LAUSR

The interactions between insect herbivores and their host plants are a truly fascinating area of study, primarily because they are multifaceted, dynamic and fundamental to explaining the diversity and abundance of both taxa. Indeed, interactions between these two disparate taxa are responsible for most of the terrestrial diversity on Earth (Futuyma & Agrawal 2009). Research into the evolutionary consequences of plant defence has provided compelling evidence of the effects of proliferation in phytochemical complexity on insect herbivore specialisation and diversity. For example, the phylogenies of Blepharida (Coleoptera: Chrysomelidae) and Bursera, the latter is a genus of approximately 100 species of plant endemic to regions from the south-western United States to Peru, indicate that host shifts by beetles have been onto species of plant with comparable suites of plant secondary metabolites (PSMs) to hosts (Becerra 1997). Interestingly, older species of Bursera have less diverse suites of PSMs than more recent species supporting an escape and radiate model of coevolutionary divergence (Becerra et al. 2009). Similar findings have been reported in the Neotropical tree genus Inga, although variation in defensive metabolites exhibits little phylogenetic signal (Kursar et al. 2009). The implications arising from these macroevolutionary and macroecological studies of herbivory are that reciprocal evolution is fundamental to the high rates of host specificity exhibited by phytophagous insects and shapes entire communities of plants and insects (Becerra 2015; Richards et al. 2015). Reciprocal adaptation and speciation of Eucalyptus and Acacia and certain groups of insect herbivores is likely to have occurred (Austin et al. 2004); however, the molecular, chemical and phylogenetic studies required to demonstrate such evolution is yet to be conducted. Consequently, we are still unable to infer the mechanistic processes that led to the huge diversity of insect herbivores supported by these two dominant plant genera. It is important that such studies are conducted because there is evidence that the divergence of some insect taxa which are functionally significant in Australian ecosystems might not be explained using the escape and radiate coevolutionary paradigm. For instance, association by colonisation has been inferred for a number of psyllids that utiliseGenista species of host onAtlantic Macaronesian islands; these associations are consistent with sequential evolution rather than contemporaneous co-speciation of psyllids and their hosts (Percy et al. 2004). Nevertheless, as woody, apparent plants, species of Eucalyptus and Acacia experience greater herbivory than non-woody species which must partly explain the high diversity of some groups of native insect herbivores (Turcotte et al. 2014). The evolutionary flipside of plant defence is insect host specificity, the expression of which involves behavioural and physiological components exhibited by adult and/or immature lifecycle stages. In general, female insects are specific for species of plant, plant module and/or age of module that sustain optimal performance of their offspring (Gripenberg et al. 2010). Host plant quality is a key concept in terms of the ‘Mother knows best’ paradigm because it determines the performance of immatures which in turn determines the fecundity of the adults (Scriber & Slansky 1981; Awmack & Leather 2002). Using the most simplistic scenario, we can define plant quality for an insect herbivore on a given species of host using a two-dimensional space delimited by the concentration of a biologically active PSM and a nutrient essential for growth (often some form of nitrogen). The PSM (generally considered non-nutritive) chosen for such a characterisation of host plant quality is either known or assumed to have an adverse impact on performance while the effect of the nutritive element selected is either known or assumed to be positive. In this way, it is possible to compare individual variation in plant quality during a growing season or among habitats for discrete generations and/or populations of that species of insect (Hemming & Lindroth 1995; Riipi et al. 2004; Audusseau et al. 2016). The availability of nutritional and non-nutritional metabolites permits comparison of the suitability of different species and chemotypes of plant used by oligophagous and polyphagous species of insect (Fox & Macauley 1977; Matsuki & MacLean 1994; Steinbauer & Matsuki 2004; Wheeler 2006). Host plant quality can also be defined using silicon or fibre (components responsible for the physical characteristics of plant modules) in place of a PSM. Plant silicon, in particular as phytoliths of silica (SiO2), has long been included among traits providing constitutive defence against insect herbivores (Reynolds et al. 2009). The resistance of grasses, in particular, to small herbivores has been attributed to their abrasiveness (caused by high silica content); however, this hypothesis may only partly explain the effect of silica against mandibulate insects (Massey et al. 2006). Specifically, the high silica content of some grasses can reduce the digestibility of tissues following ingestion by insect herbivores (Massey et al. 2006; Hunt et al. 2008). Improved understanding of the mode of regulation of silicon deposition in crop and pasture grasses has fuelled a substantial upsurge in the study of this trait on levels of herbivory (Ye et al. 2013; Frew et al. 2017, 2018; Johnson & Hartley 2017). Given the interest in the effects of silica on herbivory, the contribution by Fiona Clissold and her colleagues is especially timely. Moreover, techniques that facilitate easier quantification of plant traits are vital so that our mechanistic understanding of insect host utilisation advances more rapidly than it has in past decades. Clissold et al. (2018) describe a quick way to measure silica in very small samples of leaf/faecal material, which in combination with measures of protein and carbohydrate for the same material, will provide a