Multi-host parasite species in cophylogenetic studies

International Journal for Parasitology 35 (2005) 741–746 www.parasitology-online.com Current opinion Multi-host parasite species in cophylogenetic studies Jonathan C. Banks*, Adrian M. Paterson Bioprotection and Ecology Division, Lincoln University, P.O. Box 84, Canterbury 8150, New Zealand Received 1 March 2005; received in revised form 21 March 2005; accepted 21 March 2005 Abstract Cophylogenetic studies examine the relationship between host and parasite evolution. One aspect of cophylogenetic studies that has had little modern discussion is parasites with multiple definitive hosts. Parasite species with multiple host species are anomalous as, under a codivergence paradigm, speciation by the hosts should cause speciation of their parasites. We discuss situations such as cryptic parasite species, recent host switching or failure to speciate that may generate multi-host parasites. We suggest methods to identify which of the mechanisms have led to multi-host parasitism. Applying the suggested methods may allow multi-host parasites to be integrated more fully into cophylogenetic studies. q 2005 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Coevolution; Cophylogeny; Cospeciation; Failure to speciate; Phthiraptera; Sphenisciformes Cophylogenetic studies of relationships between parasites and their hosts have been revitalised recently. Studies have been conducted on diverse host–parasite assemblages such as chewing lice and pocket gophers (Hafner et al., 1994) and parasitic copepods and teleost fish (Paterson and Poulin, 1999). The principles underpinning host–parasite cophylogenetic studies have also been applied to other systems such as herbivore–plant interactions (Futuyma and McCafferty, 1990). Most, if not all, of these studies investigate the origins of current associations, asking whether they arose by descent or by colonisation. Association by descent proposes that current host–parasite associations have arisen because each host species has inherited the association from its ancestral species. Association by colonisation proposes that host switching, a parasite switching to a host species from a lineage other than the host’s ancestor, is the predominant explanation for the parasites’ distributions (Brooks and McLennan, 1991). Cophylogenetic studies have traditionally assessed the extent of codivergence, i.e. parallel speciation in * Corresponding author. Current address: Department of Entomology, University of Illinois, 320 Morrill Hall, Urbana Champaign, 505 S. Goodwin Ave, Urbana, IL 61801, USA. Tel.: C1 217 265 8123; fax: C1 217 244 3499. E-mail address: jbanks@life.uiuc.edu (J.C. Banks). the unrelated host and parasite lineages (Clayton et al., 2003), by examining the degree of congruence between host and parasite phylogenies. Incongruent host and parasite phylogenies suggests host switching in the parasite lineage (Brooks and McLennan, 1991). However, incongruence does not necessarily imply host switching as congruence can be hidden by other cophylogenetic events, such as sorting (e.g. extinction) and duplication (intra-host speciation) events (Paterson and Banks, 2001; Clayton et al., 2003). Likewise, congruent host and parasite phylogenies do not necessarily indicate a history of cospeciation as processes other than cospeciation may generate ‘false’ congruence (Clayton et al., 2003). False congruence can arise, for example, if a parasite species has undergone a series of sequential host switches, successively colonising the host’s closest relatives and then speciating (Brooks and McLennan, 1991). While most of the events potentially affecting the distribution of parasite species on their hosts have been discussed, especially by Clay (1949), one aspect of cophylogenetic studies that has had little modern discussion is the distribution of a single parasite species on multiple definitive host species (the host on which a parasite reproduces sexually) and the influence such parasites have on coevolutionary history. Studies of host–parasite interactions have often concentrated on the chewing lice, with 0020-7519/$30.00 q 2005 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2005.03.003 742 J.C. Banks, A.M. Paterson / International Journal for Parasitology 35 (2005) 741–746 gophers and their lice being a textbook example of host– parasite cospeciation. As lice are able to coexist with their hosts for evolutionary long periods, and have few opportunities to transfer between host species, they should be host specific and show a high degree of codivergence with their hosts (Page and Hafner, 1996). Lice with multihost distributions at first appear to be somewhat anomalous within a codivergence paradigm. However, louse species infesting multiple host species are relatively common. For example, eight of the 15 species of obligate ectoparasitic chewing lice, Phthiraptera, parasitising penguins, Spheniscidae; two of the seven species of lice parasitising kiwi, Apterygidae; and 13 of the 14 louse species parasitising albatrosses, Diomedeidae, in the New Zealand region, have multiple hosts (Price et al., 2003). Some louse species have a very wide distribution indeed, for example, the louse Menacanthus eurysternus (Burmeister, 1838) parasitises 176 bird species in 100 genera and 34 families and there are only three small bird orders on which all species of lice are host specific (Price et al., 2003). In this paper, we comment on the different processes that may produce multi-host parasitism and how these processes may mask instances of cospeciation or host switching by parasites. We examine factors that enable populations of multi-host parasites to remain in genetic contact despite parasitising different host species and suggest that population genetic techniques may be required to fully explain host–parasite associations. We suggest the following explanations for the presence of multi-host parasites: (i) (ii) (iii) (iv) cryptic parasite species, misclassified (over-split) hosts, recent host switches, failure to speciate by parasite populations despite their host taxa diverging, (v) incomplete host switching (sensu Clayton et al., 2003). Note that for (i)–(iii), although the parasite populations infesting divergent host taxa appear to be the same species, the parasite populations are actually genetically isolated from each other. Parasite morpho-species may appear to have multiple host species if parasite populations are isolated on their hosts and have diverged genetically but the parasite species are morphologically conservative, i.e. cryptic species. Cryptic species can be identified relatively easily using genetic data. The lack of morphological differences in the parasites may be due to factors such as similar selection pressures on the parasite species causing conservation of morphology (Fig. 1A). Cryptic species may also be present if morphological convergence has occurred between parasite species that are not closely related (Fig. 1B), for example, if the parasite species face similar selection pressure. Once divergent parasite taxa are identified from genetic data, they should be A A I B A I B II B I C Ci I Cii A I A I B I B I D E A I A I B I B I Fig. 1. Stacked hypothetical phylogenies for host (solid lines) and parasite (broken lines) morpho species representing the processes that can produce a pattern of multi-host parasitism in cophylogenetic studies. Branch lengths are proportional to time. The parasite phylogeny is displaced to the left and is below the host phylogeny. Apparent multi-host parasites due to: (A) Cryptic species. Parasite I is in reality two species but classified as a single species due to failure to detect differences between the two populations. (B) Morphological convergence. Parasite I is two species but classified as a single species due to convergence. (Ci) Host switching, recent. Parasite I is in reality two species but classified as a single species as insufficient time has elapsed for differences to accumulate. (Cii) Host switching, ancient. Parasite I is in reality two species but classified as a single species as no differences have been identified, perhaps due to morphological similarity. True multi-host parasites due to (D) failure to speciate. Parasite I is a single species due to gene flow between parasite populations. The double-headed broken arrows designate ongoing gene flow between the parasite populations. (E) Host switching, incomplete. Parasite I has colonised a new host species but the parasite populations on the two host species are not genetically isolated from each other. treated as separate taxa in cophylogenetic analyses and may support either association by descent (Fig. 1A) or colonisation (Fig. 1C), depending on their position in the phylogeny. Cryptic louse species parasitising doves have been identified using genetic data. For example, the chewing louse, Columbicola passerinae (Wilson, 1941) parasitising the blue ground dove, Claravis pretiosa (Ferrari-Pérez, 1886), differs from C. passerinae parasitising the common ground dove, Columbina passerina (L., 1758) by 11.3% for a portion of the mitochondrial cytochrome oxidase c subunit I (COI) gene suggesting the two louse populations could be reclassified as two species (Johnson et al., 2002). The chewing louse genus Physconelloides also contains multihost louse taxa implying considerably more host specificity than apparent from a consideration of morpho-species alone (Johnson et al., 2002). Parasites may appear to have several hosts if host ‘species’ are over-split. If two host species are actually a single species then a parasite species will appear to J.C. Banks, A.M. Paterson / International Journal for Parasitology 35 (2005) 741–746 parasitise several host species. A robust alpha taxonomy for both hosts and parasites is required for cophylogenetic studies. If a parasite colonises new host species, relative levels of divergence between parasite populations established by the host switching events can range anywhere from zero, if the colonisation event occurred only recently (Fig. 1Ci), to slightly less than the host species, if colonisation of the new host species occurred just after the origin of that host species (Fig. 1Cii). We stress that these are genetic distances that have been adjusted for differences in the rates of divergence in hosts and parasites. Next, we discuss reasons why populations of a parasite species on different host species could have few genetic differences in a fast evolving region of the genome despite significant genetic differences in the hosts (Table 1). If a host switch has occurred recently, it seems likely that a newly established population would be established from a founding event of a few individuals colonising the new host species and thus there will be less genetic diversity in the newly established population compared with the parent population. Lack of diversity within a population could be used to distinguish a recent host switch from failure to speciate and incomplete host switching (both of which will also give rise to populations with no or few genetic differences). Measures of genetic diversity within populations such as allele frequencies for microsatellite loci were used to infer a founder event in the Western Australian population of the parasitoid Diaeretiella rapae (M’Intosh, 1855) (Hymenoptera: Braconidae) that parasitises several species of aphids (Baker et al., 2003). True multi-host parasites occur when there is ongoing gene flow between parasite populations infesting divergent hosts. This can occur by two mechanisms—either failure to speciate (Johnson et al., 2003) or incomplete host switching (Clayton et al., 2003). Failure to speciate occurs when parasite populations maintain genetic contact despite the hosts diverging (Fig. 1D). Failure to speciate has also been called inertia (Paterson and Banks, 2001) and cophylogeny without cospeciation (Hugot et al., 2001). Failure to speciate will generally occur between parasite populations on closely Table 1 Cophylogenetic events possible when relative genetic distances (i.e. genetic distances corrected for disparate rates of divergence in hosts and parasites) between host species are compared to the genetic distances between their parasite species (H is host, P is parasite). Events in italics are those that will give rise to apparent and true multi-host parasites Relative genetic distances Possible explanations HZP POH P!H P/H Cospeciation Duplication, Over-split host species Host switch Very recent host switch, Failure to speciate, Incomplete host switch 743 related host taxa as it requires that the parasite species are inherited from the ancestor of the sibling host species. Failure to speciate might occur when the hosts’ behavioural or life history traits such as breeding site philopatry reproductively isolate host populations while sufficient contact between host taxa occurs at other times to allow gene flow between parasite populations. Failure to speciate supports association by descent. One of several examples of possible failure to speciate in the penguin lice is Austrogoniodes waterstoni (Cummings, 1914) that parasitise blue penguins, Eudyptula minor (J.R. Forster, 1781) (Price et al., 2003). Austrogoniodes waterstoni from Australia and New Zealand do not differ for 12S rRNA gene (342 bp, Genbank accession numbers AY229913–AY229918) or cytochrome b (313 bp, Genbank accession numbers AY345911–AY345916) despite the two host populations showing relatively high levels of divergence (2.6% for third domain of the mitochondrial 12S rRNA gene (12S) and 4% for cytochrome b (cyt b) (Banks et al., 2002). The presence of birds of the Australian haplotype in southern New Zealand (Banks et al., 2002) may enable lice to transfer between the two host populations. A pattern similar to failure to speciate could be generated if a parasite colonised the sister taxon of its original host, and maintained genetic contact with the source population. This is known as incomplete host switching (Clayton et al., 2003). If an incomplete host switch has occurred, we would expect that there would be an increase in the parasite population size as the new niche is populated. Tests of population growth such as Tajima’s D, Fu and Li’s F* and D* and mismatch analyses could be applied to mitochondrial DNA of parasites on divergent hosts. These tests can distinguish failure to speciate, where the parasite population size has likely remained constant, from incomplete host switching, where the parasite population size has likely increased markedly following the colonisation event. These analyses were applied to a study of Horsfield’s bronze cuckoo, Chalcites basalis (Horsfield, 1821) a brood parasite, and suggested that a range expansion to novel host species had occurred (Joseph et al., 2002). Nested clade analysis can also help distinguish between ongoing gene flow and complete host switching to new host taxa. Parasite populations may also show less divergence than their host species if the rate of genetic change of parasites is slower than that of their hosts. Generally parasites have shorter generation times than their hosts and shorter generation times have been shown to be associated with faster rates of genetic change (Page and Hafner, 1996). However, there are a few isolated examples of parasites diverging more slowly than their hosts. Thus, multi-host parasites need to be studied as part of a larger cophylogenetic study so that relative rates of host and parasite divergence can be estimated. Genetic contact between parasite species on divergent hosts does not simply result from an absence of other 744 J.C. Banks, A.M. Paterson / International Journal for Parasitology 35 (2005) 741–746 processes. It seems likely that multi-host parasites differ from host specific parasite species in their behavioural or life history characteristics. These differences might include more opportunities to transfer between their different host taxa, and once they transfer, better survival and higher fecundity allowing the maintenance of gene flow between parasite populations parasitising divergent hosts. Thus, we predict that the hosts of multi-host parasites will have overlapping ranges, multi-host parasites will have larger populations than host specific parasites and there will be fewer barriers to multi-host parasites surviving on divergent hosts. We will now compare several characteristics of host specific and multi-host chewing lice and their penguin (Sphenisciformes) hosts. As chewing lice have poor dispersal abilities (Page and Hafner, 1996), ongoing gene flow between louse populations parasitising divergent host taxa requires contact between their host species. For example, Austrogoniodes cristati (von Kéler, 1952) parasitises all six species of crested penguin (Eudyptes). One opportunity for populations of A. cristati to maintain contact may be during the moult by their hosts. All six crested penguin species moult, at least occasionally on the Snares Islands (del Hoyo et al., 1992). Hybridisation between host species may also break down isolation between the parasites from different host species. For example, rockhopper penguins, Eudyptes chrysocome (Forster, 1781), have been reported to form mixed species pairs with erect-crested, Eudyptes sclateri, Buller, 1888, royal, Eudyptes schlegeli, Finsch, 1876, and macaroni, Eudyptes chrysolophus (Brandt, 1873) penguins although with limited breeding success (del Hoyo et al., 1992). Dispersal of young hosts from natal colonies may be another mechanism by which gene flow among parasite populations may occur. For example, blue penguin juveniles disperse widely until returning to their natal colonies after 2 or 3 years to breed (del Hoyo et al., 1992). Genetic contact between louse populations on divergent host taxa may also occur by phoresis, i.e. lice transferring between hosts by attaching to the hippoboscid flies that parasitise many bird species. We believe phoresis is unlikely to be a significant method for lice to transfer between penguin taxa as we could find no reports of hippoboscid flies in the Australasian subantarctic region. Louse abundance may be one factor that affects gene flow between louse populations. For example, we examined 40 live emperor penguins, Aptenodytes forsteri, G.R. Gray, 1844, and found one specimen of the host specific louse Austrogoniodes mawsoni, Harrison, 1937. Likewise we examined 40 live Adelie penguins, Pygoscelis adeliae (Hombron and Jacquinot, 1841) and found three of the host specific louse Austrogoniodes antarcticus, Harrison, 1937 on one bird, while on the other 39 Adelie penguins we found no lice. In contrast, we found 25 females and 40 males of the multi-host louse Austrogoniodes demersus (von Kéler, 1952) on a single African penguin, Spheniscus demersus (L., 1758), carcass. We have also seen differences in the abundance of host specific and multi-host lice on the same host species. We found a single male specimen of the host specific Austrogoniodes vanalphenae, Banks and Palma, 2003 and 11 males of the multi-host Austrogoniodes concii (von Kéler, 1952) on 16 yellow-eyed penguin, Megadyptes antipodes (Hombron and Jacquinot, 1841), carcasses. We compared the number of host specific lice and multihost lice held in the Museum of New Zealand that have been collected from individual penguins. We found that significantly fewer host specific lice were collected from individual penguins (mean number of host specific lice collected per hostZ4.9, SDZ4.1, n (hosts)Z66, mean multi-host lice collected per hostZ8.0, SDZ10.1, n (hosts)Z93, Student’s t-test PZ0.02). We suggest that the higher abundance of multi-host lice might increase the probability of transfer between divergent host taxa, which in turn may help parasites maintain gene flow between populations. We also examined lice held in the Museum of New Zealand collected from penguins that had the potential to be infested by both host specific and multi-host lice and calculated their prevalence, i.e. the number of host individuals on which multi-host and host specific lice were present or absent (Clayton et al., 2003). Multi-host lice infested significantly more penguin hosts than did host specific lice (Fisher exact test, c2Z15.48, 1 df, P!0.01). We also compared the mean intensity, i.e. the number of multi-host lice and host specific lice on an individual (Clayton et al., 2003), collected from penguin host species that could be parasitised by either multi-host or host specific lice. Significantly more multi-host lice than host specific lice (paired t-test, tZ2.417, 11 df, PZ0.04) had been collected from penguins. While we acknowledge that these results could have been confounded by factors such as unequal host population size and differences in sampling effort, the results suggest that multi-host lice parasitised more host individuals than did host specific lice and multi-host lice were more numerous than host specific lice. Krasnov et al. (2004) found a similar result for fleas parasitising small mammals. The higher prevalence and intensity of multi-host lice may increase the opportunities for gene flow to occur between populations of multi-host lice infesting divergent host taxa. Gene flow between parasite populations will also be facilitated if there are weak barriers to parasite survival on their alternative host taxa. For example, chewing lice evade host grooming by inserting themselves between feather barbs or by holding on to feather shafts (Clayton et al., 2003). When lice were experimentally transferred to a new host species, survival of lice was significantly reduced in proportion to the mean difference in feather barb size of the new and original hosts. Lice that did survive shifted their microhabitat to feathers of mean barb diameter similar to that of their original host. Thus there may be fewer barriers to the maintenance of genetic contact between parasite populations if the hosts are physically similar. J.C. Banks, A.M. Paterson / International Journal for Parasitology 35 (2005) 741–746 The presence of multi-host lice is not as simple as contact between host taxa allowing gene flow between louse populations. Some host taxa are parasitised by both host specific and multi-host lice. One example is the rockhopper penguin, which is often subdivided into three subspecies, the western, E. c. chrysocome, the eastern, E. c. filholi and Moseley’s, E. c. moseleyi. All three rockhopper subspecies are parasitised by A. cristati, but within the rockhopper penguin species, the host specific Austrogoniodes keleri, Clay, 1967 is present only on the western rockhopper subspecies. Austrogoniodes concii is only present on Moseley’s rockhopper penguin and Austrogoniodes hamiltoni, Harrison, 1937 is present only on the eastern rockhopper penguin, although both A. concii and A. hamiltoni parasitise other penguin species (Price et al., 2003). The inability of multi-host louse species such as A. concii and A. hamiltoni to parasitise populations within a species suggests that multiple factors affect the host specificity of parasites. Studying how the behavioural and life history characteristics of both hosts and parasites affect the distribution of parasites may be of value in predicting which parasites are likely to switch to new hosts and could be of use in identifying risks to economically or culturally important species. Generally cophylogenetic studies have concentrated on host specific parasites as multi-host parasite species are problematic for analyses (Johnson et al., 2003). Several methods of analysing cophylogenetic data cannot deal with multi-host parasites or else multi-host parasites make the methods unwieldy (Johnson and Clayton, 2003). Although multi-host parasites reduce congruence between host and parasite phylogenies (Johnson and Clayton, 2003), they can still contribute to association by descent. Suggestions to deal with multi-host parasites include creating ‘dummy lineages’ with a phylogeny matching the host phylogeny (Page and Charleston, 2002). This method assumes that multi-host parasites are an unresolved clade that more data will resolve. However, creating dummy lineages for multihost parasite populations is justified only if the populations are cryptic species. Alternatively, it has been assumed that parasites have multiple hosts because of recent host switching (Brooks et al., 2004). Others have applied the principle of parsimony to resolve multi-host parasites suggesting it is more parsimonious to infer failure to speciate when multi-host parasites are present on closely related hosts (Hugot et al., 2001). However, there are potentially several reasons to explain the distribution of parasites on multiple hosts. For parasite species with multiple host taxa, and with few genetic differences between populations, the application of population level genetic techniques will be useful. If our goal is to explain why we see particular host–parasite associations, it will be necessary to consider the information carried by parasites with multiple hosts. Recognition of the processes that can result in multi-host parasites, much like the recognition of sorting 745 and duplication events, is necessary if the distribution of parasites on their hosts is to be better understood. Acknowledgements Lincoln University and a Kelly Tarlton’s Antarctica Scholarship supported this work. Thanks to R. Palma, Museum of New Zealand for confirming our identifications of louse specimens. Comments by R. Cruickshank, K. Johnson, M. Kennedy, C. Vink, R. Page and two anonymous reviewers improved this paper. References Baker, D.A., Loxdale, H.D., Edwards, O.R., 2003. Genetic variation and founder effects in the parasitoid wasp, Diaeretiella rapae (M’Intosh) (Hymenoptera: Braconidae: Aphidiidae) affecting its potential as a biological control agent. Mol. Ecol. 12, 3303–3311. Banks, J.C., Mitchell, A.D., Waas, J.R., Paterson, A.M., 2002. An unexpected pattern of molecular divergence within the blue penguin (Eudyptula minor) complex. Notornis 49, 29–38. Brooks, D.R., McLennan, D.A., 1991. Phylogeny, Ecology and Behaviour: A Research Program in Comparative Biology. The University of Chicago Press, Chicago. Brooks, D.R., Dowling, A.P.G., van Veller, M.P.G., Hoberg, E.P., 2004. Ending a decade of deception: a valiant failure, a not-so-valiant failure, and a success story. Cladistics 20, 32–46. Clay, T., 1949. Some problems in the evolution of a group of ectoparasites. Evolution 3, 279–299. Clayton, D.H., Al-Tamimi, S., Johnson, K.P., 2003. The ecological basis of coevolutionary history. In: Page, R.D.M. (Ed.), Tangled Trees. Phylogeny, Cospeciation, and Coevolution. University of Chicago Press, Chicago, pp. 310–341. del Hoyo, J., Elliot, A., Sargatal, J., 1992. Handbook of the Birds of the World. Lynx Edicions, Barcelona. Futuyma, D.J., McCafferty, S.S., 1990. Phylogeny and the evolution of host plant associations in the leaf beetle genus Ophraella (Coleoptera, Chrysomelidae). Evolution 44, 1885–1913. Hafner, M.S., Sudman, P.D., Villablanca, F.X., Spradling, T.A., Demastes, J.W., Nadler, S.A., 1994. Disparate rates of molecular evolution in cospeciating hosts and parasites. Science (Washington, DC) 265, 1087–1090. Hugot, J.P., Gonzalez, J.P., Denys, C., 2001. Evolution of the old world Arenaviridae and their rodent hosts: generalized host-transfer or association by descent?. Infect. Genet. Evol. 1, 13–20. Johnson, K.P., Clayton, D.H., 2003. Coevolutionary history of ecological replicates: comparing phylogenies of wing and body lice to columbiform hosts. In: Page, R.D.M. (Ed.), Tangled Trees. Phylogeny, Cospeciation, and Coevolution. University of Chicago Press, Chicago, pp. 262–286. Johnson, K.P., Williams, B.L., Drown, D.M., Adams, R.J., Clayton, D.H., 2002. The population genetics of host specificity: genetic differentiation in dove lice (Insecta: Phthiraptera). Mol. Ecol. 11, 25–38. Johnson, K.P., Adams, R.J., Page, R.D.M., Clayton, D.H., 2003. When do parasites fail to speciate in response to host speciation?. Syst. Biol. 52, 37–47. Joseph, L., Wilkie, T., Alpers, D., 2002. Reconciling genetic expectations from host specificity with historical population dynamics in an avian brood parasite, Horsfield’s Bronze-Cuckoo Chalcites basalis of Australia. Mol. Ecol. 11, 829–837. 746 J.C. Banks, A.M. Paterson / International Journal for Parasitology 35 (2005) 741–746 Krasnov, B.R., Poulin, R., Shenbrot, G.I., Mouillot, D., Khokhlova, I.S., 2004. Ectoparasitic ‘Jacks-of-all-trades’: relationship between abundance and host specificity in fleas (Siphonaptera) parasitic on small mammals. Am. Nat. 164, 506–516. Page, R.D.M., Charleston, M.A., 2002. Treemap versus BPA (again): A Response to Dowling. University of Glasgow, Glasgow, p. 26. Page, R.D.M., Hafner, M.S., 1996. Molecular phylogenies and host– parasite cospeciation: gophers and lice as a model system. In: Harvey, P.H., Leigh-Brown, A.J., Maynard Smith, J., Nee, S. (Eds.), New Uses for New Phylogenies. Oxford University Press, Oxford, pp. 255–270. Paterson, A.M., Banks, J.C., 2001. Analytical approaches to measuring cospeciation of host and parasites: through a glass darkly. Int. J. Parasitol. 31, 1012–1022. Paterson, A.M., Poulin, R., 1999. Have chondracanthid copepods co-speciated with their teleost hosts?. Syst. Parasitol. 44, 79–85. Price, R.D., Hellenthal, R.A., Palma, R.L., 2003. World checklist of chewing lice with host associations and keys to families and genera. In: Price, R.D., Hellenthal, R.A., Palma, R.L., Johnson, K.P., Clayton, D.H. (Eds.), The Chewing Lice. World checklist and Biological Overview Illinois Natural History Survey Special Publication, vol. 24, pp. 1–448.