Plants have (re)adapted independently to saline sites several times. We have begun to construct phylogenetic trees for several salt marsh species to determine their closest relatives and to determine intraspecific variation. In the process, we have repeatedly noticed that species within the salt marshes are variable, in particular possessing different numbers of chromosome sets (polyploidy). The aim of this project is to create a phylogenetic tree for further species of the North German salt marshes (from the genera Triglochin maritimum, Elymus spec., Silene flos-cocculli, Lotus corniculatus/L.tenuis, Potentilla anserina, Centaurium spp., Tripleurospermum maritimum/T. inodorum) and to use it for the analysis of different questions of evolution. Where do the northern German salt marsh species come from? Are the closest relatives also already adapted to salt? What characteristics do the salt marsh species have in common that they did not get from their ancestors? These questions will be investigated using a variety of different methods.

Methods: Flow cytometry, microscopy (root anatomy), computer based phylogenetic and comparative analyses.

Flavonoids are a widely distributed class of compounds in the genus Veronica. In the last 50 years a number of species have been studied, so it is known that mainly flavones such as luteoline and apigenine occur, which are related to oxidative stress but also antibacterial activity of plants. In addition, anthocyanins are present, especially delphinides but also cyanides, which are responsible for the blue to red coloration of flowers. Comparison of the data with phylogenetic analyses of recent years has shown that there are some substances that are characteristic of certain groups within the genus Veronica. For example, flowers of the subsection Cymbalaria are white, suggesting an absence of anthocyanins. Cyanidins have only been recorded in the New Zealand section Hebe. The subgenus Pseudolysimachium differs from all other subgenera in the production of spicosides, 6-OH-luteolin acylated with phenolic acids such as caffeic acid. Finally, the relationship of the morphologically very different subgenera Pocilla and Pentasepalae was supported by the presence of 8-0H-flavones exclusively in these two subgenera. Genomic and transcriptomic data, now available for many of these groups, may now allow us to find the genetic basis for this variation in flavonoid presence.

Methods: Analysis and comparison of genomic and transcriptomic data.

Veronica subgenus Beccabunga is a group of species that can be found adjacent to or inside bodies of water, like rivers and lakes. Species delimitation is challenging because plants can look very different if they grow in water or on dry soil. Thus, we need a DNA based analysis to disentangle useful characters from those having evolved in parallel or phenotypically plastic ones. We aim to elucidate the phylogenetic relationships in the subgenus, as well as population structure and hybridization patterns by analyzing High-Throughput Sequencing (HTS) data. In addition, we want to review herbarium specimens in order to find morphological characters that may help us in delimiting closely related species.

Methods: DNA extraction, Nanopore sequencing, genotyping-by-sequencing, bioinformatics analysis methods, microscopy

n a previous master's thesis, we found that Veronica spicata and V. longifolia often form hybrids with each other, and we also performed this hybridization with different origins in the greenhouse. The hybrids have now germinated and grown. This raises the question of how the hybrid differs in relation to the parents. For this purpose, DNA sequencing should first be used to prove that they are hybrids. After that, morphologically the hybrid should be described in relation to the parents. Since the parents differ very much in location, V. longifolia is in marshy meadows, V. spicata in dry grasslands, it should be investigated how well the hybrid tolerates drought.

Methods: DNA sequencing, morphological description, greenhouse experiment.

Many species of Veronica are self-incompatible, i.e. there are gene products that allow plants to recognize pollen from their own plant (gametophytic self-incompatibility). Recent molecular biology methods, especially transcriptomics/RNA sequencing, now allow to identify the corresponding genes. Within the Plantaginaceae this has so far only been done for Antirrhinum. In this project, the aim is to sequence RNA from the pistil for V. filiformis, V. spicata and V. chamaedrys and to identify the corresponding homologous genes for Veronica.

Methods: RNA sequencing

Central Europe has virtually no new plant species of its own, because almost all species had to migrate post-glacial. An exception is the small scabious, which presumably forms an own, purely Central European species as a post-glacial hybrid. Resynthesis from the presumed parents will be attempted with crossing experiments. Experiments on the germination behavior of the seeds in climate chambers and in the field will shed light on possible reasons for the success of the new species. Furthermore, a molecular marker will be established that could provide information on the place of origin and migration history in future analyses. Create F1 hybrids, survey 3 species (ca. 30 ind.), seed 200, germination behavior qualitative.

Methods: germination experiments, morphological studies, DNA sequencing.

Rhododendron (Ericaceae) is the largest genus of woody plants (about 1200 species), with a complex and highly debated taxonomy. About 98% of the rhododendron’s species are confined to the old world except most of the beautiful Azaleas and few non- Azalean rhododendrons. Among the non-Azalean rhododendrons, Rhododendron subsect. Caroliniana has a very confused taxonomy and interesting ecology with three species. Recently, Bauer and Albach reported a new species, which led to a change in the total number of species in R. subsection Carolianiana from three to four. The newly reported species is from the Great Smoky mountains named as R. smokianum Ralf Bauer & Albach. We are interested to investigate the genetic cohesion/diversity based on the genomic data in this four species complex and whether there may actually even be more unrecognized species in the group.

Methods: GBS, phylogenomic and population genetics.

Urban gardens are nowadays a significant part of nature and thus of insect habitat. Thus, planted ornamental plants are an important resource for pollen- and nectar-collecting insects. Ornamental plants are selected and bred for attractiveness to humans. However, breeding often results in a decrease in pollen and nectar. However, this has rarely been studied thoroughly. The goal of this project is to investigate this for 1-3 species (e.g. Veronica spicata, Rhododendron spec., Geranium wallichianum). The project involves measuring nectar and pollen levels and observing insects on these plants to investigate the hypothesis that ornamental plants provide fewer resources for insects than native plants.

Methods: Nectar measurement, pollen counting, observation of flower visitors.

Flowers are necessary for sexual reproduction of plants. The longer they are open, the higher the chance of pollination and successful reproduction. However, keeping flowers open also imposes costs on the plant through increased transpiration and water loss. Therefore, there is a high variation in plants in how long flowers are open. Little is known about what influences the flowering length of a single flower. This will be investigated by studying as many species of the genus Veronica as possible. Is there variation among closely related species? Do species with longer open flowers have a higher probability of being pollinated?

Methods: Flower observation, comparative computer analysis in a phylogenetic-comparative framework.

Insects and therefore most pollinators perceive other colors than we do, because their eyes are additionally sensitive for ultraviolet radiation. So it is no wonder that many flowers display UV-signals, that are not visible for us. The shape and look of these signals, their occurrence and distribution among plant families or genera and some other factors are not known very well. The candidate will mainly systematically screen as many flowers in the Botanical Garden as possible for a potential UV-signal.

Method: comparison of self-made normal and UV-photographs.

Veronica hederifolia and V. sublobata are two closely related species that differ in ploidy (number of chromosome sets), but nevertheless occur directly next to each other also in Oldenburg. In previous studies it was determined which characteristics are suitable to distinguish the species and in which they do not differ. In the case of the characteristics, however, it is not clear what influence the environment has. For example, flowers are different when the plants are next to each other, but are the flowers of V. hederifolia different only in the sun, but also pale in the shade. Or are leaves the same when the species are next to each other in shade, but different in sun.

Methods: Growth experiments in the greenhouse, flow cytometry if necessary.

Geometric morphometrics allows the quantitative study of leaf shapes. However, these methods work best with specimens that have been collected and photographed specifically for this purpose, while preparation of data and extraction of leaf outlines from herbarium leaves may require extensive manual processing of photographs. A few pipelines have been proposed to extract leaves from photographs of herbarium leaves using object recognition algorithms. One of these pipelines is ginjinn2. Your task will be to test the usefulness of ginjinn2 for extracting leaf outlines from herbarium leaves of different speedwell species (Veronica, Plantaginaceae) and to analyze the resulting leaf outlines using geometric morphometric methods (Elliptic Fourier Descriptors). In particular, we are looking for differences in leaf shape between populations from different geographical areas.

Methods: Morphological images, computer analyses

Salt marsh plants are adapted to a location that is extreme for plants. Physiological characteristics are mostly in the foreground. Morphological characteristics are less known. This project will investigate some hypotheses about the difference between salt marsh species and their closest relatives. These include stomatal size and density (see Li & al. 1996; Maherali & al. 2009), epidermis thickness, and xylem cell size and xylem cell wall thickness (see Nassar & al. 2008, Maherali & al. 2009, Hao & al. 2013).

Methods: morphological and anatomical examination on microscope, staining and preparation of microscopic objects, collecting plants in nature, if necessary for M.Sc. thesis analysis of traits in phylogenetic comparative analysis. Experiments can also be carried out in the Botanical Garden to study the phenotypic plasticity of the traits.

The herbarium in Oldenburg contains about 20,000 specimens of dried plants of many different species from many different countries. But what else was actually dried? We are sure that partly plants were dried that contain pathogens or symbionts like fungi or insects. The aim of the project is to search the herbarium and to try to find and identify these "by-catches". Are there specific fungi or insects? The search should begin systematically with a genus of plants.

Methods: microscopy, DNA sequencing

Veronica has about 460 species, of which there are currently IUCN assessments for 19 species only, although there are still a number of endemic species that are studied for the country by which they occur (especially New Zealand). However, the lack of endangerment classification prevents effective protection of the species. The project will investigate how well an automated assessment compares to expert knowledge. Based on distribution data from GBIF, inaturalist and targeted additional information (e.g. Herbar Oldenburg), species will be classified into endangerment classes (a priori) using the IUCNN program (Zizka et al. 2022) and deep learning models. The results will be discussed with experts and problems identified. These problems could be taxonomic issues, common confounders, or historical changes in distribution and abundance. The distribution data will then be improved and the analyses performed again, resulting in a second classification (a posteriori).

Methods: distribution range modeling, conservation assessment methods.