Courtship behavior, nesting microhabitat, and assortative mating in sympatric stickleback species pairs

Abstract

The maintenance of reproductive isolation in the face of gene flow is a particularly contentious topic, but differences in reproductive behavior may provide the key to explaining this phenomenon. However, we do not yet fully understand how behavior contributes to maintaining species boundaries. How important are behavioral differences during reproduction? To what extent does assortative mating maintain reproductive isolation in recently diverged populations and how important are “magic traits”? Assortative mating can arise as a by‐product of accumulated differences between divergent populations as well as an adaptive response to contact between those populations, but this is often overlooked. Here we address these questions using recently described species pairs of three‐spined stickleback (Gasterosteus aculeatus), from two separate locations and a phenotypically intermediate allopatric population on the island of North Uist, Scottish Western Isles. We identified stark differences in the preferred nesting substrate and courtship behavior of species pair males. We showed that all males selectively court females of their own ecotype and all females prefer males of the same ecotype, regardless of whether they are from species pairs or allopatric populations. We also showed that mate choice does not appear to be driven by body size differences (a potential “magic trait”). By explicitly comparing the strength of these mating preferences between species pairs and single‐ecotype locations, we were able to show that present levels of assortative mating due to direct mate choice are likely a by‐product of other adaptations between ecotypes, and not subject to obvious selection in species pairs. Our results suggest that ecological divergence in mating characteristics, particularly nesting microhabitat may be more important than direct mate choice in maintaining reproductive isolation in stickleback species pairs.

Full text

Ecology and Evolution. 2021;11:1741–1755.
|
1741www.ecolevol.org
Received: 16 August 2020
|
Revised: 26 Nove mber 2020
|
Accepted: 14 December 2020
DOI: 10.1002 /ece3.7164
ORIGINAL RESEARCH
Courtship behavior, nesting microhabitat, and assortative
mating in sympatric stickleback species pairs
Laura L. Dean | Hannah R. Dunstan | Amelia Reddish | Andrew D. C. MacColl
This is an op en access arti cle under the ter ms of the Creative Commons Attribution L icense, which pe rmits use, dis tribution and reproduction in any medium,
provide d the original wor k is properly cited.
© 2021 The Authors . Ecology and Evolution published by John Wiley & Sons Ltd.
School of Life Sciences, University of
Nottingham, Nottingham, UK
Correspondence
Laura L. Dean, School of Life Sciences,
University of Nottingham, University Park,
Nottingham, N G7 2RD, UK .
Email: lldean18@gmail.com
Funding information
Natural Environment Research Co uncil,
Grant/Award Number: NE/L00260 4/1 and
NE/R00935X/1
Abstract
The maintenance of reproductive isolation in the face of gene flow is a particularly
contentious topic, but differences in reproductive behavior may provide the key to
explaining this phenomenon. However, we do not yet fully understand how behavior
contributes to maintaining species boundaries. How important are behavioral dif-
ferences during reproduction? To what extent does assortative mating maintain re-
productive isolation in recently diverged populations and how important are “magic
traits”? Assortative mating can arise as a by-product of accumulated differences be-
tween divergent populations as well as an adaptive response to contact between
those populations, but this is often overlooked. Here we address these questions
using recently described species pairs of three-spined stickleback (Gasterosteus ac-
uleatus), from two separate locations and a phenotypically intermediate allopatric
population on the island of North Uist, Scottish Western Isles. We identified stark
differences in the preferred nesting substrate and courtship behavior of species pair
males. We showed that all males selectively court females of their own ecotype and
all females prefer males of the same ecotype, regardless of whether they are from
species pairs or allopatric populations. We also showed that mate choice does not
appear to be driven by body size differences (a potential “magic trait”). By explicitly
comparing the strength of these mating preferences between species pairs and sin-
gle-ecotype locations, we were able to show that present levels of assortative mat-
ing due to direct mate choice are likely a by-product of other adaptations between
ecotypes, and not subject to obvious selection in species pairs. Our results suggest
that ecological divergence in mating characteristics, particularly nesting microhabitat
may be more important than direct mate choice in maintaining reproductive isolation
in stickleback species pairs.
KEYWORDS
assortative mating, behavior, court ship, evolution, gene flow, mate choice

1742
|
DEAN Et Al.
1 | INTRODUCTION
Behavior dictates the way in which an organism interacts with
other members of the same species, with other living entities, and
with its surrounding environment (Levitis et al., 2009). It is funda-
mental for sur vival and reproduction and is particularly interest-
ing in the light of evolution, because differences in reproductive
behavior can play a crucial role in advancing, maintaining, and
breaking down boundaries between species. Gene flow between
divergent populations can lead to hybridization and homogeniza-
tion of two populations, but population-level differences and the
progression of speciation can also be maintained in the face of
gene flow. The latter has been a particularly contentious topic for
deca des (Berlo ch er & Fe der, 2002; Bi rd et al., 2012; Bolnick, 2011;
Bolnick & Fitzpatrick, 2007; Dieckmann & Doebeli, 1999;
Smith, 1966; Smith et al., 2013; Via, 2001), but today, it is ac-
cep te d that, under ce rta in circumst an ce s, new sp ecies can em er ge
while exchanging genes (Foote, 2018; Papadopulos et al., 2019;
Richards et al., 2019). Divergent mating behavior and active
mate choice provide a potential mechanism for this phenome-
non (Kirkpatrick, 2001; Liu et al., 2011), particularly with regard
to reproductive isolation between recently diverged populations
because behavioral differences can arise quickly, whereas postzy-
gotic barriers to gene flow (i.e., hybrid sterility or inviability) tend
(although not always) to require a suite of mutations which take
longer to accumulate (Coyne & Orr, 1989, 1997; Rull et al., 2013).
However, our understanding of how reproductive behavioral dif-
ferences arise and how important they are for reproductive isola-
tion in wild populations is incomplete.
Behavioral differences underlie most prezygotic isolation, either
directly via active mate choice, or indirectly via temporal or eco-
logical differences during reproduction. Divergent mating behavior
can theoretically even complete the speciation process in the face
of gene flow (Dobzhansky, 1937; Rice & Hoster t, 1993; Servedio &
Noor, 2003). Classically, these differences in mating behavior are
thought to evolve as a result of direct selection for assortative mat-
ing or via reinforcement (selection against the production of hybrids
of reduced fitness) (Pfennig, 2016). However, they can also arise as
a by-product of other adaptations without any selection for assor-
tative mating, for example, one population might evolve reduced
courtship displays to avoid predation while another might solve the
same problem by nesting/courting in dense foliage, leading to re-
duced cross-population mating as a by-product of other adaptations,
a phenomenon which is often overlooked (Rice & Hoster t, 1993;
Vines & Schluter, 2006). Low-level gene flow between divergent
populations with incomplete postzygotic barriers to hybridization
is common in nature (Campagna et al., 2014; Moritz et al., 2009;
Ravinet et al., 2013; Sa-Pinto et al., 2010), but we know far less than
we should about the prevalence and role of divergent mating behav-
ior in these populations.
Contact between phenotypically and ecologically divergent
ecotypes with varying degrees of reproductive isolation occurs
throughout the Holarctic range of the three-spined stickleback
(Gasterosteus aculeatus, hereaf ter “stickleback”), largely as a result of
their early Holocene marine to freshwater radiation (Bell et al., 2004;
Jones et al., 2012; Magalhaes et al., 2016; Taylor & McPhail, 2000).
Stickleback also perform a suite of well-characterized courtship be-
haviors (Candolin, 1997; Hughes et al., 2013; Tinbergen, 1952), mak-
ing the stickleback radiation an ideal system for investigating the
role of mating behavior in maintaining species boundaries in the face
of gene flow. Assortative mating (the ability to recognize and choose
to mate with conspecific individuals) is common between reproduc-
tively isolated stickleback ecotypes and has been documented in
benthic–limnetic (Bay et al., 2017; Kozak et al., 2011), lake–stream
(Andreou et al., 2017), lava–nitella (Olafsdottir et al., 2006), and
anadromous–freshwater resident (Furin et al., 2012) species pairs.
This assortative mating can be driven by variation in factors directly
involved in mating interactions such as mating behavior (Ishikawa &
Mori, 2000), body size (McKinnon et al., 2012), nuptial coloration
(M c K innon , 1995 ) , and nes t struc ture (B l o uw & Hage n , 199 0 ), or var i-
ation in spatial and temporal aspects of courtship that result in fine-
scale segregation of phenotypes (Borzee et al., 2016; Hagen, 1967;
Pegoraro et al., 2016; Snowberg & Bolnick, 2012). Divergence in
body size is a particularly important aspect of almost all stickle-
back species pairs and body size has been implicated as a potential
“magic trait” in this system, involved in both ecological adaptation
and assortative mating (Bay et al., 2017; Conte & Schluter, 2013;
Head et al., 2013; MacColl, 2009; McKinnon et al., 2004; Nagel &
Schluter, 1998; Schluter, 1993). However, the role of mating behav-
ior in maintaining reproductive isolation and the frequency with
which assortative mating arises because it is itself either directly or
indirectly (e.g., through reinforcement) selected for remains uncer-
tain (Bolnick & Kirkpatrick, 2012; Vines & Schluter, 2006).
The island of North Uist, Scot tish Western Isles is covered by
a mosaic of interconnected freshwater and brackish lochs and
coastal lagoons, most of which have been colonized by stickleback
since the last glacial retreat 10,000–20,000 YBP (Ballantyne, 2010).
Stickleback populations on North Uist vary ex tensively in morpho-
logical and associated genetic characteristics (MacColl et al., 2013;
Magalhaes et al., 2016). The island contains isolated allopatric eco-
types and genetically and phenotypically distinct species pairs.
Anadromous stickleback migrate from the open ocean into coastal
lagoons and streams during the spring breeding season, during
which they breed sympatrically alongside lagoon resident (hereafter
“lagoon”) and freshwater resident (hereafter “freshwater”) ecotypes
that do not migrate to sea. Anadromous fish are much larger, more
heavily armored and differ from resident ecotypes in body shape and
various trophic morphological traits such as gill raker number, re-
flecting their more pelagic lifestyle (see Figure 1 for photographs of
different ecotypes). Reproductive isolation in species pairs is strong
despite low levels of gene flow (Dean et al., 2019), and there is likely
strong selection against hybrids, many of which probably attempt to
migrate to sea without the full suite of associated traits. Hybrid fish,
for example, often exhibit an intermediate lateral plate phenotype
(Dean et al., 2019), which would likely make them more vulnerable
to predation. Assor tative mating may be impor tant for maintaining

|
1743
DEAN E t Al.
reproductive isolation in lagoon–anadromous species pairs but, to
the best of our knowledge, has never previously been tested.
Here we focus on investigating the behavioral mechanisms
responsible for maintaining reproductive isolation in recently de-
scribed lagoon–anadromous species pairs. We use no-choice mating
trials to investigate ecotype-level differences in male nesting and
courtship behavior and female mate choice. Evidence that repro-
ductively isolated ecot ypes both display positive assortative mating
and differ in nesting habits and courtship behavior would suggest
that these behavioral differences may be responsible for maintain-
ing reproductive isolation in the face of gene flow. To investigate
whether the sympatric condition of species pairs is responsible for
assortative mating, we also compare levels of assortative mating in
two sympatric species pair populations with those in a naïve land-
locked allopatric population (as a best available prox y for allopatric
populations of the same ecotypes, as species pair ecotypes do not
occur allopatrically). A preference for a corresponding ecotype in
fish from an allopatric population can only have arisen as a by-prod-
uct of other adaptations and not from either direct selection on
mating preferences or reinforcement (as there is no contact between
allo patri c ecoty pes). We, the refor e, pred ic t that if it is spe cific all y se-
lected for assortative mating between sympatric ecotypes will likely
be stronger than that of the allopatric population. Taken together,
our findings shed light on the role of behavior in adaptive divergence
and in maintaining reproductive isolation in sympatr y.
2 | MATERIALS AND METHODS
2.1 | Fish collection and husbandry
In April–May 2016 and 2017, stickleback were caught from three
lochs on North Uist: two that contain sympatric species pairs of la-
goon and anadromous fish (Obse and Faik), and a third, containing an
isolated, solitary, allopatric population (Reiv, see Table 1 for detailed
loch information) using unbaited minnow traps (Gee traps, Dynamic
Aqua, Vancouver) set overnight in water 30-10 0 cm deep. For use
in mate choice trials, fish in breeding condition (males displaying full
FIGURE 1 Ecotype characteristics
and experimental design. (a) Photographs
showing examples of (female) anadromous
and lagoon fish from Obse and freshwater
fish from Reiv. (b) Mean body length for
each ecotype from each lake. Means are
calculated using all fish successfully used
in mate choice trials. (c) Diagram showing
the experimental design: all possible
trials between different ecotypes from
different lochs were conducted. Arrows
indicate pairings for mate choice trials
Loch name ID nSalinity Location
Ob nan Stearnain Obse 30 (50) [0] brackish 57° 36′6′′N; 7 °10′22 ′′W
Fairy Knoll Faik 34 (35) [0] brackish 57° 38′7′′N; 7°12′5 4′′W
na Reivil Reiv 0 (0) [33] freshwater 57°36′39′′N; 7°30′50′′W
Note: Sample sizes for mate choice trials are given for lagoon fish, anadromous fish (curved
parentheses), and freshwater fish (square parentheses). Brackish salinity classif ications
describe water with absolute conductivity 20,000–35,000 µS/cm and freshwater: absolute
conductivity < 500 µS/cm. Sampling locations are given in latitude, followed by longitude.
TABLE 1 Stickleback sampling sites

1744
|
DEAN Et Al.
nuptial coloration and heavily gravid females) were selected from
each catch and transported to a rental property on the island in aer-
ated loch water, after which they were transferred to loch specific
stock tanks containing either freshwater (dechlorinated tap water)
or ~20-30 ppt brackish water, depending on the salinity of source
lochs (Table 1), and some weed collected from source lochs for cover.
Brackish water was either pumped directly from the sea and mixed
with dechlorinated tap water to achieve the required salinity or pre-
pared using Seamix artificial sea water mix and dechlorinated tap
water. All fish were fed on washed, defrosted bloodworm once a day
and were kept in stock tanks until required for mate choice trials,
after which they were anaesthetized using an overdose of MS222
and killed in accordance with Schedule 1 of UK Home Office regula-
tions for use in other research.
2.2 | Mate choice trials
To investigate ecotype-level differences in male nesting and court-
ship behavior and female mate choice, female no-choice mating tri-
als were conducted between the lagoon and anadromous ecotypes
from the two species pairs (Obse and Faik). Ideally, to determine
whether sympatry (and therefore, contact) with other stickleback
ecotypes drives assortative mating, comparisons of the strength
of assortative mating would be made between sympatric ecotypes
and morphologically and ecologically similar, but solitary, allopatric
ecotypes. However, there are no known locations on North Uist in
which anadromous or lagoon ecotypes (which both occur in brackish
water) spawn allopatrically. As a best available proxy, an allopatri-
cally breeding freshwater population (Reiv), that was morphologi-
cally as intermediate as possible between lagoon and anadromous
ecotypes (Figure 1a, b), was selected for comparison. Our experi-
mental design included trials involving all possible pairings between
the three different ecotypes (anadromous, lagoon, and freshwater)
from three different populations (Obse, Faik, and Reiv), Figure 1c.
This allowed for testing the effects of ecotype and loch (within spe-
cies pair trials) simultaneously.
No-choice, as opposed to choice, trials were used as they gener-
ally work well in stickleback, providing a more conservative estimate
of mating preferences than choice trials (Coyne, 1993; Dougherty &
Shuker, 2015; Furin et al., 2012). Fur thermore, pilot studies showed
that lagoon males did not construct nests in the presence of anad-
romous males and were frequently bullied when males of both
ecotypes were housed together. Nest destruction between compet-
ing ecotypes is also common in a confined environment (Nagel &
Schluter, 1998), making choice trials impractical.
Trials were conducted outside so as to maintain consistent tem-
perature and lighting conditions, in 55 L clear plastic boxes filled
with fresh (for freshwater males) or brackish water (for lagoon and
anadromous males) prepared as in stock tanks. Each box contained
at least one rock for cover, some aquatic plant material collected
from stickleback source lagoons, 200 seven centimeter long black
cotton threads, which could be used as nesting material (Smith &
Wootton, 1999) (but in fact were not utilized by males in our ex-
periments), and two large petri dishes, one filled with sand, and
one with gravel collected from nearby lochs, for nesting substrate.
Brackish (lagoon and anadromous) males were also given an addi-
tional large petri dish containing mud, and a seaweed covered rock,
both collected from nearby brackish environment s to encourage
them to construct nests. Nesting substrates collected from brackish
environments were not included in freshwater boxes so as not to
alter the salinity of the water, and because these substrates are not
generally available in freshwater. All nesting substrates provided to
the dif ferent males occur naturally in their source habitat and are
distributed haphazardly within the same spawning areas in source
lakes. After acclimatization for at least 24 hr in stock tanks, males
were transferred to individual nesting boxes. Boxes were checked
daily for signs of nest construction, and a nest was deemed com-
plete when both an entry and exit hole were visible. The substrate
on which males chose to build their nests, along with their ecot ype,
was recorded for each nest in order to investigate potential micro-
habitat differences in nest location between ecotypes. Males that
failed to construct a nest within 7 days were replaced.
Following nest completion, a single heavily gravid female was
introduced to each box in a small plastic jar (with the lid off ), which
subsequently acted as a refuge for the female during the trial (male
courtship in sticklebacks can be aggressive, particularly when a
larger male and a smaller female are involved). For trials involving
females whose native salinity differed from that of the male (and
therefore the water in the trial boxes), females were acclimatized
to the same salinity as males over the 24-hr period preceding the
trial. Stickleback are naturally euryhaline and are capable of re-
sponding plastically, even to very abrupt changes in salinity (Taugbol
et al., 2014), and therefore, this was unlikely to affect fish adversely
during the trials. The behavior of both stickleback was recorded
using a wide-angle waterproof DB-power digital video camera posi-
tioned at the opposite end of the box to the nest. Trials began upon
first interaction between the male and female (which usually took
place within 10 min of the female being introduced) and lasted for
approximately 40 min. If mating had not taken place after this time,
there is an extremely low likelihood of it ever occurring (Nagel &
Schluter, 1998).
After trials were complete, females were removed from the
boxes, anaesthetized, and killed according to Schedule One pro-
cedure. If spawning did not occur during the trial, females were
stripped of their eggs to confirm readiness to spawn (eggs are
easily removed from fully gravid females when gentle pressure
is applied to the upper abdomen). Trials in which females could
not easily be stripped of their eggs were disc arded (this happened
only seven times over 107 total trials). Females were measured
for standard length because body size can be an important factor
affecting mate choice in stickleback (Nagel & Schluter, 1998). In
trials where eggs had been laid, the nest was removed from the
male's box and eggs were carefully removed. Nests were subse-
quently returned to males, who were given 24 hr to rebuild their
nests before they were of fered to a subsequent female. Males of

|
1745
DEAN E t Al.
all ecotypes were used in multiple trials, 55 different males were
used in total: 19 Obse anadromous, 8 Obse lagoon, 11 Faik anad-
romous , 6 Faik lagoon, and 11 Rei v freshwater (the individual male
used was included as a random term in all statistical modeling to
ensure differences between individual males did not affect the re-
sults). Each male was used in a maximum of three trials, separated
by at least 24 hr. The order of trials was largely determined by the
availability of females, and however, a male was never offered to
a female from the same population twice. Once males had been
used in up to three trials they were also anaesthetized and killed
according to Schedule One, and measurements of st andard leng th
were taken.
2.3 | Video analysis
Videos were visually analyzed using Behavioral Observation
Research Interactive Software (BORIS) version 6.3.6 (Friard &
Gamba, 2016). To assess differences in male courtship behavior
between ecotypes, the number of times each male performed the
following behaviors was recorded: zigzag dances (rapidly swimming
from side to side toward a female), attacks (biting or bumping a fe-
male), charges (swimming rapidly toward a female), taps (tapping at
the base of a female's tail when she is inside the nest), nest activi-
ties (adjusting the nest by fanning, or adding or removing substrate),
leads (swimming toward the female fish then leading her to the nest),
nest shows (showing the female the nest by probing the nest open-
ing with his head), and dorsal pricks (pricking the female with dor-
sal spines). All of these behaviors are previously well described and
can all form par t of the courtship ritual in stickleback (Wilz, 1970;
Wootton, 1984). The occurrence of spawning (females entering
the nest and laying their eggs) was recorded for all trials. Although
ecotypes are visually distinguishable by morphology, the observer
had no prior knowledge of the different ecotypes or expectation of
the outcome of the trials during video analysis. Trials in which the
male and female failed to interact during the entirety of filming were
discarded. Total trial times were recorded and if spawning occurred,
trials ended once spawning were complete.
2.4 | Statistical analysis
All statistical analyses were carried out in R version 3.5.2 (R Core
Team, 2018). Where linear regression models were used, all numeric
variables were centered and scaled prior to analysis, and model sim-
pli fication was con du cted st ar ting with the fullest mo de l an d re mo v-
ing the interactions followed by least significant terms sequentially
until reduced models were no longer an improvement on the most
recent fuller model. The significance of terms in the model was as-
sessed using likelihood ratio tests or F tests, as appropriate. The
goodness-of-fit of the best fitting model was then evaluated using
residual and Quantile-Quantile (Q-Q) plots, and models were trans-
formed and re-fitted if the necessar y family criteria were violated.
To test whether species pair (lagoon and anadromous) males
preferred different substrates for nest construc tion, we used a
chi-square test on the proportions of nest built on each of the five
offered substrates by lagoon versus anadromous males. To test
whether freshwater males preferred some nesting substrates over
others, we performed a separate chi-square test including only the
freshwater males and the three nesting substrates which they were
offered. Differences in the number of times male stickleback per-
formed each courtship behavior were analyzed using zero-inflated,
negative binomial generalized linear mixed models (GLMMs) imple-
mented using the glmmTMB R package (Brooks et al., 2017). The
frequencies of each behavior were corrected to reflect differences
in th e leng th of trials wh en spawni ng occurred by dividing by the du-
ration of the trial (in minutes). For all models of male behaviors, the
individual male was fitted as a random effect to control for effects
of males being used in multiple trials, and fitted predictor effects
were as follows: male ecotype (anadromous, lagoon or freshwater),
whether or not the male was displaying to a female of the same or
a different ecotype (1 or 0) and the interaction bet ween the two.
To determine the contribution of each individual level (anadromous,
lagoon, and freshwater) within the multi-level predictor variable
“male ecotype” when it had a statistically significant effect size in
the optimal GLMM’s of male mating behaviors, post hoc estimated
marginal means (EMM) were calculated for all pairwise level compar-
isons. p-values were adjusted to account for multiple testing using
the Tukey method for comparing families of three estimates. In the
case of models where complete separation occurred (which resulted
from lagoon males never performing some of the behaviors that
were measured) post hoc EMM’s could not be calculated and there-
fore the contribution of the lagoon male effect to the significance
of the male ecotype term was assessed in these cases by collapsing
anadromous and freshwater males into a single level and comparing
models on the two-level and three-level male ecotype variable using
likelihood ratio tests based on the chi-squared statistic.
To test for assortative mating and identify factors affecting
spawning probability, we used a GLMM with a binomial error struc-
ture and logit link function, implemented using the lme4 package,
version 1.1-13 (Bates et al., 2015). The occurrence of spawning
during the trials was used as a binar y response variable, with year
(2016 or 2017), absolute difference in body size (mm), female eco-
type (freshwater, lagoon, or anadromous), whether both the male
and female were of the same ecotype (0 or 1), and the interaction
between the latter two as fixed effect s in the model. To control for
the effects of individual males being used in multiple trials, the indi-
vidual male used in each trial was included as a random effect.
Our experiment s were designed using anadromous and lagoon
fish from two separate locations (lochs Obse and Faik), to test
whether assortative mating between ecotypes would be main-
tained across populations. I.e. whether an anadromous female
from one loch would prefer an anadromous male over a lagoon
male, regardless of whether the male was from the same or a dif-
ferent loch. Ther efore, we repe ate d th e GL MM analysis on the oc-
cu rre n ce of sp aw n ing usin g a re duc e d dat a set (n = 63 tria ls), whi ch

1746
|
DEAN Et Al.
excluded all trials involving males or females from the allopatric
population, Reiv, in order to test for the ef fects of loch in species
pair trials. The model was specified as above, except the fixed ef-
fect s we re as follows: year (2016 or 2017 ), fem al e ecot ype (la goon
or anadromous), whether or not females and males were of the
same ecotype (0 or 1) and whether or not females and males were
from the same loch (0 or 1), and the interaction between female
ecotype and loch.
3 | RESULTS
A total of 91 successful mating trials (in which males and females
interacted during the trial, and females either layed or were easily
stripped of their eggs) were conducted (Table 2). Overall, spawning
occurred in 24 of these 91 trials (26%). Spawning occurred in at least
one trial for every possible combination of ecotypes apart from la-
goon males with anadromous or freshwater females, although sam-
ple sizes for these combinations were low (Table 2). We found lagoon
and freshwater males to have a low propensity to build nests in com-
parison with anadromous males and this coupled with variation in
the availability of gravid females lead to variable sample sizes across
trial combinations.
3.1 | Nest location
Lagoon and anadromous males preferred to build their nests on dif-
ferent substrates (χ2 = 32.03, df = 4, p < .0001, Figure 2a). Lagoon
males showed an overwhelming preference for nesting on weed
(82%), while anadromous males preferred other substrates in 90% of
cases, particularly sand (46%), followed by gravel (20%). Freshwater
males also showed a preference for some substrates over others
(χ2 = 12.00, df = 2, p = .0025, Figure 2a), choosing to nest on sand
most frequently (67%), followed by gravel (33%), and never nesting
on rock, Figure 2a. See Figure 2b–d, for examples, of nests on mud,
sand, and weed, respectively.
3.2 | Male courtship behavior
Male courtship behavior was recorded for 81 of the 91 successful
mating trials. Ma le s of di ffere nt ecotyp es differe d in th e typ es an d
quantities of courtship behaviors they performed (Figure 3a–h,
Tabl e 3). Mal e ecoty pes dif fere d in wh et her or not the y per for me d
attacks, charges, taps, nest activities, and dorsal pricking during
courtship (Table 3). They also dif fered in the number of charges
they performed toward females, the number of times they tapped
females, the number of times they pricked females with their dor-
sal spines, and the number of times they performed nest tend-
ing activities during court ship (Table 3). These dif ferences largely
reflected a difference in the behavior of lagoon males, who were
more likely not to perform behaviors or performed fewer behav-
iors than the other two ecotypes in all c ases (see Appendix S1 and
S2 for post hoc multi-level factor comparisons). Freshwater and
anadromous males did not differ from one another in any of the
courtship behaviors we measured other than the number of nest-
ing activities per formed, with freshwater males performing more
nesting activities than anadromous males (Figure 3a–h, Appendix
S1 and S2).
The only behavior that males per formed differently, if they were
courting a female of their own ecotype or not, was the number of
zigzag dances, with all males performing more zigzag dances for fe-
males of their own ecotype (Figure 3a, Table 3). The interaction be-
tween male ecot ype and whether or not males were offered females
of the same ecotype as themselves was not significant in any of our
models (Table 3), indicating that males of all ecotypes responded
in the same way to females of the same and different ecotype as
themselves.
3.3 | Female assortative mating
Our results indicated that, overall, female stickleback exhibit posi-
tive assortative mating as spawning probability was higher when
males were of the same ecotype as females (binomial GLMM,
ecotype same: LR1 = 6.30, p = .0121, Figure 4a). This pattern re-
mained consistent in the model that only included trials within spe-
cies pairs, with females still being more likely to spawn with males
of the same ecotype as themselves (binomial GLMM, ecotype same:
LR1 = 20.02, p < .0001). In the species pairs, the female preference
for males of their own ecotype occurred irrespec tive of whether
males were from the same loch as the females (binomial GLMM,
female ecotype x loch: LR1 = 0.03, p = .8700) and this ef fect was
consistent across both lagoon and anadromous females (binomial
GLMM, female ecotype: LR1 = 0.01, p = .9064).
The probability of spawning decreased with increasing body size
differences between males and females, but this effect had large
error margins and was not significant in our model (binomial GLMM,
body size dif ference: LR1 = 1.24, p = .2648, Figure 4b).
The probability of spawning was considerably higher in 2016
than 2017 (binomial GLMM, year: LR1 = 16.12 , p < .0 001), and this
effect was still present, although slightly weaker, in the species pairs
alone (GLMM year: LR1 = 7.29, p = .007). The estimated variance
component ± SD for the random effect of individual males was
1.27 ± 1.13 in the full model and 553.7 ± 23.53 in the species pair
only model.
TABLE 2 Number of mate choice trials conducted for each
combination of ecotypes
Male ecot ype
Female ecotype
Anadromous Lagoon Freshwater
Anadromous 22 21 8
Lagoon 713 2
Freshwater 585

|
1747
DEAN E t Al.
The strength of assortative mating did not dif fer between the
three ecotypes (binomial GLMM, female ecotype x ecotype same:
LR2 = 1. 55, p = .4616, Figure 4a). There were, however, differences
between ecotypes in the overall likelihood of females spawning, re-
gardless of the ecotype of the male (binomial GLMM, female eco-
type: LR2 = 7. 0 3 , p = .0297, Figure 4a). Re-running the model with
FIGURE 2 Nest locations. (a)
Differences in the substrate on which
males of dif ferent ecotypes chose to build
their nests and examples of nests built
on (b) mud (very fine particles), (c) sand
(coarse par ticles), and (d) weed during
mate choice trials. Arrows indicate nest
entrances
FIGURE 3 Variation in male courtship
behavior. Graphs show differences in
the mean occurrence per minute of
(a) zigzags, (b) attacks, (c) charges, (d)
taps, (e) nest activities, (f) leads, (g) nest
shows, and (h) dorsal pricks performed
by males of dif ferent ecotypes (A:
anadromous, L: lagoon, F freshwater)
toward either females of the same (blue
bars) or different (pink bars) ecotypes
as themselves. For all graphs, circles
represent actual data points and error
bars represent standard errors of each
mean (SEM)

1748
|
DEAN Et Al.
TABLE 3 Differences in male mating behaviors
Male behavior
(response variable) Predictor variable df LRT p-value
Random effect
variance ± SD
N zigzags m ecotype 21. 351 6 .50 87 0.000 ± 0.00 0
ecotype same 15. 7607 .016 4 0.000 ± 0.000
m ecotype * ecotype
same
21.9216 .3 826 0.000 ± 0.000
Zigzag m ecotype 25. 97 76 .2008 0.000 ± 0.000
ecotype same 110.4620 .0053 0.000 ± 0.000
m ecotype * ecotype
same
26.7626 .1490 0.00 0 ± 0.000
N attacks m ecotype 23. 2742 .1945 1.662 ± 1.289
ecotype same 10.7847 . 3757 1.658 ± 1.288
m ecotype * ecotype
same
21.2183 .5438 1 .814 ± 1. 347
Attack m ecotype 229. 69 0 0 <.0000 1.662 ± 1.289
ecotype same 10.2546 .8805 1.658 ± 1.288
m ecotype * ecotype
same
20.1497 .9973 1. 814 ± 1 .3 47
N charges m ecotype 29.8 0 70 . 0 074 2.590 ± 1.609
ecotype same 10.0176 .8945 2.522 ± 1.588
m ecotype * ecotype
same
20.170 0 .9185 2.813 ± 1.677
Charge m ecot ype 216.4340 .0025 2 .590 ± 1.609
ecotype same 16.4355 .124 3 2. 522 ± 1.588
m ecotype * ecotype
same
22.9535 .5656 2.813 ± 1.677
N taps m ecotype 27.9 0 9 4 .019 2 2.462 ± 1.569
ecotype same 10.4551 .49991 2.587 ± 1.609
m ecotype * ecotype
same
20.1433 .93 09 2.449 ± 1. 565
Tap m ecotype 213.5020 .0091 2.462 ± 1. 569
ecotype same 12.4149 .2990 2.587 ± 1.609
m ecotype * ecotype
same
22.3 677 .6685 2.449 ± 1.565
N nest activities m ecotype 221.3000 <.0000 1.113 ± 1.0 55
ecotype same 10.464 0 .4958 1.202 ± 1.096
m ecotype * ecotype
same
22. 20 41 .3322 1.199 ± 1.0 95
Nest activities m ecotype 236. 3750 <.0000 1.113 ± 1.055
ecotype same 11.2960 . 5230 1.202 ± 1.096
m ecotype * ecotype
same
21.2200 . 874 8 1.199 ± 1. 095
N leads to nest m ecotype 22.998 3 .2233 0.923 ± 0 .961
ecotype same 11.6056 . 2 051 1.056 ± 1.028
m ecotype * ecotype
same
20.7511 .6869 0.00 0 ± 0.000
(Continues)

|
1749
DEAN E t Al.
Male behavior
(response variable) Predictor variable df LRT p-value
Random effect
variance ± SD
Leading m ecotype 26.1384 .1890 0.923 ± 0.961
ecotype same 12.0669 .3558 1.056 ± 1.028
m ecotype * ecotype
same
22.8156 . 5891 0.000 ± 0.000
N shows of nes t m ecotype 21.3400 . 5117 0.000 ± 0.000
ecotype same 10.6766 .4108 0.000 ± 0.000
m ecotype * ecotype
same
21.119 .5715 0.000 ± 0.000
Nest showing m ecot ype 25. 3119 .2568 0.000 ± 0.000
ecotype same 14.7987 .0908 0.000 ± 0.000
m ecotype * ecotype
same
26.0156 .198 0 0.000 ± 0.000
N dorsal pricks m ecotype 212.8220 .0 016 0.418 ± 0.6 47
ecotype same 10.0190 .8901 0.650 ± 0.806
m ecotype * ecotype
same
20.9132 .6334 0.0 00 ± 0.000
Dorsal pricking m ecotype 213.0080 .0112 0.418 ± 0.647
ecotype same 10.3748 . 8291 0.650 ± 0.806
m ecotype * ecotype
same
21.9696 .7414 0.0 00 ± 0.000
Note: Table shows zero-inflated negative binomial GLMM results for dif ferences in male mating behaviors between male ecotypes (m. ecot ype),
different behaviors toward females of different ecotypes (ecotype same) and differences between male ecotypes in their response to females of
different ecotypes (m ecotype * ecotype same interaction). LRT values are repor ted for comparisons of models before and after the removal of each
predictor variable, with interactions removed first followed by least significant terms. Models testing the number of times a behavior was observed
when it occurred are shown with response variables beginning N. All response variables not beginning N refer to models testing the probability of a
behavior being observed (zero or nonzero). p-values < .05 are highlighted in bold.
Abbreviations: df, degrees of freedom; ecotype same, whether or not the male was cour ting a female of the same ecot ype as himself; LRT, likelihood
ratio tes t statistics; m ecotype, ecot ype of the male used in mate choice trials; N, number of (times each cour tship behavior was performed); SD, one
standard deviation.
TABLE 3 (Continued)
FIGURE 4 (a) Differences in spawning probability across ecot ypes. Predicted probabilities of females of different ecotypes spawning
with males of the same versus dif ferent ecotypes. Error bars show associated bootstrapped 95% confidence inter vals. Predictions are based
on a binomial generalized linear mixed model with individual male as a random effect and year, female ecotype and conspecific male as fixed
effects. (b) Predicted probability of spawning (P spawning) with increasing differences in body size (Abs size diff) from a binomial generalized
linear mixed model with year, absolute size difference, female ecotype, and whether or not male and female ecotypes were the same as
predictor variables and individual male as a random effect. Bootstrapped 95% confidence inter vals are indicated by the gray ribbon

1750
|
DEAN Et Al.
lagoon and anadromous females collapsed into a single level of the
female ecotype factor (“saltwater females”) confirmed that these
two ecotypes did not differ from each other in their overall spawn-
ing probability (Appendix S2), and therefore, the effect of female
ecotype was caused by a greater probability of spawning in the al-
lopatric freshwater ecotype, compared to the sympatric lagoon and
anadromous ecotypes (Figure 4a).
4 | DISCUSSION
We investigated behavioral differences during courtship and the
presence and strength of assortative mating in newly described
saltwater stickleback species pairs. We showed that species pair
males differ in their nesting microhabitat preferences and court-
ship behavior and males of all ecotypes preferentially court females
of the same ecotype as themselves. We also identified positive as-
sortative mating in both species pair and allopatric females, but we
did not find any differences in the strength of assortative mating
between species pair and allopatric females, suggesting that female
mating preferences have arisen as a by-product of other ecological
adaptations and are not necessarily under strong selection in spe-
cies pairs. Finally, we found that fish being of the same ecotype was
a much better predictor of spawning probability than difference
in body size, implying that other characteristics that are divergent
between ecotypes must be responsible for assortative mating. Our
results indicate that divergent behavior and fine-scale segregation
of ecotypes when nesting (driven by divergent male nest substrate
preferences) may drive assortative mating and play an important role
in maintaining reproductive isolation in species pairs in the wild.
4.1 | Differences in courtship behavior and
nesting location
Court ship behavioral differences often play a key role in maintain-
ing adaptive divergence in stickleback, and this has been particu-
larly well studied in divergent benthic–limnetic (Foster, 1995; Foster
et al., 2008; Shaw et al., 2007), lake–stream (Delcourt et al., 2008;
Raeymaekers et al., 2010) and japan sea anadromous—pacific ocean
anadromous ecotypes (Kitano et al., 2007). Lagoon stickleback have
only recently been recognized as an ecotype in their own right (Dean
et al., 2019; Ravinet et al., 2015), and so their courtship behavior
has not previously been investigated. We showed that lagoon males
performed fewer of many of the classic stickleback courtship be-
haviors than males of the other t wo ecotypes (or sometimes none).
This means that sympatric lagoon and anadromous males differ
markedly in courtship behavior, which may be involved in main-
taining reproductive isolation in species pairs, just as in many other
stickleback ecotypes. Anadromous stickleback are generally more
aggressive than freshwater stickleback (McKinnon et al., 2012) and
reduced courtship behavior has also been recorded in cannibalistic
anadromous populations, in which large shoals of stickleback form,
and attack and consume fry and/or eggs in the nests of lone males
(Foster, 1995; Shaw et al., 20 07). Lagoon males performed fewer (or
none) of all courtship behaviors measured except for the number of
zigzags, where they performed a higher frequency of zigzags than
the other two ecotypes. Perhaps in a naturally sympatric setting la-
goon males experience a trade-off between the risks of being con-
spicuous during courtship and the benefits of attracting a female;
avoiding most types of obvious court ship displays to reduce aggres-
sion and possible nest raids from anadromous counterparts.
We found that lagoon males also differed from the other two
ecotypes in terms of their preferred nesting substrate, favouring
weed over the sand or gravel utilized by freshwater and anadromous
ma les . This is in dic ativ e of fi ne-s cale spat ial st r uctu r e in ne s tin g loca -
tion in species pairs, which could also be important for reproductive
isolation (Borzee et al., 2016; Hagen, 1967; Pegoraro et al., 2016;
Snowberg & Bolnick, 2012), as is the case in some benthic–limnetic
species pairs (Ridgway & McPhail, 1987). Lagoon males are perma-
nently resident in the sympatric lagoon spawning habitat, so theoret-
ically have the opportunity to establish the most favorable spawning
territories before anadromous males arrive. However, lagoon males
are unlikely to be able to compete with the larger, more aggressive
anadromous males to retain such territories. Nesting on weed may
thus allow lagoon males to avoid territorial conflict with anadromous
males if the preferred nesting substrate of anadromous males is sand
or gravel. It could also confer an advantage to lagoon males if can-
nibalism or nest destruction by anadromous males occurs in species
pair s , si nce nest s susp e nd e d on weed are le ss co ns pic uou s and prob -
ably less easy to attack than those built on other, harder substrates
such as sand or gravel (Hagen, 1967; Kynard, 1979). It is probable
that lagoon females share the preference of lagoon males for nests
built on weed rather than other substrates as our data on spawning
probability showed that lagoon females preferred lagoon males (with
nests mostly on weed), rather than anadromous or freshwater males
(with nests on sand/gravel).
Lackey and Boughman (2017) recently showed that divergence
in habitat and mate choice is the two most important premating
barriers during speciation in stickleback. They found that these two
barriers evolve early and remain strong throughout the speciation
process. We identified differences both in nesting microhabitat and
in courtship behavior which strongly support the findings of Lackey
and Boughman (2017), sug gesting that these premating barr iers may
be particularly important for maintaining reproductive isolation in la-
goon – anadromous species pairs. It is important to distinguish pairs
of ecotypes that are transiently symp atric (such as the lagoon–anad-
romous pairs studied here) from many of the other fully sympatric
pairs that exist within the st ic kleback species complex (e.g ., benthic–
limnetic). Divergent male mating behavior appears to be involved
in reproductive isolation in both continually and transiently sym-
patric pairs (McPhail & Hay, 1983; Shaw et al., 2007). Interestingly,
in freshwater–anadromous pairs (also transiently sympatric) fresh-
water males also zigzag more and bite less than anadromous males
(McPhail & Hay, 1983), suggesting similar divergent mating behav-
iors may evolve repeatedly in transiently isolated species pairs.

|
1751
DEAN E t Al.
Anadromous and freshwater males did not differ in any of the
courtship behaviors other than their propensity to perform nest
tending activities (which was only marginally significant, Appendix
S1) or preferences for different nesting substrates in our exper-
iments (although it is not possible to know whether freshwater
males would have preferred nesting on weed over sand or gravel
as low growing weed is not common in the lagoon from which we
collected freshwater fish so they were not offered weed as a nesting
substrate). Anadromous males approximate the marine ancestor of
all freshwater and lagoon ecotypes (Colosimo et al., 20 05) and this,
coupled with the fact that anadromous behavior was very similar to
that of freshwater fish, suggests the differences exhibited by lagoon
males are probably derived traits, which may have evolved as a result
of sympatry in species pairs.
Studies of mating preferences and prezygotic reproductive iso-
lation primarily focus on female mate choice (Gavrilets et al., 2001;
Gerhardt, 1994; Head et al., 2013; Laloi et al., 2011), but the po-
tent ial for male mating preferences to be im port ant is be coming in-
crea si ngly ap pre ci at ed (H ug hes et al., 20 13; Mc Kinno n et al ., 2012).
We showed that males of all ecotypes preferentially courted fe-
males of the same ecotype as themselves, performing more zigzag
dances toward females of the same ecot ype. Zigzag dances involve
rapid darts from side to side and are thus are probably the most
energetically expensive of all courtship behaviors measured here,
indicating that males make an active choice to direct costly mating
behaviors toward females of their own ecotype. This could explain
the increased probability for spawning between fish of the same
ecotype and likely contributes to reproductive isolation in lagoon–
anadromous species pairs. It also adds to the growing body of ev-
idence to suggest that mate choice in stickleback is mutual (Kozak
et al., 20 0 9), and this is true ac ros s dif feren t spe c ie s pairs . Alth o ugh
male mate choice was not a focus here, our findings identify the
role of male choice in this species pair as a topic for future work.
4.2 | Evidence for isolation based on body size
Size assortative mating is commonly involved in reproductive isola-
tion in sympatric fish species (Foote & Larkin, 1988; McKaye, 1986;
McKinnon et al., 2004; Rueger et al., 2016; Sigurjonsdottir &
Gunnarsson, 1989) and is well described in stickleback (Boughman
et al., 2005; Conte & Schluter, 2013; McKinnon et al., 2004, 2012;
Nagel & Schluter, 1998). However, we found only weak (nonsignifi-
cant) evidence that assortative mating was related to body size, which
was somewhat unexpected. One possible explanation for this is that
body size is utilized by stickleback to discriminate between ecotypes
during cour tship in saltwater species pairs, but not to select mates
within ecotypes. Mate choice experiments involving lava and nitella
stickleback ecotypes in Iceland identified a similar scenario, with no
evidence for a role for size assor tative mating between ecoty pes, al-
though there was a suggestion that both lava and nitella females pre-
ferred larger males (Olafsdottir et al., 2006). Our findings, therefore,
add to evidence which suggests that factors affecting mate choice
between ecotypes are not necessarily replicated across independent
stickleback species pairs. Interestingly, in lava and nitella pairs where
there was no evidence for size assortative mating, the two morphs
differed in nesting location and structure (Olafsdottir et al., 2006),
which is exactly the pattern we identified in our data. This suggest s
that there may be a consistent pat tern in stickleback in which dif fer-
ences in male nesting behavior, par ticularly in nesting location may
be particularly important for bet ween—ecotype mate choice when
variation in body size does not play a significant role. It may also be
that other factors that were not measured in our experiment, such as
nuptial color, are more important in driving assortative mating, but
further experiments would be necessary to determine their effects.
This provides an interesting avenue for future research.
4.3 | By-product versus adaptive assortative mating
Positive assortative mating is extremely common in taxa composed
of multiple ecotypes (Hollander et al., 2005; Jarvis et al., 2017;
Machado-Schiaffino et al., 2017) and is well described in stickle-
back (Conte & Schluter, 2013; Ingram et al., 2015; McKinnon &
Rundle, 2002), and therefore, it is not surprising that we identified
it in all ecotypes in this study. The preferences we identified could
have a genetic basis, but species pairs of benthic and limnetic stick-
leback acquire their ecotype specific preferences in a process of
sexual imprinting on their fathers during early development (Kozak
et al., 2011) (stickleback fry are cared for by males for a number of
weeks after hatching), so it is also possible that imprinting plays a role
here and the mechanism of heritability of mating preferences in Uist
stickleback would be an interesting topic of fur ther investigation.
Allopatric populations cannot hybridize, so by default, any
assortative mating that exists between totally isolated popu-
lations can only have evolved as a by-product of their isolation
(Schlu te r, 2001) . We detect ed as so r tativ e matin g in both allo patri c
and sympatric stickleback populations, which illustrates that it has
evolved at least partially as a by-product of other adaptations in
this system. Given the substantial phenot ypic adaptations, includ-
ing differences in body armor, size, and shape (Campbell, 1979;
MacColl et al., 2013; Magalhaes et al., 2016), across stickleback
populations on North Uist, some by-product assortative mating
would be expected (Dodd, 1989; Kilias et al., 1980). This mecha-
nism is also implicated in causing assortative mating in other re-
productively isolated, allopatric stickleback populations (Vines &
Schluter, 2006), which suggests that reasonably substantial assor-
tati ve mati ng , t ha t is pu rely a by-p roduc t of ot her diff erences , may
be common among stickleback. We did not detect any exaggera-
tion of assortative mating in species pairs compared with naïve
allopatric stickleback, which suggests that if a signal of selection
on assortative mating is present in species pairs it is not strong
enough to be detected in our experiments. It is also possible that
the assortative mating that we detected in species pairs arose via
other nonadaptive, but population-specific mechanisms such as
genetic drift or sexual selection.

1752
|
DEAN Et Al.
5 | CONCLUSIONS
We identified behavioral differences, both in courtship rituals and
preferred nesting substrate, which likely contribute to maintaining
reproductive isolation in newly described saltwater stickleback spe-
cies pairs. We found evidence that both species pair and allopatric
males selectively court females of the same ecotype as themselves.
We also showed that females in species pairs prefer to mate with
males of their own ecot ype, but these preferences are not driven
by the commonly implicated magic trait, body size, and are also no
stronger in sympatric species pairs versus allopatric populations.
Our results suggest that divergent behavior, particularly that which
causes fine-scale differences in nesting location, is probably impor-
tant for maintaining reproductive isolation in sympatric populations.
Our findings also highlight that reasonably strong assor tative mating
can develop as a by-product of other differences between popula-
tions, and strong selection on mate choice may not be par ticularly
important in this system.
ACKNOWLEDGMENTS
The authors would like to thank Sarah Evans and Talib Chitheer for
assistance conduc ting mate choice trials. This work was funded by
a National Environment Research Council (NERC) funded Ph.D. stu-
dentship to LLD, PI: ADCM (grant number NE/L002604/1) and a
NERC grant (NE/R00935X/1) to ADCM.
CONFLICT OF INTEREST
The authors declare no competing interests.
AUTHOR CONTRIBUTIONS
Laura L. Dean: Conceptualization (equal); Formal analysis (equal);
Investigation (equal); Methodology (equal); Writing-original draft
(lead); Writing-review & editing (equal). Hannah R. Dunstan:
Formal analysis (suppor ting). Amelia Reddish: Investigation (equal).
Andrew D. C. MacColl: Conceptualization (equal); Data curation
(equal); Formal analysis (supporting); Funding acquisition (lead);
Investigation (equal); Methodology (equal); Project administration
(lead); Resources (lead); Supervision (lead); Writing-review & edit-
ing (equal).
DATA AVAILAB ILITY STATE MEN T
All data from this manuscript are accessible in the Dryad digital re-
pository (https://doi.org/10.5061/dryad.xd254 7dg0).
ORCID
Laura L. Dean https://orcid.org/0000-0002-4711-0931
Andrew D. C. MacColl https://orcid.org/0000-0003-2102-6130
REFERENCES
Andreou, D., Eizaguirre, C., Boehm, T., & Milinski, M. (2017). Mate choice
in sticklebacks reveals that immunogenes can drive ecological spe-
ciation. Behavioral Ecology, 28, 953–961. http s://doi.org /10.1093/
behec o/arx074
Ballant yne, C. K. (2010). Extent and deglacial chronology of the last
British-Irish ice sheet: Implications of exposure dating using cosmo-
genic isotopes. Journal of Quaternary Science, 25, 515–534. https://
doi.org/10.10 02/jqs.1310
Bates, D., Ma chler, M., Bolker, B. M., & Walker, S. C. (2015). Fit ting linear
mixed-effects models using lme4. Journal of Statistical Soft ware, 67,
1–4 8 .
Bay, R. A., Arnegard, M. E., Conte, G. L., Best, J., Bedford, N. L., McCann,
S. R., Dubin, M. E., Chan, Y. F., Jones, F. C., Kingsley, D. M., Schluter,
D., & Peichel, C. L. (2017). Genetic coupling of female mate choice
with polygenic ecological divergence facilitates stickleback spe-
ciation. Current Biology, 27, 3344–3349. https://doi.org/10.1016/j.
cub.2017.09.037
Bell, M. A., Aguirre, W. E., & Buck, N. J. (2004). Twelve years of con-
temporary armor evolution in a threespine stickleback population.
Evolution, 58, 814–824. https://doi.org/10.1111/j.0014-3820.2004.
t b 0 0 4 1 4 . x
Berlocher, S. H., & Feder, J. L. (20 02). Sympatric speciation in phy-
tophagous insects: Moving beyond controversy? Annual Review
of Entomology, 47, 773–815. https://doi.org/10.1146/annur
ev.ento.47.091201.145312
Bird, C. E., Fernandez-Silva, I., Skillings, D. J., & Toonen, R. J. (2012).
Sympatric speciation in the post "modern synthesis" era of evo-
lutionary biology. Evolutionary Biology, 39, 158–180. https://doi.
org /10.1007/s1169 2-012-9183 -6
Blouw, D. M., & Hagen, D. W. (1990). Breeding ecology and evidence
of reproductive isolation of a widespread stickleback fish (gaster-
osteidae) in Nova-Scotica, Canada. Biological Journal of the Linnean
Society, 39, 195–217. https://doi.org/10.1111/j.1095-8312.1990.
t b 0 0 5 1 2 . x
Bolnick, D. I. (2011). Sympatric speciation in threespine stickleback:
Why not? International Journal of Ecology, 2011, 1–15. https://doi.
org /10.1155/2011/9428 47
Bolnick , D. I., & Fitzpatrick, B. M. (20 07). Sympatric speciation: Models
and empirical evidence. Annual Review of Ecolog y Evolution and
Systematics, 38, 459–487. https://doi.org/10.1146/annur ev.ecols
ys.38.091206.095804
Bolnick , D. I., & Kirkpatrick, M. (2012). The relationship between in-
traspecific assortative mating and reproductive isolation between
divergent populations. Current Zoology, 58, 484–492. https://doi.
org /10.1093/czoo l o/58.3.4 84
Borzee, A., Kim, J. Y., Da Cunha, M. A . M., Lee, D., Sin, E., Oh, S., Yi,
Y., & Jang, Y. (2016). Temporal and spatial differentiation in micro-
habitat use: Implications for reproductive isolation and ecological
niche specification. Integrative Zoology, 11, 375–387. https://doi.
org /10.1111/1749- 4877.122 00
Boughman, J. W., Rundle, H. D., & Schluter, D. (2005). Parallel evolution
of sexual isolation in sticklebacks. Evolution, 59, 361–373. htt ps://doi.
org/10.1111/j.0014-3820.2005.tb009 95.x
Brooks, M. E., Kristensen, K., van Benthem, K. J., Magnusson, A., Berg,
C. W., Nielsen, A., Skaug, H. J., Machler, M., & Bolker, B. M. (2017).
glmmTMB balances speed and flexibility among packages for ze-
ro-inflated generalized linear mixed modeling. R Journal, 9, 378–400.
https://doi.org/10.32614/ RJ-2017-066
Campagna, L., Kopuchian, C., Tubaro, P. L., & Lougheed, S. C. (2014).
Secondary contact followed by gene flow between divergent mito-
chondrial lineages of a widespread neotropical songbird (Zonotrichia
capensis). Biological Journal of the Linnean Society, 111, 863–868.
https ://doi.or g/10.1111/ bij.12272
Campbell, R. N. (1979). Sticklebacks [Gasterosteus aculeatus (L.) and
Pungitius pungitius (L.)] in the outer he br ides, Scotla nd . The Hebridean
Naturalist, 3, 8–15.
Candolin, U. (1997). Predation risk affects courtship and attractiveness
of competing threespine stickleback males. Behavioral Ecology and
Sociobiology, 41, 81–87. https://doi.org/10.1007/s0026 50050367

|
1753
DEAN E t Al.
Colosimo, P. F., Hosemann, K. E., Balabhadra, S., Villarreal, G., Dickson,
M., Grimwood, J., Schmutz, J., Myers, R. M., Schluter, D., & Kingsley,
D. M. (2005). Widespread parallel evolution in sticklebacks by re-
peated fixation of ectodysplasin alleles. Science, 307, 1928–1933.
https://doi.org/10.1126/scien ce.1107239
Conte, G . L., & Schluter, D. (2013). Experimental confirmation that body
size determines mate preference via phenotype matching in a stickle-
back species pair. Evolution, 67, 1477–1484. ht tps://doi .org /10.1111 /
evo.12041
Coyne, J. A. (1993). The genetics of an isolating mechanism between
two sibling species of Drosophila. Evolution, 47, 778–788 . https://doi.
org /10.2307/2410183
Coyne, J. A., & Orr, H. A. (1989). Patterns of speciation in drosophila.
Evolution, 43, 362–381. ht tps://doi.org/10.23 07/2409213
Coyne, J. A., & Orr, H. A . (1997). "Patterns of speciation in
Drosophila" revisited. Evolution, 51, 295–303. https://doi.
org /10.1111/j .1558-5646.1997.tb024 1 2. x
Dean, L. L., Magalhaes, I. S., Foote, A., D'Agostino, D., McGowan, S., &
MacColl, A. D. C. (2019). Admixture between ancient lineages, se-
lection, and the formation of sympatric stickleback species-pairs.
Molecular Biolog y and Evolution, 36 (11), 2481–2497. https://doi.
org /10.1093/molbe v/msz161
Delcourt, M., Raesaenen, K., & Hendr y, A. P. (2008). Genetic and plastic
components of divergent male intersexual behavior in Misty lake/
stream stickleback. Behavioral Ecology, 19, 1217–1224. https://doi.
org/10.1093/behec o/arn084
Die ckmann, U., & Doeb el i, M . (1999). On the origin of speci es by sympa t-
ric speciation. Nature, 400, 354 –357. ht tps://doi.or g/10 .103 8/22521
Dobzhansky, T. U. (1937). Genetics and the origin of specie s, (1st ed.).
Columbia University Press.
Dodd, D. M. B. (1989). Reproductive isolation as a consequence of adap-
tive divergence in Drosophila pseudoobscura. Evolution, 43, 1308–
1311. https://doi.org/10.2307/2409365
Dougherty, L . R., & Sh uke r, D. M. (2015). The ef fect of expe rimen tal de-
sign on the measurement of mate choice: A meta-analysis. Behavioral
Ecology, 26, 311–319. https://doi.org/10.1093/behec o/aru125
Foote, A. D. (2018). Sympatric speciation in the genomic era. Tre nd s
in Ecology & Evolution, 33, 85–95. https://doi.org/10.1016/j.
tree.2017.11.003
Foote, C. J., & Larkin, P. A. (1988). The role of male choice in the assor-
tative mating of anadromous and non-anadromous sockeye salmon
(Oncorhynchus nerka). Behaviour, 106(1-2), 43–61. https://doi.
o r g / 1 0 . 1 1 6 3 / 1 5 6 8 5 3 9 8 8 X 0 0 0 8 9
Foster, S. A. (1995). Understanding the evolution of behavior in threespine
stickleback: The value of geographic variation. Behaviour, 132(15-16),
1107–1129. htt ps://doi.or g/10.1163/15685 39 95X 0 0487
Foster, S. A., Shaw, K. A., Robert, K. L., & Baker, J. A. (2008). Benthic,
limnetic and oceanic threespine stickleback: profiles of reproductive
behaviour. Behaviour, 145(4–5), 485–508.
Friard, O., & Gamba, M. (2016). BORIS: A free, versatile open-source
event-logging software for video/audio coding and live observa-
tions. Methods in Ecolog y and Evolution, 7, 1325–1330. https://doi.
org /10.1111/2041-210x.1258 4
Furin, C. G., von Hippel, F. A ., & Bell, M. A. (2012). Partial reproduc-
tive isolation of a recently derived resident-freshwater population
of threespine stickleback (Gasterosteus aculeatus) from its puta-
tive anadromous ancestor. Evolution, 66, 3277–3286. https://doi.
org /10.1111/j .1558-5646. 2012. 01672.x
Gavrilets, S., Arnqvist, G., & Friberg, U. (2001). The evolution of fe-
male mate choice by sexual conflict. Proceedings of the Royal Society
B-Biological Sciences, 268 (1466), 531–539. https://doi.org/10.1098/
rspb.2000.1382
Gerhardt, H. C. (1994). Reproductive character displacement of female
mate choice in the gray treefrog, Hyla chrysoscelis. Animal Behaviour,
47, 959–969. https://doi.org/10.1006/anbe.1994.1127
Hagen, D. W. (1967). Isolating mechanisms in threespine sticklebacks
(Gasterosteus). Journal of the Fi sheries Board o f Canada, 24, 1637–1692.
Head, M. L., Kozak, G. M., & Boughman, J. W. (2013). Female mate pref-
erences for male body size and shape promote sexual isolation in
threespine sticklebacks. Ecology and Evolut ion, 3, 218 3–2196 . http s://
doi.org/10.1002/ece3.631
Hollander, J., Lindegarth, M., & Johannesson, K. (2005). Local adapta-
tion but not geographical separation promotes assortative mating in
a snail. Animal Behaviour, 70, 1209–1219. https://doi.org/10.1016/j.
anbeh av.2005.03.014
Hughes, L. C., Foster, S. A., & Baker, J. A . (2013). Can ecotypic differences
in male courtship behaviour be explained by visual cues provided by
female threespine stickleback? Evolutionary Ecolog y Research, 15,
43 7– 451.
In gra m, T., Jia ng, Y. X., Ra ngel, R. , & Bol nick, D. I. (2 015) . Wid espre ad po s-
itive but weak assortative mating by diet within stickleback popula-
tions. Ecology and Evolution, 5, 3352–3363. https://doi.org/10.1002/
ece3.1609
Ishikawa, M., & Mori, S. (20 00). Mating success and male courtship be-
haviors in three populations of the threespine stickleback. Behaviour,
137, 1065–1080. https://doi.org/10.1163/15685 39005 02439
Jarvis, W. M. C., Comeau, S. M., Colborne, S. F., & Robinson, B. W. (2017).
Flexible mate choice may contribute to ecotype assortative mating
in pumpkinseed sunfish (Lepomis gibbosus). Journal of Evolutionary
Biology, 30, 1810–1820. https://doi.org/10.1111/jeb.13127
Jones, F. C., Chan, Y. F., Schmutz, J., Grimwood, J., Brady, S. D., Southwick,
A. M., Absher, D. M., Myers, R. M., Reimchen, T. E., Deagle, B. E.,
Schluter, D., & Kingsley, D. M. (2012). A genome-wide SNP geno-
typing array reveals patterns of global and repeated species-pair
divergence in sticklebacks. Current Biology, 22, 83–90. https://doi.
org /10.1016/ j.cub.2 011.11.0 45
Kilias, G., Alahiotis, S. N., & Pelecanos, M. (1980). A multifactorial genetic
investigation of speciation theor y using Drosophila melanogaster.
Evolution, 34, 730–737. https://doi.org/10.2307/2408027
Kirkpatrick, M. (2001). Reinforcement during ecological speciation.
Proceedi ngs of the Royal Society B-Biological Sciences, 268 (1473) ,
1259–1263. https://doi.or g/10.1098/rspb.200 0.1427
Kitano, J., Mori, S., & Peichel, C. L. (2007). Phenotypic divergence
and reproductive isolation between sympatric forms of Japanese
threespine sticklebacks. Biological Journal of the Linnean Society, 91,
671–685 . ht tps://doi.or g/10.1111/j.1095-8312.2007.00 824.x
Kozak, G. M., Head, M. L., & Boughman, J. W. (2011). Sexual imprinting on
ecologically divergent traits leads to sexual isolation in sticklebacks.
Proceedi ngs of the Royal Society B-Biological Sciences, 278(1718),
2604–2610. https://doi.org/10.1098/rspb.2010.2466
Kozak, G. M., Reisland, M., & Boughman, J. W. (2009). Sex differ-
ences in mate recognition and conspecific preference in species
with mutual mate choice. Evolution, 63, 353–365. https://doi.
org /10.1111/j .1558-5646. 20 08.00564.x
Kynard, B. E. (1979). Nest habitat preference of low plate number morphs
in threespine sticklebacks (Gasterosteus aculeatus). Copeia, 1979(3),
525–528. https://doi.or g/10.23 07/1443234
Lackey, A. C . R., & Boughman, J. W. (2017). Evolution of reproductive
isolation in stickleback fish. Evolution, 71, 357–372. https://doi.
org /10.1111/evo .13114
Laloi, D., Eizaguirre, C., Federici, P., & Massot, M. (2011). Female choice
for heterozygous mates changes along successive matings in a liz-
ard. Behavioural Processes, 88, 149–154. htt ps://doi.or g/10.1016/j.
beproc.2011.08.011
Levitis, D. A., Lidicker, W. Z., & Freund, G. (20 09). Behavioural biologists
do not agree on what constitutes behaviour. Animal Behaviour, 78,
103–110. https://doi.org/10.1016/j.anbeh av.2009.03.018
Liu, G., Stapleton, E., Innes, D., & Thompson, R. (2011). Aggregational
behavior of the blue mussels Mytilus edulis and Mytilus trossulus:
A potential pre-zygotic reproductive isolation mechanism. Marine

1754
|
DEAN Et Al.
Ecology-an Evolutionary Perspective, 32, 480–487. https://doi.
org /10.1111/j .143 9-04 85.2011. 00 4 46 .x
MacColl, A . D. C. (2009). Par asites may contribute to 'magic trait'
evolution in the adaptive radiation of three-spined s ticklebacks,
Gasterosteus aculeatus (Gasterosteiformes: Gasterosteidae).
Biological Journal of the Linnean Society, 96, 425–433. https://doi.
org /10.1111/j .1095-8 312.20 08 .01123 .x
MacColl, A. D. C., El Nagar, A., & de Roij, J. (2013). The evolutionar y
ecology of dwarfism in three-spined sticklebacks. Journal of Animal
Ecology, 82, 642–652. https://doi.org/10.1111/1365-2656.12028
Machado-Schiaffino, G ., Kautt, A . F., Torres-Dowdall, J., Baumgarten, L.,
Henning, F., & Meyer, A . (2017). Incipient speciation driven by hy-
pertrophied lips in Midas cichlid fishes? Molecular Ecology, 26, 2348–
2362 . https: //doi.org/10 .1111/mec.140 29
Magalhaes, I. S., Agostino, D. D., Hohenlohe, P. A., & Maccoll, A. D. C.
(2016). The ecology of an adaptive radiation of three-spined stickle-
back from North Uist, Scotland. Molecular Ecology, 25, 4319–4336.
https ://doi.or g/10.1111/m ec .1374 6
McKaye, K. R. (1986). Mate choice and size assortative pairing by the
cichlid fishes of lake Jiloa, Nicaragua. Journal of Fish Biology, 29, 135–
150. https://doi.org/10.1111/j.1095-8649.1986.tb050 05.x
McKinnon, J. S. (1995). Video mate preferences of female three-spined
sticklebacks from populations with divergent male coloration.
Animal Behaviour, 50, 1645–1655. https://doi.org/10.1016/00 03-
3472(95)80018 -2
McKinnon, J. S., Hamele, N., Frey, N., Chou, J., McAleavey, L., Greene,
J., & Paulson, W. (2012). Male choice in the stream-anadromous
stickleback complex. PLoS One, 7, 8. https://doi.org/10.1371/journ
al.pone.0037951
McKinnon, J. S., Mori, S., Blackman, B. K., David, L., Kingsley, D. M.,
Jamieson, L., Chou, J., & Schluter, D. (20 04). Evidence for ecology's
role in speciation. Nature, 429, 294–298. https://doi.org/10.1038/
natur e02556
McKinnon, J. S., & Rundle, H. D. (20 02). Speciation in nature: The
threespine stickleback model systems. Trends in Ecology & Evolution,
17, 480–488 . https://doi.org/10.1016/s0169 -5347(02)02579 -x
McPhail, J. D., & Hay, D. E. (1983). Differences in male cour tship in
freshwater and marine sticklebacks (Gasterosteus aculeatus).
Canadian Journal of Zoology, 61, 292–297. https://doi.org/10.1139/
z83-039
Moritz, C., Hoskin, C. J., MacKenzie, J. B., Phillips, B. L ., Tonione, M.,
Silva, N., VanDerWal, J., Williams, S. E., & Graham, C. H. (2009).
Identif ication and dynamics of a cryptic suture zone in tropical
rainforest. Proceedings of the Royal Societ y B-Biological Sciences,
276 (1660), 1235–1244. https://doi.org/10.1098/rspb.20 08.1622
Nagel, L., & Schluter, D. (1998). Body size, natural selection, and
speciation in sticklebacks. Evolution, 52, 209–218. https://doi.
org /10.2307/2410936
Olafsdottir, G. A., Ritchie, M. G., & Snorrason, S. S. (2006). Positive as-
sortative mating between recently described sympatric morphs
of Icelandic sticklebacks. Biology Letters, 2, 250–252. https://doi.
org/10.1098/rsbl.2006.0456
Papadopulos, A. S. T., Igea, J., Smith, T. P., Hutton, I., Baker, W. J., Butlin,
R. K., & Savolainen, V. (2019). Ecological speciation in sympatric
palms: 4. Demographic analyses support speciation of Howea in
the face of high gene flow. Evolution, 73, 1996–2002. https://doi.
org /10.1111/evo .13813
Pegoraro, L., Cafasso, D., Rinaldi, R., Cozzolino, S., & Scopece, G. (2016).
Habitat preference and flowering-time variation contribute to repro-
ductive isolation between diploid and autotetraploid Anacamptis py-
ramidalis. Journal of Evolutionary Biology, 29, 2070–2082. https://doi.
org /10.1111/j eb.12930
Pfennig, K. S. (2016). Reinforcement as an initiator of population di-
vergence and speciation. Current Zoology, 62, 145–154. ht tps://doi.
org /10.1093/cz/zow033
R Core Team (2018). R: A language and environment for statis tical com put-
ing. Version 3.5.2. R Foundation for Statis tical Computing.
Raeymaekers, J. A . M., Boisjoly, M., Delaire, L., Berner, D., Raesaenen,
K., & Hendr y, A. P. (2010). Testing for mating isolation between eco-
types: Laboratory experiments with lake, stream and hybrid stick-
leback. Journal of Evolutionary Biology, 23, 2694–2708. https://doi.
org /10.1111/j .1420 -9101. 2010.0213 3. x
Ravinet , M., Hynes, R., Poole, R., Cross, T. F., McGinnity, P., Harrod, C.,
& Prodoehl, P. A. (2015). Where the lake meets the sea: Strong re-
productive isolation is associated with adaptive divergence between
lake resident and rnadromous three-spined sticklebacks. PLoS One,
10, 0122825. https://doi.org/10.1371/journ al.pone.0122825
Ravinet , M., Prodohl, P. A., & Harrod, C. (2013). On Irish stickle-
back: Morphological diversification in a secondary contact zone.
Evolutionary Ecolog y Research, 15, 271–294.
Rice, W. R ., & Hoster t, E. E. (1993). L aboratory experiments on specia-
tion - What have we learned in 40 years. Evolution, 47, 1637–1653.
htt ps://doi.org/10. 2307/2410209
Ric ha rds, E. J., Ser vedio, M. R ., & Mar ti n, C. H. (2019). Search ing for sym-
patric speciation in the genomic era. BioEssays, 41, 1900 047. https://
doi.org/10.1002/bies.20190 0047
Ridgway, M. S., & McPhail, J. D. (1987). R ival male efffects on courtship
behavior in the Enos lake species pair of sticklebacks (Gasterosteus).
Canadian Journal of Zoology, 65, 1951–1955. https://doi.org/10.1139/
z87-297
Rueger, T., Gardiner, N. M., & Jones, G. P. (2016). Size matters: Male
and female mate choice leads to size-assortative pairing in a coral
reef cardinalfish. Behavioral Ecology, 27, 1585–1591. https://doi.
org/10.1093/behec o/arw082
Rull, J., Abraham, S., Kovaleski, A., Segura, D. F., Mendoza, M., Clara
Liendo, M., & Teresa, V. M. (2013). Evolution of pre-zygotic and
post-zygotic barriers to gene flow among three cryptic species within
the Anastrepha fraterculus complex. Entomologia Experimentalis Et
Applicata, 148, 213–222. ht tps://doi.org/10.1111/eea.12094
Sa-Pinto, A., Baird, S. J. E., Pinho, C., Alexandrino, P., & Branco, M.
(2010). A three-way contact zone between forms of Patella rus-
tica (Mollusca: Patellidae) in the central Mediterr anean Sea.
Biological Journal of the Linnean Society, 100, 154–169. https://doi.
org /10.1111/j .1095-8 312.2010 .01405.x
Schluter, D. (1993). Adaptive radiation in sticklebacks - size, shape,
and habitat use efficiency. Ecology, 74, 699–709. https://doi.
org /10. 23 07/1940797
Schluter, D. (20 01). Ecology and the origin of species. Trends in
Ecology & Evolution, 16, 372–380. https://doi.org/10.1016/s0169
-5347(01)02198 -x
Servedio, M. R., & Noor, M. A. F. (2003). The role of reinforcement in
speciation: Theory and data. A nnual Review of Ecolog y Evolution and
Systematics, 34, 339–364. https://doi.org/10.1146/annur ev.ecols
ys.34.011802.132412
Shaw, K. A., Scotti, M. L., & Foster, S. A. (2007). Ancestral plasticit y
and the evolutionary diversif ication of courtship behaviour in
threespine sticklebacks. Animal Behaviour, 73, 415–422. https://doi.
org/10.1016/j.anbeh av.2006.09.002
Sigurjonsdottir, H., & Gunnarsson, K. (1989). Alternative mating tac-
tics of arctic charr, Salvelinus alpinus, in Thingvallavatn, Iceland.
Environmental Biology of Fishes, 26, 159–176. https://doi.o rg/10.1007/
bf000 04814
Smith, C., & Wootton, R. J. (1999). Parental energy expenditure of the
male three-spined stickleback. Journal of Fish Bi ology, 54, 1132–1136.
https://doi.org/10.1006/jfbi.1999.0940
Smith, J. M. (1966). Sympatric speciation. American Naturalist, 100, 6 37–
650. https://doi.org/10.1086/282457
Smith, J. A., Chenoweth, L. B., Tierney, S. M., & Schwarz, M. P. (2013).
Repeated origins of social parasitism in allodapine bees indicate that
the weak form of Emer y's rule is widespread, yet sympatric speciation

|
1755
DEAN E t Al.
remains highly problematic. Biological Journal of the Linnean Society,
109, 320–331. https://doi.org/10.1111/bij.12043
Snowberg, L. K., & Bolnick, D. I. (2012). Partitioning the ef fects of spa-
tial isolation, nest habitat, and individual diet in causing assortative
mating within a population of threespine stickleback. Evolution, 66,
3582–3594. https://doi.org/10.1111/j.1558-56 46.2012.01701.x
Taugbol, A., Arntsen, T., Ostbye, K., & Vollestad, L. A . (2014). Small
changes in gene expression of targeted osmoregulatory genes when
exposing marine and freshwater threespine stickleback (Gasterosteus
aculeatus) to abrupt salinity transfers. PLoS One, 9, e106894. https://
doi.org/10.1371/journ al.pone.0106894
Taylor, E. B., & McPhail, J. D. (2000). Historical contingenc y and eco-
logical determinism interact to prime speciation in sticklebacks,
Gasterosteus. Proceedings of the Royal So ciety B-B iological Sciences,
267, 2375–2384.
Tinbergen, N. (1952). The curious behavior of the stickleback. Scientific
American, 187, 22–27. https://doi.org/10.1038/scien tific ameri can12
52-22
Via, S. (20 01). Sympatric speciation in animals: The ugly duckling
grows up. Trends in Ecolog y & Evolution, 16, 381–390. https://doi.
org/10.1016/s0169 -5347(01)02188 -7
Vines, T. H., & Schluter, D. (2006). Strong assortative mating between
allopatric sticklebacks as a by-product of adaptation to different
environments. Proceedings of the Royal Society B-Biological Sciences,
273(1589), 911–916. https://doi.org/10.1098/rspb. 2005.3387
Wilz, K. J. (1970). C ausal and functional analysis of dorsal pricking and
nest activity in courtship of 3-spined stickleback Gas teros teus acu-
leatus. Animal Behaviour, 18, 115. https://doi.org/10.1016/00 03 -
3472(70)90 078 -3
Wootton, R. J. (1984). A functional biology of sticklebacks. University of
California Press.
SUPPORTING INFORMATION
Additional suppor ting information may be found online in the
Supporting Information section.
How to cite this article: Dean LL, Dunstan H, Reddish A,
MacColl ADC. Courtship behavior, nesting microhabitat, and
assortative mating in sympatric stickleback species pairs. Ecol
Evol. 2021;11:1741–1755. https://doi.org/10.1002/ece3.7164