View - Kowalewski, M. - Virginia Tech

PALEONTOLOGICAL SOCIETY PAPERS, V. 8, 2002initially attacking live but dying animals, orsubsisting on their wounds (facultativeectoparasitism), followed by (ii) initiating wounds(obligate ectoparasitism), and finally (iii) invadingthe host itself and causing its death (parasitoidism)(Zumpt, 1965). In lineages such as carabid beetles,parasitoidism may have developed from predation.Erwin and Erwin (1976) proposed a sequence inwhich an ancestral ground beetle species was apolyphagous predator on insects; it wascharacterized by an active larva that emerged fromeggs deposited in the general habitat of its eventualvictims, and it pupated in an area removed from itsprey. The spatial confinement of a particular preyspecies resulted in monophagy of the beetle larvalpredator, and the reduction of prey seeking resultedin metabolic savings. This in turn favoredconsumption of fewer prey items and pupationwithin the immediate host environment. Anyenvironmental change that provoked host dispersalwould favor a shift from a larval predator to a small,immobile, parasitoid living within a single host.This explanation works best for spatially confinedhosts, such as wood-boring or gall-inhabitinglarvae. There are at least 100 known originationsof parasitoidism in insects, most by way ofmycophagy, then saprophagy, and the fewestthrough predation and herbivory (Eggleton andBelshaw, 1992). Parasitoidism itself has in turnevolved into other trophic modes such asprovisioning polyphagy and predation in somecoleopteran, dipteran, and hymenopteran lineages(Eggleton and Belshaw, 1992).Parasites.—When compared to the relativelygood fossil record of parasitoids, the record of insectparasitism is poor: only about 30 percent of parasitefamilies have fossil occurrences. The reason for thisdeficiency is that, whereas the adult forms ofparasitoids are overwhelmingly free-living and socan enter the fossil record through a variety ofenvironments (many are small wasps), the adults ofmost parasitic organisms are attached to theirvertebrate hosts and are destroyed by the processesof carcass decomposition. Consequently, only inexceptional environments, such as resin flows, areinsect parasites preserved (Lewis and Grimaldi,1997). Fine-grained compression depositsoccasionally contain fossil parasites, such as theenigmatic mecopteroid Strashila from the UpperJurassic of the Siberia (Rasnitsyn, 1992), the possibleflea Saurophthirus from the Lower Cretaceous ofTransbaikalia (Ponomarenko, 1976; but seeLakshminarayana et al., 1984), and the flea Tarwiniafrom the Lower Cretaceous of New South Wales,Australia (Riek, 1970; Jell and Duncan, 1986; butsee Smit, 1972). All three taxa display ectoparasiticfeatures, including hypognathous, piercing-andsucking mouthparts, compact antennae, long legswith grasping claws, and distensible abdomens(Rasnitsyn, 1992). Both Strashila and Saurophthirusare dorsoventrally flattened and morphologicallyconvergent on bat parasites such as the hemipteranPolyctenidae, the dipteran Nycteribiidae, and eventhe dermapteran Arixeniidae (Popham, 1962;Ponomarenko, 1976), which feed on bat wingmembranes and occupy sparse to tomentose hair;the resemblance suggests a similar feeding style forthe Mesozoic fossils, possibly on pterosaurs. In theAustralian flea, of indeterminate family assignment,lateral body compression indicates life in dense furnear the body core, rather than on the wingmembrane; and they likely inhabited mammalianhosts such as monotremes, triconodonts, ormultituberculates. Perplexingly, a literal reading ofthis evidence points to an apparently missing fossilrecord, judging from the relationship of theseparasitic clades to their free-living (and sorepresented by more fossils) sister clades (Lyal,1985; Kristensen, 1999). These phylogeneticrelationships, in conjunction with the fossil recordsof the better-preserved clade, indicate splittingevents ranging from the Early Permian to the LateJurassic (Downes, 1971b; Lyal, 1985; Kinzelbachand Pohl, 1994; Fang et al., 1999). The conflictingdata suggest either that the parasite clades are goodexamples of fossil ghost taxa, or that they originatedmuch later within the crown groups of their sisterclades(e.g., Chalwatzis et al., 1996).The origins of hematophagy during theMesozoic, and the likely initial hosts for severalmajor parasitic insect clades, have generatedconsiderable discussion. The best evidence comes234