A generalist herbivore requires a wide array of plant species to maintain its populations

Highlights

• A generalist grasshopper, Celes akitanus, occurred in grasslands with higher plant species richness.

• C. akitanus exhibited non-random plant selection in its diet, despite its generalist nature.

• High plant species richness in its habitat is important for the persistence of populations of this generalist insect.

Abstract

Generalist herbivores are less susceptible to changes in the plant composition of their habitats than specialists are because generalists can consume a diverse array of plants. However, even generalists exhibit dietary choices, for example because they need to balance their nutritional intake for both growth and reproduction. In this study, we showed that an endangered generalist herbivorous grasshopper (Celes akitanus; Orthoptera: Acrididae) actively chooses which plants to include in its diet. Moreover, we found that grasshopper abundance is correlated with host plant abundance and richness. This grasshopper has been reported to occur mainly in traditionally managed grasslands that harbour more diverse plant species than other, nearby grasslands. To elucidate the links between this grasshopper and plant richness, we surveyed grasshopper abundance in grasslands under traditional and other management practices. Plant DNA barcoding of faecal samples demonstrated that this grasshopper is a generalist herbivore while also showing that it makes active dietary choices. Furthermore, although the grasshopper's host plants occurred in all grasslands, the grasshopper itself was found only in species-rich grasslands. In addition, grasshopper abundance was positively related to the abundance and richness of host plants. Our findings suggest that this endangered herbivore requires a wide array of host plants to maintain its populations.

Introduction

For herbivorous insects, diet breadth plays an essential role in a range of ecological phenomena, such as population dynamics (Eskildsen et al., 2015; Descombes et al., 2016) and interspecific interactions (Carvalheiro et al., 2010; Charles and Linklater, 2013). The risk of extinction may also be related to diet breadth (Goulson et al., 2005). For example, the abundance of specialist herbivores decreases when the proportional abundance of their host plants in their habitat decreases (Jonsen and Fahrig, 1997; Steffan-Dewenter and Tscharntke, 2000; Rand and Tscharntke, 2007), and the loss of a host plant causes local extinctions of any specialist herbivores that depend on it (Uchida and Ushimaru, 2014). Thus, specialist herbivores are susceptible to changes in the plant community within their habitats. Conversely, the loss of potential host plants is less likely for generalist herbivores than for specialists because they can use a wide range of plant species. Therefore, the risk of extinction for generalist herbivores does not seem to be associated with changes in plant community composition; rather, declines in generalist herbivorous insects are widely accepted to be caused by such changes as the loss of suitable habitat (Thomas, 1985; Warren et al., 2001) and changes in vegetation structure (Gardiner et al., 2002; Hochkirch et al., 2008).

However, under certain circumstances the composition of the plant community in the habitat may even affect the fitness of generalist herbivores. For example, similarly to specialist herbivores, a population of generalist herbivores cannot persist if the overall abundance of host plants in its habitat is low (i.e., the herbivore is relatively oligophagous). In other cases, an array of plant species might be important to a generalist's population persistence. Indeed, some generalist herbivores show a preference for certain host plants, although they can potentially consume a wide range of plant species (Ibanez et al., 2013; Huang et al., 2016). If maturation and/or reproduction is constrained by the shortage of some nutrients, a potentially generalist herbivore should still prefer the host plants that include such essential nutrients. Growth and reproduction can be controlled by controlling the balance among different nutrients consumed (Raubenheimer and Simpson, 1993; Bernays et al., 1994; Lee et al., 2008; Behmer and Joern, 2008; Behmer, 2009; Franzke et al., 2010; Polak et al., 2017), so the extinction risk of generalist herbivores may be increased not only by the loss of habitat, but also by changes in their habitats' plant communities. Therefore, to evaluate habitat quality for the effective conservation of threatened generalist herbivores it is essential to know exactly what plants the herbivore eats.

Decreases in populations of Celes akitanus (Orthoptera: Acrididae), which is considered a generalist grasshopper, might be associated with plant community changes. Populations of C. akitanus were distributed across central and northern Honshu, Japan, until the early 20th century (e.g., Uchida et al., 2016a), but at present only a few populations are known to exist in distantly separated localities. An intensive investigation of the plant and insect communities of one of the localities with extant populations (in Nagano Prefecture) revealed that C. akitanus has survived in only a few specific types of grasslands (Uchida et al., 2016a). In that locality, grasslands are subjected to four types of management (traditional, burning, mowing, and abandonment; see Materials and methods for details), and C. akitanus is found exclusively in grasslands subjected to one of these, specifically traditional management (hereinafter referred to as traditional grasslands). Given that in other localities this grasshopper species also occurs in grasslands subjected only to burning or mowing management, as well as in seaside natural grasslands (Nagahata, 2006; Okatsu et al., 2018; Uchida, unpublished data), it seems unlikely that differences in management practices are the direct cause of its exclusive occurrence in traditional grasslands in Nagano. However, grassland plant communities vary depending on management practices, and the traditional grasslands harbour plant communities of higher species richness than those in other types of grasslands. We therefore hypothesized that the characteristics of plant communities mediate the persistence of C. akitanus populations; this could be, for example, because certain host plants exist exclusively in traditional grasslands, or because the grasshopper may need to inhabit a community with high plant species richness to meet some of its physiological or life history requirements (e.g., to control the balance of nutrients).

In the present study, we examined the abundance of C. akitanus in various grasslands to explore the relationship between this grasshopper and plant species composition. In addition, to empirically assess whether this grasshopper is a generalist we determined its host plant species and the frequency of feeding on each host plant using DNA barcoding of its faeces. Based on these results, we discussed the hypothesis that the persistence of C. akitanus populations is associated with host plant composition in semi-natural grasslands.