Open Access

Adaptations, either structural, physiological, or behavioural are essential for the survival of species or individuals in changing environments. For animals that live in two distinct environments, adaptations only persist if they provide advantages in one or both media, and if the adaptations do not negatively interfere with other substantial life functions in any of the media. The ability to flexibly respond to rapid environmental changes within such adapted forms potentially provides safeguards, limiting adverse effects of any impact. This also applies to the sensory modalities of amphibious living vertebrates (e.g., seals (Pinnipedia), crocodilians (Crocodilia), penguins (Sphenisciformes)). They are evolutionarily very well adapted to the aquatic element for foraging but have also kept adaptations that enable them to live on land, where they perform major life history tasks like breeding. It is therefore likely that the hearing and vocal systems are also strongly impacted by living in water and air, as they have different acoustic properties. However, to date, the auditory perception and vocal flexibility of penguins have remained largely unexplored in both media. To address some gaps in our knowledge, this thesis focusses on three questions concerning the ability of penguins to hear in air and their vocal flexibility during the breeding season: 1) can penguins in the wild adjust vocal properties depending on the acoustic environment of their colony?; 2) can captive penguins be trained to perform hearing trials? and; 3) how well can penguins hear in air? Gentoo penguins (Pygoscelis papua) live in large social breeding groups that can include different species. Because of differences in the acoustic environment of such colonies, the vocal characteristics of Gentoo penguins may differ between colonies that include multiple species and those where only Gentoo penguins breed. To examine differences in vocalisations, the vocal parameters of ecstatic display calls (e.g., frequency, structure, duration) were compared between individual Gentoo penguins breeding in colonies with Adélie (Pygoscelis adeliae) or Chinstrap penguins (Pygoscelis antarcticus) and those breeding in monospecific colonies. Gentoo penguin calls showed a lower frequency and narrower energy distribution when they bred in proximity to Adélie penguins but their calls did not vary when they bred in proximity to Chinstrap penguins, whose vocalisations are more distinct. Even though the underlying mechanisms remain unknown, these results suggest that Gentoo penguins can adjust the parameters of their vocalisations according to the social (and presumably acoustic) environment of their colonies. To validate that these vocal adjustments result from the acoustic environment encountered by penguins in their colony, it is essential to know precisely what they can hear in air. The most sensitive method to examine the hearing capacity of animals is to train them for psychoacoustic measurements. Such training methods have, however, not yet been applied to penguins and there was considerable uncertainty whether the existing training methods (typically used for marine mammals and terrestrial birds) could be applied to penguins. Moreover, it was unknown how individual differences and environmental conditions could affect their trainability. In this thesis, I describe the successful training of four captive Humboldt penguins (Spheniscus humboldti) for psychoacoustic measurements over a 48-month period. Monitoring training progress revealed that their trainability depended on individuality, time of day, moult, pairing and daylength. Penguins have an ear morphology evolved for an aquatic lifestyle. How this morphological adaptation might influence their auditory capabilities in air was previously unknown. Using psychophysical hearing tests in the afore-mentioned Humboldt penguins, I found that their hearing range at 50 % hit rate was below 76 dB rms re 20 μPa between 0.250 kHz and 10 kHz, with the most sensitive hearing at 2 kHz (mean threshold of 15.3 dB rms re 20 μPa, SD = 10.6 dB, n = 4). After a surprisingly rapid decline of sensitivity in all animals at 0.500 kHz (mean ± SD = 64.1 dB rms re 20 μPa ± 9 dB, n = 4), three of the animals consistently detected the 0.250 kHz frequency stimulus again at comparably low sound levels (mean ± SD = 36.8 dB rms re 20μPa ± 2.4 dB, n = 3). Specific aquatic adaptations of penguins therefore do not appear to have strongly limited their hearing capacity in air compared to other aquatic birds. By combining my studies on two species, one in the wild and one in captivity, this thesis suggests that penguins may be able to adjust their vocal characteristics to the acoustic environment of their colony by adapting towards lower frequency vocalisations. This finding was supported by the unexpected sensitivity in the lower frequency range found here in the trained penguins. The range of acoustic flexibility required by each species for their ecological niche, could depend on morphological adaptations to a semi-aquatic lifestyle, and their phenotypic plasticity. This thesis provides hypotheses on the specific traits that could be involved, and that can now be further elucidated by targeted investigations.

The central nervous system (CNS) is the integration center for the coordination and regulation of all body activities of animals and the source of behavioral patterns, behavioral plasticity and personality. Understanding the anatomy and the potential for plastic changes of the CNS not only widens the knowledge on the biology of the respective species, but also enables a more fundamental understanding of behavioral and ecological patterns. The CNS of species with different sensory ecologies for example, will show specific differences in the wiring of their CNS, related to their lifestyle. Spiders are a group of mesopredators that include stationary hunting species that build webs for prey capture, and cursorial hunting species that do not build capture webs. These distinct lifestyles are associated with major differences in their sensory equipment, such as size of the different eyes. In this thesis, I aimed to answer if a cursorial mesopredator would change its behavior due to different levels of perceived predation risk, and if this behavior would be influenced by individual differences (chapter 1); how the visual pathways in the brain of the cursorial hunting jumping spider Marpissa muscosa differs from that of the nocturnal cursorial hunting wandering spider Cupiennius salei (chapter 2); to what degree the visual systems of stationary and cursorial hunting spiders differ and whether CNS areas that process vibratory information show similar differences (chapter 3); and finally if the CNS in stationary and cursorial hunting spiders shows different patterns of neuroplasticity in response to sensory input and deprivation during development (chapter 4). In chapter 1, I found that jumping spiders adjust their foraging behavior to the perceived level of risk. By favoring a dark over a light substrate, they displayed a background-matching strategy. Short pulses of acute risk, produced by simulated bird overflights, had only small effects on the behavior. Instead, a large degree of variation in behavior was due to among-individual differences in foraging intensity. These covaried with consistent among-individual differences in activity, forming a behavioral syndrome. Our findings highlight the importance of consistent amongindividual differences in the behavior of animals that forage under risk. Future studies should address the mechanisms underlying these stable differences, as well as potential fitness consequences that may influence food-web dynamics. In chapter 2, I found that the visual pathways in the brain of the jumping spider M. muscosa differ from that in the wandering spider C. salei. While the pathway of the principal eyes, which are responsible for object discrimination, is the same in both species, considerable differences occur in the pathway of the secondary eyes, which detect movement. Notably, M. muscosa possesses an additional second-order visual neuropil, which is integrating information from two different secondary eyes, and may enable faster movement decisions. I also showed that the tiny posterior median eye is connected to a first-order visual neuropil which in turn connects to the arcuate body (a higher-order neuropil), and is thus not vestigial as suggested before. Subsequent studies should focus on exploring the function of the posterior median eyes in different jumping spider species, Foraging behavior, neuroanatomy, and neuroplasticity in cursorial and stationary hunting spiders as they show considerable inter-specific size differences that may be correlated with a differing connectivity in the brain. In chapter 3, I described all neuropils and major tracts in the CNS of two stationary (Argiope bruennichi and Parasteatoda tepidariorum) and two cursorial hunting spiders (Pardosa amentata and M. muscosa). I found major differences in the visual systems of the secondary eyes between cursorial and stationary hunting spiders, but also within the groups. A. bruennichi has specialized retinula cells in two of the secondary eyes, which connect to different higher-order neuropils. P. tepidariorum has only a single visual neuropil connected to all secondary eyes, and lacks recognizable mushroom bodies. The neuroanatomy of CNS areas that process mechanosensory information on the other hand, is remarkably similar between cursorial and stationary hunting species. This suggests that the same major circuits are used for the processing of mechanosensory information in both cursorial and stationary hunting spiders. Future studies on functional aspects of sensory processing in spiders can build on the findings of our study. In chapter 4, I found that developmental neuroplasticity in response to sensory input differs between a cursorial (M. muscosa) and a stationary hunting spider (P. tepidariorum). While deprivation of sensory input leads to a volume increase in several visual and mechanosensory neuropils M. muscosa, neither sensory deprivation nor sensory enrichment had an effect on the volume of neuropils in P. tepidariorum. However, exposure to mechanical cues during development had an effect on the allometric scaling slope of the leg neuropils in both M. muscosa and P. tepidariorum. Future studies should focus on the genetic and cellular basis of developmental neuroplasticity in response to sensory input in order to explain the observed patterns.

The need for the diversification of utilised species has emerged in the present aquaculture production environment. Shifts in consumer interest, climate change-induced temperature increases, and major fish disease outbreaks have put a strain on this industry. In this context, the pikeperch (Sander lucioperca) has become a new target species for aquaculture in Central Europe. This new aquaculture focus species exhibits high numbers of offspring, fast growth, and high consumer acceptance. It can also effectively deal with higher temperatures and turbid water. However, the rate of successful rearing is still low, as various developmental transformations and environmental effects commonly lead to high mortality rates during the early ontogenetic stages. The aim of this doctoral project was thus to obtain insight into embryonic to larval developmental changes during pikeperch ontogeny. Specifically, the times of change that influence survival were of focus. Based on the available literature, particular attention was paid to general growth patterns and the connected developmental changes, the determination of myogenesis gene marker expression changes, and the support of animal welfare efforts for pikeperch rearing procedures. To achieve the aims of the study, a methodical setup consisting of morphometric and developmental observations was combined with transcriptome gene marker analysis for the different ontogenetic stages. Three developmental phases were differentiated during the embryo-larval transition. Each of these possessed distinct growth patterns with different growth rates. The intermediate threshold phase showed internal organ development that focused on digestive, neuronal, and heart tissues. Three activity phases of myogenesis were determined: during early embryonic development, before hatching, and after hatching during the larval stages. Therefore, muscle development seemed to be regulated to balance energy expenditures. Additionally, two coinciding skeletogenic phases were found. Furthermore, a cell line from whole embryos was developed to support the replacement of animals in future experimental setups. A software system for video analyses was developed to support rearing procedures in aquaculture facilities. This prototype can be used to automate the counting of specimens and thus allows for faster responses to increasing mortalities. Based on the results of this thesis project, further insights into the early development of pikeperches were obtained. This will facilitate the design and adaptation of raising and husbandry protocols, which can help to further establish pikeperch as an aquaculture species and support its application in modern recirculatory systems.

Species are the basic units of evolution and biodiversity, and the process of speciation has been one of the most important questions in biology. The evolution of species with common descent is considered to be mainly driven by natural and sexual selection. The material basis and mechanical cause of organismic evolution were recognized during the formation of the modern synthesis of the evolutionary theory in the early 20th century, providing the framework for speciation studies. During this period, the biological species concept was developed in the frame of population genetics, putting emphasis on the reproductive isolation between populations. The phylogenetic species concept developed in the 1980s, on the other hand, does not make any particular assumption about evolutionary or speciation processes. It defines species via their unique combination of character states which are compatible with phylogenetic practices. However, the aforementioned two species concepts are difficult to apply in alpha-taxonomy, where newly discovered species are largely described by the morphological (typological) species concept for practical reasons. Nevertheless, the description of morphological species provides the basis for further assessments of species delimitation via other species concepts and approaches. One of the tools for assisting the identification and discovery of animal species is DNA barcoding, which uses a standard region of mitochondrial DNA sequence as a universal DNA barcode. However, its assumption of intraspecific genetic distances being smaller than interspecific genetic distances does not always hold. Species-level poly-/paraphyly is prevalent due to the discrepancy between the phylogenies of mitochondrial DNA and species. This suggests that the application of DNA barcodes must be combined with an integrative taxonomic approach. Beside the application as a tool for assisting species identification, the information from mitochondrial DNA sequences opens up a window for looking into the complex history of species. Sexual selection is a potential mechanism driving the evolution of species. It favors traits that increase mating probability and mating success. It can result from intrasexual competition, female preference or sexual conflict. However, previous comparative studies using the degree of sexual dimorphism as a proxy for the strength of sexual selection have yielded inconsistent results as to the relationship between sexual selection and species richness. A possible cause of the inferred low association are factors other than sexual selection, which can also lead to the evolution of sexual dimorphism, such as selection for increased female fecundity. In order to assess the effect of sexual selection on speciation, the lability and evolvability of traits need to be studied that are clearly under sexual selection. The aim of this thesis is to improve the knowledge about dwarf spider (Erigoninae, Linyphiidae) diversity and taxonomy, and to assess the evolutionary patterns of dimorphic traits that are under sexual selection. I focused on the abundant and diverse male prosomal modifications in dwarf spiders that are linked to the transfer of secretions from the male to the female during courtship and mating (gustatory courtship). This approach explores the process of speciation and the role of sexual selection on species diversification. I described new erigonine species and revised the classification of known species based on phylogenetic analyses. I also applied X-ray micro-computed tomography (micro-CT) to investigate the distribution and evolutionary pattern of the gustatory glands to tease apart the evolution of prosomal shape and glandular equipment. This cumulative thesis consists of three publications: Publication 1: This publication aimed at contributing to the knowledge of erigonine diversity. The genus Shaanxinus previously contained only two species from China. I collected dwarf spiders from multiple locations in Taiwan from above-ground vegetations with a seldom applied collecting method. Inspection of the collected material resulted in the discovery of 13 Shaanxinus species. An additional species from Vietnam was described from a museum collection. I provided a revision of the genus Shaanxinus. A phylogenetic analysis using morphological characters was conducted for determining the possible generic synamomorphies. I also reconstructed the glandular distribution associated with male prosomal modifications, as well as the detailed structure of a male secondary sexual organ (pedipalp) by micro-CT. Furthermore, I conducted phylogenetic analyses based on sequences from two mitochondrial and one nuclear loci, and assessed the efficacy of different criteria in species identification using DNA barcoding. Distinction of morphologically similar species have been assisted by molecular data. The species level poly-/paraphyly found in mitochondrial DNA sequences caused the low efficacy of many distance- and tree-based species identification methods, while the nearest neighbor method showed high identification success. The non-monophyly is likely caused by instances of interspecific hybridization and recent parapatric speciation. The genus Shaanxinus thus lend itself as an ideal group for congeneric phylogeographic studies addressing the interactions between closely related species. Published in: Lin, S.-W., Lopardo, L., Haase, M. & Uhl, G. 2019. Taxonomic revision of the dwarf spider genus Shaanxinus Tanasevitch, 2006 (Araneae, Linyphiidae, Erigoninae), with new species from Taiwan and Vietnam. Organism Diversity & Evolution, 19, 211-276. Publication 2: Sexually dimorphic prosomal modifications that are related to gustatory courtship occur in many dwarf spider species. These features evolved in the context of sexual selection, which has a potential effect on species diversification. In contrast to many erigonine genera which present little variability in male prosomal traits, the genus Oedothorax presents higher diversity in male prosomal structures among species not only in the position and shapes of the modifications, but also in the degree of modification, ranging from absent to highly elaborated. This genus thus lends itself as a suitable target group for studying the effect of gustatory-courtship-related traits on species diversification. I conducted a revision of the 82 species previously belonging to this genus. Based on the result of a phylogenetic analysis, this genus was re-delimited with 10 species as Oedothorax sensu stricto, while taxonomic decisions were made for other species including synonymization with species from other genera and transferring species to other existing and newly defined genera. 25 species were deemed as “Oedothorax” incertae sedis. The reconstruction of character state evolution suggested multiple origins of specific prosomal modification types. Convergent evolution of these traits among different lineages suggests that sexual selection has played an important role in the species diversification of dwarf spiders. Published in: Lin, S.-W., Lopardo, L. & Uhl, G. 2021. Evolution of nuptial-gift-related male prosomal structures: taxonomic revision and cladistic analysis of the genus Oedothorax (Araneae: Linyphiidae: Erigoninae). Zoological Journal of the Linnean Society, XX, 1-168. Publication 3: Although sexually dimorphic traits have inspired the concept of sexual selection as the driving force of their evolution, they might also have evolved due to other ecological factors. These factors include the sexual signal adaptation to the environment as well as sexual differences in ecological relations and parental investment. In contrast, the gustatory courtship in dwarf spiders is associated with sexually dimorphic male prosomal modifications, which have clearly evolved in the context of sexual selection. Multiple origins of various external prosomal modifications have been shown in erigonine phylogeny, but the evolutionary pattern of the associated glands has not been investigated. Our phylogenetic analysis incorporated the characters related to the glandular distribution in the male prosoma as well as the external shapes yielded from X-ray micro-computed-tomography showed a single origin of gland among the investigated erigonine taxa. The internal anatomy revealed previously undetected trait lability in attachments of muscles to the cuticular structures, as well as the presence/absence and differences in glandular distribution even in species without external modification. Our finding further supports that erigonine male prosomal traits are under divergent selection, and corroborates the argument that erigonines are a suitable group for investigating the effect of sexual selection on speciation. Published in: Lin, S.-W., Lopardo, L. & Uhl, G. 2021. Diversification through gustatory courtship: an X‑ray micro‑computed tomography study on dwarf spiders. Frontiers in Zoology, 18: 51.   The results of this thesis corroborate the importance of applying phylogenetic methods and an integrative approach in the description of new species, as well as in revising taxa which might not be monophyletic. Overall, the studies contributed to a more comprehensive knowledge about erigonine species diversity, phylogeny and the possible diversifying effect of sexual selection on male traits associated with gustatory courtship.

As the effects of anthropogenic climate change become more pronounced, it is critical to understand if and how species can persist in novel environments. Range-expanding species provide a natural experiment to study this topic: by studying the factors contributing to successful colonization of new habitats, we can gain insight into what influences organisms’ adaptive potential. The wasp spider, Argiope bruennichi, has expanded its range from warm, oceanic and Mediterranean climate zones (populations in this region are referred to as “ancestral” or “core”) into a new thermal niche, the continental climate of the Baltic States and Scandinavia (referred to as “expanding” or “edge”) within the last century. Past work demonstrated that the expanding populations are European in origin, but are more diverse than the ancestral populations, due to genetic admixture. This discovery led to the following questions, which are investigated in this dissertation: (i) Was the successful colonization of colder, more continental northern climates due to phenotypic plasticity or genetic adaptation? (ii) If A. bruennichi’s establishment of northern latitudes can be attributed to genetic adaptation, did selection act on standing genetic variation, on genetic variation introduced via admixture/introgression, on specific genomic regions, or on novel mutations? (iii) Is there a role of the microbiome in the A. bruennichi range expansion? In Chapter 1, we assembled a chromosome-level genome for the species: the first such high-quality genome for a spider, which we made use of as a resource to provide the genomic context of single nucleotide polymorphisms in our primary study on genetic adaptation and phenotypic plasticity (Chapter 3). The genome assembly also opened the door to many new projects, such as the study presented in Chapter 2. In Chapter 2, the chromosome-level resolution of our assembly allowed us to identify the sex chromosomes in A. bruennichi. Due to the X1X20 sex chromosome system, where males have one copy of two X chromosomes, and females have two copies, the X chromosomes have a lower effective population size, and lower recombination rate, than autosomes. These characteristics give rise to the theoretical prediction of increased evolutionary rates in sex chromosomes. Knowing the identity of the sex chromosomes in our A. bruennichi genome assembly will allow us to test if there is stronger differentiation between populations on the X chromosomes. Chapter 3 represents the central study of this dissertation. We performed a reciprocal transplant common garden experiment to assess plasticity and adaptation in cold tolerance traits, using spiderlings from the core of the range in France, and the edge of the range in Estonia. We combined this with data on clinal variation in adult phenotypes (body size, pigmentation, and fecundity) and genotypes in a transect across the European range. This study revealed a strong signature of genetic adaptation for increased cold tolerance in edge populations, and clear genetic differentiation of ancestral and expanding populations over a very short geographic distance, despite gene flow. We provide genome-wide evidence for adaptive introgression, and conclude that the A. bruennichi range expansion was enabled by adaptive introgression, but has reached a poleward range limit. Interactions with microbes shape all aspects of eukaryotic life. Endosymbiotic bacteria have been shown to alter the thermal tolerance of arthropod hosts, and influence dispersal behavior in spiders. With this background, in Chapter 4, we asked whether the microbiome might play a role in the rapid range expansion of A. bruennichi. We characterized the microbiome in various dissected tissues of spiders from two populations. Although we found no obvious differences between populations or tissues, this study yielded the discovery of a novel, dominant, vertically transmitted symbiont with astoundingly low similarity to all other sequenced bacteria. Since that discovery, we have found evidence of the unknown symbiont in A. bruennichi populations across the Palearctic (unpublished data), making it relatively unlikely to play a role in the range expansion. By studying the establishment and subsequent differentiation of core versus edge populations of A. bruennichi following range expansion, we were able to gain insight into the evolutionary and ecological processes that allowed this species to successfully cope with novel environments. The rapidity with which local adaptation arose in A. bruennichi suggests that evolutionary adaptation to novel environments is possible over short time periods. However, this may only be possible in species with sufficient standing genetic variation, or with genetic variation introduced via admixture, as in A. bruennichi, which has important implications for our understanding of species responses in the face of ongoing global climate change.