Dr Rudiger Riesch

Personal profile

Research interests

Research in my lab aims to understand the evolutionary processes that generate biological diversity. Using integrative approaches we try to answer the questions of (a) how and why organisms diversify phenotypically, (b) what reproductive barriers are important in facilitating a reduction in gene flow between diverging populations, (c) what evolutionary forces shape these reproductive barriers, (d) what role does human-induced environmental change have on biodiversity in general, and the stability of population differentiation among diverging populations in particular, and (e) why are some species so potent invaders?



(1) Ecological speciation in extremophile poeciliids

My main research focuses on population divergence in livebearing fishes (Poeciliidae) from extreme habitats (in collaboration with Martin Plath, Ingo Schlupp and Michael Tobler). Extreme habitats are characterized by the presence of physiochemical stressors that require (of any organism tolerating them) costly adaptations that are absent in closely related species. The presence of such stressors creates complex environmental gradients and affects both the ecology of individuals and the evolutionary trajectory of populations living along the gradients. Examples for such stressors include permanent darkness (as experienced in subterranean habitats) and hydrogen sulfide (H2S) toxicity. H2S is acutely toxic to most metazoans, because it competes with oxygen in the respiratory chain, and this effect is aggravated by the extreme hypoxia also associated with the presence of this natural toxicant. Even at sublethal concentrations, H2S has several adverse effects on non-adapted fishes, including reduced egg production, smaller-sized larvae, and decreased gonadosomatic index. Thus, organisms able to cope with persistently high concentrations of H2S usually exhibit various specific adaptations. This makes H2S a source of strong divergent selection between locally adapted populations/species and populations/species that cannot tolerate it, and consequently H2S has recently been demonstrated to drive speciation events. 

El Azure, Tabasco, MexicoEl Azufre: The downstream waters are cloudy due to suspended colloidal sulfur.


Livebearing fishes inhabit a wide variety of different habitats, ranging from small creeks to large streams, freshwater lakes to coastal (brackish) lagoons, and even subterranean and toxic environments. All the adverse effects of H2S notwithstanding, several species of poeciliid fishes have been documented to thrive (and speciate) in highly toxic waters. Over the last couple of years, we have been able to identify several targets of divergent selection in extremophile poeciliids. On the life-history level, females from sulfidic habitats produce larger, but fewer offspring, and population differences have been demonstrated to have a heritable basis (e.g., Riesch et al. 2010, 2014, 2016). Morphologically, extremophile livebearers exhibit convergent evolutionary trends by developing larger heads, and thus larger branchial baskets than fish from nontoxic habitats (Riesch et al. 2016). Finally, poeciliids in toxic waters have been shown to rely on a behavioral adaptation, aquatic surface respiration (ASR), to help them cope with the extreme hypoxia associated with H2S (Tobler et al. 2009).

Poecilia sulphurariaThe sulphur molly (Poecilia sulphuraria) is a species that is endemic to toxic springs in Tabasco, México.

Lip Appendages

A large proportion of adult P. sulphuraria have dermal lip protuberances that facilitate oxygen aquisition.


My particular interest in these systems is on the role of life histories and behaviours in population divergence, reproductive isolations, and ultimately speciation. Specifically, we are investigating (i) what life-history adaptations enable these fishes to cope with H2S, (ii) how much of this divergence is due to gene-by-environment interactions (i.e., phenotypic plasticity) and how much is due to heritable differences between populations, (iii) whether the same divergent behavioural and life-history traits evolved independently along similar environmental gradients (the degree of shared vs. unique patterns of divergence), and (iv) whether local adaptation always leads to reproductive isolation from conspecifics inhabiting adjacent benign habitats.

Offspring SizeFecundity as a function of H2S concentration, with Poecilia spp. (squares) and Gambusia spp. (circles)                                                                                                        from non-toxic (white), slightly toxic (peach), and highly toxic habitats (red).


(2) Ecological speciation in Gambusia spp. from divergent predator regimes

In collaboration with Brian Langerhans (North Carolina State University), I am investigating ecology’s role in population divergence of a post-Pleistocene radiation of Bahamas mosquitofish (Gambusia hubbsi) inhabiting inland blue holes on Andros Island in the Bahamas. In some of these blue holes G. hubbsi experience relatively predator-free environments that are devoid of piscivorous fishes; in others G. hubbsi are heavily preyed upon by the bigmouth sleeper (Gobiomorus dormitor). Differences in predation regimes among blue holes drive predictable divergence in G. hubbsi body shape with fish in high-predation blue holes having smaller anterior body/head regions but larger mid-body/caudal regions compared to fish from low-predation blue holes. Furthermore, divergence in body shape is coupled with assortative mating for body shape, resulting in reproductive isolation between populations from divergent predator regimes. My projects mainly focus on life-history trait divergence across high and low predation blue holes, but I am also investigating rapid human-induced life-history evolution in Bahamas Gambusia inhabiting fragmented and unfragmented tidal creeks on six different Bahamian islands (Riesch et al. 2015).

Blue HolesAndros Island on The Bahamas with the location of blue holes some of our studies focus on.


To investigate how differences in predation and resource availability drive life-history evolution in Bahamas mosquitofish, we have examined life-history trait divergence across 7 high and 7 low predation blue holes that also vary in resource availability. However, resource availability does not co-vary with predation regime, providing an excellent opportunity to attempt to disentangle how these two selective forces might interact.

Divergence in life-history strategies across these 7 high- and 7 low-predation blue holes is largely predictable and convergent in response to differences in predation regime. In high-predation blue holes, female G. hubbsi produce more but smaller offspring, and both sexes are characterized by a higher relative lean weight (i.e., more muscle mass for more efficient fast-start response) and show a tendency for reduced body size. Moreover, at least some of these differences in life histories (i.e., offspring size and fecundity) are heritable and thus persist for several generations even under common-garden conditions (Riesch et al. 2013). These patterns of life-history divergence are largely similar to those found in guppies (Poecilia reticulata), Brachyrhaphis rhabdophora, and Brachyrhaphis episcopi, and therefore represent a very convincing case of both intraspecific and interspecific convergence in response to similar selective regimes (i.e., predation).

 Male LHs

 Female LHs

Male (top) and female (bottom) life-history differentiation in Bahamas mosquitofish from 7 high- and 7 low-predation blue holes.


(3) Population divergence in killer whales (Orcinus orca)

Although generally regarded as a single species, numerous divergent killer whale lineages are recognized worldwide, which diverged in diet, behaviour, morphology, genetics, pigmentation, and social structure (Riesch et al. 2012). From 2000 to 2011 my research focused on acoustic communication and divergence of communication signals in killer whales from the eastern North Pacific. Here, three sympatric ecotypes (residents, transients, and offshores) coexist in sympatry. Resident killer whales specialize on fish (salmon in particular), transients hunt marine mammals, and offshores probably also specialize on fish (albeit species like Pacific sleeper sharks and Pacific halibut). Killer whales produce three types of sounds: echolocation clicks are thought to function in orientation and prey detection, whereas pulsed calls and whistles are communicative signals. Despite evidence for some universal acoustic signals, the structure and frequency of use of most vocalizations differs strikingly between ecotypes. Evidence suggests that these differences are partially due to differences in hearing sensitivities of killer whale prey: marine mammals have good underwater hearing and exhibit anti-predator behavior in response to transient (i.e., marine mammal-eating) killer whale vocalizations. Salmon and other fishes, on the other hand, cannot detect killer whale sounds over significant distances.

Johnstone Strait

Johnstone Strait in British Columbia, Canada.

Northern resident

 Male northern resident killer whale surfacing in our wake waves.


My collaborative research (with Volker Deecke, John Ford, and Frank Thomsen) has mainly focused on whistle communication. Killer whale whistles are highly modulated signals that show some degree of directionality and have lower sound pressure levels and higher fundamental frequencies compared to pulsed calls. A large proportion of a killer whale's whistle repertoire is made up of stereotyped whistles that are often emitted in elaborate sequences (Riesch et al. 2008). Furthermore, whistle types of northern residents, southern residents, and transient killer whales differ drastically in structure, and while residents whistle during almost all behavioural contexts (albeit most often during socializing), transients restrict whistling to non-foraging situations (Riesch et al. 2006; Riesch & Deecke 2011).


W1T: Stereotyped whistle of northern resident killer whales.


W4T: Stereotyped whistle of northern resident killer whales.


For comparison a northern resident pulsed call: N4.

dorsal fin

Male northern resident showing his impressive dorsal fin.


(4) Ecological speciation in Timema walking-stick insects

Pathways that can lead to reproductive isolation can be categorized as occurring either before or after the formation of hybrid zygotes. Current evidence suggests that pre-zygotic isolation evolves faster than intrinsic post-zygotic isolation, and one important way in which pre-zygotic isolation may evolve is via reduced heterotypic mating due to divergence of mate preferences and traits (= sexual isolation). For example, it has been suggested that adaptation to local environments can lead to population divergence in communication signals (= sensory drive), thereby incidentally resulting in sexual isolation. However, despite the accumulation of examples of sexual isolation, for most study systems there are still fundamental, unresolved questions concerning which traits sexual isolation is actually focused upon, and what their underlying genetic basis is. In collaboration with Patrik Nosil from the University of Sheffield, my research on Timema spp. aims at resolving some of these questions by employing an integrated ecological, experimental, and genomic approach to investigate the role of chemical communication as a mechanism of sexual isolation in the group.

Timema cristinae

 Male Timema cristinae: green morph (left) and green-striped morph (right; illustration by Rosa Ribas).

Host Plants

 Some of the host plants used by Timema spp. (illustration by Rosa Ribas).


The genus Timema (Phasmatodea: Timemidae) consists of around 20 species of small, wingless, and plant-feeding walking-stick insects that are common throughout Southwestern North America (i.e., primarily California). Timema spp. present an outstanding model system for investigating the traits and genes involved in speciation, because they exhibit highly variable levels of sexual isolation between and within species. While sexual isolation has been documented between certain host-plant ecotypes (i.e., populations within species feeding on a particular species of host plant), the actual traits or sensory modalities involved in establishing sexual isolation have yet to be characterized (for more about Timema, please visit Patrik Nosil’s website at http://nosil-lab.group.shef.ac.uk). A recent description of mating behavior and courtship in Timema suggests that species recognition and mate recognition mechanisms act at relatively short-range. This implies an important role for chemical communication, for example, via cuticular hydrocarbon compounds (CHC), which are known to be important cues that serve as inter- and intra-specific recognition signals in many insects. Congruently, a recent study of CHC variation within and between several Timema species found detectable differences even between several within-species morphs (Schwander et al. 2013). Hence, chemical communication could be a central mechanism that has established and maintains reproductive isolation between species of Timema, but direct tests of this hypothesis are lacking.

Timema podura

A female Timema podura on chamise (Adenostoma fasciculatum).


My research in this system focuses on three specific questions:(1) What exactly is the extent of CHC variation in this genus? In particular, is CHC variation sex-specific, ecotype-specific and/or species-specific?(2) Do CHCs underlie sexual isolation, and/or is CHC variation host-related (i.e. ecologically based)?(3) What is the genetic basis of CHC variation? For example, how many loci are associated with CHC variability and how are they distributed across the genome?

Timema poppensis

A male Timema popoppensis on coast redwood (Sequoia sempervirens).


(5) Coexistence in a bisexual/unisexual vertebrate mating system

Among unisexual vertebrates, modern bony fishes (Teleostei) are of central interest because different modes of asexual reproduction have repeatedly evolved within this group: besides hemiclonal inheritance (hybridogenesis) and facultative parthenogenesis, several unisexual fishes are known to reproduce via sperm-dependent parthenogenesis (or gynogenesis), in which sperm is required to initiate embryogenesis, even though oocytes do not undergo meiosis, and inheritance is strictly maternal. The gynogenetic, all-female Amazon molly (Poecilia formosa), for example, resulted from a single hybridization of the two bisexual species Poecilia latipinna and Poecilia mexicana, and predominantly uses sperm from males of these two species for gynogenetic reproduction. This, however, creates a paradox: If all else is equal, unisexuals produce twice as many daughters (provided a 1:1 sex ratio in the bisexual species) and should quickly outcompete bisexuals, thereby driving ecologically similar bisexual taxa to extinction––which in turn would lead inevitably to their own extinction. Hence, the mechanisms underlying maintenance of coexistence are of considerable interest in evolutionary ecology.

Amazon molly

The Amazon molly (Poecilia formosa).


Several hypotheses have been put forth to explain the apparent coexistence of unisexual poeciliids (i.e., the Amazon molly and the Poeciliopsis complex), among them the frozen niche variation hypothesis (for details please refer to the publications by Robert Vrijenhoek), the Red Queen hypothesis (for details see, e.g., this publication), the behavioural regulation hypothesis, and the life-history regulation hypothesis. I am particularly interested in the interplay between the latter two. The behavioural regulation hypothesis basically posits that male mate choice plays an important role in regulating the mating complex, and that unisexual females may not receive sufficient sperm from ‘host’ males to continuously fertilize all of their oocytes. The life-history regulation hypothesis, on the other hand, essentially predicts coexistence if lifetime reproductive success of unisexuals was half that of their sexual counterparts. This could be mediated by male mate choice, but also by the accumulation of deleterious mutations. In collaboration with Ingo Schlupp, I have therefore been looking at life-history differences between the species and how male mate choice might translate into life-history differences (Riesch et al. 2008, 2012).

Martindale III

One of our field sites in Martindale, Texas; here P. formosa is syntopic with P. latipinna.


(6) Maternal provision in livebearing fishes

According to the Trexler-DeAngelis model of maternal provisioning (Trexler & DeAngelis 2003), postfertilization maternal provisioning is most likely to evolve in environments with constant, high levels of resource availability. The full range of maternal provisioning can be envisioned as a continuum. On one end of the continuum, all nutrients necessary for full embryonic development are packaged into the yolk prior to fertilization (i.e., lecithotrophy), while the developing embryo relies almost exclusively on a continuous supply of nutrients obtained directly from the mother during gestation (i.e., matrotrophy) on the other end of the continuum. 

Maternal provisioning

The continuum of maternal provisioning ranging from lecithotrophy via incipient matrotrophy to substantial matrotrophy as visualized by the Matrotrophy Index (MI), which tracks embryo weight during gestation. If the eggs were fully provisioned by yolk before fertilization (i.e., lecithotrophy) then we would expect the embryos to lose 25%-40% of their dry mass during development (MI between 0.60 and 0.75).


Bahamas mosquitofish, a species lacking the so-called "follicular pseudoplacenta" (a structure that is usually associated with substantial matrotrophy in poeciliids), is capable of varying degrees of maternal provisioning. In fact, different G. hubbsi populations span almost the full range of what is traditionally interpreted as the continuum between lecithotrophy and incipient matrotrophy (MI ranging from 0.70-1.10). Moreover, G. hubbsi originating from blue holes characterized by high resource availability also had higher rates of post-fertilization maternal provisioning, while abortion rates were higher in low-resource environments; thus lending strong support to two predictions of the Trexler-DeAngelis model of maternal provisioning.







View all (107) »

View all (33) »

View all (4) »

ID: 23065147