, 1977).
This situation creates intense intrasexual selection among
males and can lead to selection for behaviors (Andersson,
1994) or traits (e.g., elongate mandibles, antlers, and horns of
particular beetles and flies; for review, see Dodson, 2000;
Emlen and Nijhout, 2000) that enable males to be more com-
petitive in contests for resources or harems.
One result of intense competition among males for mates is
selection for alternative reproductive strategies and tactics in
which dominant males exhibit one phenotype and sub-
ordinate males exhibit another (for review, see Brockmann,
2001). The production of two or more distinct phenotypes
can result in a discontinuous or multimodal distribution of
characters such as weapon size (Brockmann, 2001; Emlen and
Nijhout, 2000). The evolution of two different alternative
strategies, each with its own morphology and behavior, can
evolve and is common (Shuster and Wade, 2003); however,
the evolution of three or more alternative phenotypes is uncom-
mon (Sinervo and Lively, 1997; Zimmerer and Kallman,
1989; but see Shuster, 1987).
Characters under sexual selection often show allometric
values greater than 1.0 when scaled on body size; that is, larger
individuals have disproportionately larger trait values than do
smaller individuals of a species (for review, see Andersson, 1994;
Emlen and Nijhout, 2000; but see Eberhard, 2002). For
example, positive allometry is exhibited by traits used by
females in choosing a male mate (Alatalo et al., 1988; Green,
1992; Petrie, 1988, 1992) and traits used by males in combat
Address correspondence to C.D. Kelly. E-mail: cdkelly@utm.
utoronto.ca.
Received 2 October 2003; revised 14 May 2004; accepted 1 June
2004.
Behavioral Ecology vol. 16 no. 1
International Society for Behavioral Ecology 2005; all rights reserved.
with rivals (Baker and Wilkinson, 2001; Eberhard, 2002;
Eberhard and Gutiérrez, 1991; Forsyth and Alcock, 1990;
Gould, 1974; Hanley, 2001; Palestrini et al., 2000; Simmons and
Tomkins, 1996; Tatsuta et al., 2001). Although the reason why
sexually selected traits tend to exhibit positive allometry is not
certain, Baker and Wilkinson (2001) suggested that it is because
smaller-bodied individuals have relatively little to gain from
high investment in such traits. On the other hand, Petrie (1988)
suggested that only traits used to signal competitive ability
to conspecifics should exhibit positive allometry, and there is
no reason to expect allometric slopes greater than one in
structures that function only as weapons. Empirical evidence
appears to contradict Petrie’s (1988) suggestion, however, as
Eberhard and colleagues (Eberhard and Gutiérrez, 1991;
Eberhard et al., 1998) have demonstrated positive allometry
in structures in beetles (horns) and earwigs (cerci) that appar-
ently serve strictly as weapons, although a signaling function for
these devices has not been ruled out (Eberhard, 2002).
The net effect of sexual selection, therefore, could be the
evolution of alternative reproductive phenotypes within a sex
with each phenotype expressing differently a trait that is
positively allometric and unimodal in distribution.
Hemideina tree weta (Orthoptera: Tettigonioidea: Anosto-
stomatidae) are a group of large, flightless, nocturnal insects
endemic to New Zealand (Gibbs, 2001). Anostostomatids
exhibit various forms of male weaponry, including elephan-
tine tusks (e.g., Motuweta isolata, Gibbs, 2001) and enlarged
mandibles (e.g., Hemideina tree weta and Anostostoma king
crickets, Field and Deans, 2001; Gwynne and Jamieson, 1998;
see also Hudson, 1920; Koning and Jamieson, 2001; Moller,
1985; Spencer, 1995).
In the sexually dimorphic Wellington tree weta, Hemideina
crassidens, males use their enlarged heads and mandibles in
fights for possession of shelters in tree cavities (Kelly CD,
unpublished data). Within these cavities, known as galleries,
one to several females refuge during the day (Field and
146
Figure 1
Frequency distribution of head length for F1 male H. crassidens
(n ¼ 179) raised in the laboratory until sexual maturity. Final moult
can occur at either the 8th, 9th or 10th instar. Eggs were collected
from adult female tree weta living around Wellington, New Zealand.
The overlap among distributions was not made visible in this
figure. Modified from Spencer (1995) with permission.
Sandlant, 1983; Hudson, 1920). Mating system theory predicts
that a limited number of galleries within a location will create
the opportunity for males to monopolize them and, thus,
females (Emlen and Oring, 1977). Furthermore, galleries of
large size or high quality are expected to accommodate more
females and thus create the opportunity for males to
monopolize several females simultaneously (Emlen and
Oring, 1977). Males that control galleries with several females
have the potential for higher reproductive success because
copulations occur within the cavity and at the cavity entrance
(Field and Jarman, 2001). Therefore, competition for
galleries containing more females should be more intense,
and thus, males with larger mandibles should control them.
Little is known about the reproductive behavior of H.
crassidens in the wild. Similar to its congeners, this sexually
dimorphic species is hypothesized to be resource-defense
polygynous (Field and Jarman, 2001). Another unconfirmed
aspect of H. crassidens reproductive biology is the suggestion
that males adopt alternative reproductive strategies (Field and
Deans, 2001; Spencer, 1995). In the laboratory, male H.
crassidens, but not females, mature at the 8th, 9th, or 10th
instar (Barrett, 1991; Spencer, 1995; Stringer and Cary, 2001).
A demonstration that precocially maturing males possess
smaller heads, on average, than do males that delay
maturation to the 10th instar (Figure 1) led Spencer (1995)
to suggest that an irreversible alternative mating strategy
(Brockmann, 2001) may exist with smaller males adopting
a wandering or sneaking tactic to obtain matings opportunis-
tically whereas larger males fight for control of galleries.
In this article, I present a statistical examination of sexual
dimorphism in H. crassidens. I also test three predictions to
address the hypothesis that sexual selection is driving the
enlargement of head size in male H. crassidens. First, if male
head size is under sexual selection, then I predict that traits
related to head size will scale in a positively allometric manner
against body size. Second, if laboratory-reared males exhibit
a trimodal distribution in head size because they mature at
one of three instars and head size is positively related to instar
number (Figure 1), then I predict that wild-caught males will
also exhibit a trimodal distribution in head size. I test this
hypothesis by using a novel statistical technique developed to
determine the number of instars present in samples of
fossilized arthropods (Hunt and Chapman, 2001). Third, if
males with larger heads are more likely to monopolize
galleries, then I predict that single adult males with larger
heads will reside in galleries housing larger groups of adult
females. In contrast to previous studies of Hemideina tree weta
(Gwynne and Jamieson, 1998), my observations are on a large
Behavioral Ecology
(approximately several thousand) apparently interbreeding
population of H. crassidens.
METHODS
This study was conducted February 2001, March–May 2002, and
March–April 2003 on Te Hoiere/Maud Island, New Zealand
(41
029 S, 173
549 E), a 309-ha scientific reserve free of alien
predators (e.g., rodents [Mus and Rattus spp.] and stoats
Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022
[Mustela erminea]). The endemic morepork, Ninox novaeseelan-
diae, is the only known predator of adult tree weta on the island.
On Maud Island, H. crassidens readily use bird nest-boxes as
shelter. These boxes were installed on trees in the early 1990s
by the Ornithological Society of New Zealand for a rifleman,
Acanthisitta chloris, breeding program. All nest-boxes have
since been abandoned by the birds and, in addition to H.
crassidens, are now occupied by common geckos, Hoplodactylus
maculatus. Each nest-box (hereafter referred to as artificial
gallery) measured approximately 150 3 100 3 95 mm with
a 25-mm entrance hole. Each census involved permanent
removal of all occupants and was separated by several days
to allow re-occupation. I also destructively sampled approxi-
mately 180 natural galleries by splitting open branches on
dead trees. This sampling method did not permit me to deter-
mine gallery size. I report data for only those galleries contain-
ing weta. Galleries without weta are not reported because it is
not possible to determine whether the cavity was or was not
suitable for occupation by tree weta.
In all years, I opportunistically collected adult H. crassidens
by scanning vegetation at night. In 2002 and 2003, I also
sampled natural and artificial galleries. For each censused
cavity, I counted all individuals and noted the sex and relative
age (immature versus adult) of each occupant.
By using digital calipers (Mitutoyo Digimatic), I measured
each adult weta to the nearest 0.05 mm for each adult left and
right hind tibia length, left and right femur length, left and
right mandible length, total head length, head width at eyes,
and ovipositor length (of females). Body weight was measured
to the nearest 0.1 g with a 10-g Pesola scale. All weta were
released back into the wild after being measured and
individually marked with bee tags (H. Thorne Ltd.); marking
prevented resampling individuals. I note that I did not
recapture any individuals marked in previous years, suggest-
ing that H. crassidens either live for 1 year only as an adult or
they move to areas not censused by me. This life history
contrasts with that of the mountain stone weta, H. maori, in
which adults live for more than 1 year ( Jamieson et al., 2000).
Data analysis
Before statistical testing, data were graphically inspected for
normality and heteroscedasticity by using normal quantile
plots and box plots, respectively (Quinn and Keough, 2002;
Sokal and Rohlf, 1995). All morphometric variables were
normal for females and did not require transformation;
however, the number of females in a gallery could not be
normalized. Because all male variables were not normal, I
attempted to transform each by using the Box-Cox procedure
(Sokal and Rohlf, 1995); however, only the distribution of
male mean tibia length was made normal. All data are
presented mean 6 SE.
All measurements of the head (length, width, left mandible
length, right mandible length) and body (weight, mean tibia
length, mean femur length, ovipositor length) were pairwise
correlated by using either Pearson correlation (female data
only) or Spearman rank correlation (male data only). To
correct for multiple comparisons of correlation coeffi-
cients, the sequential Bonferroni test (Rice, 1989) was used.
Kelly • Sexual selection on weaponry of male tree weta 147

I describe the allometric relationship between log head size Table 1

and log body size in males and females by using major axis Correlation coefficients of head size and body size measurements in
regression and assume equal variance in both variables both sexes of Hemideina crassidens
(McArdle, 1988). Koning and Jamieson (2001) showed that
in H. maori, femur length is a good indicator of body size, and Male Female
head length and head width are appropriate indicators of
Variables rp n rs n
sexual dimorphism. All head traits and body traits for H.
crassidens are positively correlated in both sexes (Table 1). I Left mandible length vs.
report the relationship between femur length and head width right mandible length .98 353 .60 42

Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022

and femur length and head length separately. All variables in Left mandible length vs.
this analysis were log transformed. Differences in major axis head length .99 353 .65 42
regression slopes were tested by using a procedure analogous Left mandible length vs.
to Student’s t test (Clarke, 1980). head width .97 353 .53 42
I use the approach of Eberhard and Gutiérrez (1991) to Head length vs. head
detect a nonlinear relationship between the trait of interest width .99 359 .89 386
and body size. This technique not only allows the determina- Tibia vs. femur .94 350 .89 380
tion of whether a dimorphism exists but also establishes the Body weight vs. femur .90 331 .71 365
exact body size at which a developing individual is most likely Body weight vs. tibia .89 324 .69 363
to switch from one morph to another. To determine if the Body weight vs.
relationship between head size and body size is linear, they ovipositor — — .56 351
recommend the model Ovipositor vs. tibia — — .69 366
Ovipositor vs. femur — — .70 367
Y ¼ b0 þ b1 X þ b2 X 2 þ e; ð1Þ

in which Y is the log of head length or width, X is the log of
femur length, bi is the regression coefficients, and e is the
random component with assumed normal distribution, mean
zero, and common variance. If b2 does not differ significantly
from zero, then the relationship between head size and body
size shows no significant deviation from linearity, and
consequently, determination of the body size switchpoint is
not justified (Eberhard and Gutiérrez, 1991).
Eberhard et al. (2000) and Kotiaho and Tomkins (2001)
suggest modifying Expression 1 to discriminate morphs based
on the dimorphic character itself rather on some correlate
of it, most commonly, body size. To accomplish this, they
recommend simply substituting the Y with X and X with Y
to get
Y ¼ b0 þ b1 X þ b2 Y 2 þ e; ð2Þ
where the terms are as defined in Expression 1.
If the relationship between male head size and body size is
discontinuous and a switchpoint exists, I tested whether
additional switchpoints were present as outlined in Eberhard
et al. (2000). If males exhibit a trimorphsim in head size, then
it is possible that two switchpoints exist (i.e., a separate slope
for each instar or morphotype).
It is possible for a trait to be multimodally distributed and
exhibit a linear relationship with body size if each mode
exhibits continuous variation in trait size. Therefore, males
could have alternative strategies based on morphotype
without exhibiting the clearly demarcated switchpoints seen
in beetles and earwigs (see Eberhard and Gutiérrez, 1991).
Detecting multimodality is important to this study because
adult male H. crassidens are known to exhibit trimodality in
the laboratory associated with variation in instar number. I am
testing the prediction that trimodality exists in the wild as
well. Therefore, in cases in which I found a linear relation-
ship, I identified whether trait distributions comprised
a mixture of normal distributions in both males and females
in two separate steps. First, I visually inspected normal
quantile plots to determine whether each trait was normally
distributed. Normal quantile plots are recommended (Sokal
and Rohlf, 1995) because other methods have either low
power (e.g., Kolmogorov-Smironov) or perform poorly when
there are tied data (e.g., Shapiro-Wilks; Zar, 1999). If the trait
was not normal, I used maximum-likelihood analysis of
mixture models (mixture models analysis [MMA], Hunt and
All correlations were significant at the appropriate p value (male
pcorrected ¼ .007 and female pcorrected ¼ .005) after Bonferroni
correction.
Chapman, 2001), which was developed to evaluate hypotheses
that multiple normal distributions comprised trait distribu-
tions (e.g., different arthropod instars). This statistical
approach provides an objective comparison of various
plausible hypotheses of grouping when faced with a data set
suspected of comprising more than one subdistribution.
Briefly, MMA achieves this by (1) choosing the model with
the number of groups that best fits the data (i.e. the best-
supported log-likelihood score for the hypothesized number
of groups) and (2) estimating the parameters (means,
variances, and mixing proportions) of each of the subdis-
tributions for the best-fitting model (Hunt and Chapman,
2001). Next, MMA uses parametric bootstrapping to compare
hypotheses that different numbers of groups comprise the
data. The power of a bootstrap test was calculated when the
bootstrap failed to reject the null hypothesis (Hunt and
Chapman, 2001).
All statistical analyses were performed by using JMP 5.0
(SAS Institute, 2002) for Mac OSX and MMA software pack-
age for MS-DOS available at www.paleodb.org/paleosource
(see also Hunt and Chapman, 2001).
RESULTS
Male and female morphology
A total of 110 (59 males and 51 females), 325 (143 males and
182 females), and 317 (159 males and 158 females) adult tree
weta were collected and measured in 2001, 2002, and 2003,
respectively. There were no significant within-sex differences
among years for any measured morphological trait (one-way
ANOVA, all tests p . .05). Therefore, data for the 3 years were
pooled for each sex.
There was male-biased sexual dimorphism in the four traits
related to head size and female-biased dimorphism in the
three traits related to body size (Table 2). Males showed more
variation than did females in all measured traits as reflected in
relatively high coefficients of variation (CVs) (Table 2). As in
other species of Hemideina (Field and Deans, 2001; Gwynne
and Jamieson, 1998), there was directional asymmetry in
148 Behavioral Ecology

Table 2
Sample size, mean, and CV for absolute values of head size (mm) and body size (mm) in male and female Hemideina crassidens

Males Females
Dimorphism
Trait Mean 6 SE Range CV (%) n Mean 6 SE Range CV (%) n (% difference)

Weight 3.95 6 0.07 1.757.55 32.08 332 4.75 6 0.05 2.508.20 21.11 368 þ16.84
Left mandible length 10.18 6 0.14 5.2716.87 25.64 353 5.19 6 0.08 4.336.59 9.72 42 49.02
Right mandible length 9.24 6 0.12 4.9214.82 24.80 360 4.79 6 0.06 4.085.49 7.78 42 48.16

Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022

Head length 20.88 6 0.24 12.7631.72 21.63 359 13.04 6 0.05* 10.7815.91 6.82 386 37.55
Head width 10.69 6 0.10 7.4114.78 16.77 361 8.51 6 0.03* 6.9810.33 6.37 386 20.39
Tibia length 19.52 6 0.10 15.2524.08 9.57 351 21.86 6 0.08 17.1626.37 7.16 382 þ10.70
Femur length 18.47 6 0.09 14.9122.62 9.20 359 20.83 6 0.07 17.0023.82 6.38 385 þ11.33
Ovipositor length — — — — 19.96 6 0.08 15.6224.51 7.81 373

Values are mean 6 SEM or n. Differences between the sexes in each trait were tested by using a Student’s t test or Mann-Whitney U test
(denoted by *). Positive values for dimorphism denote female-biased difference and negative values denote male-biased difference. All
tests p , .001.

mandible length, with the left mandible always being longer
than the right in both sexes.
Both head size traits were positively correlated with body
size in all 3 years for each sex (Tables 1 and 3). Because the
slopes of the relationship between each head size trait and
femur length did not differ among years for each sex, the data
were pooled by trait and by sex. The relationship between
head length and femur length was positively allometric in
males, as the slope was significantly greater than two (Figure 2
and Table 3). In females, however, the slope did not differ
from one and was therefore isometric (Figure 2 and Table 3).
The relationship between head width and femur length was
also positively allometric in males (slope . 1.0) but not in
females (Figure 2 and Table 3). Males had significantly
steeper slopes than did females for each trait (test for
heterogeneity of slopes: head length, T ¼ 9.78, df ¼ 354.50,
p , .0001; head width, T ¼ 8.35, df ¼ 355.34, p , .0001).
The relationship between head size and body size did
not differ significantly from linearity (log10 head length b2 ¼
0.99 6 0.97, t ¼ 1.02, df ¼ 354, p ¼ .31; log10 head width b2 ¼
1.17 6 0.70, t ¼ 1.66, df ¼ 356, p ¼ .10 ), nor did the rela-
tionship between body size and head size (model 2; log10 head
length b2 ¼ 0.06 6 0.09, t ¼ 0.69, df ¼ 354, p ¼ .49; log10 head
width b2 ¼ 0.09 6 0.13, t ¼ 0.68, df ¼ 356, p ¼ .50). Given
that the slopes (b2) did not differ from zero in any case,
all relationships were linear and did not warrant further anal-
ysis of body size switchpoints. A continuous relationship is
expected, despite the possibility of a trimodal distribution, if
at least one individual is present in each trait-value bin and
there are no clear breaks in the trait distribution.
As predicted, inspection of normal quantile plots for data
pooled for 3 years showed that female head length, head
width, and femur length were normally distributed (Figure
3a–c) whereas male head length, head width, and femur
length were clearly not normal (Figure 3d–f). Male head
length was not normal in each of the three sampling years
(Figure 4), suggesting that the nonnormal distribution of the
pooled data is not an artefact of biased sampling for a
particular head size in any one year. That is, all traits mea-
sured in both males and females exhibited similar distribution
patterns among years.
MMA suggests that there is a significantly higher likelihood
that the distribution for each male trait is bimodal rather than
trimodal (Table 4). I am confident that a two-group model

Table 3
Slope, SE of slope, and the intercept of the allometric relationship between log head length and log femur length for male and
female Hemideina crassidens

Year Sex n Slope SE Intercept Lower CL Upper CL r2 p

Head length
2001 female 49 1.21 0.11 0.47 0.92 1.60 0.54 ns
male 59 2.47 0.12 1.82 2.21 2.78 0.84 ,.05
2002 female 174 1.08 0.05 0.30 0.94 1.22 0.67 ns
male 141 2.43 0.06 1.76 2.30 2.58 0.89 ,.05
2003 female 158 1.07 0.05 0.30 0.94 1.22 0.61 ns
male 157 2.44 0.06 1.77 2.32 2.57 0.90 ,.05
pooled female 381 1.09 0.03 0.32 1.00 1.17 0.63 ns
male 357 2.44 0.04 1.78 2.36 2.54 0.89 ,.05
Head width
2001 female 49 1.05 0.09 0.46 0.85 1.31 0.65 ns
male 59 1.86 0.09 1.33 1.68 2.08 0.86 ,.05
2002 female 174 1.00 0.04 0.39 0.92 1.09 0.76 ns
male 141 1.88 0.05 1.36 1.79 1.99 0.91 ,.05
2003 female 158 0.99 0.05 0.38 0.88 1.11 0.65 ns
male 159 1.84 0.04 1.31 1.75 1.94 0.91 ,.05
Pooled female 381 1.00 0.03 0.39 0.94 1.07 0.71 ns
male 357 1.86 0.04 1.33 1.80 1.93 0.90 ,.05

P value denotes significance test that slope is . 1.0.

Kelly • Sexual selection on weaponry of male tree weta
Figure 2
Allometric relationships between head length (A) and head width (B) on mean femur length in male (empty circles) and female
(filled circles) H. crassidens. See Table 3 for statistical details.
better explains the distributions of both head length and
femur length (body size) better than does a one-group model.
However, I am more cautious about choosing between two
and three groups for femur length because this test was nearly
significant at the 0.05 a-level (p ¼ .072) (Table 4) and because
tests for head length and femur length each suffered from low
statistical power (0.19 and 0.17, respectively) (Table 4). Head
width trimodality was the only hypothesis rejected with high
power (Table 4). Low power suggests that even if the best-
supported three-group model were true, data such as these
would be able to reject the two-group model only about 19%
and 17% of the time (Table 4). That the mean head lengths
calculated under a three-group hypothesis (Table 5) fall
either within (intermediate ¼ 21–22 mm and large ¼ 25–26
mm) or close (small ¼ 15–16 mm) to each putative mode
identified through visual examination provides some support
for a trimodal distribution. Assuming a three-group distribu-
tion in my data, the subdistributions show considerable
overlap (Figure 3); a result consistent with Spencer’s (1995:
see also Figure 1) laboratory data (because Spencer’s raw data
were unavailable to me, I could not subject them to MMA).
This overlap explains the low statistical power of the analysis
because MMA has higher resolving power when modes are
Figure 3
Frequency distribution of head length, head width, and mean femur length in males (bottom) and females (top) for all years combined.
149
Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022
clearly demarcated (see Hunt and Chapman, 2001). That
each mode overlaps another also explains why the regressions
between head size traits and body size were linear and
continuous (see above) as opposed to having a clear break in
the distribution and, thus, a discontinuous relationship (e.g.,
beetle horns; Eberhard and Gutiérrez, 1991). When taken
together, regression analysis and MMA clearly show that head
traits are multimodally distributed but have a linear relation-
ship with body size.
Harem size and male head size
My surveys of cavities revealed up to seven adult females
residing in the same cavity with a single adult male (Figure 5).
Modal harem size was zero (range ¼ 0–5, n ¼ 29) in natural
cavities and 1.0 (range ¼ 0–7, n ¼ 79) in artificial cavities (both
years pooled). Resident male head length was significantly
positively correlated with the number of females in an artificial
cavity in 2003 (rs ¼ .34, n ¼ 55, p ¼ .011) but not in 2002 (rs ¼ .27,
n ¼ 24, p ¼ .21). Male head length was significantly positively
correlated with harem size in both natural galleries (rs ¼ .52, n ¼
29, p ¼ .004) and in artificial galleries (pooled data for 2002 and
2003, rs ¼ .33, n ¼ 79, p ¼ .003). Pooling data for both cavity
150 Behavioral Ecology

My study provides the first evidence that the multimodal

distribution of head sizes observed in the laboratory (Barrett,
1991; Spencer, 1995; Stringer and Cary, 2001) for H. crassidens
exists in nature. Unlike other Hemideina species, male H.
crassidens can mature at one of three different instars while
females mature at the 10th instar only (Barrett, 1991; Spencer,
1995; Stringer and Cary, 2001). That the bimodal, and
possible trimodal, distribution in male head length and body
size observed in the present study represents individuals

Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022

Figure 4
Frequency distribution of male head length for each of 3 years.
types over both years produced a strong positive correlation
(rs ¼ .39, n ¼ 108, p , .0001).
DICUSSION
I found female-biased body size dimorphism and male-biased
head size dimorphism in Wellington tree weta. Not only do
males possess larger heads relative to females but they also
exhibit more variation in head size. This large variation could
be attributed to the apparent trimodal distribution in male
head length which is likely a consequence of males maturing
at different instars (Spencer, 1995; Stringer and Cary, 2001)
(see also Figure 1).
maturing at a different instar is strongly supported by patterns
of head length distribution observed in laboratory-reared
male H. crassidens (Spencer, 1995). The head length of adult
males in Spencer’s (1995) study showed a clear trimodal
distribution, with each mode representing a final instar that
was positively correlated with head length (Figure 1). More
specifically, 8th instar males had a smaller modal head length
(11.0–11.5 mm, approximately 5.0% of all males) than did
10th instar males (21.0–23.5 mm, approximately 8.4% of all
males) and 9th instar males exhibited a modal head length
(15.0–15.5 mm, approximately 6.8% of all males) intermedi-
ate to the other two (Spencer, 1995) (Figure 1). That the
distribution of head sizes I observed in nature differs from
Spencer’s (1995) is expected given the greater variation in
larval feeding history in the wild compared with the
laboratory. Also, Spencer’s (1995) laboratory-reared males
may have exhibited overall smaller head sizes than those
reported here because the quality of the laboratory diet could
have been poor compared with a natural diet. Alternatively, an
intermediate mode may be rare on Maud Island because it is
selected against or alleles influencing head polymorphism
conform to Hardy-Weinberg equilibrium and renders the
intermediate mode more rare (see Shuster and Wade, 1991). I
suggest that males are trimorphic with 9th instar males being
more rare than either 8th or 10th instar individuals.
The Wellington tree weta is the only orthopteran and one
of only a few animals (e.g., the isopod, Paracerceis sculpta:
Shuster, 1987; side-blotched lizard, Uta stansburiana: Sinervo
and Lively, 1997; pygmy swordtail, Xiphophorus nigrensis:
Zimmerer and Kallman, 1989) known to exhibit trimorphism
in weaponry, body size, or ornamentation. Moreover, the
Wellington tree weta is unique in that the trimorphic
character displays continuous variation and is not discrete as
is the body morphology of male isopods, color of lizards, or
even the clearly dimorphic horns of beetles.
As predicted, male H. crassidens with larger heads, on
average, resided with larger groups of females. Smaller, 8th
instar male Wellington weta are hypothesized to assume
a sneak mating strategy whereby they opt not to control
cavities and instead acquire copulations via sneaking or forced
copulation (Spencer, 1995). However, my data show that
smaller males (, 20 mm head length) (Figure 3) do indeed
associate with females in cavities. Perhaps this is not an

Table 4
Results of the maximum likelihood analysis for the one- through three-group models for each trait in male Hemideina crassidens

Trait Bootstrap test (H1 vs. H2) Observed 2l Bootstrap results Bootstrap p Power

Head length 1 vs. 2 groups 85.32 1000/1000 ,.0001* -

2 vs. 3 groups 14.15 1712/2000 .144 0.19
Head width 1 vs. 2 groups 100.18 1000/1000 ,.0001* -
2 vs. 3 groups 8.36 322/1000 .678 0.86
Femur length 1 vs. 2 groups 16.05 977/1000 .023* -
2 vs. 3 groups 15.76 928/1000 .072 0.17

Bootstrap results are the number of bootstrap replicates that exceeded the observed value for the log-likelihood ratio (2l), out of the total
number of replicates performed. An asterisk indicates that the model with fewer groups (H1) is rejected in favor of the model with more
groups (H2) at the a ¼ 0.05 level.
Kelly • Sexual selection on weaponry of male tree weta
Table 5
Mean (mm) and variance for the two- and three-group solutions for each trait in male Hemideina
crassidens
Two group Three group
Trait x1 (variance) x2 (variance) x1 (variance)
Head length 17.20 (4.14) 25.06 (6.62) 17.24 (4.21)
Head width 9.14 (0.63) 12.13 (1.41) 9.20 (0.68)
Femur length 17.26 (1.03) 20.03 (1.02) 17.24 (1.01)
irreversible strategy (Brockmann, 2001) but simply males in
poorer condition (see below) ‘‘making the best of a bad job’’
(Dawkins, 1980), thus maturing earlier and using a condi-
tional strategy based on the prevailing environmental condi-
tions and their own phenotype (Brockmann, 2001). Whether
the phenotype of a male tree weta is determined by either
a genetic polymorphism or condition dependence remains to
be investigated.
Male H. crassidens exhibited a positively allometric relation-
ship between each head size trait and body size, whereas the
relationship in females was isometric for those same traits. In
both sexes, head length exhibited a steeper slope than did head
width, suggesting that length may be more important than is
width in male-male combat. That all relationships were also
linear suggests that males of intermediate body size produce
intermediate head sizes compared with species that exhibit
a sigmoidal or discontinuous scaling relationship (Emlen and
Nijhout, 2000). Males of species in the latter category have
either minimal or complete expression of the trait over a narrow
range of intermediate body sizes (Emlen and Nijhout, 2000).
In the congeneric H. maori, Jamieson and colleagues
(Gwynne and Jamieson, 1998; Koning and Jamieson, 2001)
have found conflicting evidence of allometry in cephalic
weaponry. Gwynne and Jamieson (1998) did not find a slope
significantly greater than 1.0 either between mandible length
and tibia length (slope ¼ 0.76) or between head width and tibia
length (slope ¼ 0.76). In contrast, Koning and Jamieson (2001)
found positive allometry and a linear scaling relationship with
femur length in male head length (slope ¼ 1.31) and width
(slope ¼ 1.26). These differences may be the result of each study
focusing on different or inappropriate characters; perhaps head
width, as in H. crassidens (the present study), is not as important
in combat as is length, and femur length may be a better
indicator of body size than is tibia length. Given that slopes were
inappropriately calculated by using model I least-squares
regression in both studies, a direct comparison of scaling
relationships between H. crassidens and H. maori is not possible.
Figure 5
The number of adult females in artificial (empty circles) and natural
(filled circles) cavities correlated with the head length of the resident
male. Data for artificial cavities pooled for both years.
151
x2 (variance) x3 (variance)
21.98 (0.001) 25.33 (5.83)
12.19 (1.20) 14.63 (0.00)
Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022
19.95 (0.97) 21.55 (0.00)
The allometric values reported here for male head length
are similar to values reported for weaponry in other insect
species. For example, in their comparative analysis of 42
earwig species, Simmons and Tomkins (1996) found that
males in 11 species showed positive allometry in forceps
length (range of model I slopes ¼ 1.47–3.50), a trait used by
males in aggressive interactions with rival males (Briceño and
Eberhard, 1995). Adult male staphylinid beetles, Leistotrophus
versicolor, use their enlarged mandibles in combat with
conspecific males over access to dung and carrion, which is
used by sexually receptive females as a foraging site (Forsyth
and Alcock, 1990). Forsyth and Alcock (1990) reported
allometric values of 2.1 and 1.8 for the relationship between
mandible length and head width and mandible length and
wing length, respectively. A sexual selection hypothesis is
supported in that species for the evolution of enlarged
mandibles because males with larger mandibles are better
fighters and mate with more females (Forsyth and Alcock,
1990). Cephalic horn length in Onthophagus beetles is
considered to be important in male-male combat (Eberhard,
1982, 1979; Emlen, 1994), and allometric values of 7.5
(Palestrini et al., 2000) and 6.23 (Eberhard et al., 1998) in
O. taurus and O. incensus, respectively, support the notion that
sexual selection is operating on this trait. Contrary to these
examples, Eberhard reported that the front tibia in the male
sepsid fly, Palaeosepsis dentatiformis, function as weapons yet do
not exhibit allometric values significantly greater than 1.0
(slope from OLS regression ¼ 0.96) and are only slightly
sexually dimorphic. He suggested that because the front tibia
also serve a highly specialized function in courtship, their
further specialization as weaponry is compromised. Similar to
that of P. dentatiformis, the front femora of males in two species
of Diptera (Chymomyza mycopelates and C. exophthalma) act as
weapons but exhibit isometric relationships with body size
(slopes from OLS regressions ¼ 0.919 and 0.745, respectively;
Eberhard, 2002). These structures do not serve in courtship
or mating and are apparently not precluded from specializa-
tion as weapons (Eberhard, 2002). Eberhard (2002) suggested
that perhaps among species not all sexually selected traits will
have allometric values significantly greater than 1.0 if the costs
of building and carrying such a structure outweigh the
benefits. Controlled laboratory studies are required to better
understand how head traits scale with body size for each adult
instar in male H. crassidens and to estimate the fitness benefits
of maturing at different instars.
I thank Darryl Gwynne, Marlene Zuk, and two anonymous reviewers
for comments on the manuscript; Steve Ward (Department of
Conservation, Marlborough Sounds, New Zealand) for logistical
support on Maud Island; and Peter Gaze and Ian Miller (Department
of Conservation, Nelson, New Zealand) for research permits. A special
thanks to Anita Spencer for generously permitting me to use
a modified figure from her thesis and to the Insect Behaviour Group
of Erindale College for valuable discussion. This research was
supported by the National Science Engineering and Research Council
(NSERC) and National Geographic Society grants to D.T.G.
152
REFERENCES
Alatalo RV, Hoglund J, Lundberg A, 1988. Patterns of variation in tail
ornament size in birds. Biol J Linn Soc 34:363–374.
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Prince-
ton University Press.
Baker RH, Wilkinson GS, 2001. Phylogenetic analysis of sexual
dimorphism and eye-span allometry in stalk-eyed flies (Diopsidae).
Evolution 55:1373–1385.
Barrett P, 1991. Keeping wetas in captivity. Wellington: Wellington
Zoological Gardens.
Briceño RD, Eberhard WG, 1995. The functional morphology of male
cerci and associated characters in 13 species of tropical earwigs
(Dermaptera: Forficulidae, Labiidae, Carcinophoridae, Pygidicra-
nidae). Washington, DC: Smithsonian Institution Press.
Brockmann HJ, 2001. The evolution of alternative strategies and
tactics. In: Advances in the study of behavior (Slater PJB, Rosenblatt
JS, Snowdon CT, Roper TJ, eds). London: Academic Press; 1–51.
Clarke MRB, 1980. The reduced major axis of a bivariate sample.
Biometrika 67:441–446.
Dawkins R, 1980. Good strategy or evolutionary stable strategy. In:
AAAS selected symposium; (Barlow GW, Silverberg S, eds). Boulder,
Colorado: Published by Westview Press for the American Associa-
tion for the Advancement of Science; 331–367.
Dodson GN, 2000. Behavior of the Phytalmiinae and the evolution of
antlers in tephritid flies. In: Fruit flies (Tephritidae): phylogeny and
evolution of behavior (Aluja M, Norrbom AL, eds). New York: CRC
Press; 175–184.
Eberhard WG, 1979. The function of horns in Podischnus agenor
(Dynastinae) and other beetles. In: Sexual selection and reproduc-
tive competition in insects (Blum MS, Blum NA, eds). New York:
Academic Press; 231–258.
Eberhard WG, 1982. Beetle horn dimorphism: making the best of
a bad lot. Am Nat 119:420–426.
Eberhard WG, 2002. Natural history and behavior of Chymomyza
mycopelates and C. exophthalma (Diptera: Drosophilidae), and
allometry of structures used as signals, weapons, and spore
collectors. Can Entomol 134:667–687.
Eberhard WG, Garcia-C JM, Lobo J, 2000. Size-specific defensive
structures in a horned weevil confirm a classic battle plan: avoid
fights with larger opponents. Proc R Soc Lond B 267:1129–1134.
Eberhard WG, Gutiérrez EE, 1991. Male dimorphisms in beetles and
earwigs and the question of developmental constraints. Evolution
45:18–28.
Eberhard WG, Huber BA, Rodriguez S. RL, Briceño RD, Salas I,
Rodriguez V, 1998. One size fits all? relationships between the size
and degree of variation in genitalia and other body parts in 20
species of insects and spiders. Evolution 52:415–431.
Emlen DJ, 1994. Environmental control of horn length dimorphism
in the beetle Onthophagus acuminatus (Coleoptera: Scarabaeidae).
Proc R Soc Lond B 256:131–136.
Emlen DJ, Nijhout HF, 2000. The development and evolution of exag-
gerated morphologies in insects. Ann Rev Entomol 45:661–708.
Emlen ST, Oring LW, 1977. Ecology, sexual selection, and the
evolution of mating systems. Science 197:215–223.
Field LH, Deans NA, 2001. Sexual selection and secondary sexual
characters of wetas and king crickets. In: The Biology of wetas, king
crickets and their allies (Field LH, ed). Wallingford: CAB Interna-
tional; 179–204.
Field LH, Jarman TH, 2001. Mating behaviour. In: The biology of
wetas, king crickets and their allies (Field LH, ed). Wallingford:
CAB International; 317–332.
Field LH, Sandlant GR, 1983. Aggression and mating behavior in the
Stenopelmatidae (Orthoptera, Ensifera), with reference to New
Zealand wetas. In: Orthopteran mating systems: sexual competition
in a diverse group of insects (Gwynne DT, Morris GK, eds). Boulder,
Colorado: Westview Press; 120–146.
Forsyth A, Alcock J, 1990. Female mimicry and resource defense
polygyny by males of a tropical rove beetle, Leistotrophus versicolor
(Coleoptera: Staphylinidae). Behav Ecol Sociobiol 26:325–330.
Gibbs G, 2001. Habitats and biogeography of New Zealand’s
Deinacridine and tusked weta species. In: The biology of wetas,
king crickets and their allies (Field LH, ed). Wallingford: CAB
International; 35–55.
Behavioral Ecology
Gould SJ, 1974. The origin and function of ‘‘bizarre’’ structures:
antler size and skull size in the ‘‘Irish Elk’’ Megaloceros giganteus.
Evolution 28:191–220.
Green AJ, 1992. Positive allometry is likely with mate choice,
competitive display and other functions. Anim Behav 43:170–172.
Gwynne DT, Jamieson I, 1998. Sexual selection and sexual dimorphism
in a harem-polygynous insect, the alpine weta (Hemideina maori,
Orthoptera Stenopelmatidae). Ethol Ecol Evol 10:393–402.
Hanley RS, 2001. Mandibular allometry and male dimorphism in
a group of obligately mycophagous beetles (Insecta: Coleoptera:
Downloaded from https://academic.oup.com/beheco/article/16/1/145/206168 by Consolidation Plus QUEEN user on 09 August 2022
Staphylinidae: Oxyporinae). Biol J Linn Soc 72:451–459.
Hudson GV, 1920. On some examples of New Zealand insects
illustrating the Darwinian principle of sexual selection. Trans Proc
NZ Inst 52:431–438.
Hunt G, Chapman RE, 2001. Evaluating hypotheses of instar-grouping
in arthropods: a maximum likelihood approach. Paleobiology 27:
466–484.
Jamieson IG, Forbes MR, McKnight EB, 2000. Mark-recapture study of
mountain stone weta Hemideina maori (Orthoptera: Anostostomati-
dae) on rock tor ‘‘islands.’’ NZ J Ecol 24:209–214.
Koning JW, Jamieson IG, 2001. Variation in size of male weaponry in
a harem-defence polygynous insect, the mountain stone weta
Hemideina maori (Orthoptera: Anostostomatidae). NZ J Zool 28:
109–117.
Kotiaho JS, Tomkins JL, 2001. The discrimination of alternative male
morphologies. Behav Ecol 12:553–557.
McArdle BH, 1988. The structural relationship: regression in biology.
Can J Zool 66:2329–2339.
Moller H, 1985. Tree wetas (Hemideina crassicruris) (Orthoptera:
Stenopelmatidae) of Stephens Island, Cook Strait. NZ J Zool 12:
1255–1269.
Palestrini C, Rolando A, Laiolo P, 2000. Allometric relationships and
character evolution in Onthophagus taurus (Coleoptera: Scarabaei-
dae). Can J Zool 78:1199–1206.
Petrie M, 1988. Intraspecific variation in structures that display
competitive ability: large animals invest relatively more. Anim Behav
36:1174–1179.
Petrie M, 1992. Are all secondary sexual display structures positively
allometric and, if so, why? Anim Behav 43:173–175.
Quinn GP, Keough MJ, 2002. Experimental design and data analysis
for biologists. Cambridge: Cambridge University Press.
Rice WR, 1989. Analyzing tables of statistical tests. Evolution 43:
223–225.
SAS Institute Inc., 2002. JMP: the statistical discovery software: release
5.0. Cary, North Carolina: SAS Institute Inc.
Shuster SM, 1987. Alternative reproductive behaviors: three discrete
male morphs in Paracerceis sculpta, an intertidal isopod from the
northern gulf of California. J Crustacean Biol 7:318–327.
Shuster SM, Wade MJ, 1991. Equal mating success among male
reproductive strategies in a marine isopod. Nature 350:608–610.
Shuster SM, Wade MJ, 2003. Mating systems and strategies. Princeton,
New Jersey: Princeton University Press.
Simmons LW, Tomkins JL, 1996. Sexual selection and the allometry of
earwig forceps. Evol Ecol 10:97–104.
Sinervo B, Lively CM, 1997. The rock-paper-scissors game and the
evolution of alternative male strategies. Nature 380:240–243.
Sokal RR, Rohlf FJ, 1995. Biometry: the principles and practice of
statistics in biological research, 3rd ed. New York: Freeman.
Spencer AM, 1995. Sexual maturity in the male tree weta Hemideina
crassidens (Orthoptera: Stenopelamatidae) (Masters). Wellington:
Victoria University of Wellington.
Stringer IAN, Cary PRL, 2001. Postembryonic development and
related changes. In: The biology of wetas, king crickets and their
allies (Field LH, ed). Wallingford: CAB International; 399–426.
Tatsuta H, Mizota K, Akimoto SI, 2001. Allometric patterns of heads
and genitalia in the stag beetle Lucanus maculifemoratus (Co-
leoptera: Lucanidae). Ann Entomol Soc Am 94:462–466.
Zar JH, 1999. Biostatistical analysis, 4th ed. Upper Saddle River, New
Jersey: Prentice Hall.
Zimmerer EJ, Kallman KD, 1989. Genetic basis for alternative
reproductive tactics in the pygmy swordtail, Xiphophorus nigrensis.
Evolution 43:1298–1307.