DISCUSSION
For a deeper understanding of the potential coevolution between lemur and plants, we tested for cophylogenetic signals in plant-lemur interactions, and detected significant, although weak, signals in frugivory, folivory, florivory and granivory interactions. In the case of frugivory, plant speciation seems to have influenced lemur body size, at the same time that lemur speciation modulated diaspore size. Weak cophylogenetic signals do not necessarily imply lack of evolutionary history shared between symbiotic partners. Indeed, cophylogenetic patterns observed for frugivory, florivory and granivory emerged from lemurs sharing similar sets of partners interacting with phylogenetically-related plants with no reciprocal trend. In the case of folivory, plant diversification was also affected by lemurs. Our results do not give support to a strong coevolutionary diversification between lemurs and plants, in agreement with previous observations for Neotropical primates and Angiosperms (Fuzessy et al. 2022). Instead, our findings reflect that lemur feeding behavior is very generalist, specially of large lemur species, and support the idea that cophylogenetic signals are modulated by processes that are not directly related to strong co-speciation events, such as vicariance, phylogenetic tracking, and diffuse coevolution (Blasco-Costa el al. 2021).
The role of vicariance in coevolution —where common biogeographic events cause subdivision of distribution ranges of interacting partners —is dependent on the geographic scale. Vicariance contributes to coevolutionary processes by promoting diversification of entire populations that become isolated from each other. Over time, as populations diverge and adapt to their respective environments and evolve distinct and specialized structures and traits, including those that are important for their close relationship, they become less compatible. In this sense, vicariance is associated with high cophylogenetic signals (Blasco-Costa el al. 2021), so in the specific case of Madagascar, especially in the context of a relatively small island, it is not expected to play a significant role. Phylogenetic tracking is also usually associated with high cophylogenetic signals. It refers to a tendency of interacting partners to track each other’s evolutionary history, such that closely related species are more likely to interact than distantly related species (Blasco-Costa et al. 2021). Despite our finding of low cophylogenetic signals in lemur-plant interactions, phylogenetic congruence arose even in a supergeneralist system where other plant-consuming animals, such as insects and other vertebrates, may also be shaping plant diversification. The patterns observed here for frugivory, florivory and granivory emerged from similar lemurs interacting with closely related plants with no reciprocal trend (i.e., plant diversification was unaffected by lemurs). In this case, generalist lemurs seem to be tracking temporal fluctuations in resource availability, and increasing the consumption of one plant species at the expense of decreased consumption of an alternative, phylogenetically-related, resource, known as switching behavior (Carnicer et al. 2008). Such low levels of CS associated with a correlation between phylogenetic distances and dissimilarities in sets of interacting plants for lemurs can be interpreted as an indicative case of diffuse coevolution. In this sense, each lemur species interacts with a set of plants, and experiences a complex suite of selective pressures derived from the set of plant species they interact with.
In the case of folivory, similar plants were also found to interact with closely related lemur species, as an indicative of coevolution. Except for the strict folivorous genera Hapalemur and Lepilemur , lemurs use a variety of resources as feeding items, such as fruits, leaves flowers, seeds, nuts, nectar, and animal prey. Regardless of the main resource, leaves constitute an important resource for most species, reinforcing the idea that lemur-plant interactions are quite generalist. A weak cophylogenetic signals associated with positive correlations between phylogenetic distances and dissimilarities in sets of interacting partners for both lemurs and plants may also be interpreted as a case of diffuse coevolution, since distinct amounts of each feeding item vary across taxa, and other coexisting frugivorous and herbivorous clades act together, thus, favoring a diffuse coevolutionary process (Erikson 2016).
The multi-specific nature of the process delineating cophylogenetic patterns, as observed here, does not necessarily mean either that lemurs and plants have not coevolved, or that coevolution has necessarily been the primary force fueling diversification (e.g. Althoff et al. 2014; Fuzessy et al. 2022). Taken together, our results agree with the notion that vertebrates and plants can influence each other evolution but also highlight that they do not share a very tight coevolutionary history (Valenta & Nevo 2020, Fuzessy at al. 2022), such as those observed in host-parasite interactions (Gandon & Michalakis 2002) or in some plant-pollinator systems (Herrera 2019). Spatio-temporal asymmetries, disruptions in the strength of the relationships between interacting partners, and shifts between periods of coevolution and independent evolution seem to be leading to adaptive changes, ultimately resulting in a weak and continuous process (Erikson 2016), in agreement with previous observations in the Neotropics (Fuzessy et al. 2022). The resemblance between Madagascar and the Neotropics is remarkable, given that the number of frugivorous mammals in Madagascar is low when compared to other tropical communities, which are dominated for frugivorous primates (Fleming et al. 1987; Goodman & Ganzhorn 1997). In Madagascar, there are very few medium-sized strictly frugivorous lemurs (mostly Varecia and some species of Eulemur ), and the proportion of fruits consumed by Malagasy lemurs is low compared with primates in other regions (Fleagle & Reed 1996). The distinctive phenology of Malagasy forests, associated with a great variability in spatio-temporal fruit availability has been proposed as the main drivers of the underrepresentation of frugivores in Madagascar (Dewar & Richar 2007). These patterns highlight potential risks associated with the current disruption of lemur-plant interactions. Despite high levels of generalism, we have shown that extant lemur-plant interactions are phylogenetically constrained, so if a species belonging to an already limited community of frugivores is lost, the probability of other species to fulfill the vacant role is reduced. This issue may be exacerbated if the species lost play the role of both dispersers and pollinators, as is the case between the black-and-white ruffed lemur,Varecia variegata , and the traveler palm, Ravenala madagascariensis (Kress et al. 1994). Such double mutualisms appear to be highly prevalent in island ecosystems, and pose additional risks for the continuing health of these forests when facing the extinction of frugivorous animal partners (Fuster et al. 2019).
Overall, lemurs consumed plant species that were more closely related than expected by chance, potentially indicating the consumption of plants with similar traits. In addition, invasive plant species have not disrupted the cophylogenetic structure of the interaction networks, suggesting a deep conservatism of interactions (or a low invasion rate in Madagascar). Indeed, we found no clear pattern of the influence of plant traits on the cophylogenetic signals. Nevertheless, to make any precise statement in this regard, we need complete and more detailed information on plant traits, including those related to fruits, leaves, and flowers, which are most frequently consumed by lemurs. Therefore, we encourage future studies to characterize Malagasy flora in terms of functional characteristics related to interactions with mutualistic and antagonistic partners (e. g. Albert-Daviaud et al. 2021).
Lemur feeding preferences and activity patterns likely shaped the congruence between their phylogeny with that of plants. Due to the unique ecology of Madagascar, lemurs evolved to occupy a wide range of niches, from night-living arboreal insectivores to ground-dwelling cathemeral folivores-frugivores, and diurnal frugivores (Gould & Sauther 2006). As expected, nocturnal lemurs and those that include a great amount of insects in their diets (and the least amount of plants), showed the largest Procrustean residuals, and contributed the least to the shared evolutionary history with plants, regardless of interaction type, and in agreement with previous studies (Fuzessy et al. 2022). Overall, diet and activity patterns have been important drivers of lemur evolution, shaping morphology, physiology, and behavior in unique ways that have allowed them to thrive in Malagasy diverse habitats. Dietary adaptations have driven the evolution of morphological and physiological features, such as specialized teeth, tongue, digestive systems, and metabolic rates (Cuozzo & Yamashita 2006; Powzyk & Mowry 2006; Pastor et al. 2021). Over evolutionary time, insectivores such asDaubetonia , for instance, have evolved specialized teeth and long fingers that are adapted for catching and handling prey, added to an enhanced night vision and specialized auditory systems to aid in hunting (Goodman et al. 1993; Sterling & McCreless 2006; Lhota et al. 2008). On the other hand, lemurs that consume great amounts of fruits and flowers have evolved color vision to help them distinguish ripe fruits and identify flowers, specially from a conspicuous background (Nevo et al. 2018).
Our study highlighted the important role of evolutionary history in structuring modern interaction patterns between vertebrates and plants. Our cophylogenetic analysis compared phylogenies of interacting partners based on observed extant interactions presented as binary lemur-plant interaction matrices. While this approach provides an important basis for investigating cophylogenetic relationships (Balbuena et al. 2013; Blasco-Costa et al. 2021), it ignores that the strength of associations among interacting partners may affect evolutionary outcomes, and thus precludes a deeper assessment of cophylogenetic processes. In the case of frugivory interactions, in addition to the congruence between plant and lemur phylogenies, we also detected a congruence between lemur phylogeny and plant morphology, and between plant phylogeny and lemur morphology. Although this result suggests that, to some extent, plant speciation shaped lemur body size, and lemur speciation somehow affected diaspore morphology, we still lack sufficient information on the role of plant traits, including those related to fruits, leaves, and flowers, in shaping the cophylogenetic patterns described. Studies characterizing Malagasy flora in terms of functional characteristics related to interactions with mutualistic and antagonistic partners are urgently required (e. g. Albert-Daviaud et al. 2021). Despite these limitations, results presented here shed unprecedented light on the potential for supergeneralist interactions, including both antagonisms and mutualisms, to shape shared evolutionary trajectories between vertebrates and Angiosperms.