Barriers to interbreeding limit gene flow between sister taxa, leading to reproductive isolation and the maintenance of distinct species. These barriers come in many forms, and can act at different stages in the reproductive process. Pre-copulatory barriers may be due to individuals discriminating against heterospecifics in mate choice decisions. These decisions may be informed through a range of sensory modalities. If a female is mated and inseminated, then there may be multiple postmating-prezygotic barriers that affect the success of heterospecific sperm in attaining fertilisations. Post-zygotic barriers can be very early acting, resulting in embryonic fatality, or may be later acting, affecting the fitness of hybrid offspring. In this thesis I investigate potential reproductive barriers between the interbreeding field crickets Gryllus bimaculatus and G. campestris. I find that females of both species show only weak preference for conspecific calling song, and may even respond phonotactically to songs typical of heterospecific males. Female G. bimaculatus are repeatable in their preferences and strength of response. G. bimaculatus females presented with synthetic songs prefer those with longer inter-pulse intervals, whereas G. campestris show no discrimination between these songs. Upon meeting, G. campestris females strongly discriminate against heterospecific males, behaving aggressively towards them. This is likely driven by females responding to close range species recognition cues, including chemoreception. The species differ in their cuticular hydrocarbon profiles, and females that are no longer able to use their antennae to receive chemosensory information reduced their aggressive behaviour towards heterospecific males. G. bimaculatus females will mate with heterospecific males, though less readily than to conspecifics. When sequentially mated to both conspecific and heterospecific males, these females will preferentially take up and store sperm from the conspecific male, and sperm from conspecific males is more likely to sire offspring than would be predicted from the proportion of sperm in storage. Eggs from inter-species mating pairs are less likely to begin embryogenesis, and are more likely to suffer developmental arrest during the early stages of embryogenesis. However hybrid embryos that survive to later stages of development have hatching success similar to that of pure-bred embryos. After mating, phonotaxis of G. bimaculatus females towards male songs follows a pattern of suppression and subsequent recovery, likely triggered through detection of seminal proteins transferred in the male ejaculate, or detection of mechanical filling of the spermatheca. This pattern of suppression and recovery of phonotaxis does not differ between females mated to conspecific or heterospecific males. Females that lay few or no eggs do not experience a refractory period.

For some behavior there are consistent differences between individuals within a population, which is called animal personality. Across species, ranging from insects to mammals, personality has been described along behavioral gradients like activity, exploration, boldness and aggression. Monoamines such as dopamine have been shown to be essential for modulating animal behavior and could therefore be important also in explaining variation in animal personality. Supporting this, the dopaminergic system affect activity (in Confused flour beetles), and aggression (in Mediterranean field crickets). However, the causality and effect of dopamine on these behaviors, and also other behavioral traits used to describe personality is currently less explored. This study experimentally investigated how increased level of dopamine affects activity, boldness, exploration and aggression in Mediterranean field crickets (Gryllus bimaculatus). I show that dopamine manipulation had no effects on measured behavior. These results indicate that increased dopamine levels do not affect the scored personality traits in Mediterranean field crickets. The causality and generality of the relationship between dopamine and behavior used to score variation in personality is thus not clear in this species.

Ziel der vorliegenden Dissertation war es, zum Verständnis der genetischen Grundlagen des akustischen Kommunikationssystems der Grille Gryllus bimaculatus beizutragen (s. auch 1.2). Dazu wurde die Expression eines Orthologs des no-on-transientA-Gens (nonA) mit Hilfe der RNA-Interferenz-Methode spezifisch herunterreguliert. Bei nonA handelt es sich um ein vielversprechendes Kandidatengen, da Punktmutationen in der codierenden Region des Gens die Eigenschaften des männlichen Balzgesangs bei Drosophila melanogaster beeinflussen. Zudem belegen Gentransfer-Experimente bei Drosophila, dass dieses Gen artspezifische Informationen des Balzgesangs enthält. Die Analyse der Gesangsdaten ergab, dass sich die Periodenlänge durch das Herunterregulieren von NONA nicht verändert. Außerdem konnte gezeigt werden, dass nonA-dsRNA-injizierte Tiere seltener 3-silbige Chirps produzieren, dafür aber mehr 4- und 5-silbige Chirps. Die Auswertung der tageszeitlichen Gesangsaktivität zeigte, dass alle Tiere signifikant am häufigsten im ersten Nachtquartal (nach Erlöschen der Beleuchtung) zirpten. Ein Effekt durch das Herunterregulieren von NONA konnte statistisch nicht belegt werden. Allerdings schien es einen Trend bei nonA-dsRNA-injizierten Tieren zu geben, gleichmäßiger über den Tag verteilt Gesangsaktivität zu zeigen. Transgene Drosophila melanogaster, deren arteigenes nonA durch das der Grille ersetzt bzw. ergänzt worden war, zeigten durchweg eine verbesserte Überlebensfähigkeit (Steigerungen zwischen 27 und 340%). Auch das positiv phototaktische Verhalten wurde durch das Grillen-NONA bei allen transformanten Fliegen verstärkt; allerdings fiel dieser Effekt eher marginal aus. Dennoch kann durchaus von einer zumindest teilweisen funktionellen Konservierung des nonA-Gens zwischen Gryllus bimaculatus und Drosophila melanogaster ausgegangen werden. / The present thesis aims to widen our understanding of the genetic background of the acoustic communication system of the cricket Gryllus bimaculatus (see also 1.2). Therefore the expression of an ortholog of the no-on-transientA (nonA) gene was specifically inhibited via RNA-interference. The nonA gene is one of the most interesting candidate genes in this context, as point mutations in the coding region of the gene affect the characteristics of the male’s calling song. Furthermore, gene transfer experiments in Drosophila showed that this gene obviously carries species-specific song information. The analysis of the calling song of nonA-RNAi-treated crickets, revealed that the duration of the syllable period was not influenced by the “knock-down” of the gene, but that the inhibition had a certain impact on the maximum number of syllables per chirp, as nonA-dsRNA-injected crickets produced significantly less 3-syllable chirps and significantly more 4- and 5-syllable chirps. Differences in the daytime calling activity between nonA-dsRNA-injected crickets and control groups could not be verified. The calling activity of all groups reached its peak in the first quarter of the night and significantly differed from the low calling activity during the remaining quarters of the day. Although the activity of all animals reached its peak during the first quarter of the night, there seems to be a trend that this rhythmical behaviour was less pronounced in nonA-dsRNA-injected crickets. Drosophila melanogaster mutants, which had been transformed with the nonA ortholog of Gryllus bimaculatus, increased their survival by 27% to 340%. In addition, the positive phototactic behaviour was slightly increased in all tested animals - this effect, however, remained marginal. Nevertheless, the nonA gene seems to be at least partly functionally conserved between Gryllus bimaculatus and Drosophila melanogaster.

Die vorliegende Dissertation verbindet Ergebnisse aus neuroethologischen, verhaltensbiologischen, quantitativ genetischen und genomischen Ansätzen bei Feldgrillen (Gryllus), um neue Erkenntnisse über die Rolle von sexueller Selektion bei Artbildung zu erlangen. Es wird gezeigt dass multivariate Gesangspräferenzen von Grillenweibchen von wenigen Merkmalen abhängen und zwischen Arten ähnlich sind, während sich Männchengesänge in allen Merkmalen unterschieden. Verschiedene Ebenen der Gesangserkennung sind durch unterschiedliche Präferenzfunktionen charakterisiert. Multivariate Präferenzen können also gleichzeitig verschiedene Indikatoren für Paarungspartnerqualität aus den Gesangsmerkmalen erkennen. Eine polygene genetische Architektur der Gesangsmerkmale und der Präferenz wurde beobachtet und weist auf eine eher langsamere Divergenz hin, obwohl gonosomale Vererbung mehrerer Gesangsmerkmale höhere Evolutionsraten zulässt. Starke Kovarianz zwischen den Merkmalen die direkt sexueller Selektion unterliegen und Merkmale, die nicht direkt von Weibchen gewählt werden, zeigen, dass indirekte Selektion teilweise für die markante Divergenz der Gesänge verantwortlich sein könnte, trotz begrenzter Divergenz der Präferenzen. Ferner zeigte ein Artvergleich der multivariaten Gesangsmerkmale, dass die Form der Präferenzfunktion die Ausrichtung der Kovarianzen und damit die erwartete Selektionsantwort der männlichen Gesänge beeinflussen kann. Simulationen ergaben starke Hinweise auf Genfluss zwischen zwei nahverwandten Arten über einen langen Zeitraum . Nur wenige Contigs zeigten hohe genetische Divergenz und hohe Raten nicht-synonymer Polymorphismen. Diese stimmten aber mit Genen überein, die experimentell nachgewiesene Funktionen in neuromuskulärer Entwicklung und im Paarungsverhalten haben. Zusammen zeigen die Ergebnisse das Potential von sexueller Selektion bei der Entstehung und Aufrechterhaltung von reproduktiver Isolation zwischen Arten. / This thesis integrates insights from neuro-ethological, behavioural, quantitative genetics, and genomic approaches in field crickets to provide novel insights in the role of sexual selection in speciation, in particular focusing on speciation with gene flow. It was shown that song preferences depend on few traits and are similar across species while the male song has diverged strongly in all traits. Because the different levels of song recognition are characterized by different types of preference functions, it is conceivable that multivariate preferences can extract various cues for mate quality from different traits simultaneously. A polygenic genetic architecture was found for song traits and preferences, probably limiting divergence rates. However, sex-chromosomal inheritance of some song traits may have allowed for somewhat higher rates. Strong covariance was found between traits that are under sexual selection and traits that are not directly selected by females. This indicates that indirect selection may be responsible in part for striking multivariate divergence in the male calling song despite limited divergence in female preferences. Furthermore, comparing multivariate song traits among species showed that the shape of the preference function can affect the orientation of trait covariance and thereby the selection responses of the male song. Coalescent simulations revealed evidence for a long history of gene flow between two closely related cricket species. Only few contigs with high genetic divergence and high rates of non-synonymous SNPs were found, but many of those that were highly diverged matched genes with experimentally proven functions in neuromuscular development and courtship behavior. Together, these findings underline the potential for sexual selection to drive reproductive isolation.

Akustische Signale dienen vielen Tierarten als Mittel zur Partnerfindung. Diese Tiere müssen ihre Entscheidung für den besten Paarungspartner durch die Integration der Parameter verfügbarer Signale treffen. Das Wahlverhalten weiblicher Grillen basiert auf der Attraktivität des zeitlichen Musters des männlichen Gesangs und auf dessen Intensität. Diese Eigenschaften korrelieren nicht zwangsweise positiv, daher wird sich kein Männchen in einer Aggregation von singenden Männchen als überragend abheben und somit wird dem Weibchen die Entscheidung erschwert. Die vorliegende Arbeit beschäftigt sich mit der Frage wie die relevanten Merkmale des männlichen Gesangs während der Entscheidungsfindung des Weibchens integriert werden. 6 Arten von Feldgrillen wurden in nicht-Wahl- und Wahlexperimenten hinsichtlich ihrer Antworten und Präferenzen für männliche Signale, welche sich in Pulsrate, Modulationstiefe, Intensität, Anordnung der Chirps während der Wiedergabe und der zeitlichen Verschiebung zweier Signale zueinander unterschieden, getestet. Des Weiteren wurde durch Transitivitätstests untersucht, ob dem Wahlverhalten weiblicher Grillen rationale oder vergleichende Entscheidungsmechanismen zugrunde liegen. Zusammenfassend zeigen die Ergebnisse, dass die Verarbeitungsmechanismen, welche der Entscheidungsfindung bei weiblichen Feldgrillen zugrunde liegen, in den untersuchten Arten ähnlich sind. Die Lokalisierung des Signals ist nicht unabhängig von der Erkennung. Neben einem generellen Verarbeitungsschema wurden zwischen den einzelnen untersuchten Arten charakteristische Unterschiede in der Mustererkennung, der Verarbeitung der Signalintensität und der Intensitätsgewichtung festgestellt. Jedoch wiesen nah verwandte Arten ähnlichere Präferenzen auf, als nicht nah verwandte Arten. Die Tests zur Transitivität der Präferenzen weisen auf vergleichende Entscheidungs-mechanismen hin und widersprechen einer rationalen Partnerwahl. / In many animals acoustic signals serve for mate attraction. They thus need to make a decision about which is the best mating partner in a choice situation by integrating the cues of the available signals. Choice behaviour in female crickets is based on the attractiveness of the temporal pattern of a male’s calling song and on its intensity. The parameters of a calling song are not necessarily positively correlated and no male will emerge as superior in an aggregation of singing males. The present thesis addresses the question on which cues decision making is based and how these relevant cues of a male’s song are integrated during the decision process of a female. To this aim no-choice and choice experiments with 6 species of field crickets were conducted. Experiments systematically varied pulse rate, modulation depth, intensity, chirp/trill arrangement and temporal shifts of synchronously presented signals. Furthermore, tests for transitivity of preferences examined if female choice behavior is based on rational or comparative decision making mechanisms. In summary, the results reveal that sensory processing underlying female decisions is similar in the 6 species studied here. Incoming signals are analyzed separately in bilaterally paired networks with parallel pathways for signal attractiveness and signal intensity. A gain-control mechanism fuses the outcome of both pathways and signal intensity is weighted by pattern attractiveness. Thus localization is not independent from signal recognition. Despite this general scheme remarkable characteristic differences between species were observed in pattern recognition, processing of signal intensity and weighting of signal intensity. Closely related species exhibited more similar preferences than unrelated species. Furthermore the tests for transitivity of preferences indicated that females use comparative decision making mechanism and contradicted rational mate choice.

Darwinian sexual selection predicts that males with the most extravagant secondary sexual traits suffer elevated mortality. Although correlative evidence has generally not borne this idea out, recent research, including a field cricket study, showed that investment in sexually selected traits is costly to survival. I investigated male survival, ornamentation (song) and mating success in a North American grylline, Gryllus pennsylvanicus, to test the generality of previous work and highlight the importance of ecology differences to resource allocation. As the calling songs of older male G. pennsylvanicus are highly attractive to females, in Chapter 2 I tested whether male age correlated with calling song and found a weak but statistically significant correlation, thus leaving open the possibility that choosy females use an age-based indicator mechanism. In Chapter 3, I tested the condition dependence of male survival and calling effort. In contrast to previous work, I found that high condition males both called more and lived longer than low condition males, although there was no trade-off between survival and calling effort. The substantial condition dependence of calling effort suggests that calling effort is under strong directional selection. In Chapter 4 I tested whether female mating preferences resulted in strong selection on male calling effort. I also tested for the condition dependence of female mating preferences. I found that female choosiness was condition-dependent, but the rank of preferred male songs (preference function) was not. Both low and high condition females preferred high calling effort over low calling effort song. In Chapter 5 I tested for evidence of nonlinear selection on male survival that might explain the nonlinear pattern of male investment in survival seen in Chapter 3 (i.e. male survival leveled-off with increasing condition). I found that socially experienced females, but not virgin and naive females, exerted linear selection on male age. I discuss these and the other results of my thesis in the context of previous work on field crickets and condition-dependent ornamentation. Finally, Appendix A reports results that confirm ancient Chinese cultural knowledge that large headed male crickets are more successful in male-male combat.

Darwinian sexual selection predicts that males with the most extravagant secondary sexual traits suffer elevated mortality. Although correlative evidence has generally not borne this idea out, recent research, including a field cricket study, showed that investment in sexually selected traits is costly to survival. I investigated male survival, ornamentation (song) and mating success in a North American grylline, Gryllus pennsylvanicus, to test the generality of previous work and highlight the importance of ecology differences to resource allocation. As the calling songs of older male G. pennsylvanicus are highly attractive to females, in Chapter 2 I tested whether male age correlated with calling song and found a weak but statistically significant correlation, thus leaving open the possibility that choosy females use an age-based indicator mechanism. In Chapter 3, I tested the condition dependence of male survival and calling effort. In contrast to previous work, I found that high condition males both called more and lived longer than low condition males, although there was no trade-off between survival and calling effort. The substantial condition dependence of calling effort suggests that calling effort is under strong directional selection. In Chapter 4 I tested whether female mating preferences resulted in strong selection on male calling effort. I also tested for the condition dependence of female mating preferences. I found that female choosiness was condition-dependent, but the rank of preferred male songs (preference function) was not. Both low and high condition females preferred high calling effort over low calling effort song. In Chapter 5 I tested for evidence of nonlinear selection on male survival that might explain the nonlinear pattern of male investment in survival seen in Chapter 3 (i.e. male survival leveled-off with increasing condition). I found that socially experienced females, but not virgin and naive females, exerted linear selection on male age. I discuss these and the other results of my thesis in the context of previous work on field crickets and condition-dependent ornamentation. Finally, Appendix A reports results that confirm ancient Chinese cultural knowledge that large headed male crickets are more successful in male-male combat.

No description available.

Animal personality can be defined as a set of physiological and behavioral characteristics that differ between individuals, but are consistent over time and across situations. The evolution of individual differences in behavior that are consistent over time and situations is still not clear. Our understanding of why animals have personality can be improved by investigating the underlying physiological mechanisms of animal behavior. Serotonin is a key monoamine that serves as a physiological modulator of animal behavior. Selective serotonin reuptake inhibitors are a group of chemicals that increase levels of serotonin in the brain. Fluoxetine is one such chemical and is used to treat depression in humans. In the field cricket (Gryllus bimaculatus), increased levels of serotonin have been linked to higher activity and boldness, which are both personality traits. In the current study, the effects of induced serotonin on activity, exploration, boldness and aggression was investigated. My results show that injecting fluoxetine causes substantial changes in behavioral traits used to describe personality in field crickets. This result is opposite to previous studies, as serotonin induced individuals were less active, less explorative, and won less fights, compared to control individuals. This could be due to serotonin existing naturally within the circulatory system of the field cricket, whereas fluoxetine is a manufactured chemical intended for human receptors, or that fluoxetine has a similar effect in modulating personality in field crickets as in humans. Since fluoxetine acts similarly in field crickets as in humans, an increased understanding of the effects of induced serotonin on different behaviors in field crickets could be beneficial for treating psychological illnesses.

The frequency, duration, and co-occurrence of several environmental stressors are increasing globally. Multiple stressors may have compounding or interactive effects on animals, resulting in either additive or non-additive costs, but animals may mitigate these costs through various strategies of resource conservation or shifts in resource allocation. Thus, through two related factorial experiments, I measured a range of traits—from those related to life history and behavior to underlying physiology— to investigate the nature of costs (additive, non-additive, or neither additive nor non-additive), cost-mitigating strategies (resource conservation or allocation), and life-history strategy related to multiple stressors. First, I leveraged life-history strategy differences in the sand field cricket, Gryllus firmus, to investigate the individual and interactive effects of food and water limitation on fitness-related traits. Gryllus crickets exhibit a wing dimorphism mediating two distinct life-history strategies—long-winged crickets invest into flight capability while short-winged crickets do not. My results indicate that traits vary in their sensitivity to environmental stressors and stressor-stressor interactions (e.g., flight muscle). I only found support for non-additive costs or single-stressor costs of water and food limitation to fitness-related traits. Water availability had a larger effect on traits than food availability, affected more traits than food availability (wing dimorphism), and mediated the effects of food availability. Second, I investigated the role of life-history strategy in cost-mitigating strategies and further examined the costs (additive or non-additive) of multiple stressors to fitness-related traits, physiology, and behavior. I used the variable field cricket, Gryllus lineaticeps, to examine the costs of a simulated heat wave and water limitation. These stressors resulted primarily in single-stressor or non-additive costs to important traits (e.g., survival, final body mass, and total water content), extensive shifts in resource allocation priorities (e.g., reduced prioritization of body mass), and a limited capacity to conserve resources (heat wave reduced energy use only when water was available). Further, life-history strategy influenced the emergency life history stage (ELHS) because wing morphology and stressor(s) interacted to influence gonad and body mass, boldness behavior, and immunocompetence. Together, these two studies demonstrate that water availability and life-history strategy should be incorporated into future studies integrating important conceptual frameworks of stress (multiple-stressor framework and ELHS) across a suite of traits—from survival and life history to behavior and physiology.