Why so many polyphagous fruit flies (Diptera: Tephritidae)? A further contribution to the ‘generalism’ debate

 ,, – – Biological Society, 2016, .. With 3 figures. Biological Journal Journal of of the the Linnean Linnean Society, Society,2017, 2016,120, 245–257. REVIEW Biogeographical and co-evolutionary origins of Why so many polyphagous fruit flies (Diptera: scarabaeine dung beetles: Mesozoic vicariance versus Tephritidae)? A further contribution to the ‘generalism’ Cenozoic debate dispersal and dinosaur versus mammal dung 1 Scarab Department of ZoologySciences, & Entomology, University of Pretoria, P/B X20, SchoolResearch of Earth,Group, Environment and Biological Queensland University of Technology (QUT), Hatfield Brisbane,0028, Qld, South 4001, Africa Australia 2 Plant Biosecurity Cooperative Research Centre, LPO Box 5012, Bruce, ACT, 2617, Australia Received 22 June 2016; revised 28 July 2016; accepted for publication 28 July 2016 Received 5 May 2016; revised 8 July 2016; accepted for publication 8 July 2016 The subfamily Scarabaeinae (Coleoptera: Scarabaeidae) may have originated in Gondwanaland, through Mesozoic vicariance dispersal that in association dinosaurinsects dung, are or through Cenozoic‘improbability’. dispersal in association with It has beenorsuggested generalist with herbivorous an evolutionary However, many mammal We review evidence from age-calibrated phylogenies, fossiltwo records, biogeographical species ofdung. tephritid fruit flies are generalists utilising larval hosts across or more plant families.patterns, Focusingand on ecological calibrated phylogenies for Scarabaeoidea Cretaceous origin for the genus associations. Bactrocera, it Fossil is argued that this high frequency of generalism ispredict due to afour evolutionary drivers Scarabaeinae, although would rely on other scarabaeoid groups the given the validity which both decrease the age needestimates for specialism and promote generalism. Decreasing need fordoubtful specialism, there of is Mesozoic fossils. clockhost calibrated phylogenies an earlyimpact Cenozoic firstly no ‘scarabaeine’ co-evolutionary armsMolecular race between plant and herbivorefor as Scarabaeinae Bactrocera dopredict not negatively on origin coeval with modern the mammal diversification. fossil evidence suggests thethe exploitation of dinosaur plant fitness; this removes need for specialism to Trace counter plant defences. Secondly, endemic hosts of most dung by fossorial insects, although scarabaeine fossils only validated for the Cenozoic. Bactrocera are plants with vertebrate-dispersed fleshyarefruits. Fruit morphology and theirAlthough display we are discuss highly fossil evidence for dinosaur dungdispersers; as a faunal resource, thedoorigin of modern scarabaeines fromhost an location earlier constrained to attract those seed Bactrocera thus not need to specialize to increase Mesozoic vicariant or dispersal fauna remainssignal. unsupported. phylogram topography is capacity as many hosts produce a common Thirdly,Although in nativeclock-constrained, rainforest ecosystems Bactrocera are rare consistent with early Cenozoic palaeoclimatic and unlikely. palaeoecological events, Eocene are marine barriers would demand and so specialization to avoid competition seems Promoting generalism braconid parasitoids, which dispersal to cues explain the their distributional of Scarabaeinae. Inconsistencies between classification and utilise plant to find hosts. If a origins fly can use a novel host plant then it moves into enemy-free space, so phylogeny analysis, although earlier radiation of basally-derived tribestohas promoting complicate generalism.biogeographical From this multi-part hypothesis, a groupsouthern of testable predictions emerge. If shown be probably been for followed by later diversification and the global dispersal of both basally and an terminally-derived correct then the generalism/specialism debate frugivorous tephritids would provide example where taxa, primarily landevolutionary links. © 2016 The Linnean of London, Biological Journal of the Linnean generalism mayvia be restored a ‘probable’ outcome. © 2016Society The Linnean Society of London, Biological Journal Society, 2016, 00, 000–000. of the Linnean Society, 2016, 2017, 00, 120,000–000. 245–257. KEYWORDS: specialization – Gondwanaland – Scarabaeinae fossil record – Scarabaeinae KEYWORDS: dung Bactrocera – Dacinaedrivers – enemy-free space – frugivory – herbivory – polyphagy – specialism. phylogeny – tracks. INTRODUCTION INTRODUCTION Some 98% of insect herbivores are specialists, feeding Questions onorthe evolutionary of dung beetles on only one a small numberorigins of related plants (Jaein nike, the 1990).subfamily Because of Scarabaeinae their frequency,(Coleoptera: the theory Scarabaeidae) focused on evidence obtained supporting the have evolution of specialism is well develfrom foodmature specialization (Scholtz&&Southwood, Chown, 1995), oped and (Strong, Lawton 1984; fossil records (Chin & Gill, 1996; Krell, 2000), Bernays, 1988; Futuyma & Moreno, 1988; spatial Funk, distribution patterns Scholtz & Philips, which 2002; 2010), but this is not(Davis, the case for generalism Davis, 2009), and temporal frameworks derivedevent, from has been dismissed as a transitory evolutionary classification 1991a) ora passing morphological e.g. It appears (Cambefort, that generalism is only means (Philips, & Scholtz, 2004;a human Tarasovmis& to an end Pretorius [of specialism], or as simply G enier, 2015) andtogether molecular phylogeniesspecialist (Monaghan take of clustering unrecognized linet al., 2007; & is Scholtz, 2010; Ahrens, Schwarzer eages, e.g. . Sole . .so it with specialism vs generalism, *Corresponding author. E-mail: *E-mail: such that the more molecular markers that are & Vogler,the 2014; Gunter et ofal.,generalism, 2016). Twocertainly evolutionapplied, phenomenon in ary biogeographical havetoresulted. Differterms of polyphagy, hypotheses is perhaps seen be largely illuent studies support spatiotemporal in sory (both quotes from Loxdale & Harvey, origins 2016). NevMesozoic Mesozoic or Cenozoic ertheless, vicariance, recent studies havedispersal demonstrated that dispersal with trophic not origins on dinosaur or mamgeneralism is perhaps as rare as has been previmal The present examines the all merits ouslydung. considered (Foristerreview et al., 2015), nor are genof hypotheses with regard to waiting trophic, spatial, tempoeralists simply specialists to happen as ral, and fossil evidence. generalist lineages can arise from other generalists Food specialization leddiet to the dominance of (Hardy & Otto, 2014),has and breadth can even dung-feeding anddaughter has clearly played an important increase in the species of more specialist role in driving the(Mao, evolution of the subfamily. Early ancestral species Schuler & Berenbaum, 2007; opinion considered scarabaeine to bea estabDay, Hua & Bromham, 2016). beetles Developing more lished 1972) of and associated withevolution dinosaurs robust (Halffter, understanding what drives the of (Je (...truncated)