Herbivore digestion as environmental filter - which seed traits help species survive?

doi:10.21203/rs.3.rs-1746037/v1

Download PDF

Research Article

Herbivore digestion as environmental filter - which seed traits help species survive?

https://doi.org/10.21203/rs.3.rs-1746037/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Herbivorous animals are one of the vectors for seed dispersal of open-landscape plant species. The plant species are expected to be adapted to this type of dispersal and researchers presume they have specific seed traits. To find which traits help seeds survive the passage through digestion of wild herbivores, we conducted a comprehensive feeding experiment with almost forty species of plants and four species of herbivorous mammals. We fed specified numbers of seeds to the animals, collected the dung and germinated the dung content. We explored whether seed morphological traits and seed nutrient contents are good predictors of seed survival after passage through the herbivore digestive system. We also tested how the seed survival differed after the passage through different herbivore species. We found that species survival and germination success was positively correlated to seed nitrogen content and negatively to seed elongation. However, when we tested species from main families separately, i.e., legumes, grasses, and all other species, these trends changed directions. This suggests that seed dispersal by free-ranging wild herbivores is more a random process than driven by seed traits.

dry grasslands

endozoochory

wild herbivores

seed dispersal

seed traits

Effective dispersal of plants is crucial for the connection between different habitat patches, survival of plant populations and the metapopulation dynamics, and their genetic diversity (Lozada-Gobilard et al., 2021). Large animals, e.g., mammals, are potential drivers for both short- and long-distance dispersal (Cain et al., 1998). A simplified model of endozoochorous dispersal can be expressed by this equation:

$${N}_{dispersed}={N}_{shoots}\ast {N}_{seeds/shoot}\ast {p}_{eaten}\ast {p}_{survival}$$

The effectivity of dispersal as measured by the number of dispersed and germinated seeds (${N}_{dispersed}$) is dependent on the characteristics of both parties: the dispersed plant species, here the number of shoots and the number of seeds per shoot (${N}_{shoots}$ and ${N}_{seeds/shoot}$), and the dispersal vector (animal), which influences the probability the seed is eaten in the first place (${p}_{eaten}$). Both plants and animal vector characteristics affect the probability with which the seed survives the passage through the digestive tract (${p}_{survival}$).

From characteristics influencing survival in the digestive tract, seed traits are the main driver. Contrary to epizoochory, where the morphology of dispersed propagule clearly affects its ability to attach and remain in fur (Albert et al., 2015; Tackenberg et al., 2006), in endozoochory the traits supporting successful passage through the digestive tract are less clear. The propagule must be able to survive rough mechanical and chemical conditions, e.g., molar mill and acidic environment in the stomach, taking into account it is actually eaten in the first place. Based on previous research, successful endozoochorously dispersed seeds are small and without appendages (to escape the molar mill), and exhibit long survival in the soil bank (which is similarly destructive as the chemical conditions in the stomach; Pakeman et al., 2002, Albert et al.. 2015). The effect of seed shape received contradictory support with better survival of round seeds (Mouissie et al., 2005) as well as elongated seeds (Cosyns et al., 2005) and no effect of seed shape whatsoever (D’hondt and Hoffmann, 2011).

However, what is rarely taken into account are the constraints of the herbivore species, e.g., how morphological constraints of the animal body influence dispersal (Illius and Gordon, 1992) - small animals are more likely to disperse only small seeds, whereas large-bodied animals are able to ingest both small and large seeds (Chen and Moles, 2015). The body size influences another aspect of the digestive system. That is the speed by which particles travel through the tract: large animals have alonger digestive system, and it takes longer for the particles to pass through it (Clauss et al., 2006). Hand in hand with feeding preferences go the feeding strategies and nutrient demands of herbivore species. Unlike hind-gut fermenters, ruminants are able to digest phytic acid from the seeds thus disintegrating the seeds in the process. Furthermore, Cervids (e.g., deer) have higher phosphorus demands because they store it in their antlers, whereas other ruminants (e.g., Bovids) excrete the excess phosphorus in the dung (Sitters et al., 2014). All of these - speed of passage, digestive type, and nutrient demands - directly affect the probability with which seeds survive the travel through the digestive tract.

Furthermore, the effect of seed traits on survival in the digestive tract is usually tested on the species composition found in dung samples in comparison with available vegetation (Albert et al., 2015). This approach is used to suggest which plant traits help the species to be eaten and successfully survive the passage. But the probability of being eaten (${p}_{eaten}$) is a function of several factors including plant frequency in the landscape (Lepková et al., 2018), seed production (Bruun and Poschlod, 2006), and feeding preferences of the herbivore (Hofmann, 1989). The feeding preferences are further influenced by a number of plant traits, e.g., the nutritional value of foliage (Hejcmanová et al., 2016). When species found in dung are compared with the available vegetation, all these filters are included and direct deduction of traits enabling endozoochory is thus questionable. Different type of experiments is available to study the traits influencing seed survival per se - comparing germination rates of seeds fed directly to animals in known quantities.

In this paper, we aim to address the question of which seed traits drive the survival and germination success of species after passage through the digestive tract of four species of wild animals: red deer, sika deer, mouflon, and wild boar. Specifically, we asked two main questions: (i) which seed morphological traits and nutrient contents drive the species’ ability to survive the passage through the digestive tract? We hypothesized that the best surviving species have a round shape, mucilaginous surface, and low amount of nutrients. (ii) Does the survival success differ between herbivore species? We tested which of our hypotheses is valid for our set of herbivores: (a) body size has a positive effect on germination rate (red deer > sika deer > wild boar > mouflon). (b) Length of digestive tract has a negative effect on germination rate (mouflon > wild boar > sika deer > red deer). (c) Rumination has a negative effect on germination rate (wild boar > others). We chose thirty-eight species of plants from the landscape of dry grassland, both previously found in dung and from the same area but absent from dung (Lepková et al., 2018). A predefined number of seeds was fed to four different herbivores and their germination rate was established with a greenhouse experiment.

We selected the plant species from the vegetation of dry grasslands in the Doupov Mountains where we studied endozoochorous dispersal by deer and boars before (Lepková et al., 2018). Species were picked based on two criteria: frequency of species in dung samples (> 5%) and vegetation (> 20%). Not all species were available on the market resulting in a list of 39 species from which 23 species were previously known as dispersed in dung. One species - Knautia arvensis - did not germinate from control pots or dung samples. It was included in the feeding experiment, but it was excluded from all analyses. The analysed number of species was, therefore, thirty-eight.

All seeds were obtained from the commercial source Planta naturalis (plantanaturalis.com, Markvartice, CZ) from the harvest of 2015. Random amounts of seeds were weighed and counted to find out an exact number of seeds per gram. The obtained seed numbers per gram were then used to weigh the amount of seeds to be fed. Plant species were divided into four groups according to their weight: extra heavy (more than 10 mg), heavy (1–10 mg), medium (0,1–1 mg) and light (less than 0,1 mg) (data from D3 and LEDA databases; Hintze et al. (2013) and Kleyer et al. (2008); respectively). Total numbers of seeds in fed mixtures reflected their absolute weight so that animals were not fed too large quantities of large and heavy seeds (for numbers of seeds fed to one individual, see Table 1). Not enough seeds were available for Vicia tetrasperma, and we used 400 and 200 seeds for deer and mouflon/wild boar, respectively. Weight data from the databases were used because this influenced the amount of seeds ordered from the commercial supplier.

Table 1

The number of seeds fed to each animal. The numbers of individuals per animal species are in brackets. Thus, the number of seeds of a light plant species fed to all individuals of red deer was 30 000.
Herbivore	Extra heavy	Heavy	Medium	Light	V. tetrasperma
Red deer (3) Sika deer (3)	500	1 000	5 000	10 000	400
Mouflon (7) Wild boar (1)	250	500	2 000	5 000	200

The experiment was conducted in forest ZOO Malá Chuchle in Prague, Czech Republic. Four species of ungulates were used for the experiment: three individuals of red deer (Cervus elaphus), three individuals of sika deer (Cervus nippon), seven individuals of mouflon (Ovis musimon), and one young individual of wild boar (Sus scrofa). Even though three of the chosen herbivore species are ruminants with a similar digestive system, both species of deer belong to the group of intermediate feeders, whereas mouflon is a more selective browser (Hofmann, 1989). Compared to the ruminants, wild boar is an omnivore who chooses plant-based forage primarily during the vegetation season (Schley and Roper, 2003). Different digestive systems and different feeding behaviors lead to differences in the processing of forage in the digestive system (Baker and Hobbs, 1987).

For the feeding experiment, we divided the plant species into two groups to make the identification of seedlings easier and not to feed the animals such large quantities of seeds. The groups of plant species were fed separately in two different phases: on 4-Dec-2016 and on 20-Feb-2017. All individuals of one animal species were held in an enclosure together (pen in the case of boar in the first phase). The seeds were fed in a mixture of oat, barley, carrot and apples with lukewarm water and all animals were fed all species of plants. The animals were fed alfalfa hay and their usual forage (fruit, dry bread, and cereals) for three days before the experiment, during the experiment, and until the end of the dung collection. All dung was removed from the paddocks before the experiment and it was used as a control sample to check for germination from dung before the feeding started. After the feeding, the dung was collected every day for four days, with few exceptions (see Table 2) so as not to disturb the untamed animals. After collections, the dung samples were stratified in a freezer for two months and then crumbled and air-dried.

Table 2

Results of the generalized linear mixed-effect model testing the effect of herbivore species on the germination success of ingested seeds of 38 plant species. Post-hoc pairwise comparison was performed using package emmeans and revealed that survival of seeds after passage through red deer differed from all other animal species. The remaining animal species did not differ from each other.
contrast	estimate	SE	df	t-ratio	p-value
Mouflon – Red Deer	-1.8836353	0.2131468	146	-8.8372682	0.0000000
Mouflon – Sika Deer	-0.5307128	0.2309078	146	-2.2983751	0.1031114
Mouflon – Wild Boar	-0.3608758	0.2631232	146	-1.3715091	0.5192164
Red deer – Sika Deer	1.3529225	0.1812927	146	7.4626422	0.0000000
Red deer – Wild Boar	1.5227594	0.2143950	146	7.1025896	0.0000000
Sika Deer – Wild Boar	0.1698370	0.2413735	146	0.7036271	0.8955137

Seed survival was checked by a greenhouse emergence experiment over a period of two and half years. The germination pots were filled with perlite, covered by non-woven fabric and the air-dried crumbled dung samples were put on top. This allowed the samples to be mixed without mixing the substrate as well and burying the seeds at the bottom of the pot. The emergence took place under natural light conditions from May and June 2017 for the first and second phases, respectively. Pots were watered as necessary with rain-water and shifted around the greenhouse at random. In the last months of the experiment, when germination decreased, samples were left to dry out several times and mixed to let seeds from lower layers to germinate. Three different controls were set: (i) pots with dung samples collected from each enclosure before the feeding started, (ii) pots with no seeds and no samples to control for seed rain (small amount of potting soil was used on top of the non-woven fabric as substrate), and (iii) pots with 500 seeds per species to establish germination rate without passing through the guts (mixed with small amount of potting soil). In all cases, pots were filled with perlite and covered with non-woven fabric. All controls were stratified for the same period as the samples.

Seed traits

We measured seed nutrient reserves as the content of nitrogen (N), phosphorus (P), carbohydrates, and oils. Carbohydrates content was calculated as the sum of fructans and starch. N content was measured by flow injection analysis after Kjeldahl mineralization. P content was measured by flow injection analysis after perchloric acid mineralization. Starch and fructans were measured with the enzymatic procedure Megazyme (McCleary et al. 1994). Oil content was measured by Soxhlet extractor (International Organization for Standartization 2016). See Mašková and Herben (2021) for more details. To calculate the index of shape (Thompson et al., 1993) we used seed dimension from LEDA database (Kleyer et al., 2008). The index is the variance in dimensions and it ranges from 0 (perfectly round seeds) to 1 (elongated seeds). The following equation was used to calculate the index:

$$\frac{\sum {\left({x}_{i}-\overline{x}\right)}^{2}}{n}$$

To make the understanding of the index simpler, it is further referred to as seed elongation. We used the information about the presence of mucilaginous surface from D3 database (Hintze et al., 2013).

Data analyses

No plant species germinated better after passing through the digestive system than the unpassed controls. We thus used the germination rate of controls as the maximum possible germination potential for seeds fed to animals and corrected the number of fed seeds for this germination potential.

We used a generalized linear mixed-effect model (GLMM) to test the effect of seed traits on species germination rate. Two models were built: a model not accounting for phylogeny using package glmmTMB (version 1.1.2.9000; Brooks et al., 2017) and a model with phylogenetic correction using package phyloglmm (developer version from GitHub, DOI: 10.5281/zenodo.2639887). We used beta-binomial distribution to account for overdispersion and animal species as a random factor. To test for the differences between animal species, we used GLMM from the same package glmmTMB with betabinomial distribution (Brooks et al., 2017). For the tests with betabinomial distribution, we made a matrix of successes (germinated seeds) and failures (germinated seeds subtracted from the fed seeds). Furthermore, we separated species into three major groups and tested these separately: Fabaceae (9 species), Poaceae (14 species), and all other species (15). All models were made in the same manner (GLMM with betabinomial distribution and animal species as random factor).

We used the RLQ method (Kleyer et al., 2012) from the package ade4 (Thioulouse and Dray, 2007) to test the correlation between seed traits, seed germination rate, and animal species. These three types of information were all stored separately in three matrices. The RLQ analysis consists of several steps: first, individual matrices were analysed using principal component analysis (seed trait matrix and matrix of “environmental” variables, here the species of animal) and correspondence analysis (species germination rate matrix). The final RLQ model was then tested using fourth-corner statistics (function forthcorner.rlq, package ade4). All analyses were performed in R (version 4.1.1., R Core Team, 2021). Nomenclature of plant species follows Kubát et al. (2002).

None of the studied species showed increased germination after passage through the digestive tract, and eight species did not germinate from dung samples at all. The germination rate ranged from 0–39% (percentage of germinated seedlings from the total number of fed seeds corrected for germination rate of seeds in the control lot). The most successful species belonged to the Fabaceae family: Vicia cracca, Trifolium repens, Securigera varia. However, even in the species with the highest germination rate, it varied between animals: the best surviving Vicia cracca showed a germination rate of almost 40% after passage through red deer but less than 6% after passage through mouflon. Plant species survived the passage through red deer better than through other herbivores (Fig. 1.).

Seed traits and nutrient contents

The effect of seed traits on germination rate was tested both with and without phylogeny. Accounting for phylogeny resulted in a non-significant effect of any of the tested variables. Without accounting for phylogeny, the germination rate was positively affected by the amount of nitrogen and negatively by seed elongation, i.e., round seeds had better survival than elongated seeds (see Table 3, first column). In all variables, the plant species with zero or extremely low germination rates were distributed along the entire measured gradient. However, no model with zero inflation argument improved the fit.

Table 3

Results of the GLMM of all species, only *Fabaceae*, only *Poaceae*, and all other species. P-values for the significant variables are shown with arrows indicating the direction of effect (↑ for positive effect and ↓ for negative effect). A different trend is visible for the most important driver of species germination success - the seed elongation with a negative effect in both *Fabaceae* and *Poaceae*, and a positive effect in the remaining species.
	All species		Fabaceae		Poaceae		Other species
Seedmass	ns.	–	0.001	↑	ns.	–	0.01	↑
P	ns.	–	0.01	↑	0.05	↓	< 0.001	↑
N	0.01	↑	< 0.001	↓	ns.	–	< 0.001	↓
Starch	ns.	–	ns.	–	0.02	↑	ns.	–
Oil	0.06	–	ns.	–	ns.	–	ns.	–
Mucilage	ns.	–	—	—	ns.	–	ns.	–
Elongation	0.01	↓	< 0.001	↓	0.08	–	< 0.001	↑

The trait analyses have also been done separately for the families with the highest number of species: Fabaceae and Poaceae, and for the remaining species. Different results were found for individual groups of species. In some cases, we found a positive effect of a trait or nutrient in one group and a negative in another (Table 3, Fig. 2 for red deer). The effect of seed phosphorus content which was non-significant in the test for all species, was significant for all groups separately: a positive effect was found in Fabaceae and other species, and negative in Poaceae. The seed nitrogen content had a significantly positive effect when tested for all species but negative in both Fabaceae and the group of other species, and non-significant for Poaceae. Last but not least, the effect of seed elongation was negative (better survival of round seeds) in all but the group of other species where it had a positive effect (better survival of elongated seeds). Contrary to our predictions, none of the tests revealed any relationship between seed germination rate and the presence of the mucilaginous surface.

Differences between herbivores

There was no significant relationship between seed traits and the species survival after passage through different species of animals as tested by RLQ and fourth corner analysis. However, there were significant differences in the survival of seeds after passage through different species of herbivores (GLMM, p < 0.001). Post hoc testing revealed that after passage through red deer, germination was significantly higher than after passage through any other tested herbivore (Table 2, Fig. 3).

Interestingly, red deer dung samples collected after four days from the feeding trials still contained seeds of some species suggesting the retention time was underestimated. This was the case for Alopecurus pratensis, Trifolium arvense, and Clinopodium vulgare, and at a lower rate for Hypericum perforatum, Plantago media, Potentilla argentea, and Securigera varia (Fig. 1). However, this has not been the case for other animal species, even though in sika deer, the collection was omitted on day 3 in phase 2.

None of the species showed increased germination after passing through the digestive tract. In agreement with our predictions, we found that round seeds were more likely to survive the passage through the digestive tract. This held true when tested for all species together as well as for Fabaceae separately. However, when Fabaceae and Poaceae were excluded, the trend was opposite. The seed germination rate was higher after passage through red deer compared to any other herbivore supporting hypothesis (a) “body size has a positive effect on germination rate.” Contrary to our prediction, the seed nutrient content had different effect based on the plant species group and the nutrient in question. Contrary to our previous results, we did not find support for the hypothesis that seeds with the mucilaginous surface will survive the passage through the digestive tract better.

Only one tested trait showed a sufficiently significant effect on the germination success of all species: the seed elongation (index of shape). We found that, in general, round seeds survived better than elongated seeds. When tested separately for Fabaceae, Poaceae, and all other species, this held true for all but the last where survival and subsequent germination success were better for the species with elongated rather than round seeds. This could also be an explanation for contrasting support in literature: better survival of round seeds was reported by Mouissie et al. (2005), better survival of elongated seeds in Cosyns et al. (2005), and no significant relationship between seed survival and shape in D’hondt and Hoffmann (2011). The cited studies used a range of plant species but only in the last case did the authors take phylogenetic relationships into account. Even though seed traits are often phylogenetically conserved (Moles et al., 2005), analyses with phylogeny might not receive significant effects (Bello et al., 2015), as was also our case. This suggests that both the tested traits and the environmental filter (here, the passage through herbivore guts) are phylogenetically conserved (Bello et al., 2015). Furthermore, in the cited studies (Cosyns et al., 2005; Mouissie et al., 2005; D’hondt and Hoffmann, 2011), the seed shape measurements used were made on seeds with no appendages, whereas in the presented study we used data from the LEDA database where seeds were measured with appendages. When grazing, herbivores feed on vegetation and the seeds they consume are not cleaned of glumes or pericarps. However, such structures can provide extra protection for seeds in the digestive tract, and therefore, they should be included in the measurements.

Our results are partially supporting our hypothesis that seed survival and subsequent germination success will be directly influenced by seed roundness and the presence of mucilaginous surface. The effect of shape has been addressed above, but we have found no effect of mucilaginous surface on seed survival and germination rate, which is in contrast to our previous findings (Lepková et al., 2018). The mucilaginous surface is a rarely occurring trait which is believed to have a connection to seed survival in the digestive tract (Hintze et al., 2013). Some species exhibiting mucilage were very common in field-collected dung, e.g., Poa pratensis and Veronica chamaedrys, but here, all tested species with mucilage showed very low germination rates (less than 1%). However, the low germination rate of frequently dispersed species is common throughout the studied species set. Two reasons are plausible: (i) seeds from the commercial supplier were of insufficient quality. This is not probable because the species showed successful germination in the control pots. (ii) The animals in the field ingest numbers of seeds several orders of magnitude higher than what we fed them in the experiment. This suggests preferential grazing which has nothing to do with seed traits or their nutritional content but can be driven by other traits of mother plants which influence palatability.

Previous research showed that even a very similar setting of feeding experiment does not guarantee similar outputs. For example, in a multi-species study with a number of plant species fed to cattle, Cosyns et al. (2005)d hondt and Hoffmann (2011) found vastly different results in the species of plants included in both studies, e.g., the relative germination rate of Agrostis capillaris was 17 and 54%, respectively. In the presented study, the measured germination rate for the same plant species was effectively zero, no matter the herbivore species. However, this species is one of the most common grasses dispersed by wild herbivores (present in 35% and 20% of deer and wild boar dung samples, respectively; Lepková et al., 2018). This example only emphasizes the high discrepancy between different types of experiments and also between similar experiments but under the influence of naturally behaving animals (for the effect of animal personality on seed dispersal see Zwolak and Sih, 2020).

The most successful species were members of the Fabaceae family (similarly to Gardener, McIvor, and Jansen, 1993), which is in contrast to often used laboratory experiments (Milotić and Hoffmann, 2016) and shows the importance of testing with real animals. The success of the Fabaceae family is often explained by their mechanical characteristics (e.g., thick seed coat; Gardener, McIvor, and Janes, 1993). In the case of the presented study, the best surviving Fabaceae had round and large seeds. The effect of seed mass is counterintuitive and unexpected as endozoochorous seeds are usually small (Bruun and Poschlod, 2006; Albert et al., 2015). This result has been clearly driven by one species: Vicia cracca with the highest germination success, and when the species was excluded, the seed mass became non-significant.

Our experiment revealed a complex relationship between seed survival in the guts and the content of available nutrients. In both cases of nitrogen and phosphorus, we found diverging results when groups of plant species were tested separately (Table 3). Both legumes and grasses are known for specific amounts of seed nutrients (Mašková and Herben, 2021), and we expected these nutrient contents to affect the seed survival, or more precisely, the effectivity with which animals extract the nutrients and thus kill the seeds. However, we revealed relationships contrary to our predictions. In Fabaceae and the group of other species (legumes and grasses excluded), we found a significant positive effect of phosphorus content on seed survival even though ruminant herbivores are preadapted to digest the phytic acid in which phosphorus is stored (Klopfenstein et al., 2002). Only the effect of phosphorus on the survival of grasses was significantly negative and, as a result, the overall test of all species did not show a statistically significant effect. This suggests the herbivore species included in this study are more adapted to the digestion of graminoid seeds, which are more common in their diet in Central Europe (Spitzer et al., 2020).

Differences between herbivore species

Four herbivores exhibiting different feeding and digestive behavior were used for the experiment, and differences based on body size, length of the digestive system, and/or feeding style were expected. Since at the time of the feeding experiment, only one individual of omnivorous wild boar was available, comparing herbivores and omnivores is outside the scope of the presented study. Our results support our hypothesis (a), which states that the driving force is the size of the animal. Dung samples from red deer, as the largest animal (Anděra and Horáček 2005), showed the highest germination rates of seedlings. However, the effect of body size is contradictory to other literature using feeding experiments (Simao Neto et al., 1987; Cosyns et al., 2005).

Experiments with multiple animal species are rare, and the majority of published data are on domestic animals (Bonn, 2004; Cosyns et al., 2005). Therefore, it is difficult to further address the effect of body size on seed survival. Chen and Moles (2015) performed a meta-analysis on the relationship between seed size, seed dispersal, and disperser size. They found out that in large ungulates in particular the relationship with seed size is negative, i.e., the large animals primarily ingest small-seeded species. The relationship is even more complicated because ruminants spit large seeds which are not digested at all (Castañeda et al., 2018). This complexity can also be the reason for the non-significant effect of seed size in our dataset.

Speed of passage

Since some species were still germinating in significant numbers from samples from the fourth day of collection, we must assume the retention time was longer than 96 hours (but see Cosyns et al., 2005). Our results show that Fabaceae had a slower passage through the guts (Gardener, McIvor, and Jansen, 1993). Combined with the fact that Fabaceae also had the highest measured survival success, we can assume the survival was still underestimated. This is in contrast to our prediction that long passage through the digestive tract shall be more destructive for the seeds. However, this result also means that seeds, which stay in the digestive tract for this long, can be dispersed further away from the mother plant which can even compensate for losses during the passage (Janzen, 1984).

Our experiment with almost forty species of plants and four herbivores did not reveal a single seed trait that predicts seeds survival when passing through the herbivore digestive system. The most important trait was the seed shape, but we found contrasting effect on species from different families. It is possible an important piece of knowledge may be missing to understand the survival of seeds better, e.g., seed coat thickness. However, based on our comprehensive experiment, we conclude, that digestion by wild, free-ranging herbivores does not act as environmental filter and the seed survival after passage through digestive system is driven by chance rather than seed traits.

Acknowledgements

We thank Tomáš Herben for his input in the experiment planning, data collection and analyses, and for his feedback to the manuscript. We thank Toášs Peterka, Adam Veselý, the entire PEE group, and zookeepers in Malá Chuchle. We thank the Charles University for financial support (grant no. 325215) and Prague Municipal Forests for enabling us to perform the feeding experiment.

The authors declare that the experiment has been performed and complies with laws of the Czech Republic.

Funding: This research has been supported by Charles University Grant Agency (GAUK, grant no. 325215).

Conflict of interest: The authors declare no conflict of interest.

Ethics approval: All applicable institutional and national guidelines for the care and use of animals were followed. Ethics approval for this study was not required according to Act No. 246/1992 Coll. Of the Legislation of Czech Republic.

Consent to participate: Not applicable

Consent for publication: Not applicable

Availability of data and material: All data produced from this study are provided in this manuscript.

Code availability: Not applicable

Authors' contributions: BL planned and performed the feeding and germination experiments. TM collected the data on seed nutrient content. BL analyzed the data and wrote the manuscript, TM provided feedback for data analyses and revised the manuscript. BL and TM both approved the final version.

Anděra M, Horáček I (2005) Poznáváme naše savce. Sobotáles, Praha, II. edition
Albert A, Auffret AG, Cosyns E et al (2015) Seed dispersal by ungulates as an ecological filter: a trait-based meta-analysis. Oikos 124:1109–1120. https://doi.org/10.1111/oik.02512
Baker DL, Hobbs NT (1987) Strategies of digestion: digestive efficiency and retention time of forage diets in montane ungulates. Can J Zool 65:1978–1984. https://doi.org/10.1139/z87-301
de Bello F, Berg MP, Dias ATC et al (2015) On the need for phylogenetic ‘corrections’ in functional trait-based approaches. Folia Geobotanica 349–357. https://doi.org/10.1007/s12224-015-9228-6
Bonn S (2004) Dispersal of plants in the Central European landscape – dispersal processes and assessment of dispersal potential exemplified for endozoochory. PhD thesis
Brooks ME, Kristensen K, van Benthem KJ et al (2017) glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R J 9:378–400. https://doi.org/10.32614/rj-2017-066
Bruun HH, Poschlod P (2006) Why are small seeds dispersed through animal guts: large numbers or seed size per se? Oikos 113:402–411. http://www.jstor.org/stable/40234823
Cain ML, Damman H, Muir A (1998) Seed Dispersal and the Holocene Migration of Woodland Herbs. Ecological Monographs 68:325–347. https://doi.org/10.1890/0012-9615(1998)068%5B0325:SDATHM%5D2.0.CO;2
Castañeda I, Fedriani JM, Delibes M (2018) Potential of red deer (Cervus elaphus) to disperse viable seeds by spitting them from the cud. Mammalian Biology 90:89–91. https://doi.org/10.1016/j.mambio.2017.10.004
Chen S-C, Moles AT (2015) A mammoth mouthful? A test of the idea that larger animals ingest larger seeds. Global Ecology and Biogeography. https://doi.org/10.1111/geb.12346. n/a–n/a
Clauss M, Hummel J, Streich WJ (2006) The dissociation of the fluid and particle phase in the forestomach as a physiological characteristic of large grazing ruminants: An evaluation of available, comparable ruminant passage data. Eur J Wildl Res 52:88–98. https://doi.org/10.1007/s10344-005-0024-0
Cosyns E, Delporte A, Lens L, Hoffmann M (2005) Germination success of temperate grassland species after passage through ungulate and rabbit guts. J Ecol 93:353–361. https://doi.org/10.1111/j.1365-2745.2005.00982.x
D’hondt B, Hoffmann M (2011) A reassessment of the role of simple seed traits in mortality following herbivore ingestion. Plant Biol (Stuttgart Germany) 13:118–124. https://doi.org/10.1111/j.1438-8677.2010.00335.x
Gardener CJ, McIvor JG, Janes A (1993a) Survival of seeds of tropical grassland species subjected to bovine digestion. J Appl Ecol 30:75–85. https://doi.org/10.2307/2404272
Gardener CJ, McIvor JG, Jansen A (1993b) Passage of Legume and Grass Seeds Through the Digestive Tract of Cattle and Their Survival in Faeces. J Appl Ecol 30:63–74. https://doi.org/10.2307/2404271
Hejcmanová P, Pokorná P, Hejcman M, Pavlů V (2016) Phosphorus limitation relates to diet selection of sheep and goats on dry calcareous grassland. Appl Veg Sci 19:101–110. https://doi.org/10.1111/avsc.12196
Hintze C, Heydel F, Hoppe C et al (2013) D3: The Dispersal and Diaspore Database - Baseline data and statistics on seed dispersal. Perspect Plant Ecol Evol Syst 15:180–192. https://doi.org/10.1016/j.ppees.2013.02.001
Hofmann RR (1989) Evolutionary steps of ecophysiological adaptation and diversification of ruminants: a comparative view of their digestive system. Oecologia 78:443–457. https://doi.org/10.1007/BF02352565
Illius AW, Gordon IJ (1992) Modelling the nutritional ecology of ungulate herbivores: evolution of body size and competitive interactions. Oecologia 89:428–434. DOI: 10.1007/BF00317422
International Organization for Standardization (2016) Occupational health and safety management systems – Requirements with guidance for use (ISO/DIS Standard No. 45001)
Janzen DH (1984) Dispersal of Small Seeds by Big Herbivores: Foliage is the Fruit. Am Nat 123:338–353. https://doi.org/10.1086/284208
Kleyer M, Bekker RM, Knevel IC et al (2008) The LEDA Traitbase: A database of life-history traits of the Northwest European flora. J Ecol 96:1266–1274. https://doi.org/10.1111/j.1365-2745.2008.01430.x
Kleyer M, Dray S, de Bello F et al (2012) Assessing species and community functional responses to environmental gradients: Which multivariate methods? J Veg Sci 23:805–821. https://doi.org/10.1111/j.1654-1103.2012.01402.x
Klopfenstein TJ, Angel R, Cromwell G et al (2002) Animal Diet Modification to Decrease the Potential for Nitrogen and Phosphorus Pollution. Council For Agricultural Science And Technology 21:15
Lepková B, Horčičková E, Vojta J (2018) Endozoochorous seed dispersal by free-ranging herbivores in an abandoned landscape. Plant Ecol 219. https://doi.org/10.1007/s11258-018-0864-9
Li M, Bolker B(2021) phyloglmm: Machinery for Phylogenetic. GLMMs. R Package Version 0.1.0.9001. DOI: 10.5281/zenodo.2639887
Lozada-Gobilard S, Jeltsch F, Zhu J (2021) High matrix vegetation decreases mean seed dispersal distance but increases long wind dispersal probability connecting local plant populations in agricultural landscapes. Agric Ecosyst Environ 322:107678. https://doi.org/10.1016/j.agee.2021.107678
Mašková T, Herben T (2021) Interspecific differences in maternal support in herbaceous plants: CNP contents in seeds varies to match expected nutrient limitation of seedlings. Oikos 130:1715–1725. https://doi.org/10.1111/oik.08186
McCleary B, Solah V, Gibson TS(1994) Quantitative measurement of total starch in cereal flours and products. Journal of Cereal Science, 20, 51–58. International Organization for Standartization (2016). Occupational health and safety management systems - Requirements with guidance for use (ISO/DIS Standard No. 45001)
Milotić T, Hoffmann M (2016) How does gut passage impact endozoochorous seed dispersal success? Evidence from a gut environment simulation experiment. Basic Appl Ecol 17:165–176. https://doi.org/10.1016/j.baae.2015.09.007
Moles A, Ackerly D, Webb C et al (2005) A Brief History of Seed Size. Science 307:576–580. https://doi.org/10.1126/science.1104863
Mouissie AM, Veen GFC, Van Der Veen C, Van Diggelen R (2005) Ecological correlates of seed survival after ingestion by Fallow Deer. 19:284–290
Pakeman RJ, Digneffe G, Small JL (2002) Ecological correlates of endozoochory by herbivores. Funct Ecol 16:296–304. https://doi.org/10.1046/j.1365-2435.2002.00625.x
R Core Team (2021) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/
Schley L, Roper TJ (2003) Diet of wild boar Sus scrofa in Western Europe, with particular reference to consumption of agricultural crops. Mammal Rev 33:43–56. https://doi.org/10.1046/j.1365-2907.2003.00010.x
Simao Neto M, Jones RM, Ratcliff D (1987) Recovery of Pasture Seed Ingested by Ruminants. 1. Seed of Six Tropical Pasture Species Fed to Cattle, Sheep and Goats. Aust J Exp Agric 27:239–246. https://doi.org/10.1071/EA9870239
Sitters J, Maechler MJ, Edwards PJ et al (2014) Interactions between C: N: P stoichiometry and soil macrofauna control dung decomposition of savanna herbivores. Funct Ecol 28:776–786. https://doi.org/10.1111/1365-2435.12213
Spitzer R, Felton A, Landman M et al (2020) Fifty years of European ungulate dietary studies: a synthesis. Oikos 129:1668–1680. https://doi.org/10.1111/oik.07435
Tackenberg O, Römermann C, Thompson K, Poschlod P (2006) What does diaspore morphology tell us about external animal dispersal? Evidence from standardized experiments measuring seed retention on animal-coats. Basic Appl Ecol 7:45–58. https://doi.org/10.1016/j.baae.2005.05.001
Thioulouse J, Dray S (2007) Interactive multivariate data analysis in R with the ade4 and ade4TkGUI packages. J Stat Softw 22:1–14. https://doi.org/10.18637/jss.v022.i05
Thompson K, Band SR, Hodgson JG (1993) Seed Size and Shape Predict Persistence in Soil. Funct Ecol 7:236–241. https://doi.org/10.2307/2389893
Zwolak R, Sih A (2020) Animal personalities and seed dispersal: A conceptual review. Funct Ecol 34(7):1294–1310. https://doi.org/10.1111/1365-2435.13583

ESM1.xls

Download PDF

Reviewers agreed at journal
29 Aug, 2022
Reviewers invited by journal
29 Aug, 2022
Editor assigned by journal
18 Jun, 2022
First submitted to journal
15 Jun, 2022

You are reading this latest preprint version

Herbivore digestion as environmental filter - which seed traits help species survive?

Status:

Version 1

Abstract

Figures

Introduction

Material And Methods

Seed traits

Data analyses

Results

Seed traits and nutrient contents

Differences between herbivores

Discussion

Differences between herbivore species

Speed of passage

Conclusions

Declarations

Acknowledgements

References

Supplementary Files

Status:

Version 1