Can Seed Hardness Be an Ecological Filter in Seed Dispersal by Rodents?

Abstract

Seed hardness has been acknowledged as one of the most significant physical traits influencing seed consumption and caching by animals. From an evolutionary perspective, a hard seed shell should potentially be regarded as a dispersal and predator escape strategy rather than a maladaptive trait of plants. However, to our knowledge, this prediction has not been adequately tested in the context of seed dispersal. Here, we offered seeds with hard shells to the main seed consumers widely distributed in two distinct forest ecosystems to test our hypothesis that a hard seed shell might act as an ecological filter to deter predators but attract obligatory seed dispersers. Our studies demonstrated that seeds with the hardest shells consistently and directly deterred small-bodied seed consumers that have been proven to be either larder-hoarders or scatter-hoarders in the two forests. Nevertheless, rodents with the largest body size and seed handling capacity seemed to be effective seed dispersers targeting hard seed shells. The deterrence to seed predators and the attraction to an effective seed disperser reflects the evolutionary significance of seed hardness in the seed dispersal syndrome. Our studies in different forest ecosystems strongly suggest that a hard seed shell is not an evolutionary dead end in plant–animal interactions. On the contrary, the outcome of a hard seed shell in the seed dispersal syndrome is of evolutionary importance for plant–animal mutualistic interactions in various forest ecosystems.

1. Introduction

Seed dispersal appears to be the most crucial life history strategy of plants. During the past half-century, seed dispersal by animals has become an important subject within the realm of plant–animal interactions [1,2,3]. An increasing amount of the literature has shown that plants with large-sized seeds mainly depend on seed dispersal agents (such as rodents and birds) which consume their fruits or seeds to transport seeds far from the parent trees for the successful establishment and regeneration of seedlings [4,5,6,7]. To achieve successful seed dispersal, plants have therefore evolved various adaptive seed characteristics to guarantee successful movement and recruitment in appropriate habitats [8,9,10].

However, the manipulation of seeds by animals will result in diverse seed fate pathways in terms of the decision-making processes of animals [11,12]. The connection between food hoarding animals and seeds fluctuates between antagonism and conditional mutualism along a continuum [13]. Seeds can be dispersed by animals that potentially function as seed dispersers or be destroyed by those serving as pure seed predators [14,15,16], although seed predators will drive the evolution of seed defenses and favor an alternative seed dispersal mutualism [17]. Previous evidence has indicated that there is a conditional mutualism between plants and seed consumers [18]. Whether seed consumers act as antagonistic seed predators or mutualistic seed dispersers often depends on the physical and chemical characteristics of seeds (e.g., seed size/mass, nutrition contents, shell thickness, chemical defenses of seeds and seed crop size) as well as the functional traits of dispersers [5,19,20,21]. It has been discovered that large-sized seeds are more likely to be scatter-hoarded rather than eaten in situ at both inter- and intraspecific levels [22], though a few studies reveal distinct patterns (for instance, [20]). Recently, Cao et al. [23] offered empirical proof that medium-sized seeds might possess higher seed dispersal fitness compared with large and small seeds, suggesting the uncertainty of seed size in influencing seed dispersal fitness.

Apart from the size or mass of seeds, the hardness of seed shells has been regarded as one of the most significant physical characteristics influencing the removal and caching of seeds by animals [24]. A hard seed shell implies that seed-consuming animals need to invest a considerable amount of time to open and consume the physically protected seeds, as proposed by the seed handling time hypothesis [25,26]. Thus, seeds with hard or thick shells are more prone to be removed and then scatter-hoarded compared with those with thin or soft shells [27]. Nevertheless, as handling costs increase, seeds become less appealing to animals due to the decreased profitability and predation risks caused by physical barriers [28]. With the rise in seed hardness and seed handling cost, small-sized animals with weak masseters might encounter difficulties in cracking seeds with extremely hard shells and subsequently refuse to select or consume them. For instance, the scatter-hoarding animal, the Siberian chipmunk, Tamias sibiricus, completely refuses to select and eat nuts of Manchurian walnut Juglans mandshurica with extremely hard shells [29]. Another extreme case is that animals face challenges in detaching the convoluted cotyledons from the hard and tightly fitting shell of black walnut (J. nigra) [30]. Therefore, hard seed shells not only safeguard the seeds from animals that are not efficient agents of seed dispersal but may also incidentally discourage the legitimate seed dispersers [29,31].

However, from an evolutionary perspective, the hard shell of physically defended seeds should not merely be regarded as a maladaptive trait of plants. On the contrary, the evolution of seed hardness ought to be seen as a strategy for dispersal and predator evasion [32]. The partner selection between food hoarding animals and plants is largely dependent on the optimal matching of functional traits [33]. From the viewpoint of seed dispersal, the advantages of seed hardness might be evolutionarily significant and help account for the presence of plants with extremely hard seed coats in various forest ecosystems. Hence, it is highly likely that seeds with the hardest shell in a given ecosystem may function as an ecological filter and eventually retain one or two legitimate seed dispersers to make up for the loss of the majority of seed consumers. In this situation, it can be anticipated that the interactions between plants with hard shells and seed consumers will form a pairwise coevolutionary pattern, although most of them are manifested as diffuse coevolutionary patterns. Nevertheless, existing studies have mainly focused on seed dispersal networks or the functional response of seed consumers to interspecific variations in seed traits [34,35,36,37,38]. To the best of our knowledge, few studies have been carried out to illustrate how rodent communities respond to a given seed species [11,39], especially for those having hard seed shells. Therefore, the evolutionary significance of hard shells still needs to be emphasized in the seed dispersal syndrome. Here, we hypothesized that hard-coated seeds may exclude smaller or weaker-jawed rodents but select for larger-sized rodents to promote their seed dispersal, representing an evolutionary importance for plant–animal mutualistic interactions. To verify this hypothesis, we offered seeds with the hardest seed shells to the main seed consumers that were widely distributed in two distinct forest ecosystems in China. Our aim was to answer the question whether the hard shell plays a crucial role in determining seed predation and seed dispersal. Moreover, we would like to know whether hard shells will ecologically filter out all predators and reserve an effective scatter-hoarder to ensure effective seed dispersal.

2. Materials and Methods

2.1. Seed Species

This study was conducted in 2017. Nuts of Manchurian walnut Juglans mandshurica were collected in a temperate forest in Dailing, Heilongjiang province in northeastern China (45°58′ N, 129°08′ E). We collected nuts of Lithocarpus harlandii and Choerospondias axillaris in a subtropical evergreen forest in Dujiangyan, Sichuan province of southwestern China (31°04′ N, 103°42′ E). We chose C. axillaris because rodents were expected to participate in secondary dispersal of its nuts after endozoochory of frugivore (e.g., Paguma larvata and Muntiacus spp.) [40]. Nuts collected in this study were typically dependent on rodents for seed dispersal and represented the plant species bearing large seeds with the hardest shell in each of the three ecosystems (Table 1; Figure S1a,b).

2.2. Study Animals

To trap animals for enclosure experiments, SJL601 steel-framed live-traps (9 cm × 10 cm × 25 cm, manufactured by Sichuan Shujile Company, Chengdu China) baited with peanuts and carrots were placed in forests at 5 m intervals along four transects at 8:00 a.m. following the previous procedures [41]. All traps were pre-baited for 1 day and protected from predators by wrapping with steel mesh. Only adult rodents were chosen for the behavioral experiments. The captured animals were transported by vehicle to the animal housing room in 1 h. Animals were individually housed in the steel-frame cages (H × W × L = 20 cm × 40 cm × 50 cm) and maintained at a temperature of 20 ± 3 °C. Photoperiod in the animal room was set at 14:10 h light:dark. Bedding materials, tap water and peanuts were provided ad libitum.

Animal capture and handling in this study adhered to the ASAB/ABS Guidelines for the Use of Animals in Research. The authorization to capture, handle and maintain animals was issued by the Qufu Normal University. No animals died in the behavioral experiments and all individuals were released at the exact point of capture once experiments finished following health inspection.

In Dailing of Heilongjiang province, we captured 6 Sciurus vulgaris, 8 Tamias sibiricus, 8 Apodemus peninsulae and 6 Myodes rufocanus. We captured 15 Leopoldamys edwardsi, 10 A. draco, 15 A. chevrieri, 11 Niviventer confucianus and 7 N. fulvescens in Dujiangyan of Sichuan province. The captured animals represented the most abundant rodent species that handle and eat seeds in the study areas. Sciurus vulgaris and L. edwardsi represented the largest rodent species responsible for seed dispersal in the two forest ecosystems, respectively (Table 1; Figure S2).

2.3. Behavioral Protocols in Artificial Enclosure

After one week of acclimation in the housing room and 2 days in enclosures, the behavioral experiments were conducted in artificial enclosures measuring 10 m × 10 m, which were established in the treeless area near the forest [5]. Housing will not change behavioral activity of rodents. In each enclosure, an artificial nest made of wood and a bowl of drinking water were placed at one corner. In Dailing of Heilongjiang province, 20 tagged nuts of J. mandshurica were presented to each caged animal in each enclosure. Nuts were offered to S. vulgaris and T. sibiricus at 07:00 in the morning as they are diurnal animals, while seeds were provided to A. peninsulae and M. rufocanus at 19:00 in the evening since they are considered nocturnal rodents. As L. edwardsi, A. draco, A. chevrieri, N. confucianus and N. fulvescens are nocturnal rodents in Dujiangyan of Sichuan province, 40 nuts of L. harlandii were given to each individual at 19:00 in the evening, respectively. The same procedure was repeated using nuts of C. axillaris. After 10–13 h of handling by rodents, the focal animals were removed from the enclosures and returned to the cages in the animal housing room. Each animal was tested once for 1–2 days depending on animal species, which is supposed to be long enough for rodents to handle the seeds. Then, seed fates in each enclosure were checked and classified into intact in situ (IIS), eaten (E), scatter-hoarded (SH) and larder-hoarded (LH). By doing so, we expected to test which animal was deterred by hard seed shell and which animal was selected for legitimate seed dispersers of seeds with the hardest shell in each ecosystem.

2.4. Statistical Analyses

Statistical analyses were performed with the SPSS 16.0 Package. Proportions of seeds handled by rodents were arcsine-square-root transformed before statistical analysis to obtain variances and to approximate normal distributions of the model residuals. The proportion of seeds remaining intact in situ (IIS) was calculated for each rodent species in each study site. Then, a general linear model (GLM) was used to determine if rodents handled seeds differently in response to seed hardness. Accordingly, we calculated the proportions of E, SH and LH and ran the same analyses using GLM. The proportions of seeds with different fates (i.e., IIS, E, SH and LH) were dependent variables in each of the models. We also ran linear regression to see the correlation between rodent body mass and seed fates.

3. Results

Our results showed that S. vulgaris, with the largest body size in Dailing of Heilongjiang province, harvested more nuts of J. mandshurica released at seed stations in artificial enclosures than T. sibiricus, A. peninsulae and M. rufocanus (F = 144.982, df = 3, p < 0.001) (Figure 1). Even T. sibiricus, a with medium body size, were unable to select and consume nuts of J. mandshurica. Sciurus vulgaris was observed to be the only species that can eat nuts in situ among the four rodent species (F = 72.215, df = 3, p < 0.001) (Figure 1). As expected, we witnessed that S. vulgaris scatter-hoarded significantly more nuts of J. mandshurica than T. sibiricus, A. peninsulae and M. rufocanus (F = 94.187, df = 3, p < 0.001). Small-bodied A. peninsulae and M. rufocanus larder-hoarded a small proportion of nuts of J. mandshurica in the artificial nests (F = 13.121, df = 3, p < 0.001) (Figure 1).

In Dujiangyan of Sichuan province, the largest rodent species L. edwardsi harvested more nuts of L. harlandii than small-bodied A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 24.774, df = 4, p < 0.001) (Figure 2). Leopoldamys edwardsi was the only species that was capable of eating nuts of L. harlandii among the five species (F = 2.971, df = 4, p = 0.027) (Figure 2). More nuts of L. harlandii were scatter-hoarded by L. edwardsi than by A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 12.872, df = 4, p < 0.001) (Figure 2). Seeds of L. harlandii were also more likely to be larder-hoarded by L. edwardsi than by A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 2.420, df = 4, p = 0.060) (Figure 2).

The same patterns were observed in nuts of C. axillaris after being manipulated by L. edwardsi, A. draco, A. chevrieri, N. confucianus and N. fulvescens (Figure 3). Leopoldamys edwardsi, with the largest body size, removed more nuts of C. axillaris from the seed source than A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 17.630, df = 4, p < 0.001) (Figure 3). Moreover, only L. edwardsi was capable of eating nuts of C. axillaris among the five rodent species (F = 5.858, df = 4, p = 0.001) (Figure 3a). Importantly, L. edwardsi scatter-hoarded significantly more nuts of C. axillaris than A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 3.883, df = 4, p = 0.009) (Figure 3). Hard nuts of C. axillaris were more likely to be larder-hoarded by L. edwardsi than by A. draco, A. chevrieri, N. confucianus and N. fulvescens (F = 5.038, df = 4, p = 0.002) (Figure 3).

The proportions of L. harlandii seeds remaining at seed stations were negatively correlated to the body mass of rodent species (r² = 0.87, p = 0.02), indicating that large-sized rodents dispersed more seeds than small-bodied ones. Moreover, the proportions of scatter-hoarded L. harlandii seeds were positively correlated to the body mass of rodent species (r² = 0.91, p = 0.01), showing that large-sized rodents were effective seed dispersers. Similarly, large-sized rodents tended to disperse and scatter-hoard more C. axillaris seeds than those with a small body mass (r² = 0.97, p = 0.002; r² = 0.93, p = 0.008). Moreover, the proportions of L. harlandii seeds scatter-hoarded were positively correlated to the body mass of the rodent species (r² = 0.91, p = 0.01), indicating that large-sized rodents were effective seed dispersers. However, we failed to detect clear correlations between body mass and the proportions of J. mandshurica seeds remaining at seed stations (r² = 0.53, p = 0.27). A marginally positive correlation between body mass and seed scatter-hoarding was observed (r² = 0.89, p = 0.058), possibly due to the medium-sized T. sibiricus refusing to handle the hard-shelled seeds.

4. Discussion

Our systematic research conducted in large enclosures clearly demonstrated that, among the seeds in the two forests, those with the hardest shell consistently and directly prevented the access of small-sized seed consumers, which have been proven to be either larder-hoarders or ineffective scatter-hoarders [29,33,42]. Nevertheless, rodents with the largest body size and superior seed-handling ability seemed to be the most efficient seed dispersers for nuts with hard shells in the two forest ecosystems [31,43]. Therefore, these results provide strong support to our hypothesis that hard-coated seeds exclude smaller or weaker-jawed rodents but select for larger-sized rodents to promote their seed dispersal, which represents an evolutionary importance for plant–animal mutualistic interactions. The predation pressure exerted by large-bodied seed dispersers might have resulted in the evolution of nuts with hard shells, thereby preventing predation by small-sized seed predators. The absence of small-bodied seed predators and even ineffective seed dispersers of hard-shelled seeds seems to be significantly compensated by the legitimate large-bodied seed dispersers. The deterrence of seed predators and the attraction to legitimate seed dispersers reflect the evolutionary significance of the hard seed shell as an ecological filter in the seed dispersal syndrome. Different from soft-shelled seeds that can be consumed by a majority of seed predators, seeds with the hardest shells are anticipated to form a pairwise coevolutionary interaction with large-bodied scatter-hoarding rodents to achieve their synzoochory, that is, the dispersal of seeds by seed-caching animals [9,44]. Compared with seed coat thickness, seed mass appears to not be a decisive role in determining the possibility of seed dispersal. This prediction can be supported by the fact that small-sized A. peninsulae are able to eat and transport nuts of J. mandshurica with the largest seed mass, while medium-sized T. sibiricus fail to handle J. mandshurica seeds. On the contrary, the body mass of rodents plays a decisive role in affecting seed removal and scatter-hoarding, as observed by the correlations between body mass and seed fates. Our findings also imply that the hard seed shell might also function as an adaptive mechanism to directly deter seed predators but favor obligatory seed dispersers, which has been coined as the ‘directed deterrence hypothesis’ positing that the production of secondary metabolites functions to deter vertebrates that are likely to destroy seeds but show no impacts on seed-dispersing vertebrates [45].

Additional evidence in support of our prediction is derived from the study carried out by Wang et al. [2] in a tropical forest in Xishuanbanna, Yunnan province, China. It was revealed that the physically well-protected nuts of Castanopsis mekongensis (with a shell thickness of 1.07 mm) are mainly dispersed and cached by large-sized Rattus surifer, which is the dominant scatter-hoarding species in the tropical continuous forests [11,46]. Moreover, a field survey indicated that L. edwardsi, having the largest body size, is likely the sole known rodent species that potentially disperses the nuts of L. harlandii with extremely hard shells [47]. The observations made by Lai et al. [48] provided field evidence that L. edwardsi is highly accountable for the seed dispersal and scatter-hoarding of C. axillaries nuts. On the contrary, the nuts of C. axillaris with tough/thick shells had the lowest mean percentage of removal at shrubland sites where only small-sized rodents (such as Rattus rattus flavipectus and Niviventer fulvescens) were found [49]. Zhang et al. [50] presented evidence that S. davidianus is highly responsible for caching the nuts of J. mandshurica both in the field and enclosures. In similar forest ecosystems in northeastern China, Wang et al. [51] recently demonstrated that S. vulgaris was possibly the sole seed disperser and scatter-hoarder of J. mandshurica nuts in the field, despite the existence of other small-sized rodents like T. sibiricus, A. peninsulae and M. rufocanus. Even the Japanese squirrels (S. lis) require time to obtain the most suitable feeding method for promptly opening the hard shell of the Japanese walnut (J. ailanthifolia) [52]. In North America, it was stated that only one or no vertebrates compete with fox squirrels (S. niger) and gray squirrels (S. carolinensis) for black walnuts (J. nigra) and hickory nuts (Carya spp.) with thick shells [53]. Field evidence indicated that nearly all seedlings of walnut and hickory originated from nuts buried by squirrels. One field study conducted in Panama, Central America, revealed an exclusively mutualistic interaction between large-sized Central American agoutis (Dasyprocta punctata) and cocosoid palm Astrocaryum standleyanum with 10 g hard stony endocarps and large seeds [54]. However, medium-sized scatter-hoarding animals such as the spiny rat (Proechimys semispinosus) and the squirrel (Sciurus granatensis) make little contribution to the seed dispersal of A. standleyanum, perhaps due to their inability to handle the hard seeds. Another study by Jansen et al. [4] carried out in French Guiana, North America, confirmed the obligatory mutualism between Carapa procera with hard seeds averaging at 21 g and the main scatter-hoarding animals being red acouchy (Myoprocta acouchy) and red-rumped agouti (Dasyprocta leporina). In the African savanna, the tree squirrel (Paraxerus cepapi) seems to be the only secondary disperser of marula (Sclerocarya birrea) seeds, which have a highly lignified endocarp [55]. Thus, our large-scale studies in the two distinct forest ecosystems, along with the global evidence, strongly suggest that a hard seed shell is not an evolutionary dead end in the interactions between seeds and seed-dispersing animals. On the contrary, the result of a hard shell in the seed dispersal syndrome is of evolutionary significance for plant–animal mutualistic interactions and seems to be much more common than previously acknowledged in various forest ecosystems.

Although established caches will be recovered or pilfered by scatter-hoarding rodents and other sympatric competitors, cache owners will inevitably forget a small proportion of caches, which have a potential to germinate into seedlings [56]. Pilferers usually show inefficiency in stealing caches made by scatter-hoarders because they do not have any spatial memory to locate caches [57]. Moreover, hard shells may prevent the emission of seed odor from caches, which further reduces cache pilferage by pilferers. In this scenario, caches in the field are expected to have a higher survival possibility than in the enclosures, in turn resulting in potential seedling establishment. Therefore, the interaction between seeds with hard shells and legitimate seed dispersers might represent a recurrent convergence of an interspecies mutualism.

However, nuts and rodents in an obligatory mutualism might encounter the potential issue that plants could lose their seed dispersers when their partners are endangered or even extinct [58,59,60]. An empirical phenomenon is that pairwise coevolution with the dodo Raphus cucullatus has led to the near extinction of Calvaria major, which bears large single-seeded drupes and completely relies on the dodo for seed dispersal and seedling recruitment [61]. Due to intense hunting and human disturbance, S. vulgaris, the obligatory seed disperser of J. mandshurica, is currently mainly distributed in primitive Korean pine Pinus koraiensis forests but is rarely seen in the secondary forests where T. sibiricus, A. peninsulae and M. rufocanus are relatively common [62]. Similarly, L. edwardsi and another potential seed disperser Callosciurus erythraeus of L. harlandii and C. axillaris have become locally very scarce in recent years in Dujiangyan [1,33]. The distribution of L. harlandii and C. axillaris aligns well with the occurrence of L. edwardsi in the subtropical forests [1]. It seems that there is a close connection between the distribution of trees with hard shells and the abundance of legitimate large-sized seed dispersers, suggesting that seed coat hardness has the potential to structure plant communities by favoring species whose seed traits align with the dominant rodent dispersers. Hence, comprehending how hard seeds co-evolve with large-sized animals is of great significance in anthropogenically disrupted forests. The findings we present in this study might offer an understanding of the necessity to well protect and conserve the compulsory seed dispersers of hard seeds in the seed dispersal syndrome.

5. Conclusions

Our study, by offering seeds with hard shells to seed-eating rodents in two distinct forest ecosystems, clearly shows that hard-shelled seeds deter small-bodied seed predators but select for large-sized scatter-hoarders to ensure their seed dispersal. Seed coat hardness serves as a critical ecological filter by mediating rodent–seed interactions, influencing dispersal outcomes and shaping plant recruitment. This trait-based selection underscores the role of biotic interactions in driving ecological and evolutionary processes.

Author Contributions

X.Y. conceived and designed the experiments. L.J. performed the experiments. L.J. and X.Y. analyzed the data. X.Y. wrote the manuscript; other authors provided editorial advice. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The animal study protocol was approved by the Ethics Committee of Qufu Normal Unicersity (protocol code 2017023, 20 March 2017).

Data Availability Statement

Data access can be requested from the corresponding author.

Acknowledgments

The authors thank the anonymous reviewers for their comments on the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Table

Figure 1. Seed fates of Juglans mandshurica manipulated by (a) Sciurus vulgaris (6), (b) Apodemus peninsulae (8), (c) Myodes rufocanus (6) and (d) Tamias sibiricus (8) in artificial enclosures in Dailing of Heilongjiang province. IIS, E, LH and SH represent seeds intact in situ, eaten, larder-hoarded and scatter-hoarded by rodents for 10 h, respectively.

Figure 2. Seed fates of Lithocarpus harlandii manipulated by (a) Leopoldamys edwardsi (15), (b) Apodemus chevrieri (15), (c) A. draco (10), (d) Niviventer confucianus (11) and (e) N. fulvescens (7) in artificial enclosures in Dujiangyan of Sichuan province. IIS, E, LH and SH represent seeds intact in situ, eaten, larder-hoarded and scatter-hoarded by rodents for 13 h, respectively.

Figure 3. Seed fates of Choerospondias axillaris manipulated by (a) Leopoldamys edwardsi (15), (b) Apodemus chevrieri (15), (c) A. draco (10), (d) Niviventer confucianus (11) and (e) N. fulvescens (7) in artificial enclosures in Dujiangyan of Sichuan province. IIS, E, LH and SH represents seeds intact in situ, eaten, larder-hoarded and scatter-hoarded by rodents for 13 h, respectively.

Table 1

Rodent body mass, seed mass and seed coat thickness in this study.

Rodent Species	Mean Body Mass (g)	SD	Seed Species	Seed Mass (g)	SD	Seed Coat Thickness (mm)	SD
Sciurus vulgaris	273	14	J. mandshurica	13.6	1.28	3.2	0.65
Apodemus peninsulae	30.51	1.34	L. harlandii	4.3	0.77	1.43	0.20
Myodes rufocanus	31.61	2.65	C. axillaris	1.77	0.32	1.59	0.30
Tamias sibiricus	97.9	2.98
Leopoldamys edwardsi	386	3.47
Apodemus chevrieri	50	2.24
Apodemus draco	23	0.71
Niviventer confucianus	87	3.18

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d17030150/s1, Figure S1: Fresh weight (a) and seed coat thickness (b) of nuts used in this study; Figure S2: Body mass of rodent species captured from Dailing of Heilongjiang province, and Dujiangyan of Sichuan province.

References

1. Yang, X.; Yan, C.; Zhao, Q.; Holyoak, M.; Zhang, Z. Ecological succession drives the structural change of seed-rodent interaction networks in fragmented forests. For. Ecol. Manag.; 2018; 419, pp. 42-50. [DOI: https://dx.doi.org/10.1016/j.foreco.2018.03.023]

2. Wang, Z.; Cao, L.; Zhang, Z. Seed traits and taxonomic relationships determine the occurrence of mutualisms versus seed predation in a tropical forest rodent and seed dispersal system. Integr. Zool.; 2014; 9, pp. 309-319. [DOI: https://dx.doi.org/10.1111/1749-4877.12086]

3. Lai, Y.; Dou, H.; Zhou, Y.; Wang, Z.; Hou, L.; Cao, L. Impact of delayed seed drop on rodent-mediated seed dispersal and predation of pine. Integr. Zool.; 2024. [DOI: https://dx.doi.org/10.1111/1749-4877.12912]

4. Jansen, P.A.; Hemerik, L.; Bongers, F. Seed mass and mast seeding enhance dispersal by a neotropical scatter-hoarding rodent. Ecol. Monogr.; 2004; 74, pp. 569-589. [DOI: https://dx.doi.org/10.1890/03-4042]

5. Zhang, H.; Chu, W.; Zhang, Z. Cultivated walnut trees showed earlier but not final advantage over its wild relatives in competing for seed dispersers. Integr. Zool.; 2017; 12, pp. 12-25. [DOI: https://dx.doi.org/10.1111/1749-4877.12242]

6. Zeng, D.; Swihart, R.K.; Zhao, Y.; Si, X.; Ding, P. Cascading effects of forested area and isolation on seed dispersal effectiveness of rodents on subtropical islands. J. Ecol.; 2019; 107, pp. 1506-1517. [DOI: https://dx.doi.org/10.1111/1365-2745.13122]

7. Okawa, R.; Saitoh, T.; Noda, T. Interactive effects of two rodent species on the seed dispersal of Japanese walnut. Sci. Rep.; 2023; 13, 18098. [DOI: https://dx.doi.org/10.1038/s41598-023-44513-9]

8. Bartlow, A.W.; Agosta, S.J.; Curtis, R.; Yi, X.; Steele, M.A. Acorn size and tolerance to seed predators: The multiple roles of acorns as food for seed predators, fruit for dispersal and fuel for growth. Integr. Zool.; 2017; 13, pp. 251-266. [DOI: https://dx.doi.org/10.1111/1749-4877.12287]

9. Gómez, J.M.; Schupp, E.W.; Jordano, P. Synzoochory: The ecological and evolutionary relevance of a dual interaction. Biol. Rev.; 2019; 94, pp. 874-902. [DOI: https://dx.doi.org/10.1111/brv.12481]

10. Cui, J.; Zhang, Y.; Guo, J.; Wu, N.; Zhou, Y. Conflicting selection pressures on seed size and germination caused by carnivorous seed dispersers. Integr. Zool.; 2023; 18, pp. 799-816. [DOI: https://dx.doi.org/10.1111/1749-4877.12743]

11. Cao, L.; Yan, C.; Wang, B. Differential seed mass selection on hoarding decisions among three sympatric rodents. Behav. Ecol. Sociobiol.; 2018; 72, 161. [DOI: https://dx.doi.org/10.1007/s00265-018-2576-5]

12. Cheng, J.; He, H.; Zheng, L.; Zhang, C.; Wang, X.; Hu, X.; Niu, H.; Zhang, H. Bold rats (Niviventer confucianus) are more effective in seed dispersal: Evidence both under enclosure conditions and in the field. Integr. Zool.; 2024. [DOI: https://dx.doi.org/10.1111/1749-4877.12906] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39313867]

13. Zwolak, R.; Crone, E.E. Quantifying the outcome of plant–granivore interactions. Oikos; 2012; 121, pp. 20-27. [DOI: https://dx.doi.org/10.1111/j.1600-0706.2011.19849.x]

14. Sawaya, G.M.; Goldberg, A.S.; Steele, M.A.; Dalgleish, H.J. Environmental variation shifts the relationship between trees and scatterhoarders along the continuum from mutualism to antagonism. Integr. Zool.; 2018; 13, pp. 319-330. [DOI: https://dx.doi.org/10.1111/1749-4877.12311]

15. Zwolak, R.; Sih, A. Animal personalities and seed dispersal: A conceptual review. Funct. Ecol.; 2020; 34, pp. 310-1294. [DOI: https://dx.doi.org/10.1111/1365-2435.13583]

16. Loayza, A.P.; Luna, C.A.; Calvio-Cancela, M. Predators and dispersers: Context-dependent outcomes of the interactions between rodents and a megafaunal fruit plant. NPG; 2020; 10, 6106. [DOI: https://dx.doi.org/10.1038/s41598-020-62704-6]

17. Siepielski, A.M.; Benkman, C.W. A seed predator drives the evolution of a seed dispersal mutualism. Proc. R. Soc. B Biol. Sci.; 2008; 275, pp. 1917-1925. [DOI: https://dx.doi.org/10.1098/rspb.2008.0451]

18. Babale, A.; Thia, J.A.; Elena, M.; Pierre-Michel, F.; Chapman, H.M. Forest disturbance and seasonal food availability influence a conditional seed dispersal mutualism. Biotropica; 2018; 50, pp. 750-757.

19. Quitián, M.; Santillán, V.; Bender, I.M.A.; Espinosa, C.I.; Homeier, J.; Bhning-Gaese, K.; Schleuning, M.; Neuschulz, E.L. Functional responses of avian frugivores to variation in fruit resources between natural and fragmented forests. Funct. Ecol.; 2019; 33, pp. 399-410. [DOI: https://dx.doi.org/10.1111/1365-2435.13255]

20. Yadok, B.G.; Forget, P.M.; Gerhard, D.; Aliyu, B.; Chapman, H. Seed nutrient content rather than size influences seed dispersal by scatterhoarding rodents in a West African montane forest. J. Trop. Ecol.; 2020; 36, pp. 174-181. [DOI: https://dx.doi.org/10.1017/S0266467420000127]

21. Liu, H.; Zhang, J.; Wang, B. Contrasting seed traits of co-existing seeds lead to a complex neighbor effect in a seed-rodent interaction. Oecologia; 2023; 201, pp. 1017-1024. [DOI: https://dx.doi.org/10.1007/s00442-023-05365-2]

22. Wang, B.; Chen, J. Seed size, more than nutrient or tannin content, affects seed caching behavior of a common genus of Old World rodents. Ecology; 2009; 90, pp. 3023-3032. [DOI: https://dx.doi.org/10.1890/08-2188.1]

23. Cao, L.; Wang, Z.; Yan, C.; Chen, J.; Guo, C.; Zhang, Z. Differential foraging preferences on seed size by rodents result in higher dispersal success of medium-sized seeds. Ecology; 2016; 97, pp. 3070-3078. [DOI: https://dx.doi.org/10.1002/ecy.1555]

24. Cerri, M.; Reale, L. Anatomical traits of the principal fruits: An overview. Sci. Hortic.; 2020; 270, 109390. [DOI: https://dx.doi.org/10.1016/j.scienta.2020.109390]

25. Hadj-Chikh, L.Z.; Steele, M.A.; Smallwood, P.D. Caching decisions by grey squirrels: A test of the handling time and perishability hypotheses. Anim. Behav.; 1996; 52, pp. 941-948. [DOI: https://dx.doi.org/10.1006/anbe.1996.0242]

26. Vander Wall, V.S. How plants manipulate the scatter-hoarding behaviour of seed-dispersing animals. Phil. Trans.; 2010; 365, pp. 989-997. [DOI: https://dx.doi.org/10.1098/rstb.2009.0205]

27. Cheng, J.; Yan, X.; Zhang, J.; Zhang, C.; Zhang, M.; Wei, S.; Wang, J.; Luo, Y. Seed traits and burial state affect plant seed secondary dispersal mediated by rodents. Heliyon; 2024; 10, e32612. [DOI: https://dx.doi.org/10.1016/j.heliyon.2024.e32612]

28. Yu, F.; Jiao, G.; Niu, K.; Yi, X. Seed removal and dispersal of five sympatric seed species in Xiaoxingan Mountains. J. Northeast For. Univ.; 2011; 33, pp. 21-26.

29. Yi, X.; Wang, Z.; Liu, C.; Liu, G. Seed trait and rodent species determine seed dispersal and predation: Evidences from semi-natural enclosures. iForest; 2015; 8, 207. [DOI: https://dx.doi.org/10.3832/ifor1185-008]

30. Smith, C.C.; Follmer, D. Food preferences of squirrels. Ecology; 1972; 53, pp. 82-91. [DOI: https://dx.doi.org/10.2307/1935712]

31. Steele, M.A.; Yi, X. Squirrel-seed interactions: The evolutionary strategies and impact of squirrels as both seed predators and seed dispersers. Front. Ecol. Evol.; 2020; 8, 259. [DOI: https://dx.doi.org/10.3389/fevo.2020.00259]

32. Paulsen, T.R.; Hgstedt, G.; Thompson, K.; Vandvik, V.; Eliassen, S. Conditions favouring hard seededness as a dispersal and predator escape strategy. J. Ecol.; 2014; 102, pp. 1475-1484. [DOI: https://dx.doi.org/10.1111/1365-2745.12323]

33. Yang, X.; Zhao, K.; Yan, C.; Wang, B.; Zhang, Z. Impacts of moderateness and cobalancedness in seed mass and coat thickness on seed dispersal effectiveness of Quercus variabilis by rodents. Plant Ecol.; 2023; 224, pp. 283-297. [DOI: https://dx.doi.org/10.1007/s11258-023-01296-3]

34. Jamille, D.A.B.; Guimares, P.R.; Peres, C.A.; Carvalho, G.; Cazetta, E. Data from: Local extinctions of obligate frugivores and patch size reduction disrupt the structure of seed dispersal networks. Ecography; 2018; 41, pp. 1899-1909.

35. Emer, C.; Galetti, M.; Pizo, M.A.; Guimarães, P.R.; Moraes, S.; Piratelli, A.; Jordano, P.; Saavedra, S. Seed-dispersal interactions in fragmented landscapes—A metanetwork approach. Ecol. Lett.; 2018; 21, pp. 484-493. [DOI: https://dx.doi.org/10.1111/ele.12909]

36. Wang, M.; Yang, X.; Yi, X. Effects of seed traits on the cache size of a scatter-hoarding rodent, Leopoldamys edwardsi. Behav. Ecol. Sociobiol.; 2023; 77, 105. [DOI: https://dx.doi.org/10.1007/s00265-023-03379-7]

37. Wang, Z.; Tai, W.; Zhang, X.; Liu, S.; Niu, Y.; Chen, W.; Li, N. Importance of plant and fruit traits on the structure of bird seed dispersal networks in different disturbed habitats. Integr. Zool.; 2024; 19, pp. 753-762. [DOI: https://dx.doi.org/10.1111/1749-4877.12822]

38. da Silva, L.P.; Coutinho, A.P.; Ramos, J.A.; Heleno, R.H. Anthropogenic forests simplify seed-but not pollen-dispersal networks. Integr. Zool.; 2025; 20, pp. 160-170. [DOI: https://dx.doi.org/10.1111/1749-4877.12868]

39. Li, L.; Li, X.; Jin, Y.; Zhang, H.; Bu, F.; Zhang, R.; Wu, X.; Yuan, S.; Fu, H. Group effects of desert rodent communities on plant seed dispersal. Integr. Zool.; 2025; 20, pp. 199-207. [DOI: https://dx.doi.org/10.1111/1749-4877.12897]

40. Chen, J.; Deng, X.; Bai, Z.; Yang, Q.; Chen, G.; Liu, Y.; Liu, Z. Fruit characteristics and Muntiacus muntijak vaginalis (Muntjac) visits to individual plants of Choerospondias axillaris. Biotropica; 2001; 33, pp. 718-722.

41. Yi, X.; Steele, M.A.; Stratford, J.A.; Wang, Z.; Yang, Y. The use of spatial memory for cache management by a scatter-hoarding rodent. Behav. Ecol. Sociobiol.; 2016; 9, 70. [DOI: https://dx.doi.org/10.1007/s00265-016-2161-8]

42. Gu, H.; Yang, X.; Dirzo, R.; Zhang, Z. Differences in mutualistic or predatory interactions between tree and rodent species as revealed by using a double-duplex passive integrated transponder tagging technique. Acta Oecol.; 2021; 12, pp. 12-25. [DOI: https://dx.doi.org/10.1016/j.actao.2021.103747]

43. Gu, H.; Zhao, Q.; Zhang, Z. Does scatter-hoarding of seeds benefit cache owners or pilferers?. Integr. Zool.; 2017; 12, pp. 477-488. [DOI: https://dx.doi.org/10.1111/1749-4877.12274]

44. Yunlong, Z.; Xifu, Y.; Haifeng, G.; Yuwei, T.; Zhibin, Z. Effects of seed coat thickness on rodents’ seed dispersal behavior and seed fate. Acta Theriol. Sin.; 2021; 41, 667.

45. Levey, D.J.; Tewksbury, J.J.; Cipollini, M.L.; Carlo, T.A. A field test of the directed deterrence hypothesis in two species of wild chili. Oecologia; 2006; 150, pp. 61-68. [DOI: https://dx.doi.org/10.1007/s00442-006-0496-y]

46. Cao, L.; Wang, B.; Yan, C.; Wang, Z.; Zhang, H.; Geng, Y.; Chen, J.; Zhang, Z. Risk of cache pilferage determines hoarding behavior of rodents and seed fate. Behav. Ecol.; 2018; 29, pp. 984-991. [DOI: https://dx.doi.org/10.1093/beheco/ary040]

47. Xiao, Z.; Zhang, Z. Nut predation and dispersal of Harland Tanoak Lithocarpus harlandii by scatter-hoarding rodents. Acta Oecol.; 2006; 29, pp. 205-213. [DOI: https://dx.doi.org/10.1016/j.actao.2005.10.004]

48. Lai, X.; Guo, C.; Xiao, Z. Trait-mediated seed predation, dispersal and survival among frugivore-dispersed plants in a fragmented subtropical forest, southwest China. Integr. Zool.; 2014; 9, pp. 246-254. [DOI: https://dx.doi.org/10.1111/1749-4877.12046]

49. Hau, C.H. Tree seed predation on degraded hillsides in Hong Kong. For. Ecol. Manag.; 1997; 99, pp. 215-221.

50. Zhang, H.; Steele, M.A.; Zhang, Z.; Wei, W.; Yu, W. Rapid sequestration and recaching by a scatter-hoarding rodent (Sciurotamias davidianus). J. Mammal.; 2014; 95, pp. 480-490. [DOI: https://dx.doi.org/10.1644/13-MAMM-A-178]

51. Wang, J.; Yan, Q.; Yan, T.; Song, Y.; Sun, Y.; Zhu, J. Rodent-mediated seed dispersal of Juglans mandshurica, regulated by gap size and within-gap position in larch plantations: Implication for converting pure larch plantations into larch-walnut mixed forests. For. Ecol. Manag.; 2017; 404, pp. 205-213. [DOI: https://dx.doi.org/10.1016/j.foreco.2017.08.033]

52. Tamura, N. Population differences and learning effects in walnut feeding technique by the Japanese squirrel. J. Ethol.; 2011; 29, pp. 351-363. [DOI: https://dx.doi.org/10.1007/s10164-011-0267-z]

53. Stapanian, M.A.; Smith, C.C. A model for seed scatterhoarding: Coevolution of fox squirrels and black walnuts. Ecology; 1978; 59, pp. 884-896. [DOI: https://dx.doi.org/10.2307/1938541]

54. Jansen, P.A.; Hirsch, B.T.; Emsens, W.J.; Zamora-Gutierrez, V.; Wikelski, M.; Kays, R. Thieving rodents as substitute dispersers of megafaunal seeds. Proc. Natl. Acad. Sci. USA; 2012; 109, pp. 12610-12615. [DOI: https://dx.doi.org/10.1073/pnas.1205184109] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22802644]

55. Midgley, J.J.; Gallaher, K.; Kruger, L.M. The role of the elephant (Loxodonta africana) and the tree squirrel (Paraxerus cepapi) in marula (Sclerocarya birrea) seed predation, dispersal and germination. J. Trop. Ecol.; 2012; 28, pp. 227-231. [DOI: https://dx.doi.org/10.1017/S0266467411000654]

56. Bogdziewicz, M.; Crone, E.E.; Zwolak, R. Do benefits of seed dispersal and caching by scatterhoarders outweigh the costs of predation? An example with oaks and yellow-necked mice. J. Ecol.; 2020; 108, pp. 1009-1018. [DOI: https://dx.doi.org/10.1111/1365-2745.13307]

57. Hou, L.; Dou, H.; Hu, M.; Cao, L. Effects of cache distribution and size on cache foraging by animals. Anim. Behav.; 2024; 211, pp. 91-98. [DOI: https://dx.doi.org/10.1016/j.anbehav.2024.03.005]

58. Rogers, H.; Donoso, I.; Traveset, A.; Fricke, E. Cascading impacts of seed disperser loss on plant communities and ecosystems. Ann. Rev. Ecol. Evol. Syst.; 2021; 52, pp. 641-666. [DOI: https://dx.doi.org/10.1146/annurev-ecolsys-012221-111742]

59. Lamperty, T.; Brosi, B.J. Loss of endangered frugivores from seed dispersal networks generates severe mutualism disruption. Proc. R. Soc. B; 2022; 289, 20220887. [DOI: https://dx.doi.org/10.1098/rspb.2022.0887]

60. Mendes, S.B.; Olesen, J.M.; Memmott, J.; Costa, J.M.; Timóteo, S.; Dengucho, A.L.; Craveiro, L.; Heleno, R. Evidence of a European seed dispersal crisis. Science; 2024; 386, pp. 206-211. [DOI: https://dx.doi.org/10.1126/science.ado1464]

61. Temple, S.A. Plant-animal mutualism: Coevolution with dodo leads to near extinction of plant. Science; 1977; 197, pp. 885-886. [DOI: https://dx.doi.org/10.1126/science.197.4306.885]

62. Muha, C.; Yunga, W.; Qin, W.; Chenggong, M.; Guodong, B.; Feng, T.; Nan, Z. Effects of rodents on seed dispersal patterns of different trees in Wangyedian forest. Ecol. Environ. Sci.; 2022; 31, pp. 1118-1123.