Predatory interactions between two global aquatic invaders beyond their native ranges: An experimental approach

173 Predatory interactions between two global aquatic invaders beyond their native ranges: An experimental approach Andrey N. Reshetnikov1, Arina O. Raldugina1, Dmitriy V. Grinchenko2, Artem A. Kidov2, Nikita G. Platonov1, Andrey B. Petrovskiy1 1 Severtsov Ecology and Evolution Institute, Moscow, Russia 2 Russian State Agrarian University–MTAA, Moscow, Russia Corresponding author: Andrey N. Reshetnikov (anr[email protected]) Copyright: © Andrey N. Reshetnikov 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 The widespread distribution of invasive species inevitably leads to the emergence of a new category of biotic relationships: interspecific predatory interactions between invasive species. We assessed the vulnerability of different life stages of the clawed frog, Xenopus laevis (eggs, hatchlings, tadpoles, newly-metamorphosed froglets, and adults), to predation by the Eastern mosquitofish, Gambusia holbrooki, and evaluated whether adult clawed frogs prey upon juvenile and adult mosquitofish. The results confirm differences in the palatability of different ontogenetic stages of the clawed frog by mosquitofish, as well as the low palatability of mosquitofish for adult frogs, with the exception of fish juveniles, which are relatively protected in the light but highly vulnerable under low-light conditions. Therefore, these fish-amphibian interactions are complex and can be defined as unequal bidirectional predation: the mosquitofish readily eliminates its early-stage opponent, but may become prey for its adult individuals. The revealed existence of a vulnerable mosquitofish stage makes it difficult to predict the unconditional suppression of clawed frog populations by this fish species when these two global aquatic invaders become syntopic in new regions. In any case, mosquitofish are likely to dominate in urban water bodies in locations with constant artificial lighting. We emphasise the importance of studying interactions at all ontogenetic stages, as well as taking into account the lifestyle of the organisms studied when analysing the mechanisms of predatory interactions between any pairs of invasive species in new areas of the planet. Graphical abstract Academic editor: Filipe Ribeiro Received: 30 December 2024 Accepted: 17 June 2025 Published: 7 October 2025 Citation: Reshetnikov AN, Raldugina AO, Grinchenko DV, Kidov AA, Platonov NG, Petrovskiy AB (2025) Predatory interactions between two global aquatic invaders beyond their native ranges: An experimental approach. In: Anastácio P, Ribeiro F, Chainho P (Eds) Invasions in Aquatic Systems. NeoBiota 102: 173–189. https://doi. org/10.3897/neobiota.102.145644 NeoBiota 102: 173–189 (2025) DOI: 10.3897/neobiota.102.145644 Advancing research on alien species and biological invasions A peer-reviewed open-access journal NeoBiota ecosystems change on the global scale; assessment of newly established predatory interactions has to include diverse ontogenetic stages and take into account the lifestyle of the studied organisms American Eastern mosquitofish in the dark only Global invaders encounter in regions beyond their native distributions complex system of invader-invader interactions African clawed frog 174 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Key words: Gambusia holbrooki, Global invasions, invasion ecology, predatory interactions, Xenopus laevis Introduction The widespread distribution of invasive species inevitably leads to the emergence of a new category of biotic relationships: interspecific interactions between invasive species. Such interactions may be neutral, unidirectional or bidirectional; the establishment of some invaders may facilitate or prevent the invasions of other organisms (Jackson 2015). There are comparatively few studies of interactions between invaders, but their importance is increasing as the biota change (Sheppard et al. 2018; Arismendi et al. 2020). Some invasive organisms are now globally distributed. One such species is the Eastern mosquitofish (Gambusia holbrooki Girard, 1859), which originates from North America (Francis and Hardwick 2012). This fish and a closely related one (G. affinis (Baird & Girard, 1853)) were long considered to be a single biological species until 1988 (Wooten et al. 1988). It is assumed that they have very similar biology and behaviour. Many contemporary reviews combine the description of both species into single papers (Pyke 2005; Walton et al. 2012). The main reason for the human-mediated geographical spread of G. holbrooki, along with G. affinis, is its use for mosquito control (Pyke 2005). This small-sized omnivorous fish has a generalist diet which includes a wide range of animal prey: from ciliates and microcrustaceans to small-sized aquatic vertebrates (Pyke 2008). Invasive mosquitofish populations demonstrate an undesirable impact on native ecosystems (Hurlbert et al. 1972; Hinchliffe et al. 2017). Today, populations of G. holbrooki exist on all continents except Antarctica; its invaded range includes at least 92 countries (Reshetnikov et al. 2018). This fish invader ranks fifth among 551 non-native fish species in terms of the number of river basins colonised on a global scale; the most invaded areas are Australia, where G. holbrooki ranks 1st among other fish invaders, and the Palearctic, where it ranks 4th (Bernery et al. 2022). Similar to the fish species presented above, the clawed frog (Xenopus laevis (Daudin, 1802)) is a global invader (Measey et al. 2012). This anuran amphibian belongs to the family Pipidae and originates from Southern Africa. After metamorphosis, the adult clawed frogs remain in the water and exhibit an aquatic lifestyle. Its larvae are filtrators whereas adults consume diverse animal prey including comparatively large invertebrates and even vertebrates such as amphibians and fish (Courant et al. 2017). From the 1930s to the 1960s, this amphibian was widely used to test pregnancy (Elkan 1938) and gradually became extremely popular for various scientific laboratory investigations and as a pet. As a result, X. laevis is exported outside its native range in Southern Africa (Weldon et al. 2007), bred in captivity and sold in pet shops worldwide (Gurdon 1996). This amphibian species has been introduced into many new countries outside its native distribution (Measey et al. 2012). In new areas, this frog interacts with native biota, affecting local species through competition, predation, and transmission of infectious diseases (Amaral and Rebelo 2012). The clawed frog has been noted as an asymptomatic agent in the spread of the pathogen of the emerging amphibian fungal disease (Batrachochytrium dendrobatidis) and ranavirus (Fisher and Garner 2007; Reeder et al. 2012; Peñafiel-Ricaurte et al. 2023). To date, X. laevis has been found on all continents except Australia and Antarctica (Rödder et al. 2017). 175 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders The climatic niches of this frog and the aforementioned mosquitofish, used for mosquito control, overlap as shown by species distribution models; regions of actual sympatry of these global invaders have already been revealed on all the continents where the frog has naturalised (Ihlow et al. 2016; Jourdan et al. 2021). Furthermore, there is evidence that both the clawed frog and the mosquitofish exhibited climatic niche enlarging as they colonise new areas (Rödder et al. 2017; Jourdan et al. 2021). Throughout its vast invaded range, the mosquitofish inevitably encounters other aquatic invaders, resulting in the establishment of new trophic interactions. For example, the invasive mosquitofish co-occurs and interacts with the cane toad (Rhinella marina (Linnaeus, 1758)) in Australia, the Louisiana crayfish (Procambarus clarkii (Girard, 1852)) in Europe, the frog X. laevis in South America and other continents (Komak and Crossland 2000; Anastácio et al. 2011; Lobos 2020). Experimental studies have shown that the mosquitofish avoids feeding on the eggs and tadpoles of the cane toad, but the brief hatchling ontogenetic stage has been considered vulnerable to fish predation (Komak and Crossland 2000). The mosquitofish can consume newly hatched Louisiana crayfish, but is eaten by large adult crayfish individuals (Anastácio et al. 2011). Thus, unidirectional and bidirectional predatory interactions between mosquitofish and other invaders are possible. However, many aspects of the alien-alien interactions remain unknown. For example, it has been suggested that mosquitofish in Chile may negatively affect populations of the introduced clawed frog but detailed analyses indicate that the clawed frog population began to decline prior to the date of documented mosquitofish population growth (Lobos 2020). In addition, ontogenetic stages of the clawed frog that are vulnerable to mosquitofish predation and the potential consumption of the mosquitofish by this invasive amphibian remain unknown. Both the mosquitofish and the clawed frog are classic invaders with early sexual maturity, high fecundity, and a broad diet. They are tolerant to a wide range of environmental conditions, and readily colonise heavily modified anthropogenic landscapes (Pyke 2008; Francis and Hardwick 2012). Both the clawed frog and the mosquitofish have similar habitat requirements, preferring shallow, well-vegetated warmed areas of aquatic sites with still or slowly-moving water (Pyke 2008; Francis and Hardwick 2012). However, the possibility of longterm coexistence of populations of these two invaders in the same water body is questionable, as they are active predators with significant potential to impact on syntopic hydrobionts (Lafferty and Page 1997; Hamer 2022). The aim of this study is to investigate the predatory interactions between two globally invasive species: the Eastern mosquitofish (Gambusia holbrooki) and the African clawed frog (Xenopus laevis). We hypothesised that different ontogenetic stages of the frog are vulnerable to varying degrees of predation by mosquitofish and that mosquitofish themselves can be preyed upon by adult frogs under certain conditions. To this end, we conducted a series of controlled laboratory experiments to assess the palatability, understood as ability to consume, of X. laevis eggs, hatchlings, tadpoles, and metamorphs to mosquitofish. In parallel, we evaluated the palatability of juvenile and adult G. holbrooki to adult clawed frogs. Given that X. laevis is a predominantly nocturnal predator, we also conducted additional trials under low-light conditions to explore its predation efficiency during the dark phase of the diel cycle. 176 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Materials and methods Animals and their maintenance The Eastern mosquitofish, Gambusia holbrooki, were collected on 13 April 2022 with a deep-net from a pond (43.4263N; 39.9731E) on the Black Sea coast of the Caucasus in Krasnodar Krai at south of the Russian Federation. The collection of G. holbrooki is not regulated in Russia as this fish is recognised as invasive (Reshetnikov et al. 2018). The sizes of the fish were: a total length (TL) of 33.7 ± 0.8 (25.0–41.0) mm and a body mass (m) of 0.33 ± 0.03 (0.11–0.63) g; N = 20 (mean values and their standard errors are given throughout the article). The fish were transported to the laboratory. Prior to the experiment, the fish were kept in 45 L aquaria with density 1 fish individual per liter. We used aged tap water with the following values of hydrochemical characteristics: oxidation-reduction potential (ORP) 126.1 ± 1.4 mV; pH 7.14 ± 0.01; concentration of dissolved oxygen (DO) 8.59 ± 0.04 mg/L (Aqua TROLL® 600, the USA); ammonium concentration of < 1.0 NH3/NH4 mg/l (NILPA®, Russia). Water was changed partially every other day. For preliminary acclimation after transportation to the laboratory, fish were kept in 45 L aquaria at least 10 days before the start of experimental work. We also used artificially bred clawed frog X. laevis. Prior to the experiment, eggs and hatchlings were kept in aquaria with aged tap water (hydrochemical parameters as presented above) at water temperature of 4–5 °C. This low temperature was used to delay the development of eggs and hatchlings in order to have available materials for the experiment by the right date. After a period of keeping at low temperature, eggs and hatchlings remained alive and undamaged, which was confirmed by us: some individuals did not participate in the experiments and later underwent normal development and reached sexual maturity. The metamorphs were kept in 45 L aquaria filled to 35 L (10 individuals per aquarium) and fed with Chironomidae larvae ad libitum every other day. Adult clawed frogs were kept under the same conditions but 14 individuals per 45 L aquarium with frequent water changes. Larvae, metamorphs, and adults were kept at a water temperature of 19–21 °C. Sizes (length and mass) of different ontogenetic stages are presented in the Table 1. All the measurements of length were made with an accuracy of 1 mm. The mass was measured with 2 mg accuracy with balances KERN® CM 50-C2N (KERN & Sohn GmbH, Germany), excepting adult frogs, which were measured with balances VK-300.1 (Massa-K®, RF) with 100 mg accuracy. Amphibian sizes were determined for 15 individuals of each of the stages studied. Experimental design Each randomly chosen fish individual was placed in a separate aquarium (N = 28), 39.5 × 28.7 × 24.5 cm, containing 10 L of aged tap water (appr. 9 cm water column). Aquaria with females and males were alternated: 2 aquaria with females, 2 aquaria with males, etc. The experimental fish were visually isolated from each other using white heavy (235 g/m2) paper dividers between aquaria. The artificial photoperiod was 06:00–23:00. Illumination sources (LED lamps, cold light 6500 K) were placed 45 cm above the water and produced 623 ± 34 lx on the water surface. Under these conditions, the fish began to actively consume the food (larvae of Diptera, Chironomidae) within the first day after being 177 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Table 1. Size characteristics of different ontogenetic studies of the clawed frog Xenopus laevis, used in the experiments: Mean Length ± Standard Error (L, 1 mm), Length Range (minimum-maximum, mm), Mean Weight (W, 2 mg, 10 mg), Weight Range (mm), and number of measured individuals (n). Stage Mean L, mm Range L, mm Mean W, Mg Range W, mg n Egg 2.0 2–2 4 4–4 15 Embryo 1.5 1.5–1.5 4 4–4 15 Hatchling 7.0 7–7 4 4–4 15 Early tadpole 7.9 ± 0.2 7–9 8 8–8 15 Metamorph 15.8 ± 0.4 13–19 417 ± 32 220–760 15 Adult 36.7 ± 0.5 33–43 5000 ± 200 3500–8100 15 placed in the aquarium. Nevertheless, the fish were left to acclimatise for five days. Hydrochemical parameters on the second day of acclimation (before partial water change) were: ORP 113.6 ± 1.3 mV; pH 7.28 ± 0.04; DO 8.50 ± 0.04 mg/L (Aqua TROLL® 600, USA); < 1.0 NH3/NH4 mg/l (NILPA®, Russia). All experiments were carried out in a thermostatically controlled laboratory room from December 2022 to November 2023 at a water temperature of 19–21 °C. This temperature is within the temperature range of tolerance and feeding activity of both species studied: mosquitofish and clawed frog (Casterlin and Reynolds 1980; El-Boray 2014; Wilson et al. 2000; our observations). To standardise the level of food motivation, fish were not fed for 24 h prior to the experiment. When a potential food was added to the aquarium, the food was visually isolated from the fish by an opaque plastic screen (9.5 × 25.5 cm) to prevent the fish from responding to the experimenter’s hand movements. The screen was then removed from the aquarium with a vertical upward motion. i. Palatability of clawed frog to mosquitofish. In the first experiment, 28 mosquitofish (14 females and 14 males; one individual per aquarium) were offered eggs of the clawed frog. All fish individuals were assigned randomly to two test subgroups. During the first round of the experiment, one half of the experimental fish (first subgroup: odd-numbered aquaria) were offered one amphibian egg (2 mm; 4 mg). Remaining fish (second subgroup: even-numbered aquaria) were offered the control food, two individuals of Chironomidae larvae (each worm was 10.5 ± 0.3 (8–14) mm, data presented from measurements of 20 specimens). The mass of a pair of insect larvae individuals was 6.0 ± 0.0 (6.0–6.0) mg. After 2 h, the uneaten food or its remains were removed and the experiment was continued after another 1 h. In the second round of the experiment, fish of the first subgroup were offered control food whereas fish of the second subgroup were offered amphibian egg. The next day, fish in each group were again offered amphibians and control food twice (first and second round), but in reverse order. Such design of the experiment was applied in order to exclude the potential effect of the sequence of offering the main and control prey species. For statistical analysis we summed the data obtained over two days of an experiment, transforming the results for each certain type of prey to a binary system: consumed / not consumed. During each round, no adjacent aquaria had the same treatment to minimise any bias that might have arisen from aquarium position. No additional food was offered to the fish at the end of the test in this experiment. 178 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders In the same way to the same well-acclimated group of fish individuals, we offered embryos, hatchlings, tadpoles (stage 25 according to Gosner 1960; stage 45 following Nieuwkoop and Faber 1994), and newly-metamorphosed froglets of the clawed frog, each in a separate experiment with comparison with control food. The sizes of the used frog stages are presented in Table 1. The sequence of experiments did not correspond to the order, from egg to froglets, listed above. In an experiment with a particular prey species/stage/condition, each individual fish participated once. Testing of individual tadpoles in palatability trials is an appropriate approach, because there are no differences in the survival of clawed frog tadpoles between individual vs. group offering to G. holbrooki (Lobos 2020). Beyond the main result of this experiment (assessment of the palatability of different prey types), we had the opportunity to compare the consumption of prey items in the first and second rounds and confirmed no significant differences in prey palatability between rounds (Suppl. material 1: fig. S1). As part of the above-described series of experiments, a separate trial was used to assess possible differences in the palatability of two close but principally different stages of the clawed frog: early embryo (jelly capsule was removed mechanically by an experimenter) and hatchling (after leaving the capsule on its own). For this purpose, a direct comparison experiment was conducted: embryo vs. hatchling. In this experiment, 2 individuals of chironomid larvae were offered to the fish just after the trial to monitor the presence of food motivation. The presence of food motivation was assumed to be confirmed if chirinomidae larvae were consumed without residue. As males of mosquitofish are noticeably smaller than females, this could theoretically affect their ability to consume some types of prey. The sizes of males were: TL = 30.1 + 0.6 (25–33) mm; m = 176 + 8 (110–215) mg (80.6 and 38.8% of female length and mass respectively). We used results of the above-described series of the experiments to compare consumption of objects by males and females. For statistical analysis, males and females were coded as 1 and 0 respectively. ii. Palatability of mosquitofish to adult clawed frogs. We also conducted experiments to assess the ability of adult clawed frogs to catch and consume mosquitofish. The sizes of the adult frogs are presented in Table 1. Frogs were placed randomly one per aquarium with 10 L of aged tap water. Hydrochemical parameters of water on the second day of acclimation (before partial water change) were: ORP = 105.4 ± 1.6 mV; pH 7.28 ± 0.06; DO = 8.61 ± 0.03 mg/L (Aqua TROLL® 600, the USA); < 1.0 NH3/NH4 mg/l (NILPA®, Russia). The experiments were conducted at a water temperature of 19–21 °C according to protocol of the main experiment described above, i.e. over two days with two rounds per day. As before, chironomid larvae were used as control prey. First, we tested palatability of adult mosquitofish for frogs (2 h offering of the potential prey). As consumption of adult fish prey was not revealed in the first experiment, we conducted an additional experiment with prolonged offering during 6 h to test if additional time is important for prey palatability. To confirm the importance of fish defensive behaviour in escaping frog attacks and exclude version of possible tastelessness, we conducted an additional experiment in which adult frogs were offered live vs. dead adult mosquitofish during 2 h. At the end of the observation period (2 h) and after removal of the remaining fish individuals from the aquaria, the chironomidae larvae were offered to the frogs to confirm their food motivation. 179 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders We also tested the assumption that juvenile small fish individuals might be less able to avoid predators than large fish adults. The sizes of the offered juvenile fish were: TL = 19 ± 0.6 (17–23) mm; m = 70 ± 10 (36–136) mg; N = 20. The fish juveniles were offered in comparison with control food. iii. Palatability of mosquitofish to adult clawed frogs under low-light conditions. Considering that the mosquitofish are diurnal and the adult clawed frogs are nocturnal, we have assumed that a frog predator could have an advantage when hunting under low-light conditions. Frogs were left to acclimate in aquaria during 5 days. We conducted an appropriate trial, offering adult mosquitofish vs. control prey during 2 h under low-light conditions (<0.01 lux). Results on prey consumption were recorded at the end of the offering period (2 h). To test the assumption that frogs will be more successful predators after prolonged low-light period, we also applied modification of the experiment in a separate trial with the 6 h time of offering. Results were recorded at the end of 6 h of offering. We also hypothesised that small-sized juvenile mosquitofish individuals may have been less protected (comparing to adult fish) from frog predation under lowlight conditions. To test this hypothesis, we conducted a separate experiment with 2 h offering juvenile fish vs. control prey under low-light conditions. Data analysis Non-parametric statistic criteria were used. Cochran’s Q test (for multiple comparisons) and Sign Z test (for pairwise comparisons) were applied for assessment of comparative palatability. We used package stats from base R (R Core Team 2024) for Spearman correlation test, package coin (Hothorn et al. 2006) for Sign test and Cochran’s Q test, package ggplot2 (Wickham 2016) for plots. The level of significance was accepted as 0.05. We present unadjusted p values. All conclusions in the Result section remain valid after applying the Holm-Bonferroni correction for multiple comparisons, except where specifically stated. Results Palatability of clawed frog for mosquitofish In all cases, control prey (Chironomidae) was consumed immediately after the first attack, i.e., during the first few seconds after the start of the trial. Mosquitofish readily consumed clawed frog eggs and hatchlings; the mean number of these prey items consumed was 1.5 ± 0.1 (0–2) and 1.6 ± 0.1 (0–2) respectively. The tadpole was the most palatable stage; the fish consumed all tadpoles offered (Fig. 1). Newly-metamorphosed froglet (metamorph) was the least consumed stage: 0.11 ± 0.06 (0–1). The prey item significantly influenced the degree of palatability (Q = 82.5; N = 28; df = 4; p < 0.001). When transformed to a binary scale (consumed/not consumed – summary for two experimental rounds), the palatability level of the least edible stage (froglet) was significantly different from the palatability of the egg (Z = -4.90; N = 28; p < 0.001). The differences in palatability between newly-metamorphosed froglets and other stages are even greater (Fig. 1). Except for froglets, differences between other stages are not significant (Z ≤ 1.5; NS). Fish consumed eggs, hatchlings, and tadpoles as a whole. 180 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Unlike other stages, fish did not consume whole newly-metamorphosed froglets, but some froglets were fatally injured by the fish and died. The injured froglets were partially eaten; their mass was decreased on 12–44 mg or 8% of body mass. It is noteworthy that these newly-metamorphosed froglets did not attempt to attack the adult fish but tried to escape from the fish predator. All fishes that consumed amphibians remained alive with no signs of stress. Both females and males of mosquitofish readily consumed eggs, hatchlings, and tadpoles of the clawed frog (Suppl. material 1: fig. S2), but females tended to consume a slightly greater number of frog eggs compared to males (Spearman correlation test: rho = -0.44; N = 28; S = 5273.0; p = 0.02; P2crit = 0.017 following Holm-Bonferroni correction). In our experiment, a small number of newly-metamorphosed froglets were injured (Suppl. material 1: fig. S2). Fish males did not injure this type of prey; all froglets were injured and partially consumed by females only (Suppl. material 1: fig. S2). In a trial with direct testing of palatability of the clawed frog embryo comparing with the hatchling, fish consumed a high number of the offered embryo (Suppl. material 1: fig. S3). The level of palatability was similar to that of the hatchling and did not differ significantly (Z = -1.41; N = 28; p = 0.50). Palatability of mosquitofish for adults of the clawed frog In the experiment with one-year-old adult frogs (as predator) and adult mosquitofish (offered), the fish did not demonstrate any predatory behaviour. In contrast to the fish, each frog performed multiple but unsuccessful attempts to catch the offered fish. Frogs attacked potential prey from a short distance of 1–2 cm. Despite obvious food motivation, no frogs caught the adult fish during 2 h. We repeated this experiment using a longer time period, 6 h offering, and obtained an identical Figure 1. Palatability (means ± SE) of different ontogenetic stages of the clawed frog Xenopus laevis for the mosquitofish Gambusia holbrooki. 181 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders result: the frogs failed to catch the fish during the prolonged period. The difference in the observed palatability of the control prey (Chironomidae) and the adult mosquitofish is significant (Z = 5.29; N = 28; p < 0.001). In an additional experiment, adult frogs were offered live vs. dead adult mosquitofish for a period of 2 h. In this experiment, frogs did not eat live adult mosquitofish but consumed all the dead fish offered. The difference in palatability between live and dead fish is significant (Z = 5.29; N = 28; p < 0.001). The sizes of the consumed fish were TL = 33.9 ± 0.6 (27–44) mm, m = 404 ± 27 (142–890) mg, N = 56. These fish were 71–123% in length and 2–18% in mass of the appropriate frog predator. The frogs swallowed fish either from the head or tail end. Parts of large fish individuals were seen sticking out of the frog’s mouth. All frogs that consumed fish remained alive with no negative consequences. In experiments with adult frogs and juvenile mosquitofish, one out of 20 frogs consumed two fish individuals and two frogs consumed one fish each during two rounds. In all cases, the prey was consumed whole. One frog (L = 32 mm, m = 3.3 g) consumed fish prey representing 53% of its length and 2.3% of its mass over two days and the other two frogs consumed fish individuals representing 58 and 62% of frog’s length and 1.6 and 2.2% of its mass, respectively. The difference in palatability of test and control prey is significant (Z = 4.12; N = 20; p < 0.001) (Fig. 2b). Palatability of mosquitofish for adults of the clawed frog under lowlight conditions No adult mosquitofish were eaten by frogs during a 2-hour trial under low-light conditions; the difference in consumption of test and control prey is significant: Z = 5.29; N = 28; p < 0.001. In another experiment, during a longer time period (6 h) under low-light conditions, two out of 28 frogs had consumed mosquitofish: one prey fish per amphibian individual (Fig. 2a). The difference between the consumption of test and control prey is also significant: Z = 5.10; N = 28; p < 0.001. In this experiment, there were equal numbers of females and males. Both the consumed fish individuals were males. Their sizes were 29 mm, 252 mg and 28 mm, 238 mg. Their length was 72–83% and their mass 4–5% of a frog predator. Contrary to the above-described results, the frogs effectively caught and consumed juvenile mosquitofish under low-light conditions: most of the fish were eliminated during 2 h period (Fig. 2c). All adult frogs tested consumed juvenile fish (one or two individuals per frog) as well as control prey during the two test rounds, showing no differences in palatability between these two prey items. All the studied predatory interactions are summarised in the Table 2. Discussion In this way, we have demonstrated reciprocal predation of two studied aquatic invaders of different systematic affiliations. The results confirm differences in palatability of various ontogenetic stages of the clawed frog for mosquitofish, as well as low palatability of adults and high palatability of juveniles of this fish for adult stage of the frog. Vulnerability of juvenile fish to frog predation depends on illumination conditions. These fish-amphibian interactions are complex and their understanding requires more detailed analysis. 188 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Lobos G (2020) Vulnerability of Xenopus laevis to Gambusia holbrooki: Can the larval phase of the African clawed frog be the Achilles heel in its invasive potential? Aquatic Invasions 15: 529–541. https://doi.org/10.3391/ai.2020.15.3.11 Manteifel YB, Reshetnikov AN (2002) Avoidance of noxious tadpole prey by fish and invertebrate predators: Adaptivity of a chemical defence may depend on predator feeding habits. Archiv für Hydrobiologie 153: 657–668. https://doi.org/10.1127/archiv-hydrobiol/153/2002/657 Measey GJ, Rödder D, Green SL, Kobayashi R, Lillo F, Lobos G, Rebelo R, Thirion J-M (2012) Ongoing invasions of the African clawed frog, Xenopus laevis: A global review. Biological Invasions 14: 2255–2270. https://doi.org/10.1007/s10530-012-0227-8 Moreira FD, Marques R, Sousa M, Rebelo R (2017) Breeding in both lotic and lentic habitats explains the invasive potential of the African clawed frog (Xenopus laevis) in Portugal. Aquatic Invasions 12: 565–574. https://doi.org/10.3391/ai.2017.12.4.12 Nieuwkoop PD, Faber J (1994) Normal table of Xenopus Laevis (Daudin). Garland Publishing Inc, New York, 252 pp. Peñafiel-Ricaurte A, Price SJ, Leung WTM, Alvarado-Rybak M, Espinoza-Zambrano A, Valdivia C, Azat C (2023) Is Xenopus laevis introduction linked with Ranavirus incursion, persistence and spread in Chile? PeerJ 11: e14497. https://doi.org/10.7717/peerj.14497 Pepin P, Shears TH (1995) Influence of body size and alternate prey abundance on the risk of predation to fish larvae. Marine Ecology Progress Series 128: 279–285. https://doi.org/10.3354/ meps128279 Pyke GH (2005) A review of the biology of Gambusia affinis and G. holbrooki. Reviews in Fish Biology and Fisheries 15: 339–365. https://doi.org/10.1007/s11160-006-6394-x Pyke GH (2008) Plague minnow or mosquito fish? A review of the biology and impacts of introduced Gambusia species. Annual Review of Ecology, Evolution, and Systematics 39: 171–191. https://doi.org/10.1146/annurev.ecolsys.39.110707.173451 R Core Team (2024) R: A language and environment for statistical computing. Version 4.4.0. R Foundation for Statistical Computing. Vienna, Austria. https://www.R-project.org/ Reeder NMM, Pessier AP, Vredenburg VT (2012) A reservoir species for the emerging amphibian pathogen Batrachochytrium dendrobatidis thrives in a landscape decimated by disease. PLoS ONE 7: e33567. https://doi.org/10.1371/journal.pone.0033567 Reshetnikov AN (2008) Does rotan Perccottus glenii (Perciformes: Odontobutidae) eat the eggs of fish and amphibians? Journal of Ichthyology 48: 336–344. https://doi.org/10.1134/ S0032945208040061 Reshetnikov AN, Zibrova MG, Dgebuadze YY (2018) Gambusia holbrooki (Girard, 1859): The Most Dangerous Invasive Species of Russia (TOP-100). KMK, Moscow, 538–545. Reynolds SJ (2009) Impact of the introduced poeciliid Gambusia holbrooki on amphibians in southwestern Australia. Copeia 2009: 296–302. https://doi.org/10.1643/CH-08-101 Rödder D, Ihlow F, Courant J, Secondi J, Herrel A, Rebelo R, Measey GJ, Lillo F, de Villiers FA, de Busschere C (2017) Global realized niche divergence in the African clawed frog Xenopus laevis. Ecology and Evolution 7: 4044–4058. https://doi.org/10.1002/ece3.3010 Segev O, Mangel M, Blaustein L (2009) Deleterious effects by mosquitofish (Gambusia affinis) on the endangered fire salamander (Salamandra infraimmaculata). Animal Conservation 12: 29–37. https://doi.org/10.1111/j.1469-1795.2008.00217.x Sheppard CS, Carboni M, Essl F, Seebens H, DivGrass Consortium, Thuiller W (2018) It takes one to know one: Similarity to resident alien species increases establishment success of new invaders. Diversity & Distributions 24: 680–691. https://doi.org/10.1111/ddi.12708 Walton WE, Henke JA, Why AM (2012) Gambusia affinis (Baird and Girard) and Gambusia holbrooki Girard (Mosquitofish): A Handbook of Global Freshwater Invasive Species. Earthscan, London, 261–273. 189 NeoBiota 102: 173–189 (2025), DOI: 10.3897/neobiota.102.145644 Andrey N. Reshetnikov et al.: Interactions of two global aquatic invaders Weldon C, Villiers ALD, Du Preez LH (2007) Quantification of the trade in Xenopus laevis from South Africa, with implications for biodiversity conservation. African Journal of Herpetology 56: 77–83. https://doi.org/10.1080/21564574.2007.9635553 Wickham H (2016) Data analysis. Springer, 189–201. https://doi.org/10.1007/978-3-319-24277-4_9 Wilson RS, James RS, Johnston IA (2000) Thermal acclimation of locomotor performance in tadpoles and adults of the aquatic frog Xenopus laevis. Journal of Comparative Physiology. B, Biochemical, Systemic, and Environmental Physiology 170: 117–124. https://doi.org/10.1007/ s003600050266 Wooten MC, Scribner KT, Smith MH (1988) Genetic variability and systematics of Gambusia in the southeastern United States. Copeia 1988: 283–289. https://doi.org/10.2307/1445867 Supplementary material 1 Additional images Authors: Andrey N. Reshetnikov, Arina O. Raldugina, Dmitriy V. Grinchenko, Artem A. Kidov, Nikita G. Platonov, Andrey B. Petrovskiy Data type: pdf Explanation note: fig. S1. Comparative consumption (means ± SE) of different ontogenetic stages of the clawed frog Xenopus laevis by Gambusia holbrooki (N = 28) in 1st and 2nd rounds of the experiment. fig. S2. Comparative consumption (means ± SE) of different ontogenetic stages of the clawed frog Xenopus laevis by females (N = 14) and males (N = 14) of the mosquitofish Gambusia holbrooki. fig. S3. Comparative consumption (means ± SE) of embryos and hatchlings (direct comparison) of the clawed frog Xenopus laevis for the mosquitofish Gambusia holbrooki (N = 28). 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.102.145644.suppl1