Capturing variation in floral shape: a virtual3D based morphospace for Pelargonium

Abstract

Background Variation in floral shapes has long fascinated biologists and its modelling enables testing of evolutionary hypotheses. Recent comparative studies that explore floral shape have largely ignored 3D floral shape. We propose quantifying floral shape by using geometric morphometrics on a virtual3D model reconstructed from 2D photographical data and demonstrate its performance in capturing shape variation. Methods This approach offers unique benefits to complement established imaging techniques (i) by enabling adequate coverage of the potential morphospace of large and diverse flowering-plant clades; (ii) by circumventing asynchronicity in anthesis of different floral parts; and (iii) by incorporating variation in copy number of floral organs within structures. We demonstrate our approach by analysing 90 florally-diverse species of the Southern African genus Pelargonium (Geraniaceae). We quantify Pelargonium floral shapes using 117 landmarks and show similarities in reconstructed morphospaces for nectar tube, corolla (2D datasets), and a combined virtual3D dataset. Results Our results indicate that Pelargonium species differ in floral shape, which can also vary extensively within a species. PCA results of the reconstructed virtual3D floral models are highly congruent with the separate 2D morphospaces, indicating it is an accurate, virtual, representation of floral shape. Through our approach, we find that adding the third dimension to the data is crucial to accurately interpret the manner of, as well as levels of, shape variation in flowers.

Variation in flower size and shape of Impatiens capensis is correlated with urbanization in Montreal, Canada

“…We used a virtual 3D geometric morphometric approach (van de Kerke et al., 2020 ) to analyze simultaneously the front and profile shapes of the flowers. There was much more variation involved in the positioning of the landmarks for the front shapes (15.4%) compared to the profile shapes (3.2%), suggesting that it is more difficult to quantify the front shapes.…”

“…To analyze the frontal and profile shapes simultaneously and thus have a better idea of floral variation in 3D, the mean 2D coordinates of each individual, front and profile, were combined in a single 3D array, keeping the two sets of coordinates orthogonal (van de Kerke et al., 2020 ) to analyze both views simultaneously. To combine the datasets, a uniform z coordinate was added to all pictures in profile view (i.e.…”

Using geometric morphometrics to determine the “fittest” floral shape: A case study in large‐flowered, buzz‐pollinated Meriania hernandoi

“…Accordingly, most studies employing geometric morphometric approaches in flowers have used a 2D approach based on herbarium specimens and photos and only focused on subsets of floral organs (i.e., corolla, style; Chen et al, 2018;Rozov et al, 2018;Kriebel et al, 2020), impeding our ability to understand the joint functioning of floral organs. Recent methodological advances in the acquisition of 3D flower models, such as the development of protocols specifically tailored for preparing flowers for CT scanning (Staedler et al, 2013(Staedler et al, , 2018, (affordable) photogrammetry (Leménager et al, 2022), or virtual 3D modeling (van de Kerke et al, 2020), hold great potential for advancing the field (Figure 5). All approaches can accurately recover 3D shape aspects of externally visible organs (i.e., often the corolla), while CT scanning holds the advantage of also capturing internal organs (i.e., stamens hidden in corolla tube, nectaries at floral base).…”

“…(1) Any study will start with pollination experiments to quantify aspects of pollinator behavior and morphological fit between flower and pollinator. For flowers with pollen dispensed openly (which would be washed out in ethanol [etOH]; most flowers with longitudinal anther dehiscence; example flower Melittis melissophyllum, Lamiaceae; © Leon Schönenberger), the next step after the pollination experiment is (2A) to immediately obtain 3D flower models without chemical fixation of the flower (i.e., through recent advances in photogrammetry [Leménager et al, 2022] or virtual 3D modelling [van de Kerke et al, 2020]) and subsequently (3A) to collect reproductive organs separately to estimate how much pollen remains in anthers (male performance) and how much pollen has been deposited on stigmas (female performance). In flowers that do not openly dispense pollen (e.g., orchid flowers with pollinia; flowers with poricidal anthers, example flower Meriania hernandoi, Melastomataceae), flowers may be readily collected in ethanol in the field (2B) to later generate 3D models (3B) using, e.g., photogrammetry or CT scans (Staedler et al, 2013(Staedler et al, , 2018, and subsequent removal of reproductive organs (3C).…”

“…We further explored whether pollen is removed evenly across stamens or whether bees preferentially buzz lateral or central stamens. While we cannot fully distinguish between age-and shaperelated effects on pollination performance, our study provides a powerful template for linking state-of-the-art 3D geometric morphometrics, hitherto mostly used in macroevolutionary studies of flower evolution (Dellinger et al, 2019c;Kerke et al 2020;Artuso et al, 2021), with measures of pollen transfer. We hence conclude this article by discussing the advantages of using 3D geometric morphometrics and provide a workflow for this refined approach of studying what aspects of the floral phenotype influence pollination performance.…”