Functional trait interactions drive seed buoyancy and dispersal strategies in Echinocystis lobata

301 Functional trait interactions drive seed buoyancy and dispersal strategies in Echinocystis lobata Adam Szulc1, Patryk Czortek2, Pawel Marciniuk3 1 DoctoralSchool,UniversityofSiedlce,ul.Żytnia39,08-110Siedlce,Poland 2 FacultyofBiology,BiałowieżaGeobotanicalStation,UniversityofWarsaw,Sportowa19,17–230Białowieża,Poland 3 InstituteofBiologicalSciences,UniversityofSiedlce,ul.BolesławaPrusa14,08-110Siedlce,Poland Correspondingauthor:AdamSzulc([email protected]) Copyright: © Adam Szulc et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International – CC BY 4.0). Research Article Abstract Hydrochory (water dispersal) is a major driver of plant invasions in riparian landscapes but both the diaspore traits that determine flotation and the contribution of water-mediated transport to the colonization success of non-native species remain insufficiently understood. Echinocystis lobata, an alien vine widely naturalized along European rivers, provides a suitable model for testing how seed morphology governs buoyancy and, consequently, hydrochorous dispersal potential. The main aim of this study was to examine how functional traits influence the buoyancy of E. lobata seeds, which can be classified as normal seeds, freely released from fruits, and trapped seeds, retained within fruits. We investigated three hypotheses under controlled laboratory conditions: (H1) seed mass, particularly in interaction with thickness, determines sinking probability; (H2) surface area affects flotation mainly through interactions with mass; and (H3) trapped seeds differ functionally from normally released seeds. Using a generalized linear mixed-effects model, we demonstrated that buoyancy is governed by non-linear interactions among mass, thickness, and surface area, with distinct patterns for trapped versus normal seeds. Thin normal seeds floated longer as mass increased, whereas in trapped seeds higher mass accelerated sinking. Surface area influenced flotation only when combined with mass and trapping status. Normal seeds remained buoyant for up to 14 days, while trapped seeds sank sooner (≤11 days). These findings highlight that variability in diaspore traits generates a broad spectrum of dispersal outcomes, supporting both local deposition and long-distance hydrochoric transport. Understanding how trait-vector interactions shape dispersal enhances invasion risk prediction and informs management strategies, such as removing reproductive plants near waterways before fruit maturation. Overall, our results demonstrate that hydrochory, amplified by diaspore heteromorphism, may be a key driver increasing the invasiveness of E. lobata in European river valleys. Key words: Hydrochory, invasive alien plants, nautohydrochory, riparian ecosystems, seed dimorphism Introduction Understanding how water-mediated dispersal shapes the spread of non-native plants provides essential insight into the mechanisms facilitating rapid range spread of many herbaceous non-native species (Boedeltje et al. 2004; van den Broek et al. 2005; Jansson et al. 2005; Carthey et al. 2016). Invasive alien plants are particularly successful because they often combine multiple dispersal Academic editor: Ingo Kowarik Received: 20 June 2025 Accepted: 19 November 2025 Published: 11 December 2025 Citation: Szulc A, Czortek P, Marciniuk P (2025) Functional trait interactions drive seed buoyancy and dispersal strategies in Echinocystis lobata. NeoBiota 104: 301–318. https://doi. org/10.3897/neobiota.104.162267 NeoBiota 104: 301–318 (2025) DOI: 10.3897/neobiota.104.162267 Advancing research on alien species and biological invasions A peer-reviewed open-access journal NeoBiota 302 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis strategies that jointly enhance their colonization capacity (Sádlo et al. 2018). Human activities further reinforce these processes by promoting long-distance transport across geographic and climatic barriers, thereby increasing ecological success (Richardson et al. 2000). Such interactions between natural dispersal modes and anthropogenic influences create a powerful synergy that accelerates invasion dynamics (O’Loughlin et al. 2017). While the initial release of diaspores from parent plants initiates propagule flow, subsequent redistribution by secondary vectors strongly influences invasion outcomes (Vander Wall et al. 2005; Schupp et al. 2010; Kowarik and Säumel 2008). Secondary dispersal determines whether diaspores remain close to the maternal plant or reach distant sites, thereby altering colonization probability, gene flow, and the ability to occupy new environments (Vander Wall et al. 2005; Schupp et al. 2010). In riparian systems, secondary dispersal is especially important because seeds shed on land may later be transported by water to entirely new locations (Jansson et al. 2005). Once deposited, they can establish populations far away from their source (Andersson and Nilsson 2002). These secondary movements significantly extend dispersal distances, facilitate escape from predators and pathogens, and promote establishment in heterogeneous habitats (Hämäläinen et al. 2017). As a result, they are central to both persistence and spread of alien plant species (Vander Wall et al. 2005). Hydrochory, particularly dispersal by floating on the surface of water, represents one of the most effective secondary pathways operating in rivers and floodplains (Sádlo et al. 2018; Boedeltje et al. 2004; Parolin 2005). Through this mechanism, riparian corridors act not only as habitats but offer routes of rapid spread (Gurnell et al. 2008; Anđelković et al. 2022). Propagules transported downstream are frequently deposited on geomorphic features such as bars, benches, and floodplains (van den Broek et al. 2005). These habitats, characterized by recurrent disturbance and high resource availability, provide ideal conditions for germination and seedling establishment (Jansson et al. 2005). At the same time, seasonal floods redistribute diaspores over wide areas, maintaining dynamic seed banks and enabling repeated colonization events (Andersson and Nilsson 2002). Hydrochory encompasses a range of dispersal processes. Not only light diaspores, but also heavier fruits or vegetative fragments, can be carried by water, depending on their morphology and local hydrodynamic conditions (Boedeltje et al. 2004). Traits such as seed mass, shape, surface area, and the presence of air-filled tissues strongly determine buoyancy and retention time in water. For example, samaras of Ailanthus altissima display clear adaptations enhancing flotation and dispersal efficiency (Boedeltje et al. 2004; Carthey et al. 2016). In urban rivers, such samaras are effectively transported downstream, contributing to the spread of exotic trees such as Acer negundo and Ailanthus altissima (Kowarik and Säumel 2007; Säumel and Kowarik 2013). Other studies emphasize that seed-coat structures such as air cavities, thickened walls, or fibrous surfaces can modulate buoyancy and prolong hydrochoric dispersal in invasive species (Carthey et al. 2016; Najberek et al. 2020). Seeds with fibrous surfaces or specialized cavities may remain afloat for extended periods, increasing the probability of reaching suitable deposition sites (Carthey et al. 2016; Najberek et al. 2020). Small modifications of seed coat morphology can substantially prolong flotation and enhance dispersal success (Boedeltje et al. 2004). Nautohydrochorous dispersal of seeds is strongly influenced by fruits’ morphology, with structural traits such as shape, internal cavity (or hollow chamber), and dehiscence mechanisms determining whether seeds 303 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis are transported individually by water or released explosively, thereby affecting dispersal distance and spatial distribution (Niwa 2024). However, despite these insights, the interplay between seed morphology and dispersal potential remains insufficiently understood. This is particularly true for species that produce dimorphic diaspores. Studies on hydrochory have shown that in Cyclachaena xanthiifolia, an invasive North American species with heterocarpy, smaller, lighter, and moderately flattened seeds with a small surface area sink after about 12 h, whereas larger, heavier, strongly flattened seeds with a wide rim and greater surface area float for up to 24 h, confirming the influence of seed shape and surface area on buoyancy (Lhotská and Slavík 1969). In such plants, different propagules may vary in mass, shape, or protective structures, which in turn influences their likelihood of being transported and successfully establishing (Carthey et al. 2016). A striking example of a riparian invader illustrating these processes is the North American vine Echinocystis lobata. This fast-growing annual climber has spread rapidly across Europe and Asia, where it has colonized numerous river valleys (Klotz 2009). Genetic analyses reveal that European populations are structured, reflecting multiple introductions and regional differentiation (Jocienė et al. 2023). Such genetic variation may interact with environmental context to shape dispersal performance. For instance, fruits produced by plants growing in more natural, less disturbed habitats tend to be larger, heavier, and contain more seeds than those produced in disturbed areas (Kostrakiewicz-Gierałt et al. 2022). These differences in fruit morphology translate into variation in buoyancy, flotation duration, and ultimately dispersal distance. As a consequence, even within the same species, propagule dispersal potential is not fixed but context-dependent, reflecting both genetic background and environmental influences (Boedeltje et al. 2004; van den Broek et al. 2005). Dispersal in E. lobata has traditionally been attributed to barochory and blastochory, with seeds dropping from drying capsules or being released explosively from dehiscent fruits (Silvertown 1985; Najberek et al. 2020). However, a significant proportion of diaspores remain enclosed within fibrous fruits, which delay release and facilitate secondary transport by wind or water (Dylewski et al. 2018). Approximately one third of fruits contain trapped seeds that are less accessible to rodents and may persist longer in riparian habitats. These fruits, buoyant due to their fibrous structure, can float downstream before releasing seeds (Dylewski et al. 2018). Hydrochory therefore appears central to the invasion success of E. lobata, as both buoyancy and flotation duration directly determine colonization potential (Kowarik and Säumel 2007; Fryirs and Carthey 2022). In this way, rivers and valleys act as key corridors for the spread of E. lobata in Europe, linking distant populations and facilitating continuous range spread (Zając et al. 2011). The coexistence of normal and trapped seeds reflects a diversified dispersal strategy. Substantial variation in seed morphology, including size, mass, and shape, likely affects flotation potential and transport distance (Carthey et al. 2016). Nevertheless, how functional differences between propagule types influence the dispersal potential of E. lobata in riverine landscapes remains poorly known. These knowledge gaps are critical for understanding E. lobata invasion ecology. Moreover, mechanistic insights into seed buoyancy can enhance invasion risk predictions and guide targeted control (Zając et al. 2011). By identifying vulnerable stretches of rivers and clarifying the role of different seed traits in dispersal, management actions can be made more spatially coordinated and cost-effective (Patinet et al. 2023). 304 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis This study examines how seed traits shape dispersal performance in E. lobata. We focus on the ways in which morphological features such as mass, thickness, surface area, and shape interact to determine buoyancy. We ask whether seeds retained within fibrous capsules differ functionally from freely released seeds in their flotation potential, and how fruit morphology regulates the timing of seed release into aquatic systems. By analyzing these relationships, we aim to clarify the mechanisms by which hydrochory contributes to invasion success in this species. We hypothesized that: H1 - seed mass would strongly influence sinking probability, particularly through its interaction with thickness; H2 - surface area would affect flotation mainly in combination with other traits and exert only a limited independent effect; H3 - trapped seeds would differ functionally from normal seeds, with buoyancy shaped by unique interactions among mass, thickness, and surface area. These hypotheses were tested in controlled laboratory conditions during which we examined the buoyancy of normal and trapped seeds after measuring their functional traits: mass, thickness, and surface area. Material and methods Study species Echinocystis lobata (Michx.) Torr. & Gray., commonly known as wild cucumber or balsam apple, belongs to the Cucurbitaceae family. It is a monotypic genus, the name derived from Greek echinos (hedgehog) and kystis (bladder), referring to its spiny fruits. The species originates from North America, where it is widespread in the eastern United States, except for California and parts of the southeast, and southern Canada, mainly in Ontario and Quebec (Choate 1940; Stocking 1955; Alex 1998; Gerrath et al. 2008; Callahan and Lincoln 2022). It occurs in moist habitats such as riverbanks, wetlands, forest edges, and anthropogenic sites. It had been also introduced to some parts of the western United States (Callahan and Lincoln 2022). Invasion history E. lobata was introduced to Europe in the late 19th and early 20th centuries as an ornamental plant and soon escaped cultivation (Domin 1942; Heine and Tschopp 1953; Slavík and Lhotská 1967; Dajdok and Kącki 2003; Zając et al. 2011). It spread rapidly across central and eastern Europe, extending from Central Europe to the Russian Pacific coast (Probatova 1987). In many countries, it is now recognized as a troublesome invasive species (Tokarska-Guzik 2005; Klotz 2009). In Poland, first wild populations were noted in the early 20th century, likely introduced from both Germany and Ukraine (Tokarska-Guzik 2005). By the mid-1960s it had established in the Carpathians, and by the 1990s its range expanded rapidly, colonizing riparian zones across the country (Zając et al. 2011; Zając and Zając 2019). Today, it is widespread in moist habitats, particularly river valleys, and often forms dense, monospecific stands (Maćkowiak and Dylewski 2014). Morphology and reproduction The plant is a vigorous annual vine, climbing with branched tendrils (Fig. 1). Its fruits are inflated, spiny capsules containing 1 to 6 seeds, most commonly four (Slavík and Lhotská 1967). Fruits usually have two or three chambers (locules), 305 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis although occasionally a single-chambered fruit develops at the tops of the shoots toward the end of the growing season. (Fig. 2). As fruits dry, the distal part of the pericarp ruptures, releasing seeds. However, fibrous internal structures often retain seeds, delaying their release (Silvertown 1985; Dylewski et al. 2018). A unique trait of E. lobata is the production of additional trapped seeds within the fibrous framework of the fruit. On average, 34% of fruits contain such seeds, which remain protected from predation and may be released only after decomposition of the capsule (Dylewski et al. 2018). Trapped seeds differ slightly in weight and shape from normally released seeds but show similar germination potential. Their delayed dispersal enhances persistence in riparian habitats (Silvertown 1985; Dylewski et al. 2018). Recent observations indicate that stem anatomy and climbing behavior of E. lobata vary with support availability; shoots grow fastest and display pronounced nutation when suitable supports are dense, and their hollow pith with reinforcing sclerenchyma and collenchyma tissues facilitates efficient climbing (Dołkin-Lewko and Zajączkowska 2024). Figure 1. Reproductive and vegetative structures of Echinocystis lobata. a. Inflorescences; b. Seedlings; c. Plants climbing the willows along the Liwiec River; d. Ripening fruit. 306 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis Location of seed collection points Six populations of E. lobata distributed in Poland in the Bug and Liwiec river valleys were selected for the study (Fig. 3). Four sites are located in the Bug valley, three of which are in the Lower Bug Valley mesoregion (Skuszew [52°35.85'N, 21°29.78'E], Szumin [52°36.36'N, 21°37.22'E], Wywłoka [52°37.02'N, 21°40.63'E]) and one in the Podlasie Bug River Gorge mesoregion (Drażniew [52°22.43'N, 22°42.56'E]) (Kondracki 2002). Two sites are located in the Liwiec valley. The Liwiec River is a left-bank tributary of the Bug, and E. lobata sites are situated near its mouth (Koszelanka [52°33.94'N, 21°35.96'E]) and (Barchów [52°30.73'N, 21°38.54'E]). These sites are located in the transition zone between the Lower Bug Valley and the Wołomin Plain mesoregions. All sites occur within the zone of regular spring flooding and occasional summer floods caused by heavy rainfall. Seed measurement methods Mature fruits were collected between October and December 2024. After harvest, fruits were classified into two categories: (i) fruits with normal seeds, i.e., seeds freely released after capsule dehiscence, and (ii) fruits with trapped seeds, i.e., seeds retained by fibrous structures inside the capsule and not released during natural drying. At each site, 100 fruits containing trapped seeds were collected first. An equal number of fruits with normal seeds was then gathered so that the Figure 2. Fruit diversity in Echinocystis lobata. a. Two-loculus fruit with one trapped seed; b. Two-loculus fruit with two trapped seeds; c. Three-loculus fruits, with one trapped seed (right); d. Three-loculus fruits, with one trapped seed (left) and two trapped seeds (right); e. One-loculus fruit; f. The interior of a normal fruit with two loculus (most common type); g. Fruits in various stages of ripeness. Red arrows indicate barely visible trapped seeds. 307 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis quantities of trapped and normal seeds from each site were the same. In total, 681 normal seeds and 681 trapped seeds were tested across six populations. Abnormally developed seeds (immature or malformed) were excluded from the analyses. Preliminary experiments conducted in 2023 demonstrated that both trapped seeds and seeds remaining afloat on the water surface were capable of germination. Trapped seeds were identified by shining a flashlight into the fruit chamber, which made the retained seeds visible through the capsule walls. They were then manually extracted using a scalpel. Each seed was weighed, labeled with an individual number, placed in a separate envelope, and stored in sealed plastic containers. Storage took place under ambient outdoor conditions for approximately three to four months, i.e., from harvest until the buoyancy experiment. We measured the following seed traits: length, width, thickness (mm), and fresh weight (g). Measurements were made using an electronic caliper and a laboratory scale. Seed surface area (mm2) was calculated using the ellipse formula (πab), where a and b are half the length and width, respectively. Additionally, we calculated the shape coefficient (width/length), where values <0.5 indicated elongated, lanceolate to oval seeds, and values >0.5 indicated oval to nearly spherical seeds. Buoyancy experiment The buoyancy experiment was conducted outdoors between February and March 2025, coinciding with the period of natural river flooding due to ice floes or snowmelt. Temperatures during the experiment ranged from 1 to 7°C, preventing ice formation in the containers. Seeds were placed individually into transparent plastic containers (100 ml) filled with unfiltered river water. River water was used to account for its natural ionic composition and organic matter, which may influence Figure 3. Location of the studied Echinocystis lobata sites (red points) against the background of the species distribution map in Poland (white points) in the ATPOL grid (Zając and Zając 2019). 308 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis flotation. Prior to the experiment, the water was stored under the same conditions as the seeds. To avoid confusion, seeds were introduced one by one. The number of floating seeds was recorded every 24 hours over a 14-day period. To simulate natural turbulence, containers were gently shaken once per day during monitoring. Statistical methods We conducted all statistical analyses using R 4.3.1 (R Core Team 2025). To evaluate how seed traits shape s in shaping the probability of sinking, we fitted a generalized linear mixed-effect model with binomial error distribution. The model included five interaction terms among five predictors. Specifically, we examined (i) the effect of seed mass on seed area, (ii) the interaction between thickness and mass, and (iii) the influence of seed type (normal vs. trapped) on area, thickness, and mass. Model diagnostics based on a non-parametric dispersion test indicated no signs of underor overdispersion (dispersion parameter = 0.995, P = 0.976). Study site was included as a random factor to account for population-level variation. We calculated marginal (Rm2) and conditional (Rc2) coefficients of determination, reflecting variance explained by fixed effects alone and by both fixed and random effects, respectively. The difference (Rc2 - Rm2) indicates the proportion of variance attributable to population identity. In interpreting the model results, we placed greater emphasis on effect sizes than on P values. This decision reflects growing concern that P-values are highly sensitive to sample size: large samples may yield statistically significant results for negligible effect sizes, while small samples may fail to detect effects that are biologically meaningful (Wasserstein and Lazar 2016). Therefore, model predictions were visualized as marginal responses, i.e. predicted values while holding other predictors constant. To evaluate our model describing the probability of seed sinking, we compared it with a corresponding intercept-only (null) generalized linear mixed-effect model using the corrected Akaike Information Criterion (AICc). A lower AICc value for the full model relative to the null model indicated that inclusion of the explanatory predictors provided a better supported description of the data. Results Trapped seeds were generally smaller and lighter than normal ones (Table 1), sank faster, and remained afloat for a shorter time (Fig. 4). Our results show that seed morphology and trapping status jointly affect buoyancy and sinking probability in Echinocystis lobata (Fig. 5). Table 1. Functional traits of trapped and normal seeds: length, width, thickness and mass. Location n Trapped seeds Normal seeds Length [mm] Width [mm] Thickness [mm] Mass [g] Length [mm] Width [mm] Thickness [mm] Mass [g] Barchów 119 15.89 8.20 3.97 0.27 17.57 8.73 3.97 0.32 Drażniew 117 16.15 8.43 3.75 0.27 17.82 8.80 3.80 0.33 Koszelanka 107 16.27 8.35 4.18 0.29 17.95 8.47 4.10 0.32 Skuszew 117 16.26 8.32 4.26 0.30 18.46 8.78 3.92 0.34 Szumin 111 15.97 8.78 3.74 0.27 17.35 8.56 3.72 0.30 Wywłoka 110 16.41 8.56 4.02 0.29 18.44 8.94 3.90 0.32 mean 16.16 8.44 3.99 0.28 17.93 8.71 3.90 0.32 309 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis Functional traits and buoyancy of normal and trapped seeds We characterized the functional traits of normal and trapped seeds (Table 1). Trapped seeds were generally smaller than normal ones, particularly in length (mean 1.77 mm) and width (mean 0.27 mm). The two seed types also differed in mass due to their size differences. Trapped seeds had a lower mean mass (0.04 g) compared to normal seeds. They also had a slightly greater mean thickness (0.09 mm), although this difference was minimal. The shape coefficient (width/length) differed slightly between seed types. Normal seeds had a mean coefficient of 0.49, indicating more elongated to oval shapes, whereas trapped seeds had a higher mean value of 0.53, corresponding to more oval to nearly spherical forms. The proportion of floating normal seeds was 12.3% and trapped amounted to 17.8%. After the first day, 2.6% of normal seeds and 0.7% of trapped seeds floated. After 48 hours, the proportion of floating normal seeds dropped to 1.6%, and trapped seeds to 0.6%. Trapped seeds sank faster (up to 11 days), while normal seeds remained up to 14 days (Fig. 4). Predictors and interactive effects of seed traits on sinking probability A generalized linear mixed model revealed that seed mass, thickness, and their interactions with trapping status significantly influenced sinking probability (Table 2), explaining 43% of its variance, while the random effect of study site contributed only modestly (11%). An effect of seed size on the probability of sinking was markedly influenced by seed mass (Fig. 5a–c). For seeds with a lower mass (≤ 0.25 g), the probability of sinking increased from 9% at a size of 0.38 cm2 to 45% at 0.77 cm2 (Fig. 5a). For medium-weight seeds (0.26 < seed mass ≤ 0.35 g), it increased from 3% at a size of 0.38 cm2 to 27% at 0.77 cm2 (Fig. 5b), while for heavier seeds (> 0.35 g) it increased from 1% at a size of 0.38 cm2 to 16% at 0.77 cm2 (Fig. 5c). The relationship between seed mass and the probability of sinking was subtly influenced by seed thickness (Fig. 5d, e). For thin seeds (≤ 0.30 cm), the probability of sinking decreased from 75% at a mass of 0.12 g to 5% at 0.52 g (Fig. 5d). For thicker seeds (> 0.30 cm), it decreased from 52% at a mass of 0.12 g to 0% at 0.52 g (Fig. 5e). Figure 4. Percentage and buoyancy time of floating seeds divided into normal and trapped seeds. 316 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis Chmura D, Dajdok D, Mazurska K, Tokarska-Guzik B (2018) Plan działań na rzecz zwalczania kolczurki klapowanej Echinocystis lobata (F. Michx.) Torr. & A. Gray. Generalna Dyrekcja Ochrony Środowiska, Warszawa. Choate HA (1940) Dormancy and germination in seeds of Echinocystis lobata. American Journal of Botany 27: 156–160. https://doi.org/10.1002/j.1537-2197.1940.tb14663.x Dajdok Z, Kącki Z (2003) Kenophytes of the Odra riversides. In: Zając A, Zając M, Zemanek B (Eds) Phytogeographical Problems of Synanthropic Plants. Institute of Botany, Jagiellonian University, Cracow, 131-136. Dołkin-Lewko A, Zajączkowska U (2024) Growth strategies and climbing behavior of the invasive vine wild cucumber (Echinocystis lobata). Acta Biologica Cracoviensia. Series Botanica 66(2): 1–10. https://doi.org/10.24425/abcsb.2024.150384 Domin K (1942) Echinocystis lobata Torr. et Gray, nova ceská adventivní rostlina z čeledi tykvovitých. Věda Přírodní 21: 25. Dylewski Ł, Maćkowiak Ł, Myczko Ł (2018) Physical defence of the wild cucumber Echinocystis lobata in an invasive range changing seed removal by rodents. Plant Ecology 219: 863–873. https:// doi.org/10.1007/s11258-018-0842-2 Fryirs KA, Carthey AJR (2022) How long do seeds float? The potential role of hydrochory in passive revegetation management. River Research and Applications 38: 1139–1153. https://doi. org/10.1002/rra.3989 Fumanal B, Chauvel B, Sabatier A, Bretagnolle F (2007) Variability and Cryptic Heteromorphism of Ambrosia artemisiifolia Seeds: What Consequences for its Invasion in France? Annals of Botany 100(2): 305–313. https://doi.org/10.1093/aob/mcm108 Gerrath JM, Guthrie TB, Zitnak TA, Posluszny U (2008) Development of the axillary bud complex in Echinocystis lobata (Cucurbitaceae): Interpreting the cucurbitaceous tendril. American Journal of Botany 95: 773–781. https://doi.org/10.3732/ajb.2007362 Guja LK, Merritt DJ, Dixon KW, Wardell‐Johnson GW (2010) Buoyancy, salt tolerance and germination of coastal seeds: Implications for oceanic hydrochorous dispersal. Functional Plant Biology 37(12): 1175–1186. https://doi.org/10.1071/FP10166 Gurnell A, Goodson J, Moggridge H, Thompson K (2008) Propagule deposition along river margins: Linking hydrology and ecology. Functional Ecology 22(5): 1045–1056. https://doi.org/10.1111/ j.1365-2745.2008.01358.x Hämäläinen A, Broadley K, Droghini A, Haines JA, Lamb CT (2017) The ecological significance of secondary seed dispersal by carnivores. Ecosphere 8(2): e01685. https://doi.org/10.1002/ ecs2.1685 Heine H, Tschopp W (1953) Echinocystis lobata (Michx.) Torr. & Gray. Flora Europaea, Vol. 1. Cambridge University Press, 123–124. Jansson R, Zinko U, Merritt DM, Nilsson C (2005) Hydrochory increases riparian plant species richness: A comparison between a free-flowing and a regulated river. Journal of Ecology 93(6): 1094–1103. https://doi.org/10.1111/j.1365-2745.2005.01057.x Jocienė L, Galanina O, Ielciu I, Juškaitytė E, Krokaitė E, Kupčinskienė E, Rekašius T (2023) Genetic differentiation of invasive Echinocystis lobata in Europe. Biological Invasions 25: 1123–1138. https://doi.org/10.3390/d15101084 Klotz S (2009) Echinocystis lobata (Michx.) Torr. & Gray, wild cucumber (Cucurbitaceae, Magnoliophyta). In: Hulme PE, Nentwig W, Pyšek P, Vilà M (Eds) Handbook of Alien Species in Europe. Invading Nature Springer Series in Invasion Ecology, vol. 3. Springer, Dordrecht, The Netherlands, 347 pp. https://doi.org/10.1007/978-1-4020-8280-1_13 Kondracki J (2002) Geografia regionalna Polski. PWN, Warszawa. Kostrakiewicz-Gierałt K, Barabasz-Krasny B, Bomanowska A, Dajdok Z, Gudžinskas Z, Kucharczyk M, Maćkowiak Ł, Majk J, Możdżeń K, Pliszko A, Podgórska M, Rasimavičius M, Rewicz A, 317 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis Stachurska-Swakoń A, Szczęśniak E, Wójcik T (2022) The relationships of habitat conditions, height level and geographical position with fruit and seed traits in populations of invasive vine Echinocystis lobata (Cucurbitaceae) in Central and Eastern Europe. Forests 13: 256. https://doi. org/10.3390/f13020256 Kowarik I, Säumel I (2007) Biological flora of Central Europe: Ailanthus altissima. Perspectives in Plant Ecology, Evolution and Systematics 8: 207–237. https://doi.org/10.1016/j.ppees.2007.03.002 Kowarik I, Säumel I (2008) Water dispersal as an additional pathway to invasions by the primarily wind-dispersed tree Ailanthus altissima. Plant Ecology 198: 241–252. https://doi.org/10.1007/ s11258-008-9398-x Lhotská M, Kopecký K (1966) Zur Verbreitungsbiologie und Phytozönologie von Impatiens glandulifera Rovle an den Flusssystemen der Svitava, Svratka und oberen Odra. –. Preslia 38: 376–385. https://www.preslia.cz/article/10651 Lhotská M, Slavík B (1969) Zur Karpobiologie, Karpologie und Verbreitung der Art Iva xanthiifolia Nutt. in der Tschechoslowakei. Folia Geobotanica et Phytotaxonomica 4: 415–434. https://doi. org/10.1007/BF02854699 Lopez OR (2001) Seed flotation and postflooding germination in tropical terra firme and seasonally flooded forest species. Functional Ecology 15: 763–771. https://doi.org/10.1046/j.02698463.2001.00586.x Maćkowiak Ł, Dylewski Ł (2014) Occurrence of Echinocystis lobata in the Grabarski Canal valley and its phytosociological range. Biodiversity Research and Conservation 29: 1–8. Najberek K, Olejniczak P, Berent K, Dylewski Ł, Kostrakiewicz-Gierałt K (2020) The ability of seeds to float with water currents contributes to the invasion success of Impatiens balfourii and I. glandulifera. Journal of Plant Research 133: 649–664. https://doi.org/10.1007/s10265-020-01212-0 Niwa H (2024) Estimation of potential seed dispersal regions based on floating and ballochory of Euphorbia adenochlora capsules. Ecological Solutions and Evidence 5: e70000. https://doi. org/10.1002/2688-8319.70000 O’Loughlin LS, McGeoch MA, Richardson DM (2017) Secondary invasion: When invasion success is contingent on prior invasion. Biological Invasions 19(7): 2063–2075. https://doi.org/10.1002/ ece3.3315 Parolin P (2005) Dispersal of pioneer trees in Amazonian floodplains. Ecotropica (Bonn) 11: 13–30. Patinet M, Branquart E, Monty A (2023) Invasive alien aquatic and riparian plant species. Best management practice guide. LIFE RIPARIAS project, 188 pp. https://www.riparias.be/774 Probatova N (1987) Adventive species of the family Cucurbitaceae in the flora of the Soviet Far East. Botanical Journal 72: 1129–1137. R Core Team (2025) R: A language and environment for statistical computing. https://www.R-project.org [Accessed 29 Aug 2025] Richardson DM, Pyšek P, Rejmánek M, Barbour MG, Panetta FD, West CJ (2000) Naturalization and invasion of alien plants: Concepts and definitions. Diversity and Distributions 6: 93–107. https://doi.org/10.1046/j.1472-4642.2000.00083.x Sádlo J, Chytrý M, Pergl J, Pyšek P (2018) Plant dispersal strategies: A new classification based on the multiple dispersal modes of individual species. Preslia 90: 1–22. https://doi.org/10.23855/ preslia.2018.001 Säumel I, Kowarik I (2013) Propagule morphology and river characteristics shape secondary water dispersal in tree species. Plant Ecology 214: 1257–1272. https://doi.org/10.1007/s11258-0130249-z Schupp EW, Jordano P, Gómez JM (2010) Seed dispersal effectiveness revisited: A conceptual review. The New Phytologist 188: 333–353. https://doi.org/10.1111/j.1469-8137.2010.03402.x Silvertown J (1985) Survival, fecundity and growth of wild cucumber, Echinocystis lobata. Journal of Ecology 73(3): 841–849. https://www.jstor.org/stable/2260151 318 NeoBiota 104: 301–318 (2025), DOI: 10.3897/neobiota.104.162267 Adam Szulc et al.: Trait interactions drive seed dispersal in Echinocystis Slavík B, Lhotská M (1967) Chorologie und Verbreitungsbiologie von Echinocystis lobata (Michx.) Torr. et Gray mit besonderer Berücksichtigung ihres Vorkommens in der Tschechoslowakei. Folia Geobotanica et Phytotaxonomica 2: 255–282. https://doi.org/10.1007/BF02854618 Stocking KM (1955) Some considerations of the genera Echinocystis and Echinopepon in the United States and northern Mexico. Madroño 13: 84–101. https://www.biodiversitylibrary.org/ page/47864117#page/104/mode/1up Tokarska-Guzik B (2005) The establishment and spread of alien plant species (kenophytes) in the flora of Poland. Wydawnictwo Uniwersytetu Śląskiego, Katowice, Poland. van den Broek T, Diggelen R, Bobbink R (2005) Variation in seed buoyancy of species in wetland ecosystems with different flooding dynamics. Journal of Vegetation Science 16: 579–586. https:// doi.org/10.1658/1100-9233(2005)16[579:VISBOS]2.0.CO;2 Vander Wall SB, Kuhn KM, Beck MJ (2005) Seed removal, seed predation, and secondary dispersal. Ecology 86: 801–806. https://doi.org/10.1890/04-0847 Wasserstein RL, Lazar NA (2016) The ASA’s statement on p-values: Context, process, and purpose. The American Statistician 70: 129–133. https://doi.org/10.1080/00031305.2016.1154108 Zając A, Zając M (2019) Atlas rozmieszczenia roślin naczyniowych w Polsce: Dodatek - Distribution atlas of vascular plants in Poland: Appendix. Institute of Botany, Jagiellonian University, Kraków, Poland. Zając A, Tokarska-Guzik B, Zając M (2011) The role of rivers and streams in the migration of alien plants into the Polish Carpathians. Biodiversity Research and Conservation 23: 43–56. https://doi.org/10.2478/v10119-011-0012-z Supplementary material 1 Functional traits of Echinocystis lobata seeds Authors: Adam Szulc, Patryk Czortek, Pawel Marciniuk Data type: xlsx Explanation note: The table presents key functional traits of Echinocystis lobata seeds, including morphological and physiological characteristics. Traits include seed mass, length, width, height, entrapped status, and time of floating. Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/neobiota.104.162267.suppl1