TY - JOUR
AU - Poppinga,, Simon
AB - Abstract Background and Aims The endangered aquatic carnivorous waterwheel plant (Aldrovanda vesiculosa) catches prey with 3–5-mm-long underwater snap-traps. Trapping lasts 10–20 ms, which is 10-fold faster than in its famous sister, the terrestrial Venus flytrap (Dionaea muscipula). After successful capture, the trap narrows further and forms a ‘stomach’ for the digestion of prey, the so-called ‘sickle-shaped cavity’. To date, knowledge is very scarce regarding the deformation process during narrowing and consequent functional morphology of the trap. Methods We performed comparative analyses of virtual 3D histology using computed tomography (CT) and conventional 2D histology. For 3D histology we established a contrasting agent-based preparation protocol tailored for delicate underwater plant tissues. Key Results Our analyses reveal new structural insights into the adaptive architecture of the complex A. vesiculosa snap-trap. In particular, we discuss in detail the arrangement of sensitive trigger hairs inside the trap and present actual 3D representations of traps with prey. In addition, we provide trap volume calculations at different narrowing stages. Furthermore, the motile zone close to the trap midrib, which is thought to promote not only the fast trap closure by hydraulics but also the subsequent trap narrowing and trap reopening, is described and discussed for the first time in its entirety. Conclusions Our research contributes to the understanding of a complex, fast and reversible underwater plant movement and supplements preparation protocols for CT analyses of other non-lignified and sensitive plant structures. Carnivorous plant, functional morphology, plant movement, snap-trap, micro-CT INTRODUCTION In contrast to its larger and much more famous phylogenetic sister, the Venus flytrap (Dionaea muscipula), the carnivorous underwater dwelling waterwheel plant (Aldrovanda vesiculosa, Droseraceae) has in many respects been less well investigated scientifically. The most comprehensive works on this endangered aquatic ‘enigma’ date back to Darwin (1875) and, in particular, Ashida (1934). More recent studies have focused on the underlying modes of triggering and actuation of the very fast closure of its tiny snap-traps (Iijima and Sibaoka, 1981, 1983, 1985; Poppinga and Joyeux, 2011; Westermeier et al., 2018), and on prey spectra as well as on predator–prey interactions (Akeret, 1993; Horstmann et al., 2019; Poppinga et al., 2019). However, apart from the benchmark publication of Ashida (1934), the question of how prey, once trapped, is further processed and which configurational trap changes occur remains poorly investigated. Darwin (1875) referred to A. vesiculosa as a ‘miniature, aquatic Dionaea’, but it has become clear recently that the snap-trap mechanics of these monotypic genera are very different (Forterre et al., 2005; Poppinga and Joyeux, 2011; Westermeier et al., 2018; Sachse et al., 2020). As in Dionaea, the A. vesiculosa snap-trap is indeed a highly modified leaf blade. It consists of two trap lobes connected by a midrib, which bends upon triggering of sensitive, filamentous hairs in the trap interior. By kinematic coupling, the lobes become displaced by midrib bending without changing their own curvatures, and the trap closes. Closure is initiated by release of water from cells of the inner epidermis (Iijima and Sibaoka, 1983), which entails the release of stored prestress in the trap (Westermeier et al., 2018). Due to the physiological processes of trap triggering and movement initiation, trap closure duration is temperature-dependent and can be as short as 10 ms in a sufficiently warm regime (Westermeier et al., 2018). The cells, which are thought to be responsible for the turgor change-based initial motion, are located in the so-called motor zone next to the midrib (Ashida, 1934). Triggering occurs by mechanical perturbations from prey, or artificially by mechanical irritation or even by chemical and electrical means (Czaja, 1924; Ashida, 1934). Within a trap lobe, two prominent regions can be distinguished: the central or three-layered region close to the midrib (comprising two epidermal layers and a mesophyll layer) and the marginal region (comprising only two thin layers of epidermal cells, termed one-layered region by Ashida, 1934) towards the periphery (Fig. 1C, D). Within the central region are the trigger hairs (10–20 on each lobe) as well as numerous glands responsible for digestion (Ashida, 1934; Adlassnig et al., 2012). Four to nine traps are radially arranged around the stalk in a whorl-like manner, which explains the common name ‘waterwheel plant’ (Fig. 1A, B). Each trap is surrounded by bristles emerging from the respective petioles. The orientation of a trap relative to its bristles determines the designation of the two lobes. The one facing away from the bristles is termed the free-side lobe, the other one the bristle-side lobe (Ashida, 1934). In contrast to D. muscipula, the trap lobes of A. vesiculosa do not change their curvature during fast trap closure (Fig. 1D, first two images). However, if the trap receives a very strong stimulus or receives multiple stimuli after closure (e.g. by struggling prey), it changes from the closed to a narrowed and then, later on, to a very narrowed state (Fig. 1D, last two images) (terminology sensuAshida, 1934). In the vast majority of the traps, the free-side lobe progressively inverts its curvature in such a manner that the central regions form a tightly closed ‘capsule’ (the so-called ‘sickle-shaped cavity’) and the marginal regions touch each other form-fittingly (Fig. 1D, last two images). Why one lobe shows a distinct curvature inversion during trap narrowing whereas the other one does not has not yet been explained. We initially hypothesized that different lobe radii are responsible for this effect and investigated this more in detail during the present study. Fig. 1. Open in new tabDownload slide Aldrovanda vesiculosa morphology and trap movements. (A) Stereo microscopy image of an A. vesiculosa shoot with whorls of snap-traps. (B) Stereo microscopy image of the first whorl, as seen from the apex, which possesses juvenile, not yet opened narrowed snap-traps. (C) Light microscopy image of the inner, untreated trap surface, showing the transition from the central region with numerous glands to the marginal region. (D) Snap-trap movements. The first image shows the open trap in the ready-to-catch state. After triggering, the trap closes within 28 ms at 13 °C (second image). After further stimulation, the trap changes to the narrowed state (third image, 12 min after closure) and finally reaches the very narrowed state (fourth image, 30 min after closure). The sickle-shaped cavity has formed, where prey is digested. During narrowing, only the free-side lobe inverts its curvature. Abbreviations: cr = central region, br = bristle-side lobe, eb = enclosure boundary, fr = free-side lobe, g = glandular hair, mar = marginal region. Fig. 1. Open in new tabDownload slide Aldrovanda vesiculosa morphology and trap movements. (A) Stereo microscopy image of an A. vesiculosa shoot with whorls of snap-traps. (B) Stereo microscopy image of the first whorl, as seen from the apex, which possesses juvenile, not yet opened narrowed snap-traps. (C) Light microscopy image of the inner, untreated trap surface, showing the transition from the central region with numerous glands to the marginal region. (D) Snap-trap movements. The first image shows the open trap in the ready-to-catch state. After triggering, the trap closes within 28 ms at 13 °C (second image). After further stimulation, the trap changes to the narrowed state (third image, 12 min after closure) and finally reaches the very narrowed state (fourth image, 30 min after closure). The sickle-shaped cavity has formed, where prey is digested. During narrowing, only the free-side lobe inverts its curvature. Abbreviations: cr = central region, br = bristle-side lobe, eb = enclosure boundary, fr = free-side lobe, g = glandular hair, mar = marginal region. Ashida (1934) noted similarities between narrowing and the way new traps develop. Initially, in the juvenile non-opened stage, before the traps attain the open stage for the first time, their shape corresponds to the narrowed stage (Fig. 1B). He also observed that the distances between the midrib and the motor zone differ for each trap lobe with the motor zone being ‘a little larger in the free-side lobe than in the bristle-side lobe’. Stomach formation probably helps the plant in digesting the caught animal by forming a tightly sealed small ‘micro-environment’, thereby increasing the effectiveness of the digestive fluid by preventing dilution. It might also help in suffocating the prey, as known from the suction bladders of carnivorous aquatic Utricularia (Adamec, 2007). The duration of stomach formation is determined by temperature and stimulus strength (Ashida, 1934). Interestingly, the process of lobe inversion is reversible, as the traps are able to reopen again (Ashida, 1934). Czaja (1924) observed opening of traps fed with small prey pieces after 5–6 d, which is comparable to reopening times in the Venus flytrap (Volkov et al., 2011). However, when A. vesiculosa traps are fed with comparably very large prey, they will not open again (Czaja, 1924). As with any other underwater, delicate plant structure, experiments and anatomical investigations on A. vesiculosa snap-traps are challenging to perform because the tissues react sensitively to water loss and to the application of chemicals (Ashida, 1934). Furthermore, they are difficult to handle due to their small sizes, which may for example impede the creation of correctly aligned high-quality transverse sections for anatomical studies. In this context, computed tomography (CT) has received increasing popularity in recent years because it is possible to gain insights into the internal structures of organisms without the need of damaging the sample. Moreover, it allows for fast, digital 3D histological characterization (Ketcham et al., 2001; Rueckel et al., 2014; Hesse et al., 2019). The methodical advantages and resulting importance are reflected in the increasing number of publications on 3D characterization of biological samples in recent years (to name a few covering investigations on plants: Dhondt et al., 2010; Tracy et al., 2010; Blonder et al., 2012; Chomicki et al., 2018; Klepsch et al., 2018; Mathers et al., 2018; Staedler et al., 2018; Bunk et al., 2019; Holmlund et al., 2019; Le et al., 2019; Pierantoni et al., 2019; van den Bulcke et al., 2019; Reich et al., 2020). In the present study, we tested two different approaches as to their applicability to visualize the A. vesiculosa snap-trap anatomy in great detail. First, we apply classical hand-sectioning histology, and second, we perform micro-CT (µCT) analyses on A. vesiculosa traps treated according to different preparation protocols found in the existing literature. The visualization of soft plant tissues via µCT is limited because of the low intrinsic contrast inherent to non-lignified and/or non-mineralized plant organs. Despite recent advantages made in preparation protocols for soft tissues of animals and plants based on staining with contrasting agents (Metscher, 2009; Staedler et al., 2013), the predisposition of A. vesiculosa traps to react to chemicals, especially because of its active trapping mechanism, poses a further challenge. Put simply, most chemical treatments cause trap deformation. Yet, for accurate measurements, as little morphological damage and artefacts as possible are required. Therefore, after having tested the preparation protocols given by Staedler et al. (2013, 2018), we refined them and chose to follow two novel approaches, as described in detail below. Our findings provide novel insights into A. vesiculosa trap functional morphology, both quantitatively by measuring trap lobe radii as reference for morphometrical parameters, and qualitatively by highlighting previously unknown anatomical details. Our refined preparation method for fresh, delicate plant tissues can be easily adopted for related anatomical studies, such as small structures of other aquatic plants or delicate structures of terrestrial plants. MATERIALS AND METHODS Plant material Single, adult traps of tropical Aldrovanda vesiculosa (L.) (originating from Girraween Lagoon, close to Darwin, NT, Australia; cf. Westermeier et al., 2018), which were in the narrowed state, were cut from the plant. Plants were cultivated according to Adamec (1999) outdoors in the Botanic Garden Freiburg during the sampling time in summer. Hand-sectioning and light microscopy For light microscopy analyses, traps were fixed (90 parts 50 % isopropanol and 10 parts 99.5 % glycerine), embedded in Technovit 7100 (Heraeus Kulzer GmbH, Germany) and sectioned using a custom-made sliding microtome at 5-µm thickness. For better visualization, the sections were stained with 10 % toluidine-blue. Light microscopy was performed with a BX61 (Olympus Life Science Corp., Germany), equipped with a DP71 digital camera. In addition, traps were sectioned by hand with a razorblade and directly viewed in a drop of water using the same light microscope. Scanning electron microscopy (SEM) For SEM analysis, traps were fixed in 100 % methanol (Neinhuis and Edelmann, 1996), critical point (CP)-dried using an LPD 030 dryer (Bal-Tec/Leica Mikrosysteme Vertrieb GmbH, Germany), gold-sputtered (Cressington Sputter Coater 108 auto, Germany) and scanned with a 435VP scanning electron microscope (LEO Electron Microscopy, UK). µCT preparation For CT analyses, two approaches were followed. The first involved rapid fixation in 100 % methanol (Neinhuis and Edelmann, 1996), followed by 3 d of infiltration with 1 % phosphotungstic acid (PTA) in 100 % methanol. This procedure is based on the approaches given by Westermeier et al. (2017), who used rapid fixation with methanol for CP drying on another aquatic carnivorous genus (bladderworts, Utricularia spp.) of similar size as Aldrovanda, and by Staedler et al. (2013), who increased µCT visibility of soft plant tissue by infiltration with contrasting agents. Eventually, the traps were CP-dried. Furthermore, to avoid movement of the sample during scanning, the traps were embedded in a drop of epoxy glue (Conrad Electronic AG, Switzerland), which was mounted firmly onto the sample holders (metal stubs). To remove air bubbles from the epoxy, the samples were placed in a desiccator for 60 min before the glue could harden. In the second CT protocol, traps were immediately transferred to the fixation medium FM (60 % ethanol, 17 % glycerin, 23 % aqua dest). After 3 d, the samples were transferred to FM containing 1 % PTA. After infiltration with PTA for 3 d, single traps were washed with FM and transferred to plastic pipette tips, which were filled with FM. The end of the tip had been previously sealed by heating the plastic. The remaining opening was sealed with parafilm. The containers were firmly mounted onto sample holders made of plastic pipette tips and a brass stub. µCT scanning The samples were scanned with the absorption-based SkyScan 1272 desktop scanner (Bruker, USA). The scanning parameters are summarized in Table 1. The scanned data were reconstructed with the software NRecon (Bruker, version 1.7.1.0) and exported to bmp format. For correct positioning in order to identify and measure targeted trap structures, straight alignment of the traps was achieved by using the software DataViewer (Bruker; version 1.5.4.0). The aligned traps could then be digitally cut halfway along the length of the midrib, resulting in a transverse section at the very centre. These 2D sections were analysed and parameters measured using Fiji/ImageJ (version 2.0; Schindelin et al., 2012) and the resulting radii of the free-side and bristle-side lobe were statistically analysed with R Studio (R Development Core Team, 2015). Altogether, ten measurements were tested for normality using the Shapiro–Wilk-test, for homogeneity of variances using the Levene’s test and subsequently tested for statistical differences with a t-test. Table 1. Summary of CT-scanning parameters. Fixation . Infiltration (3 d) . CP drying . Scan medium . Acceleration voltage (kV) . Source current (µA) . Exposure time (ms) . Image pixel size (µm) . Camera binning . 360° scanning . Scan duration (h) . FM FM + 1 % PTA no FM 50 200 650 1.8–2.5 2 × 2 no (except one) 1:22-1:38 100 % methanol 100 % methanol + 1 % PTA yes Epoxy glue 50 200 786 2.0 2 × 2 yes 2:55 Fixation . Infiltration (3 d) . CP drying . Scan medium . Acceleration voltage (kV) . Source current (µA) . Exposure time (ms) . Image pixel size (µm) . Camera binning . 360° scanning . Scan duration (h) . FM FM + 1 % PTA no FM 50 200 650 1.8–2.5 2 × 2 no (except one) 1:22-1:38 100 % methanol 100 % methanol + 1 % PTA yes Epoxy glue 50 200 786 2.0 2 × 2 yes 2:55 Open in new tab Table 1. Summary of CT-scanning parameters. Fixation . Infiltration (3 d) . CP drying . Scan medium . Acceleration voltage (kV) . Source current (µA) . Exposure time (ms) . Image pixel size (µm) . Camera binning . 360° scanning . Scan duration (h) . FM FM + 1 % PTA no FM 50 200 650 1.8–2.5 2 × 2 no (except one) 1:22-1:38 100 % methanol 100 % methanol + 1 % PTA yes Epoxy glue 50 200 786 2.0 2 × 2 yes 2:55 Fixation . Infiltration (3 d) . CP drying . Scan medium . Acceleration voltage (kV) . Source current (µA) . Exposure time (ms) . Image pixel size (µm) . Camera binning . 360° scanning . Scan duration (h) . FM FM + 1 % PTA no FM 50 200 650 1.8–2.5 2 × 2 no (except one) 1:22-1:38 100 % methanol 100 % methanol + 1 % PTA yes Epoxy glue 50 200 786 2.0 2 × 2 yes 2:55 Open in new tab Segmentation, surface generation and volume rendering To visualize the reconstructed scans, the software AVIZO (Thermo Scientific, USA; version 9.2) was used by which, for example, trapped animals and remaining air bubbles could be segmented from the trap. After segmentation, volume-rendered 3D models were computed, from which trap volumes can be calculated. To approximate the volume decrease at different stages of trap narrowing, the trap volume and midrib length of a closed but not yet narrowed trap was measured with AVIZO and taken as reference. In our ecotype, measured trap proportions were very constant: midrib lengths were 1.35 ± 0.05× greater than trap radii (n = 10). The midrib length was defined between the points where the trap lobes merge into the midrib. Trap volumes of narrowed traps were then calculated in relation to this reference trap: according to the respective midrib lengths, their potential trap volumes before narrowing were calculated proportionally to the reference trap, and compared to the actual narrowed trap volumes. RESULTS Preliminary experiments on fixation and scanning approaches according to the protocols of Staedler et al. (2013) revealed the high sensitivity of traps towards chemical treatments. Commonly, traps became deformed upon fixation by acquiring intermediate states of closure, or the two trap lobes showed independent, different deformation behaviour. Altogether, the traps did not maintain their original geometry. Because of the morphological alterations observed in the traps, these treatments were consequently not followed up except for methanol fixation required for CP drying, which allows visualization at great detail while accepting artefacts. Upon methanol fixation traps show certain artefacts in the area of the motor zone, which are explained in detail below. 2D analyses with light microscopy and SEM Our light microscopy thin-section analyses revealed anatomical details on the (sub-)cellular level (Fig. 2A, B). However, sectioning was challenging because the traps often break away from the embedding agent. An undulation of the motor zone as reported by Ashida (1934) is not visible. In contrast, such deformation can be seen in the SEM images (Fig. 2C, D). In fresh hand-sections, structures such as button-shaped glands, trigger hairs and teeth at the trap margins are clearly visible (Fig. 2E, F). As observable in Fig. 2, inside the midrib a central vascular bundle supplies the trap. Moreover, there are numerous button-shaped digestion glands aligned along the midrib inside the trap. Fig. 2. Open in new tabDownload slide Aldrovanda vesiculosa trap morphology. (A) Light microscopy image of a transverse microtome section of a narrowed trap embedded in Technovit. The central and marginal regions, glands and trigger hairs on the inner trap surface, and the midrib are visible. (B) Light microscopy image of a transverse microtome section of the midrib. (C) SEM image of a trap with prominent deformation of the motor zone upon methanol treatment. Numerous vesicles on the outer epidermis of the motor zone are visible (arrowheads). (D) View into the trap interior with SEM. The distribution of glands and trigger hairs in the trap interior can be observed. In the area of the motor zone, the trap is dented to a greater extent on the trap outside (lower epidermis) than on the inside (indicated by arrows, compare to non-dented trap of A). The needle-like structures are filamentous algae. (E) Light micrograph of an untreated fresh hand-section of the inner trap surface. The midrib, glands and trigger hairs can be seen. (F) Light micrograph of a fresh hand-section of the trap lobe margin, where numerous ‘teeth’ are located. (G) Transverse CT reconstruction of a narrowed trap fixated and scanned in FM with 1 % PTA staining. The midrib and central and marginal regions can be seen. The trap does not show deformation resulting from preparation. (H) Transverse CT reconstruction of a narrowed trap fixed in methanol, CP-dried and scanned in epoxy-glue with 1 % PTA staining. The different trap regions, glands and the midrib are visible. The lower epidermis of the motor zone is undulated and vesicles are present. Scale bars: A, G, H = 500 µm; B, E, F = 100 µm; C, D = 200 µm. Abbreviations: cen = central region, g = gland, mar = marginal region, mr = midrib, mz = motor zone, t = tooth, th = trigger hair, v = vesicle. Fig. 2. Open in new tabDownload slide Aldrovanda vesiculosa trap morphology. (A) Light microscopy image of a transverse microtome section of a narrowed trap embedded in Technovit. The central and marginal regions, glands and trigger hairs on the inner trap surface, and the midrib are visible. (B) Light microscopy image of a transverse microtome section of the midrib. (C) SEM image of a trap with prominent deformation of the motor zone upon methanol treatment. Numerous vesicles on the outer epidermis of the motor zone are visible (arrowheads). (D) View into the trap interior with SEM. The distribution of glands and trigger hairs in the trap interior can be observed. In the area of the motor zone, the trap is dented to a greater extent on the trap outside (lower epidermis) than on the inside (indicated by arrows, compare to non-dented trap of A). The needle-like structures are filamentous algae. (E) Light micrograph of an untreated fresh hand-section of the inner trap surface. The midrib, glands and trigger hairs can be seen. (F) Light micrograph of a fresh hand-section of the trap lobe margin, where numerous ‘teeth’ are located. (G) Transverse CT reconstruction of a narrowed trap fixated and scanned in FM with 1 % PTA staining. The midrib and central and marginal regions can be seen. The trap does not show deformation resulting from preparation. (H) Transverse CT reconstruction of a narrowed trap fixed in methanol, CP-dried and scanned in epoxy-glue with 1 % PTA staining. The different trap regions, glands and the midrib are visible. The lower epidermis of the motor zone is undulated and vesicles are present. Scale bars: A, G, H = 500 µm; B, E, F = 100 µm; C, D = 200 µm. Abbreviations: cen = central region, g = gland, mar = marginal region, mr = midrib, mz = motor zone, t = tooth, th = trigger hair, v = vesicle. 3D-CT analyses of liquid-scanned samples At 1.8–2.5-µm voxel size, structures such as large button-shaped glands (Fig. 2G, H) and, occasionally, trigger hairs and teeth at the trap margin can be observed (only in volume rendering mode). There are no observable artefacts resulting from the chemical treatment (i.e. no deformation of the motor zone) (Fig. 2G). Distance measurements of the trap lobe radii at the central trap part return 2.52 ± 0.6 mm on average for the bristle-side lobe and 2.37 ± 0.7 mm on average for the free-side lobe (n = 10). The data are normally distributed, the homogeneity of variance is given and the trap lobe radii are not statistically different (d.f. = 9, P > 0.18). Regarding potential anatomical differences between the free-side and bristle-side lobe, none could be detected either in the classical light microscopy sections or in the CT reconstructions or volume renderings. Figure 3 shows three traps in the narrowed states with different caught prey animals. The very large, half-digested mosquito larva (Fig. 3A; Supplementary Data, Video S1) has largely impeded the trap narrowing process. Here, the free-side lobe is only partially flexed. The trapped female copepod is shorter but thicker than the prey in Fig. 3C (Fig. 3B; Video S2). Moreover, the caudal setae rise beyond the enclosure boundary. The trap is more narrowed than the trap in Fig. 3A. The long and comparably thin Chironomidae larva fits well into the sickle-shaped cavity of the trap (Fig. 3C; Video S3). Fig. 3. Open in new tabDownload slide Lateral and apical views of rendered CT scans of liquid-scanned Aldrovanda vesiculosa traps containing prey, stained with 1 % PTA. (A) Trap with half-digested mosquito larva (Supplementary Data Video S1). (B) Trap with female copepod (Video S2). (C) Trap with a Chironomidae larva (Video S3). Scale bars = 1 mm. Fig. 3. Open in new tabDownload slide Lateral and apical views of rendered CT scans of liquid-scanned Aldrovanda vesiculosa traps containing prey, stained with 1 % PTA. (A) Trap with half-digested mosquito larva (Supplementary Data Video S1). (B) Trap with female copepod (Video S2). (C) Trap with a Chironomidae larva (Video S3). Scale bars = 1 mm. Under the assumption that trap volume is directly proportional to midrib length, measurements from an individual trap showed in the closed but not yet narrowed state a trap volume of 8.7 mm3 at a midrib length of 3.5 mm. The trap in Fig. 3A has only 40.5 % (2.5 mm3) of the theoretical original trap volume (6.2 mm3) at a midrib length of 2.5 mm. The further narrowed trap in Fig. 3B possesses a trap volume of 15.3 % (1.1 mm3) of the theoretical original trap volume (7.2 mm3) at a midrib length of 2.9 mm. Finally, the very narrowed trap in Fig. 3C has a trap volume of 9.3 % (0.6 mm3) of the theoretical original trap volume (6.5 mm3) at a midrib length of 2.6 mm. 3D-CT analyses of CP-dried samples Below a scanning resolution of 2.0 µm, details such as trigger hairs, glands and teeth at the trap margin become visible (clearly visible only in Supplementary Data Video S4 and described in great detail by Lloyd (1942)). Treatment with methanol and contrasting agent leads to deformation of the trap lobes, especially in the one-layered region (see gaps between the lobes in Fig. 4A; Video S4). Moreover, the motor zone becomes markedly deformed (Fig. 4A–C; Video S4), which can be localized via the 3D volume rendering. The motor zone can be found close to the midrib with greatest width at the trap centre (Fig. 4B, C; Video S4). Here, the motor zone becomes dented with the outer trap surface deforming to a greater degree than the trap inner surface (Fig. 4A; Video S4). Furthermore, numerous large vesicles can be seen on the lower outer trap surface (Fig. 4B; Video S4). Moreover, the volume rendering highlights the distribution of the trigger hairs on the inner trap surface. Notably, they are predominantly oriented along the midrib and along the enclosure boundary (i.e. at the transition between the central and marginal region). The motor zone possesses comparatively few trigger hairs and is sparsely covered by button-shaped multicellular glands (Fig. 4C; Video S4). It is impossible to locate distinct boundaries, which would allow the motor zone area to be measured. The midrib is densely covered with button-shaped glands. The teeth at the margin are unicellular trichomes which alternate by size to intercross when the trap is closed, probably preventing prey escape (Lloyd, 1942). Fig. 4. Open in new tabDownload slide Volume rendering of CP-dried Aldrovanda vesiculosa trap, stained with 1 % PTA and scanned in epoxy-glue (Supplementary Data Video S4). (A) Transverse cut showing the deformation of the motor zone with vesiculation (indicated by arrowheads), trigger hairs and glands can be seen. The gap between the lobes is indicative of the treatment with methanol and contrasting agent, which leads to a deformation of the trap lobes. (B) View of the outer surface of a trap with the midrib and the vesicles (see arrowheads; ablation of the outer epidermal cell wall upon methanol dehydration) on the motor zone being clearly visible. (C) View into trap interior in which the glands on the midrib trigger hairs next to the midrib and the deformation of the motor zone can be seen. The motor zone is highlighted by yellow dashed lines. Abbreviations: mr = midrib with glands on the trap inside, th = trigger hair. Scale bars = 500 µm. Fig. 4. Open in new tabDownload slide Volume rendering of CP-dried Aldrovanda vesiculosa trap, stained with 1 % PTA and scanned in epoxy-glue (Supplementary Data Video S4). (A) Transverse cut showing the deformation of the motor zone with vesiculation (indicated by arrowheads), trigger hairs and glands can be seen. The gap between the lobes is indicative of the treatment with methanol and contrasting agent, which leads to a deformation of the trap lobes. (B) View of the outer surface of a trap with the midrib and the vesicles (see arrowheads; ablation of the outer epidermal cell wall upon methanol dehydration) on the motor zone being clearly visible. (C) View into trap interior in which the glands on the midrib trigger hairs next to the midrib and the deformation of the motor zone can be seen. The motor zone is highlighted by yellow dashed lines. Abbreviations: mr = midrib with glands on the trap inside, th = trigger hair. Scale bars = 500 µm. DISCUSSION In this study, we present a refined preparation protocol for the study of soft plant tissues with µCT. We scanned narrowed A. vesiculosa snap-traps in liquid fixation medium because the exact preservation of turgescent tissues cannot be guaranteed with standard preparation methods (CP-drying or freeze-drying), as already mentioned by Korte and Porembski (2011). Other liquid scanning media such as 70 % ethanol or FAA (formalin–acetic acid–alcohol) as used by Staedler et al. (2013) did not show the required combination of preservation of sample morphology, and sufficient contrast between sample and scanning medium. Contrast could be enhanced by staining with 1 % PTA, as shown by Metscher (2009) for animal tissue and Staedler et al. (2013) for plants. The liquid-based approach with FM caused no observable preparation-based morphological changes and, thus, allowed for precise analyses of the trap architecture. This allowed us to calculate trap volumes at different stages of trap narrowing. Depending on the midrib length, closed traps possess trap volumes of 6.2–8.7 mm3 and can theoretically show as little as ~10 % of the original trap volume at the very narrowed stage (Fig. 3A). Such a reduction in trap volume is probably important for efficient digestion of the prey. Preparation with FM and 1 % PTA infiltration may also be used to study other delicate, aquatic plant samples in future approaches, and may be also used for delicate structures of terrestrial plants. Because the image quality is much higher after CP drying and staining with a contrasting agent, we additionally performed scans of CP-dried samples (embedded in a drop of epoxy glue; see Staedler et al., 2013) for visualizing morphological details in greater detail. Such details can only be visualized because (1) the contrast between glue and samples is higher than with FM, and (2) because the sample is maximally stabilized by the glue, thereby minimizing movement artefacts during scanning. However, treatment with methanol for CP-drying leads to deformation artefacts before the samples become embedded in glue. More specifically, with agar-embedded sections, Ashida (1934) observed an ‘undulation of the outer epidermis’ in the region of the motor zone, as the outer epidermis collapses (e.g. bulges inside and the trap becomes dented as seen from outside) and the inner epidermis remains almost unaffected. This typical behaviour is also observable in our CP-dried samples, but in 3D. Until now, it was not possible to visualize the entire motor zone both for the trap inside and outside at the same time and in its entirety (Fig. 4). Remarkably, in the area of the motor zone, next to the deformation of the outer epidermis, vesiculation on the outer trap surface occurs upon methanol fixation (Figs 2C and 4A, B). According to Ashida (1934), this is caused by the ablation of the outermost cell walls due to the great extensibility of the cell walls of the motor zone. It could also be due to tensile prestress that the tissue adjacent to the midrib experiences in the open position and which is suddenly released when the traps close (Westermeier et al., 2018). Sudden turgor change due to methanol fixation in these highly responsive cells certainly affects the cell wall mechanics. Future biomechanical studies could investigate potential differential material properties between the outer and the inner epidermis and the mesophyll, especially before and after trap closure. In our scan we observe similar vesiculation, which makes the motor zone (in combination with its deformation) easily recognizable. Yet, the area boundaries of the motor zone are difficult to locate, and the best defining characteristics remain the vesicles on the trap exterior and the undulation of the outer epidermis. As the vesicles boarder the midrib, we could not observe different distances between the motor zone and the midrib for the two trap lobes, which were reported by Ashida (1934). Notwithstanding this, the cell deformations exclusive to the motor zone indicate its importance for the trapping mechanism in A. vesiculosa. Remarkably, a similar vesiculation behaviour can be seen in epidermal cells of snap-tentacles of the related, terrestrial carnivorous Drosera glanduligera (see fig. 2A in Poppinga et al., 2012). Although not further noted and discussed in their paper, the effect of drying preparation on the tentacle’s motor zone is striking and very similar to that observed here with A. vesiculosa. It would be interesting to similarly investigate and compare the motor zones of other plant structures, for example in Mimosa or Stylidium. Another morphological feature that is clearly visible in scans after CP-drying are the trigger hairs and their distribution within the trap. Trigger hairs are oriented along the midrib and the transition between central and marginal region (the enclosure boundary). This distribution is probably optimized towards (1) maximal freedom of movement of the motor zone, and (2) the chances of triggering by prey, while (3) maintaining relative safety against unnecessary trap closures (e.g. by detritus or water currents). Potentially, those trigger hairs closer to the trap margin (at the enclosure boundary) serve in sensing larger prey with long appendages such as the antennae of daphniids, which have recently been described to potentially be the primary structures by which these animals trigger the traps (Poppinga et al., 2019). In contrast, those trigger hairs parallel to the midrib are possibly advantageous for sensing smaller and fast moving prey, which must move deep enough into the trap to prevent escape during fast trap closure (see movie S15 of a copepod escape in Poppinga et al., 2019). Besides analyses with CT, we also studied trap anatomy using hand-sections, microtome sections and light microscopy. Our study can be used to illustrate the advantages and disadvantages of both methods, i.e. 2D classical hand-sectioning and digital slicing via CT (3D). While the latter allows higher throughput and comparatively easy analysis of the morphology at any desired arbitrary orientation, light microscopy delivers much more detailed histological information at (sub-)cellular level. Aldrovanda possesses different glandular trichomes, which have been described in detail by Lloyd (1942) and Juniper et al. (1989). All approaches revealed the distribution of multicellular button-shaped glands for digestion very well. It is noticeable that the beaded alignment along the midrib in the trap interior coincides with proximity to the central vascular bundle within the midrib. This probably helps in accelerating the process of digestion, both for enzyme secretion and for later absorption of nutrients (Juniper et al., 1989; Adlassnig et al., 2012; Matušíková et al., 2018). The cellular structure of these glandular hairs as well of the trigger hairs can be best studied via SEM and microtome sections rather than CT. Yet, neither with 2D nor with 3D investigation was it possible to find a distinctive morphological feature, which might explain the differing lobe behaviours during the narrowing movement. Our initial hypothesis, i.e. that the bristle-side lobe is longer than the free-side lobe (see Fig. 1D), could not be confirmed statistically. In this context, it would be worth investigating the cells at the transition zones themselves. During ontogeny, the cell wall undergoes modulations which regulate its shape formation and, ultimately, its mechanics. It is conceivable that, as a result of the ‘packaging’ in the plant’s apex from where new leaves emerge, the cells from the bristle-side and free-side lobe are slightly different. Potentially, the cells at the transition zones show heterogeneous distributions of cell wall material properties or variations in wall thickness, for example regarding local variations in pectin composition and cellulose reinforcements (Green, 1962; Chebli and Geitmann, 2017). In the ready-to-catch, open trap the central regions experience tensile prestress, the midrib compressive prestress and the marginal regions no prestress (Westermeier et al., 2018). From the state of equilibrium within the trap, the exact character of prestress cannot be derived for the cells at the transition zone, but these cells probably have their cell wall arrangement according to the direction of forces. Conceivably, in the closed trap the midrib potentially further relaxes during narrowing (see the differing midrib curvature states and resulting trap narrowing in Fig. 3), which results in increasing tensile stresses in the central regions (Westermeier et al., 2018). This mechanical stress would be differently ‘absorbed’ by the cells in the transition zone because of the alternating cell wall material stiffness. Another potential explanation could be the rearrangement of microtubules in the direction of maximal mechanical stress, which is known for simpler cell shapes (Williamson, 1990; Hejnowicz et al., 2000; Hamant et al., 2008). As a result, the cellular architecture would adapt and hence, so too would the global configuration of the trap. Furthermore, it must also be noted that regional biochemical changes could affect the organization of the primary cell wall, resulting in loosening or softening and, eventually, cell shape changes (Zhang et al., 2019). To find evidence for the presence and origin of such biochemical cues, which chemically remodel the cell wall in A. vesiculosa, would be a highly challenging but interesting topic for future investigations. It would also be interesting to perform quantitative mechanophysical studies on the cells of the transition zone which might be a better approach than morphological studies alone. It thus remains enigmatic why the free-side lobe always changes its curvature during the process of narrowing. The traps with different prey animals not only illustrate the opportunistic foraging behaviour of A. vesiculosa (Horstmann et al., 2019), but also show the coupling between midrib bending deformation and the degree of free-side trap lobe curvature inversion and trap volume decrease. The more the midrib is bent, the more the trap is narrowed. Eventually, the volume of the sickle-shaped cavity may account for ~10 % of the original trap volume (Fig. 3A). Note that the volume calculations are based on a single reference trap, under the assumption that trap volumes are correlated with trap lengths. In our Australian ecotype, traps possess very similar proportions, but other Aldrovanda ecotypes might be more plastic and hence trap volumes might deviate. To gain more information on Aldrovanda’s potential ‘gape’, future studies should focus on the variances of minimal (narrowed) and maximal (closed condition immediately after trap closure) trap volumes of Aldrovanda. In Fig. 3A (Supplementary Data Video S1), the trapped prey is larger than the central region and prevents full narrowing so that the free-side lobe is only partially flexed. This can also be speculated for Fig. 3B where potentially the caudal setae of the copepod which protrude from the sickle-shaped cavity and rise above the enclosure boundary might prevent full trap narrowing. Yet, the question remains how an animal larger than the sickle-shaped cavity (as is the case with the mosquito larva, Fig. 3A) could be entrapped in its entity by the trap (cf. Akeret, 1993; Horstmann et al., 2019; Poppinga et al., 2019). It remains speculative whether the prey animal was successively and passively moved into the trap, or if its own movements navigated it deeper inside. Answering these questions will be an interesting topic for future studies on the narrowing behaviour of A. vesiculosa with respect to predator–prey interactions. Altogether, the present study contributes to our understanding of A. vesiculosa trap morphology and provides an updated protocol for handling of delicate, aquatic plant material. Busse et al. (2018) had already remarked that ‘three-dimensional histology of soft-tissue samples has proven to provide crucial benefits for the understanding of tissue structure.’ For A. vesiculosa, this definitely holds true. SUPPLEMENTARY DATA Supplementary data are available online at https://academic.oup.com/aob and consist of the following. Video S1: Volume rendering of liquid-scanned A. vesiculosa containing a half-digested mosquito larva. Video S2: Volume rendering of liquid-scanned A. vesiculosa containing a female copepod. Video S3: Volume rendering of liquid-scanned A. vesiculosa containing a Chironomidae larva. Video S4: Volume rendering of critical-point dried A. vesiculosa, stained with 1 % PTA and scanned in epoxy glue. ACKNOWLEDGMENTS A.W. thanks Sandra Caliaro for help with microtome-sectioning.T.S. and S.P. initiated and supervised the research project. A.W. designed the study, performed the experiments and data analyses, wrote the first draft of the manuscript and supervised the preliminary experiments on preparation protocols by N.H. All authors added to the study design, and wrote and revised the manuscript. FUNDING A.W. and T.S. thank the German Research Foundation (Deutsche Forschungsgemeinschaft DFG) for funding within the Transregional Collaborative Research Centre (SFB/Transregio) 141 ‘Biological Design and Integrative Structures’/project A04. A.W. and T.S. are further grateful for funding within the collaborative project ‘Bio-inspirierte Materialsysteme and Verbundkomponenten für nachhaltiges Bauen im 21ten Jahrhundert’ (BioElast), which is part of the ‘Zukunftsoffensive IV Innovation und Exzellenz – Aufbau und Stärkung der Forschungsinfrastruktur im Bereich der Mikro- und Nanotechnologie sowie der neuen Materialien’, funded by the State Ministry of Baden-Wuerttemberg for Sciences, Research and Arts. S.P. gratefully acknowledges the financial support by BASF SE, Ludwigshafen, Germany, and the Ministry of Science Research and Arts of the State of Baden-Württemberg, Germany, who supported this research within the frame of the Academic Research Alliance JONAS (‘Joint Research Network on Advanced Materials and Systems’) established jointly with BASF SE and the University of Freiburg, Germany. Literature cited Adamec L . 1999 . The biology and cultivation of red Australian Aldrovanda vesiculosa . Carnivorous Plant Newsletter 28 : 128 – 132 . Google Scholar OpenURL Placeholder Text WorldCat Adamec L . 2007 . Oxygen concentrations inside the traps of the carnivorous plants Utricularia and Genlisea (Lentibulariaceae) . Annals of Botany 100 : 849 – 856 . Google Scholar Crossref Search ADS PubMed WorldCat Adlassnig W , Koller-Peroutka M, Bauer S, Koshkin E, Lendl T, Lichtscheidl IK. 2012 . Endocytotic uptake of nutrients in carnivorous plants . The Plant Journal 71 : 303 – 313 . Google Scholar Crossref Search ADS PubMed WorldCat Akeret B . 1993 . Ein neuer Fundort von Aldrovanda vesiculosa L. in der Nordschweiz und einige Bemerkungen zu Stratiotes aloides L . Botanica Helvetica 103 : 193 – 199 . Google Scholar OpenURL Placeholder Text WorldCat Ashida J . 1934 . Studies on the leaf movement of Aldrovanda vesiculosa L . Memoirs of the College of Sciience, University of Kyoto, Series B 9 : 141 – 244 . Google Scholar OpenURL Placeholder Text WorldCat Blonder B , De Carlo F, Moore J, Rivers M, Enquist BJ. 2012 . X-ray imaging of leaf venation networks . The New Phytologist 196 : 1274 – 1282 . Google Scholar Crossref Search ADS PubMed WorldCat Bunk K , Krassovitski S, Speck T, Masselter T. 2019 . Branching morphology and biomechanics of ivy (Hedera helix) stem-branch attachments . American Journal of Botany 106 : 1143 – 1155 . Google Scholar Crossref Search ADS PubMed WorldCat Busse M , Müller M, Kimm MA, et al. 2018 . Three-dimensional virtual histology enabled through cytoplasm-specific X-ray stain for microscopic and nanoscopic computed tomography . Proceedings of the National Academy of Sciences of the United States of America 115 : 2293 – 2298 . Google Scholar Crossref Search ADS PubMed WorldCat Chebli Y , Geitmann A. 2017 . Cellular growth in plants requires regulation of cell wall biochemistry . Current Opinion in Cell Biology 44 : 28 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat Chomicki G , Staedler YM, Bidel LPR, Jay-Allemand C, Schönenberger J, Renner SS. 2018 . Deciphering the complex architecture of an herb using micro-computed x-ray tomography, with an illustrated discussion on architectural diversity of herbs . Botanical Journal of the Linnean Society 186 : 145 – 157 . Google Scholar Crossref Search ADS WorldCat Czaja AT . 1924 . Reizphysiologische Untersuchungen an Aldrovandia vesiculosa L . Pflügers Archiv für die gesamte Physiologie des Menschen und der Tiere 206 : 635 – 658 . Google Scholar Crossref Search ADS WorldCat Darwin , C . 1875 . Insectivorous plants . London : John Murray . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dhondt S , Vanhaeren H, Van Loo D, Cnudde V, Inzé D. 2010 . Plant structure visualization by high-resolution x-ray computed tomography . Trends in Plant Science 15 : 419 – 422 . Google Scholar Crossref Search ADS PubMed WorldCat Forterre Y , Skotheim JM, Dumais J, Mahadevan L. 2005 . How the venus flytrap snaps . Nature 433 : 421 – 425 . Google Scholar Crossref Search ADS PubMed WorldCat Green PB . 1962 . Mechanism for plant cellular morphogenesis . Science 138 : 1404 – 1405 . Google Scholar Crossref Search ADS PubMed WorldCat Hamant O , Heisler MG, Jönsson H, et al. 2008 . Developmental patterning by mechanical signals in Arabidopsis . Science (New York, N.Y.) 322 : 1650 – 1655 . Google Scholar Crossref Search ADS PubMed WorldCat Hejnowicz Z , Rusin A, Rusin T. 2000 . Tensile tissue stress affects the orientation of cortical microtubules in the epidermis of sunflower hypocotyl . Journal of Plant Growth Regulation 19 : 31 – 44 . Google Scholar Crossref Search ADS PubMed WorldCat Hesse L , Bunk K, Leupold J, Speck T, Masselter T. 2019 . Structural and functional imaging of large and opaque plant specimens . Journal of Experimental Botany 70 : 3659 – 3678 . Google Scholar Crossref Search ADS PubMed WorldCat Holmlund HI , Pratt RB, Jacobsen AL, Davis SD, Pittermann J. 2019 . High-resolution computed tomography reveals dynamics of desiccation and rehydration in fern petioles of a desiccation-tolerant fern . The New Phytologist 224 : 97 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Horstmann M , Heier L, Kruppert Set al. ., 2019 . Comparative prey spectra analyses on the endangered aquatic carnivorous waterwheel plant (Aldrovanda vesiculosa, Droseraceae) at several naturalized microsites in the Czech Republic and Germany . Integrative Organismal Biology 1 : oby012 . Google Scholar Crossref Search ADS WorldCat Iijima T , Sibaoka T. 1981 . Action potential in the trap-lobes of Aldrovanda vesiculosa . Plant Cell Physiology 22 : 1595 – 1601 . Google Scholar Crossref Search ADS WorldCat Iijima T , Sibaoka T. 1983 . Movements of K+ during shutting and opening of the trap-lobes in Aldrovanda vesiculosa . Plant Cell Physiology 24 : 51 – 60 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Iijima T , Sibaoka T. 1985 . Membrane potentials in excitable cells of Aldrovanda vesiculosa trap-lobes . Plant Cell Physiology 26 : 1 – 13 . Google Scholar OpenURL Placeholder Text WorldCat Juniper BE , Robins RJ, Joel DM. 1989 . The carnivorous plants. London : Academic Press Ltd . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ketcham RA , Carlson WD. 2001 . Acquisition, optimization and interpretation of x-ray computed tomographic imagery: applications to the geosciences . Computers & Geosciences 27 : 381 – 400 . Google Scholar Crossref Search ADS WorldCat Klepsch M , Zhang Y, Kotowska MM, et al. 2018 . Is xylem of angiosperm leaves less resistant to embolism than branches? Insights from microCT, hydraulics, and anatomy . Journal of Experimental Botany 69 : 5611 – 5623 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Korte N , Porembski S. 2011 . Anatomical analysis of turgescent and semi-dry resurrection plants . Microscopy Research and Techiques 74 : 364 – 369 . Google Scholar Crossref Search ADS WorldCat Le TDQ , Alvarado C, Girousse C, Legland D, Chateigner-Boutin AL. 2019 . Use of x-ray micro computed tomography imaging to analyze the morphology of wheat grain through its development . Plant Methods 15 : 84 . Google Scholar Crossref Search ADS PubMed WorldCat Lloyd FE . 1942 . The carnivorous plants. Waltham : Chronica Botanica . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mathers AW , Hepworth C, Baillie AL, et al. 2018 . Investigating the microstructure of plant leaves in 3D with lab-based x-ray computed tomography . Plant Methods 14 : 99 . Google Scholar Crossref Search ADS PubMed WorldCat Matušíková I , Pavlovič A, Renner T. 2018 . Biochemistry of prey digestion and nutrient absorption. In: Ellison AM, Adamec L. eds. Carnivorous plants. Physiology, ecology, and evolution. Oxford : Oxford University Press , 207 – 220 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Metscher BD . 2009 . MicroCT for developmental biology: a versatile tool for high-contrast 3D imaging at histological resolutions . Developmental Dynamics 238 : 632 – 640 . Google Scholar Crossref Search ADS PubMed WorldCat Neinhuis C , Edelmann HG. 1996 . Methanol as a rapid fixative for the investigation of plant surfaces by SEM . Journal of Microscopy 184 : 14 – 16 . Google Scholar Crossref Search ADS WorldCat Pierantoni M , Brumfeld V, Addadi L, Weiner S. 2019 . A 3D study of the relationship between leaf vein structure and mechanical function . Acta Biomaterialia 88 : 111 – 119 . Google Scholar Crossref Search ADS PubMed WorldCat Poppinga S , Hartmeyer SR, Seidel R, Masselter T, Hartmeyer I, Speck T. 2012 . Catapulting tentacles in a sticky carnivorous plant . PLoS ONE 7 : e45735 . Google Scholar Crossref Search ADS PubMed WorldCat Poppinga S , Joyeux M. 2011 . Different mechanics of snap-trapping in the two closely related carnivorous plants Dionaea muscipula and Aldrovanda vesiculosa . Physical Review. E, Statistical, Nonlinear, and Soft Matter Physics 84 : 041928 . Google Scholar Crossref Search ADS PubMed WorldCat Poppinga S , Smaij J, Westermeier AS, et al. 2019 . Prey capture analyses in the carnivorous aquatic waterwheel plant (Aldrovanda vesiculosa L., Droseraceae) . Scientific Reports 9 : 18590 . Google Scholar Crossref Search ADS PubMed WorldCat R Development Core Team 2015 . R: a language and environment for statistical computing. Vienna : R Foundation for Statistical Computing . http://www.R-project.org. Reich D , Berger A, von Balthazar M, et al. 2020 . Modularity and evolution of flower shape: the role of function, development, and spandrels in Erica . The New Phytologist 226 : 267 – 280 . Google Scholar Crossref Search ADS PubMed WorldCat Rueckel J , Stockmar M, Herzen J. 2014 . Spatial resolution characterization of a X-ray microCT system . Applied Radiation and Isotopes 94 : 230 – 234 . Google Scholar Crossref Search ADS PubMed WorldCat Sachse R , Westermeier A, Mylo M, et al. 2020 . Snapping mechanics of the Venus flytrap (Dionaea muscipula) . Proceedings of the National Academy of Sciences of the United States of America 117 : 16035 – 16042 . Google Scholar Crossref Search ADS PubMed WorldCat Schindelin J , Arganda-Carreras I, Frise E, et al. 2012 . Fiji: an open-source platform for biological-image analysis . Nature Methods 9 : 676 – 682 . Google Scholar Crossref Search ADS PubMed WorldCat Staedler YM , Kreisberger T, Manafzadeh S, et al. 2018 . Novel computed tomography-based tools reliably quantify plant reproductive investment . Journal of Experimental Botany 69 : 525 – 535 . Google Scholar Crossref Search ADS PubMed WorldCat Staedler YM , Masson D, Schönenberger J. 2013 . Plant tissues in 3D via x-ray tomography: simple contrasting methods allow high resolution imaging . PLoS One 8 : e75295 . Google Scholar Crossref Search ADS PubMed WorldCat Tracy SR , Roberts JA, Black CR, McNeill A, Davidson R, Mooney SJ. 2010 . The X-factor: visualizing undisturbed root architecture in soils using X-ray computed tomography . Journal of Experimental Botany 61 : 311 – 313 . Google Scholar Crossref Search ADS PubMed WorldCat Van den Bulcke J , Boone MA, Dhaene J, et al. 2019 . Advanced x-ray CT scanning can boost tree ring research for earth system sciences . Annals of Botany 124 : 837 – 847 . Google Scholar Crossref Search ADS PubMed WorldCat Volkov AG , Pinnock MR, Lowe DC, Gay MS, Markin VS. 2011 . Complete hunting cycle of Dionaea muscipula: consecutive steps and their electrical properties . Journal of Plant Physiology 168 : 109 – 120 . Google Scholar Crossref Search ADS PubMed WorldCat Westermeier AS , Fleischmann A, Müller K, et al. 2017 . Trap diversity and character evolution in carnivorous bladderworts (Utricularia, Lentibulariaceae) . Scientific Reports 7 : 12052 . Google Scholar Crossref Search ADS PubMed WorldCat Westermeier AS , Sachse R, Poppinga Set al. ., 2018 . How the carnivorous waterwheel plant (Aldrovanda vesiculosa) snaps . Proceedings of the Royal Society B 285 : 20180012 . Google Scholar Crossref Search ADS PubMed WorldCat Williamson R . 1990 . Alignment of cortical microtubules by anisotropic wall stresses . Australian Journal of Plant Physiology 17 : 601 – 613 . Google Scholar OpenURL Placeholder Text WorldCat Zhang T , Tang H, Vavylonis D, Cosgrove DJ. 2019 . Disentangling loosening from softening: insights into primary cell wall structure . The Plant Journal 100 : 1101 – 1117 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Functional–morphological analyses of the delicate snap-traps of the aquatic carnivorous waterwheel plant (Aldrovanda vesiculosa) with 2D and 3D imaging techniques
JF - Annals of Botany
DO - 10.1093/aob/mcaa135
DA - 2020-10-30
UR - https://www.deepdyve.com/lp/oxford-university-press/functional-morphological-analyses-of-the-delicate-snap-traps-of-the-x9Bz3AUt3j
SP - 1099
EP - 1107
VL - 126
IS - 6
DP - DeepDyve
ER -