Abstract
Biological invasions can favour rapid changes in intraspecific competitive mechanisms such as cannibalism by imposing novel evolutionary pressures. For example, cane toad (Rhinella marina) tadpoles are highly cannibalistic on eggs and hatchlings in their invasive range in Australia, but not in their native range in South America. Whether such changes in cannibalism occur in invasive populations of other amphibian species is unknown. To explore this question, we collected wild-laid egg clutches of Japanese common toads (Bufo japonicus) from native and invasive populations in Japan, and conducted laboratory experiments to examine cannibalism responses. Contrary to the Australian system, we found that invasion has been accompanied by reduced cannibalistic tendency of B. japonicus tadpoles. This reduction has occurred despite invasive-range B. japonicus eggs/hatchlings being more vulnerable than native-range B. japonicus eggs/hatchlings to cannibalism by native-range conspecific tadpoles, and to predation by native-range frog tadpoles. Our findings thus support the idea that biological invasions can generate rapid changes in rates of cannibalism, but also show that decreases as well as increases can occur. Future work could investigate the proximate cues and selective forces responsible for this rapid decrease in rates of cannibalism by tadpoles in an invasive B. japonicus population.
Similar content being viewed by others
Introduction
Biological invasions offer a unique opportunity to examine rapid adaptive changes induced by exposure to new challenges (e.g.1,2,3,4). A species extending its range into new, unoccupied areas may encounter a suite of novel predators, prey, parasites, competitors and abiotic extremes5,6. In response to the opportunities and risks posed by those new interactions, invaders sometimes exhibit remarkably rapid evolved shifts in morphology, physiology, and behaviour7.
Some of the best examples of rapidly evolved changes in invasive species come from studies on a South American anuran (the cane toad, Rhinella marina) that was introduced to Australia in 1935 in an ill-considered attempt to control insect pests8. Although the toad invasion of Australia has been in progress for less than a century, the animals have already expanded their range by thousands of kilometres9. In the process of that accelerating range expansion, cane toads have accumulated a wide range of heritable differences in traits that affect locomotor ability and resilience to novel abiotic challenges, both within invasive-range populations (range core vs. range edge) and between native-range versus invasive-range populations10,11,12,13.
One of the most spectacular examples involves the evolution of cannibalism. In the cane toad’s native range in South America, tadpoles rarely consume conspecific eggs or hatchlings14. In contrast, tadpoles from all Australian populations tested to date are voracious cannibals, actively searching out and consuming hatchings as they emerge from the egg string14,15. This predation is species-specific; cane toad tadpoles ignore the eggs of other anuran species unless they are simultaneously exposed to the chemical cues (bufadienolide toxins) that are present in conspecific eggs and hatchlings16. Cannibalism rate is positively correlated with population density in other taxa17, and the evolution of intense cannibalism in Australia has been attributed to higher abundances of cane toads in the invasive range than in the native range, placing a selective premium on traits that maximise intraspecific competitive ability15. Consistent with that hypothesis, laboratory and outdoor mesocosm experiments show that cannibalism enhances larval viability of invasive cane toads in Australia by reducing competitor abundance18.
The greatest challenge to broadly understanding the evolution of cannibalism in invasive amphibian species is the paucity of data. For example, tadpoles of the American bullfrog (Rana (Lithobates) catesbeiana) are known to cannibalise conspecific eggs in the species’ invasive range in China19, however, whether this represents altered cannibalism rate post-invasion is unknown due to the lack of comparable cannibalism data from the species’ native range. To our knowledge, only one amphibian species (the cane toad) has been tested for differences in cannibalism in native versus invasive populations14. To assess the generality of these results, we need to examine other invasive amphibian species. To add to this area of research, we compared rates of cannibalism by a toad species (Japanese common toad, Bufo japonicus) that has been translocated from the island of Honshu (native range) to the northern island of Hokkaido (invasive range). The translocation occurred around 100 years ago, similar to the timescale of the invasion of Australia by cane toads20,21. Like Australia, Hokkaido had no native bufonid species prior to the invader’s arrival20,21. These similarities between the two bufonid species which share life histories22, provide a unique opportunity to test whether (1) invasion affects intensity of cannibalism among amphibians in the same ways, and (2) if any such changes are due to evolved shifts in vulnerability of eggs and hatchlings versus predatory behaviour of tadpoles.
Materials and methods
Study species and area
Japanese common toads lay their eggs in lentic waterbodies and the tadpoles develop in the aquatic environment23,24,25. The shared use of waterbodies as breeding sites with other amphibian species across their range makes B. japonicus eggs and hatchlings vulnerable to predation by a variety of anuran tadpoles and salamander larvae20,24,26,27,28. Cannibalism by conspecifics is also possible because, although the breeding period of B. japonicus is brief (up to 3 weeks on Honshu26,27), eggs hatch after 1 week20, giving tadpoles of early clutches a short window of time to prey upon the offspring of later-breeding conspecifics. Hatchlings typically experience high levels of predation pressure by aquatic predators prior to attaining locomotor capacity20,28. The eggs and hatchlings of B. japonicus contain maternally-invested toxins29,30; such toxins often act as a deterrent against aquatic predators31. However, amphibians that are native to Honshu and hence, have a long history of sympatry with B. japonicus, readily consume invasive-range B. japonicus hatchlings and can tolerate those toxins, including predatory frog tadpoles (Rana ornativentris) and salamander larvae (Hynobius nigrescens)20,21.
In contrast, B. japonicus is toxic to amphibians within its invasive range, including on Hokkaido, as these species have no evolutionary history of exposure to bufonids20,21. On Hokkaido, tadpoles of Rana pirica consume invasive B. japonicus hatchlings, but are almost always killed by the toxins20,21,28,30. Native Hokkaido salamander larvae (Hynobius retardus) also consume invasive B. japonicus hatchlings, with survival rates varying from 6 to 77% among populations20,28. Predation rates by these native amphibian species on Hokkaido can be high. For example, 90% of H. retardus larvae that were offered an invasive-range (Hokkaido) B. japonicus hatchling consumed the hatchling, whereas 49% of R. pirica tadpoles consumed the hatchling28.
Experimental design
We conducted laboratory experiments using wild-collected clutches to investigate predation by tadpoles on eggs and hatchlings for two B. japonicus subspecies (B. j. formosus and B. j. japonicus) and three species of Ranidae (Rana japonica, R. ornativentris, R. dybowski). In the experiments, both Bufo tadpoles and Rana tadpoles were tested with eggs and hatchlings of Bufo and Rana species. The two Bufo subspecies did not differ significantly in their predatory responses or vulnerability to predation within the native range (see Results). Therefore, we combined data for these two subspecies, and refer to them collectively as B. japonicus for brevity.
Three of our study species (B. japonicus, R. japonica, R. ornativentris) are native to the main islands of Honshu, Shikoku, and Kyushu. Bufo japonicus (specifically, B. j. formosus) has been introduced to Hokkaido as well as to Sadogashima Island and the Izu Islands32. Rana dybowski is native to Tsushima Island, located between Kyushu and the Korean Peninsula, and is not sympatric with B. japonicus (indeed, this island has no native Bufonid species).
Collection and husbandry of eggs and tadpoles
We hand-collected egg clutches from waterbodies for use in experiments (1 to 3 clutches per species). Egg clutches were identified to species level by one of us (TH) based on experience. Native-range eggs were collected from Honshu (Tochigi Prefecture: B. japonicus, R. japonica, R. ornativentris; Okayama Prefecture: B. japonicus, R. ornativentris; Wakayama Prefecture: R. ornativentris), Kyushu (Miyazaki Prefecture: B. japonicus) and Tsushima Island (Nagasaki Prefecture: R. dybowski). Invasive-range B. japonicus eggs were collected from Hakodate, Sapporo and Tsukigata on Hokkaido. All eggs were collected early on the morning of deposition. Most were collected from separate ponds, although eggs of B. japonicus and R. japonica were sometimes collected from the same pond. For frogs, we collected entire egg clutches because it was difficult to collect a sub-sample of eggs without damaging the egg mass. For toads, we collected a subsample of eggs because egg strings could be easily broken into segments.
Eggs were transported in plastic containers (14 cm × 18 cm × 8 cm, filled with pond water) by car, train or plane to the laboratory (Seto Marine Research Station of Kyoto University, and Rakuno Gakuen University) where they hatched. Once hatched, tadpoles were reared in 120 L tanks (66 cm × 86 cm × 34 cm; 1 clutch of ~ 50 tadpoles per tank) filled with aged water. Water was changed every 3 days, and tadpoles were fed a diet of algal pellets (Hikari Algal Wafers, Kyorin) ad libitum daily until used in experiments. We used the early-collected clutches to generate tadpoles as predators, and the later-collected clutches to obtain eggs and hatchlings as prey.
Laboratory experiments
We conducted a series of laboratory experiments to investigate (1) predation among native-range toads and frogs, (2) predation among native-range and invasive-range toads and frogs, and (3) cannibalism among native-range and invasive-range toads. Experiments were conducted using 1000 ml plastic containers filled with 750 ml aged water. In each experiment, a single tadpole (mid-developmental stage: stage 30 to 3533) was randomly allocated to either a control container or an egg treatment container. Tadpoles in egg treatment containers were offered either 5 or 10 anuran eggs (depending on availability) that were placed in the centre of containers. Tadpoles in control containers were fed cat food ad libitum daily for sustenance to ensure any tadpole mortality in control containers was not confounded by lack of food. All eggs were added to egg treatment containers within 24 to 36 h after collection, and thus were of comparable developmental stages for all experiments.
We directly observed predation on eggs and hatchlings during experiments, and recorded the number of eggs or hatchlings eaten by each tadpole every 24 h for a period of 72 h (at which time eggs had developed through the hatchling stage and into free-swimming tadpoles and were no longer vulnerable to predation). We recorded water temperature in each container daily using a thermometer (Takara Thermister D619). We also recorded tadpole mortality daily.
Experiment 1. Predation among native-range toads and frogs
We assessed predation rates by tadpoles on eggs and hatchlings for native-range toad (Bufo) and native-range frog (Rana) species to determine the role that tadpole phylogeny and egg/hatchling phylogeny play in the outcome of predator–prey interactions within the native range (i.e., predation by Bufonidae tadpoles vs. Ranidae tadpoles, predation on Bufonidae eggs/hatchlings vs. Ranidae eggs/hatchlings). For these experiments, we tested one toad species (B. j. japonicus and B. j. formosus combined as B. japonicus, as described above) and three frog species (R. japonica, R. ornativentris, R. dybowski; Table 1).
Experiment 2. Predation among native-range and invasive-range toads and frogs
We tested whether the effect of tadpole phylogeny (Bufonidae vs. Ranidae) on predation varies between native versus invasive populations of B. japonicus eggs/hatchlings. We did this in two ways. First, we combined the data for all native-range tadpoles tested with invasive-range B. japonicus eggs/hatchlings (B. japonicus tadpoles N = 45, R. japonica tadpoles N = 18, R. ornativentris tadpoles N = 30) and compared these data to predation by native-range tadpoles on native-range B. japonicus eggs/hatchlings (listed in Table 1). Secondly, we specifically compared the predatory responses of tadpoles of two native-range species to native-range B. japonicus eggs/hatchlings versus invasive-range B. japonicus eggs/hatchlings (respectively, B. japonicus tadpoles N = 30 vs. 45, R. ornativentris tadpoles N = 45 vs. 30).
Experiment 3. Cannibalism among native-range and invasive-range toads
We investigated whether propensity for cannibalism by B. japonicus tadpoles varies with invasion history by comparing consumption of invasive-range B. japonicus eggs/hatchlings by native-range B. japonicus tadpoles (N = 45) versus invasive-range B. japonicus tadpoles (N = 30).
Statistical analyses
General methods
We first discuss methods applicable to all statistical analyses that we conducted. Following this, we discuss statistical methods specific to each experiment.
We analysed predation data in R34 as a binomial response to treatment (egg/hatchling eaten vs. not eaten) using logistic regression35 and quasi-binomial models to account for data over-dispersion (mixed effects models: package MASS:glmmPQL36 followed by Anova (package car37)). We conducted post-hoc multiple comparisons among treatments using Tukey tests adjusted with the Holm method (package multcomp38). Water temperature and time were included as covariates in all models. Because water temperature values are continuous data, we centred these data on mean values for the dataset in question prior to analysis. Container was included as a random effect in all models to account for non-independence of observations of the same container over time.
In some instances, there was zero predation in all replicates within a treatment, resulting in models failing to reach convergence. When this occurred, we assigned a single egg/hatchling to have been eaten in that treatment to obtain a conservative estimate of treatment effect35. For models that included an interaction term, we removed the interaction term when it was non-significant and re-ran the model to obtain final estimates. We retained all main effects in the final models, regardless of their statistical significance. Tadpoles that died in egg/hatchling treatment containers (N = 3 toad tadpoles) were not included in statistical analyses assessing predation rate. Specific details for data analysis for each experiment are listed below.
Experiment 1. Predation among native-range toads and frogs
We analysed overall effects of phylogeny on predation using fixed effects of tadpole phylogeny (Bufonidae (toad) vs. Ranidae (frog)), egg/hatchling phylogeny (Bufonidae vs. Ranidae) and their interaction (tadpole phylogeny × egg/hatchling phylogeny). We then conducted a multiple comparisons test to identify differences among all four combinations of the tadpole plus egg/hatchling treatments (i.e., toad tadpole plus toad eggs/hatchlings, toad tadpole plus frog eggs/hatchlings, frog tadpole plus toad eggs/hatchlings, frog tadpole plus frog eggs/hatchlings).
Experiment 2. Predation among native-range and invasive-range toads and frogs
We combined the data for all native-range tadpoles tested with invasive-range B. japonicus eggs/hatchlings and compared these data to predation by native-range tadpoles on native-range B. japonicus eggs/hatchlings. For this analysis, we used egg/hatchling population (native-range vs. invasive-range), tadpole phylogeny (Bufonidae vs. Ranidae) and their interaction (egg/hatchling population × tadpole phylogeny) as fixed main effects.
We then conducted a multiple comparison test to assess differences in predation among all four of the tadpole plus egg/hatchling treatments (i.e., native-range toad tadpole plus native-range toad eggs/hatchlings, native-range toad tadpole plus invasive-range toad eggs/hatchlings, native-range frog tadpole plus native-range toad eggs/hatchlings, native-range frog tadpole plus invasive-range toad eggs/hatchlings).
For native-range tadpoles of B. japonicus and R. ornativentris, we conducted separate analyses to compare predation by these tadpoles on native-range B. japonicus eggs/hatchlings versus invasive-range B. japonicus eggs/hatchlings. These analyses used B. japonicus egg/hatchling population (native-range vs. invasive-range) as a fixed effect.
Experiment 3. Cannibalism among native-range and invasive-range toads
We analysed cannibalism on invasive-range B. japonicus eggs/hatchlings by native-range B. japonicus tadpoles versus invasive-range B. japonicus tadpoles using the fixed effect of B. japonicus tadpole population (native-range vs. invasive-range).
Ethics approval
All procedures were approved by Rakuno Gakuen University Animal Care Committee (permit #DH22D8). This study was carried out in compliance with the ARRIVE guidelines, and all methods were carried out in accordance with relevant guidelines and regulations.
Results
Comparison of two subspecies of toads from the native range
We tested native-range B. j. formosus tadpoles versus native-range B. j. japonicus tadpoles as predators on eggs/hatchlings of three anuran taxa (B. j. japonicus, R. japonica, R. ornativentris). There was no significant effect of B. japonicus subspecies on rate of egg/hatchling consumption by tadpoles (logistic regression: B. j. japonicus eggs/hatchlings t = 0.000, df = 13, p = 1.000, R. japonica eggs/hatchlings t = − 0.855, df = 22, p = 0.402, R. ornativentris eggs/hatchlings t = -1.929, df = 22, p = 0.067). There was similarly no difference in vulnerability to predation for native-range B. j. formosus eggs/hatchlings versus native-range B. j. japonicus eggs/hatchlings: none of these eggs or hatchlings were eaten by any native-range Bufo or Rana tadpoles. On this basis, we combined data for the two B. japonicus subspecies for subsequent analyses and refer to them as B. japonicus (identified as native-range vs. invasive-range).
Effect of covariates on predation by tadpoles on eggs/hatchlings
Time was a significant covariate in most predation models, while the effect of water temperature on consumption of eggs/hatchlings by tadpoles was more variable (Tables 2, 3, 5, 7, 8).
Experiment 1. Predation among native-range toads and frogs
Both tadpole phylogeny and egg/hatchling phylogeny were significant predictors of rates of predation for native-range species (Table 2, Fig. 1). Overall, frog tadpoles were more likely to eat eggs/hatchlings (of all kinds tested) than were toad tadpoles (Table 2, Fig. 1), and frog eggs/hatchlings were more likely to be eaten by tadpoles (of all kinds tested) than were toad eggs/hatchlings (Table 2, Fig. 1). The tadpole phylogeny x egg/hatchling phylogeny interaction was non-significant (Chi-square = 2.54, df = 1, p = 0.11).
Overall, predation varied significantly among the four native-range tadpole plus egg/hatchling treatment groups (Table 3, Fig. 2). No toad or frog tadpoles ate any toad eggs/hatchlings, whereas predation rate by toad tadpoles on frog eggs/hatchlings was low (proportion eggs/hatchlings eaten = 0.05, Fig. 2: respectively A, B, C). The greatest predation rate was by frog tadpoles on frog eggs/hatchlings, with frog tadpoles eating more frog eggs/hatchlings (proportion eaten = 0.37) than toad eggs/hatchlings (proportion eaten = 0.00, Table 4 and Fig. 2: D vs. B). Predation by frog tadpoles on frog eggs/hatchlings was also more common than predation by toad tadpoles on either frog eggs/hatchlings (Table 4 and Fig. 2: D vs. C) or toad eggs/hatchlings (Table 4 and Fig. 2: D vs. A).
Multiple comparison of tadpole plus egg/hatchling combinations for native-range species. A = toad (Bufo) tadpole vs. Bufo eggs/hatchlings, B = frog (Rana) tadpole vs. Bufo eggs/hatchlings, C = Bufo tadpole vs. Rana eggs/hatchlings, D = Rana tadpole vs. Rana eggs/hatchlings. Data plotted are number of eggs and hatchlings consumed after 72 h.
Experiment 2. Predation among native-range and invasive-range toads and frogs
Overall, B. japonicus egg/hatchling population (native-range vs. invasive-range) was a significant predictor for the rate of predation by native-range tadpoles (Table 5, Fig. 3). Tadpole phylogeny (Bufonidae vs. Ranidae) was also a significant main effect (Table 5). The tadpole phylogeny × egg/hatchling population interaction was non-significant (Chi-square = 2.059, df = 1, p = 0.1513).
Specific comparisons among the tadpole plus egg/hatchling treatments showed that tadpole phylogeny was not a significant predictor of predation by native-range tadpoles on native-range B. japonicus eggs/hatchlings (Table 6 and Fig. 3: B vs. A) but was a significant predictor for predation by native-range tadpoles on invasive-range B. japonicus eggs/hatchlings (Table 6 and Fig. 3: D vs. C). Native-range frog tadpoles ate more invasive-range B. japonicus eggs/hatchlings than did native-range toad tadpoles (Table 6 and Fig. 3: D vs. C).
In addition to these phylogenetic effects, both native-range frog tadpoles and native-range toad tadpoles ate more invasive-range B. japonicus eggs/hatchlings than native-range B. japonicus eggs/hatchlings (Table 6 and Fig. 3: respectively, B vs. D, A vs. C).
Individual species comparisons showed that native-range toad tadpoles and frog tadpoles both ate more invasive-range B. japonicus eggs/hatchlings than native-range B. japonicus eggs/hatchlings (Table 7, Figs. 4 and 5) (Table 7).
Experiment 3. Cannibalism among native-range and-invasive range toads
Native-range B. japonicus tadpoles ate more invasive-range B. japonicus eggs/hatchlings than did invasive-range B. japonicus tadpoles (Table 8, Fig. 6).
Tadpole mortality
We did not formally analyse tadpole mortality due to low mortality rates and an obvious lack of treatment effects. During our experiments, 5 B. japonicus tadpoles in control containers died, 1 B. japonicus tadpole offered frog eggs/hatchlings died without any evidence of predation, and 2 B. japonicus tadpoles offered invasive-range B. japonicus eggs/hatchlings died after eating either 0 or 1 egg/hatchling.
Discussion
Within native-range populations, overall rates of predation (both intra- and interspecific) were higher in frogs (Ranidae) than in toads (Bufonidae), demonstrating that frog tadpoles likely play a greater role as predators in structuring native larval anuran communities. For B. japonicus, cannibalism was less common in the invasive-range population on Hokkaido than in native-range populations. This post-invasion trend contrasts with another bufonid species (the cane toad) in Australia, and the harlequin ladybird in Europe, both of which exhibit higher rates of cannibalism in invasive populations compared to native populations14,17. Although cannibalism may be beneficial for species as they colonise new environments14,17, our results demonstrate that invasions do not impact cannibalism responses of all species in the same manner. For B. japonicus, invasive populations exist not only on Hokkaido, but also on Sadogashima Island and the Izu Islands32. Whether these other invasive B. japonicus populations also exhibit reduced propensity for cannibalism is unknown and warrants further investigation.
The lower frequency of cannibalism in the invasive B. japonicus population on Hokkaido was due to characteristics of the toad tadpoles and not the toad eggs/hatchlings. Tadpoles of invasive-range B. japonicus had a lower propensity for cannibalism (on eggs and hatchlings from the invasive-range population) than did the tadpoles of native-range B. japonicus. In contrast, the eggs and hatchlings of invasive-range B. japonicus were more prone to being eaten, by both native-range B. japonicus tadpoles and native-range frog tadpoles, than eggs and hatchlings of native-range B. japonicus. That is, invasive-range B. japonicus tadpoles were less likely to exhibit cannibalistic behaviour, but invasive-range B. japonicus eggs/hatchlings were more prone to being cannibalised (and eaten by heterospecific tadpoles), compared to native-range B. japonicus populations.
We found that native-range tadpoles (B. japonicus, R. japonica, R. ornativentris) consumed eggs and hatchlings of invasive-range B. japonicus without ill effect, despite these life history stages possessing toxins20,21,28,29,30. This accords with previous studies that found native-range amphibian species on Honshu (R. ornativentris tadpoles, H. nigrescens larvae) consume invasive-range B. japonicus hatchlings without dying20,21. This ability to tolerate B. japonicus toxins is likely due to co-evolutionary adaptation on Honshu20,21. However, as far as we are aware, our study is the first to examine cannibalism responses on eggs/hatchlings for either native-range or invasive-range B. japonicus tadpoles.
Our experimental design does not identify the proximate cues involved in the decreased rates of cannibalism in the invasive population of B. japonicus on Hokkaido. Detailed studies in Australian cane toads have shown that toxins released from eggs close to the time of hatching attract cannibalistic tadpoles, and also induce foraging responses by those tadpoles16,39,40. Invasive American bullfrog tadpoles in China also are attracted to chemical signals from conspecific eggs, although the precise cue remains unknown19. A similar situation to cane toads is plausible with B. japonicus, whose eggs and hatchlings also contain toxins20,21,28,29,30. The decrease in rate of cannibalism in the invasive population of B. japonicus thus might reflect a lower attraction response by conspecific tadpoles to such cues, or a change in the strength of toxin cues such as a decrease in toxin content, a change in toxin composition, or a decrease in rate of toxin release by eggs/hatchlings. Laboratory experiments exposing predators to specific scent cues rather than entire eggs/hatchlings could test those ideas (cf.39,40), and direct measures of the types and amounts of toxins in eggs and hatchlings from each B. japonicus population also would be informative (cf.40,41). Adult B. japonicus exhibit geographic variation in toxin (bufadienolide) composition within their native range42. Because the toxins in B. japonicus eggs and hatchlings are maternally-invested29,30, similar geographic variation presumably also occurs in the toxin composition of B. japonicus eggs and hatchlings. However, whether the toxin composition of B. japonicus in the invasive range differs from that of native-range populations remains to be determined. Interestingly, the fact that native-range B. japonicus tadpoles readily consumed eggs and hatchlings of invasive-range B. japonicus suggests that factors other than geographic variation in toxin composition may also be involved in causing lower rates of cannibalism by invasive-range B. japonicus tadpoles.
The evolutionary forces responsible for the decline in propensity for cannibalism in invasive B. japonicus tadpoles warrant further study, given that invasive cane toad tadpoles in Australian exhibit the opposite pattern14,15. The colder climate of Hokkaido (the invasive range) than the native range may affect temporal overlap in contact between conspecific tadpoles and eggs/hatchlings. For example, if breeding is restricted to a brief period in spring, and tadpoles metamorphose during the following summer, then there would be few opportunities for older tadpoles to consume newly-laid eggs and hatchlings. In contrast, a longer breeding season may generate within-year contact between tadpoles and eggs/hatchlings, increasing opportunities for cannibalism. Field studies could clarify that phenology in native versus invasive ranges.
The prevalence of cannibalism in populations also covaries with density and the occurrence of other predators15,43. For example, the evolution of cannibalism in Australian cane toads has been attributed to higher densities in the invasive range than in the native range, coupled with a scarcity of alternative predators on eggs and hatchlings in the invasive range14. If similar selective pressures are responsible for patterns of cannibalism for B. japonicus in Japan, we would predict lower densities and/or higher alternate predation pressure in the invasive range of Hokkaido versus the native range. To examine these ideas, we need field data on densities of toads in their native range and invasive range, and estimates of rates of egg and hatchling mortality due to intraspecific versus interspecific predation. Field enclosures in natural waterbodies may provide a way to measure these variables14,15. The availability and nutritional value of alternative food sources for tadpoles also deserve attention, as these factors can influence cannibalism rates44.
Although we found a decrease rather than increase in rates of cannibalism in the invasive population of B. japonicus on Hokkaido, contrary to the situation with cane toads in Australia14,15, it is striking that in both cases a recent invasion (~ 100 years) has initiated a substantial shift in the importance of cannibalism. That similarity supports the idea that invasions impose rapid shifts on the selective forces involved in intraspecific competition generally, and cannibalism specifically. The proximate cues and adaptive significance of such shifts are amenable to studies in the laboratory and in the field, rendering cannibalism in toads an ideal study system in which to examine rapid evolutionary change.
Data availability
Data is available from Dryad Digital Repository https://datadryad.org/stash/share/Szhq4YDLypxYit13FGv-J6JQ6LAdtf3lCKsTBp3oZAk.
References
Huey, R. B., Gilchrist, G. W., Carlson, M. L., Berrigan, D. & Serra, L. Rapid evolution of a geographic cline in size in an introduced fly. Science 287, 308–309 (2000).
Butin, E., Porter, A. H. & Elkinton, J. Adaptation during biological invasions and the case of Adelges tsugae. Evol. Ecol. Res. 7, 887–900 (2005).
Prentis, P. J., Wilson, J. R., Dormontt, E. E., Richardson, D. M. & Lowe, A. J. Adaptive evolution in invasive species. Trends Plant Sci. 13, 288–294 (2008).
Whitney, K. D. & Gabler, C. A. Rapid evolution in introduced species, ‘invasive traits’ and recipient communities: Challenges for predicting invasive potential. Divers. Distrib. 14, 569–580 (2008).
Moran, E. V. & Alexander, J. M. Evolutionary responses to global change: Lessons from invasive species. Ecol. Lett. 17, 637–649 (2014).
Borden, J. B. & Flory, S. L. Urban evolution of invasive species. Front. Ecol. Environ. 19, 184–191 (2021).
Rollins, L. A., Richardson, M. F. & Shine, R. A genetic perspective on rapid evolution in cane toads (Rhinella marina). Mol. Ecol. 24, 2264–2276 (2015).
Shine, R. Cane Toad Wars (University of California Press, 2018).
Urban, M. C., Phillips, B. L., Skelly, D. K. & Shine, R. A toad more travelled: The heterogeneous invasion dynamics of cane toads in Australia. Am. Nat. 171, E134–E148 (2008).
Hudson, C. M., McCurry, M. R., Lundgren, P., McHenry, C. R. & Shine, R. Constructing an invasion machine: The rapid evolution of a dispersal-enhancing phenotype during the cane toad invasion of Australia. PLoS ONE 11, e0156950 (2016).
Gruber, J., Brown, G., Whiting, M. J. & Shine, R. Is the behavioural divergence between range-core and range-edge populations of cane toads (Rhinella marina) due to evolutionary change or developmental plasticity?. R. Soc. Open Sci. 4, 170789 (2017).
Kosmala, G., Christian, K., Brown, G. & Shine, R. Locomotor performance of cane toads differs between native-range and invasive populations. R. Soc. Open Sci. 4, 170517 (2017).
Kosmala, G. K., Brown, G. P., Shine, R. & Christian, K. Skin resistance to water gain and loss has changed in cane toads (Rhinella marina) during their Australian invasion. Ecol. Evol. 10, 13071–13079 (2020).
DeVore, J. L., Crossland, M. R., Shine, R. & Ducatez, S. The evolution of targeted cannibalism and cannibal-induced defenses in invasive populations of cane toads. Proc. Natl. Acad. Sci. 118, e2100765118 (2021).
DeVore, J. L., Crossland, M. R. & Shine, R. Trade-offs affect the adaptive value of plasticity: Stronger cannibal-induced defenses incur greater costs in toad larvae. Ecol. Monogr. 91, e01426 (2021).
Crossland, M. R., Shine, R. & DeVore, J. L. Choosy cannibals: Targeted consumption of conspecific hatchlings by larval cane toads is triggered by species-specific defensive toxins. Ecol. Evol. 12, e8655 (2022).
Tayeh, A. et al. Cannibalism in invasive, native and biocontrol populations of the harlequin ladybird. BMC Evol. Biol. 14, 15 (2014).
Crossland, M. R., Hearnden, M. N., Pizzatto, L., Alford, R. A. & Shine, R. Why be a cannibal? The benefits to cane toad, Rhinella marina [= Bufo marinus], tadpoles of consuming conspecific eggs. Anim. Behav. 82, 775–782 (2011).
Xu, C. et al. Foraging responses of the larvae of invasive bullfrogs (Lithobates catesbeianus): Possible implications for bullfrog control and ecological impact in China. Asian Herpetol. Res. 8, 253–261 (2017).
Oyake, N. et al. Comparison of susceptibility to a toxic alien toad (Bufo japonicus formosus) between predators in its native and invaded ranges. Freshw. Biol. 65, 240–252 (2020).
Okamiya, H., Inoue, Y., Takai, K., Crossland, M. R. & Kishida, O. Native frogs (Rana pirica) do not respond adaptively to alien toads (Bufo japonicus formosus) 100 years after introduction. Ecol. Res. 36, 1005–1014 (2021).
Haas, A. Phylogeny of frogs as inferred from primarily larval characters (Amphibia: Anura). Cladistics 19, 23–89 (2003).
Zheng, X. et al. Effects of multiple stressors on amphibian oviposition: Landscape and local determinants in central Japan. Ecol. Ind. 128, e107824 (2021).
Haramura, T., Eto, K., Crossland, M. R., Nishikawa, K. & Shine, R. Competition between the tadpoles of Japanese toads versus frogs. Sci. Rep. 12, 1627 (2022).
Tanaka, K. & Ishikawa, K. Developmental stages of lotic-breeding toad, Bufo torrenticola, with a comparison to lentic-breeding B. japonicus formosus (Amphibia: Anura: Bufonidae). Curr. Herpetol. 41, 8–23 (2022).
Kusano, T., Maruyama, K. & Kanenko, S. Breeding site fidelity in the Japanese toad, Bufo japonicus formosus. Herpetol. J. 9, 9–13 (1999).
Kusano, T., Miura, T., Terui, S. & Maruyama, K. Factors affecting the breeding activity of the Japanese common toad, Bufo japonicus formosus (Amphibia: Bufonidae) with special reference to the lunar cycle. Curr. Herpetol. 34, 101–111 (2015).
Kazila, E. & Kishida, O. Foraging traits of native predators determine their vulnerability to a toxic alien prey. Freshw. Biol. 64, 56–70 (2019).
Okamiya, H., Takai, K. & Kishida, O. The Japanese common toad, Bufo japonicus formosus, contains toxin in the egg stage. Curr. Herpetol. 40, 103–106 (2021).
Okamiya, H., Tagami, M., Crossland, M. & Kishida, O. Are toxic effects of alien species affected by their prey? Evaluation by bioassay with captive-bred toad embryos and a vulnerable predator. Hydrobiologia 848, 4445–4452 (2021).
Üveges, B. et al. Chemical defense of toad tadpoles under risk by four predator species. Ecol. Evol. 9, 6287–6299 (2019).
Maeda, N. & Matsui, M. Frogs And Toads Of Japan. Bun-ichi Sogo Shuppan (in Japanese) (1989).
Gosner, K. L. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16, 183–190 (1960).
R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2021).
Warton, D. I. & Hui, F. K. C. The arcsine is asinine: The analysis of proportions in ecology. Ecology 92, 3–10 (2011).
Venables, W. N. & Ripley, B. D. Modern Applied Statistics with S 4th edn. (Springer, 2002).
Fox, J. & Weisberg, S. An R Companion to Applied Regression 3rd edn. (Sage, 2019).
Hothorn, T., Bretz, F. & Westfall, P. Simultaneous inference in general parametric models. Biom. J. 50, 346–363 (2008).
Crossland, M. R., Haramura, T., Salim, A. A., Capon, R. J. & Shine, R. Exploiting intraspecific competitive mechanisms to control invasive cane toads (Rhinella marina). Proc. R. Soc. B 279, 3436–3442 (2012).
Crossland, M. R., Salim, A. A., Capon, R. J. & Shine, R. Chemical cues that attract cannibalistic cane toad (Rhinella marina) larvae to vulnerable embryos. Sci. Rep. 11, 12527 (2021).
Hayes, R. A., Crossland, M. R., Hagman, M., Capon, R. J. & Shine, R. Ontogenetic variation in the chemical defenses of cane toads (Bufo marinus): Toxin profiles and effects on predators. J. Chem. Ecol. 35, 391–399 (2009).
Inoue, T. et al. Variation in bufadienolide composition of parotoid gland secretion from three taxa of Japanese toads. J. Chem. Ecol. 46, 997–1009 (2020).
Richardson, M. L., Mitchell, R. F., Reagel, P. F. & Hanks, L. M. Causes and consequences of cannibalism in noncarnivorous insects. Annu. Rev. Entomol. 55, 39–53 (2010).
Gould, J., Clulow, J. & Clulow, S. Food, not friend: Tadpoles of the sandpaper frog (Lechriodus fletcheri) cannibalise conspecific eggs as a food resource in ephemeral pools. Ethology 126, 486–491 (2020).
Acknowledgements
We thank Koshiro Eto (Kitakyushu Museum of Natural History & Human History) for assisting with collection of samples on Tsushima Island. We also thank an anonymous reviewer for comments that improved the manuscript. This work was supported by an internal grant from The Hakubi Center, Kyoto University, and the Environment Research and Technology Development Fund (4RF-1402) of the Ministry of the Environment, Japan.
Author information
Authors and Affiliations
Contributions
All authors contributed equally to design of the study. T.H. conducted the experiments. M.R.C. and T.H. analysed the data. All authors contributed equally to writing the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Crossland, M.R., Shine, R. & Haramura, T. A biological invasion reduces rates of cannibalism by Japanese toad tadpoles. Sci Rep 13, 9587 (2023). https://doi.org/10.1038/s41598-023-36743-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-36743-8
- Springer Nature Limited